COMMUNICATIONS STANDARDS REVIEW · February 1999 Vol. 10.2 Copyright © CSR 1999 1 COMMUNICATIONS STANDARDS REVIEW Volume 10, Number 2 February 1999 In This Issue The following reports

February 1999 Vol. 10.2 Copyright © CSR 1999 1

COMMUNICATIONS STANDARDSREVIEW

Volume 10, Number 2 February 1999

In This Issue

The following reports of recent standards meetings represent the view of the reporterand are not official, authorized minutes of the meetings.

Q15/16, Video Coding Experts Meeting, Nov. 3 - 6, 1998, Seoul, Republic of Korea .................................. 2Q15/16 Video Coding Experts Group Meeting Roster, Nov. 3 - 6, Seoul, Republic of Korea ............. 18

ETSI DTA Meeting #4, Nov.16 - 18, 1998, Berlin, Germany ........................................................................ 19DTA Meeting Roster, Nov. 16 - 18, 1998, Berlin, Germany.................................................................. 22

Q11/16, CSN Multimedia Terminal and Systems Experts Group, Nov. 17 - 20, 1998, Torino, Italy....... 23H.324 and H.320 System, Control and MUX .......................................................................................... 23Mobile ......................................................................................................................................................... 24Non-Conversational Services ................................................................................................................... 25Joint Meeting with Q11-14/16................................................................................................................... 25

Q12/16, Q13/16, and Q14/16 Rapporteurs Meeting, Nov. 17 - 20, 1998, Torino, Italy ................................ 29Q12/16 B-ISDN Multimedia Systems and Terminals........................................................................... 31Q13/16, Packet Switched Multimedia Systems and Terminals ............................................................ 32Q14/16 Common Protocols, MCUs and Protocols for Interworking with H.300-series Terminals .... 36

ETSI TM6 Meeting, Nov. 24 - 27, 1998, Sophia Antipolis, France.............................................................. 40VDSL.......................................................................................................................................................... 41SDSL.......................................................................................................................................................... 44TM6 Meeting Roster, Nov. 24 - 27, 1998, Sophia Antipolis, France...................................................... 49

ITU-T Q4/16 and Q23/16 Rapporteur Meetings, Dec. 7 - 9, 1998, San Diego, CA ..................................... 50Q23/16, PCM Modems............................................................................................................................... 50Q4/16, Modems........................................................................................................................................... 51Q4/16 and Q23/16 Rapporteurs Roster, Dec. 7 - 9, 1998, San Diego, CA............................................. 53

ETSI ATA Meeting, Dec. 7 - 11, 1998, Cape Town, South Africa................................................................ 54ATAc (Conventional Technologies).......................................................................................................... 56ATAe (Enhanced Signaling Systems) ...................................................................................................... 58ETSI ATA Meeting Roster, Dec. 7 - 11, 1998, Cape Town, South Africa ............................................. 60

ETSI TIPHON Meeting #11, Jan. 11 - 15, 1999, Sophia Antipolis, France ................................................ 61Working Group 1, Requirements .............................................................................................................. 64Working Group 2, Architecture................................................................................................................. 66Working Group 3, Call Control ................................................................................................................ 67Working Group 4, Naming, Addressing.................................................................................................. 69Working Group 5, Quality of Service........................................................................................................ 70Working Group 6, Verification .................................................................................................................. 72ETSI TIPHON Meeting #11 Roster, Jan. 11 - 15, 1999, Sophia Antipolis, France............................ 73

Q4/15 Rapporteurs Meeting, Jan. 18 - 22,1999, Sunriver, OR ...................................................................... 76G.972.2 (G.lite)........................................................................................................................................... 76G.977.1 (G.ploam) ...................................................................................................................................... 77G.966.1 (G.test).......................................................................................................................................... 78G.992.1 (G.dmt) .......................................................................................................................................... 79G.994.1 (G.hs) ............................................................................................................................................ 80G.995.1 (G.ref) ............................................................................................................................................ 81Q4/15 Rapporteurs Group Roster, Jan. 18 - 22,1999, Sunriver, OR..................................................... 81

Acronym Definitions ......................................................................................................................................... 831999 Standards Committee Meeting Schedules, as of January 28, 1999..................................................... 87Fax Bounce Back (Document Order Form..................................................................................................... 88

COMMUNICATIONS STANDARDS REVIEW

2 Vol. 10.2 Copyright © CSR 1999 February 1999

REPORT OF Q15/16, VIDEO CODING EXPERTS MEETINGNOVEMBER 3 - 6, 1998, SEOUL, REPUBLIC OF KOREA

The Q15/16 Rapporteur is G. Sullivan (PictureTel); technical and H.26L sessions were co-chairedby the Rapporteur and K. Hibi (Sharp, Associate Rapporteur).

Documents for this meeting and other Q15/16 information are on the ftp site managed by theQ15/16 Rapporteur:

ftp://standard.pictel.com/video-site

Documents for this meeting are in the 9811_Seo subdirectory of the ftp site.

The e-mail reflector for Q15/16 is managed by M. Zeug (Iterated Systems). Those wishing tosubscribe or unsubscribe may submit requests to:

[email protected]

The address for e-mail to be sent to all members of the reflector list is:

[email protected]

The Rapporteur discussed the importance of the disclosure of patents and of the filing ofstatements pertaining to such intellectual property claims with the ITU. Q15/16 members wereencouraged to consult guidelines and other information at:http://www.itu.int/ITU-Databases/TSBPatent/

H.263+ BITSTREAM PACKETIZATION

Q15-F-04 (T. Gardos and G. Deisher, Intel) contains the ad hoc committee report onpacketization. This activity has been a success, and was completed with the approval of RFC2429 in the IETF and a corresponding annex to H.225.0 in SG16.

LIAISON WITH OTHER ORGANIZATIONS

Liaison Q15-F-26 (MPEG) was received from ISO/IEC JTC1/SC29/WG11 (MPEG) regarding theH.262|IS 13818-2 (MPEG-2 video) and H.222.0|IS 13818-1 (MPEG-2 Systems) workplans, newedition, and publication issues.

MPEG intends to publish a new version of these Recommendations by May 2000. This newversion would include the amendments previously approved and those now in progress, unifiedinto a revised document rather than published as an original text with separately-publishedamendments. MPEG indicated that if the rapid publication of a common text for this new version(by May 2000) is not feasible, it recommends changing the publication method for theseRecommendations from “common text” to “technically aligned text” status. It is believed that theMay 2000 goal is reasonable.

MPEG also reported that it discovered a serious error in the recently approved text ofAmendment 5 of H.222.0|IS 13818-1 (MPEG-2 Systems) and wishes to retract the intent topublish this Recommendation. This issue is in the domain of the Rapporteur of Q12/16, S. Okubo;Q12/16 and SG16 management are fully aware of the issue.

This liaison statement was discussed, and the group appreciated having this update for theongoing collaborative relationship between the ISO/IEC JTC1 and ITU-T regarding thesecommon text Recommendations. No reply appeared necessary: Q15/16 will move ahead with theworkplan as communicated.

H.320 ADOPTION OF H.263+, AND H.245 CLARIFICATION

Q15-F-05 (G. Sullivan [for S. Gupta], PictureTel) reports the progress of work toward adoption ofthe H.263+ enhancements into the H.320 suite. Q15/16 experts were pleased with the progressof this activity, noting that the design for adoption appeared stable and that Determination had



been reached for revised versions of H.320, H.221, H.230 and H.242 at the September SG16meeting.

Q15-F-42 (G. Sullivan, PictureTel) indicates a need to clarify certain sections of the drafted textin regard to macroblock-level error handling issues. The current draft for use of these featureswithin H.320 had been based on content in H.245, but it was reported that the H.245 text wasunclear. Q15/16 did not immediately understand how best to clarify these issues and decided tosimply communicate the need for clarification to Q11/16 and Q14/16 and to encourage the Q15/16experts to further investigate the problem areas indicated.

Further work on this issue is to be conducted primarily by the Q11/16 experts; any needed workon the H.320 suite issue needs to be completed in time for the white document submissiondeadline for the Spring 1999 meeting of SG16.

MPEG-4 ADOPTION OF H.263 (BASELINE) COMPATIBILITY

Q15-F-07 (G. Sullivan, PictureTel) reports that the final draft of MPEG-4 will soon be producedbased on the results of the recent meeting of ISO/IEC JTC1/SC29/WG11 in Atlantic City.MPEG-4 version 1 is reported to be in its final editing stages for imminent Final DraftInternational Standard (FDIS) production.

CLARIFICATION REQUEST FOR H.263

Q15-F-27 (K. Lillevold, Intel) requests clarification regarding three issues in the H.263 text. Thegroup agreed that these comments were valuable and that in at least the first two (and probablyin all three) cases the text clarity was insufficient. Tentative answers were agreed. The creationof an implementer’s guide was suggested to document these and any other such issues as mayarise. In the meantime, the tentative answers agreed upon are:

1. BPPmaxKb table and Picture Padding:

Table 1/H.263 in Section 3.6 lists Minimum BPPmaxKb as a function of “Y Picture size inpixels.” It is not stated whether this is the size of the picture specified in the picture header(width x height) or whether this is the size of the picture after the width and height have beenpadded out to the nearest multiple of 16.

The interpretation of the group is that the size specified in the table is the size of the picturespecified in the picture header (i.e., the unpadded size).

Rationale: The 16-pixel gaps in size found in the table appear to signal the intent for this tospecify the unpadded size (a multiple of 16 since the width and height are each multiples of 4).Also, this is the most conservative answer as it ensures interoperability regardless of whichsize is assumed in a decoder design.

2. Direct bi-dir predicted MBs and picture extrapolation:

Section O.4 states: “In B and EP pictures, motion vectors over picture boundaries may be usedas described in section D.1 [..]”

Section D.1.1 states: “If PLUSPTYPE is present in the picture header, the motion vectorvalues are restricted such that no element of the 16x6 (or 8x8) region that is selected shallhave a horizontal or vertical distance more than 15 pixels outside the coded picture area.”

The potential problem is as follows and could occur for DIRECT predicted macroblocks in Bframes: If UMV is on and the new extended motion vectors are used, consider the case wherethe forward MV is 40 pixels and the B frame is in the middle of the previous and futureframes. For a DIRECT predicted macroblock at the top edge of the frame, the forward motionvector would be 20 pixels and the backward motion vector would be -20 pixels. Unless theDIRECT mode is not used for this macroblock, or unless the backward motion vector is clipped(to -15 pixels), pixels outside the 15 pixels mentioned in section D.1.1 may be accessed.

Now the question is: should the encoder make sure no such macroblocks occur, or should thedecoder be able to handle this situation gracefully?



The tentative answer is that the encoder shall make sure that no such macroblocks occur, bymanipulating mode decisions and/or motion vector values as necessary.

Rationale: Again the conservative response has been chosen, to ensure interoperabilityregardless of which interpretation is assumed in a decoder.

3. Parsability of GN/MBA field in BCM:

Under the “videomux” mode of Annex N (Reference Picture Selection), a BCM sent within anoutgoing video bitstream refers to the contents of an incoming bitstream.

Parsing of the variable length GN/MBA field within the BCM, however, is dependent upon theincoming bitstream’s image resolution, use of slices, and use of reduced resolution update. Thelength of the GN/MBA field is determined by these characteristics.

Does Annex N assume that the far-end recipient of the outgoing bitstream has an awareness ofthese characteristics of the incoming bitstream? Or does Annex N assume that the conferenceis symmetric with respect to these characteristics? If neither of these assumptions is made,then it would seem that the BCM cannot be parsed correctly in the videomux mode.

The tentative answer is that the far-end recipient of the outgoing bitstream (which containsBCM data referring to the incoming bitstream) must have an awareness of thesecharacteristics of the incoming bitstream. No symmetry is assumed.

Rationale: We are reasonably certain this was the intent of the design. No symmetry ofoperation is suggested anywhere in the text, and we believe none is necessary. The design wasfor bi-directional video use, in which a system which sending video should be aware of theparameters of the video it is sending (which determine how the data it receives in response isparsed).

TECHNICAL CONTENT PROPOSALS FOR H.263++ AND H.26L

Q15-F-09r2 (G. Sullivan, PictureTel), is the report of the ad hoc committee on H.263++development. Q15-F-10 (K. Hibi, Sharp), is the report of the ad hoc committee on H.26Ldevelopment.

Three algorithm designs with test results were presented. A fourth algorithm description wasalso considered as potentially falling well within the H.26L algorithm evaluation category due tothe close relationship of its design with those in the other proposals. Each proposal wasexamined by the group, and the analysis tables in this section summarizing the methods werecollected in Q15-F-43 (K. Hibi, Sharp). Readers are encouraged to study the proposals closely,as each has its own merits in addition to their use in a particular design.

Table 1 is a summary of the Telenor variable block-size multi-frame MC (motion compensator)and 4x4 interger transform in Q15-F-11 (G. Bjøntegaard, Telenor Satellite Systems). Table 2 isthe University of Strathclyde compression transform from Q15-F-18 and Q15-F-19 (R. J. Fryer,Univ. of Strathclyde). Table 3 is the Nokia affine MC with directional VQ, from Q15-F-24. Table4, from University of Elrnagen-Nurem and 8x8 summarize long-term memory with affine MC(Q15-F-33, Video Coding Using Long-Term Membory and Affine Motion-Compensated Precition,T. Weigand, E. Steinbach, and B. Girod, U. Erlangen-Nuremberg).

DEBLOCKING AND DERINGING FILTERS

Several methods for de-blocking and de-ringing filtering were presented, and this topic has beendesignated as an H.263++ Key Technical Area. In some cases these designs were asked to beconsidered as loop filter methods, and in others as post-processing filters (or as either one).Assessment of these methods and comments regarding them should be considered somewhatrelated.



Variable Block Size MC with 4x4 transformIntra Frame Coding 5 prediction mode for each 4x4 block

(DC, horizontal, vertical, two diagonals)Prediction mode is coded referring to neighboring blocks Codeword is assigned based on the probability of the mode

Inter Frame PredictionM.C. Prediction from more than one past frame

(no B-pictures usage is intended)Different block sizes for prediction 16x16, 8x8, 4x4

M.E. Step search from 16x16 down to 8x8, 4x4 smaller search window for 8x8 and 4x4Approximate bit usage is taken into account during the search

Residual codingTransform Use of 4x4 transform

integer transform, almost identical to DCTAttribute Use of H.261 like CBP

MB structure: 16 Y and 8 U, V CBP signals coded or not coded of four 4x4 blocks

Quantizer 32 different QPs corresponding to similar range of H.263No dead zone

Scan Double scan of 4x4 block for high QP values, Each 8 coefficients in 4x4 block are scanned separately

VLC Only one VLC for coding (VLC-D) Parameters designed to have the same statistics Similar to Annex D, but not reverse decodable Capable of fast recovery of VLC synchronization 1bit EOB, 2D-VLC, no escape coding (max. length=27)

Current implementation No long-term memory is usedRemarks on the results Improved performance esp. in Intra coding

“Back to basics” simplification approach theme to designMore improvement in higher bitrateReduced extent to ringing (despite use of de-ringing for reference)Somewhat sharper, better cleanup, more detailSome tendency toward blockinessError resilience update pattern was visible

Table 1. Telenor variable block-size multi-frame MC & 4x4 integer transform, Q15-F-11.

One comment made in the viewing of demonstration material for these filter designs was that thedemonstration conditions often seemed to consist of material coded at a sufficiently high qualitylevel that filtering was less necessary than would be seen under more adverse conditions. Thusdifferences in quality between filter designs may have been less apparent than in a test thatwould stress filtering more severely by coding at a lower quality level.

The group noted that if a new method were to be adopted into a Recommendation as a loop filter(as was Annex J for H.263+), then it would be necessary to consider the same rare feedbackeffects reported in April 1998 in Tampere (Q15-D-41 and Q15-D-61) concerning interaction ofhigh fidelity encoding with IDCT round-off error when using Annex J. Any future potentialadoption of loop filter technology by Q15/16 would be expected to analyze the IDCT feedbackissue and its impact on the filter design.

Nonlinear 3x3 artifact reduction using potential functions

Q15-F-13 (M. Shen, J. Kim, and C.-C. Jay Kuo, USC) presents a technique as either a loop filteror an outside-the-loop post-processing filter which used a non-iterative 3x3 region-of-support,classification between smooth and non-smooth regions, and minimization of a “potential function”(something like a distortion measure) along with clipping performed by the H.263+ Annex JUpDownRamp memoryless nonlinearity function. The complexity of the method was estimated assomewhat more complex but similar in spirit to the Annex J de-blocking filter or the de-ringingfilter of section 10 of test model number 10 (TMN10). Some subjective and objective quality



results were shown, but the group would need further information to properly assess the merits ofthis technique (such as side-by-side real-time D-1 comparisons with the Annex J + Section 10 ofTMN10 approach).

SCT (Strathclyde Compression Transform) - Variable Dimension Vector QuantizationIntra Frame Coding Identical to the anchorInter Frame Prediction

M.C. Decomposition of image Based on quad-tree decomposition of picture But, not restricted by quad-tree structureVQ and form of MC Variable Dimension VQ 32x32, 16x16, 8x8, 4x4 blocks code book size for each dimension is 256 Prediction from previous frame (no B-pictures) Different motion vectors for luma and chroma Zero motion with changing brightness

M.E. Best-first rule for selecting block to be codedCoding Index of code book is not entropy coded

8 bits for coding the index 6 bits for mean and 6 bits for gain of the vectorEntropy coding for brightness, block positions, etc.

Residual coding Any other residual coding means can be combinedNot used in the current implementation

Current implementation Off-line codecReal-time encoder with different VQ code book

Remarks on the results Better than anchor in high QP values (lower bitrate)Blocky, but with detail (no deblocking filter used)Certain isolated frames much better than othersSky areas especially good on some frames

Table 2. Strathclyde compression transform, Q15-F-18 and Q15-F-19.

Loop filtering for de-blocking, de-ringing, and corner-outlier artifacts

Q15-F-20 presents a technique which had been presented previously at the Whistler Q15/16meeting in July 1998 (Q15-E-22). This technique adaptively filters the areas of the picture usingclassifications on the values of coding parameters (e.g., on IDCT coefficient values).

For de-blocking, the technique uses a set of directional flags to indicate whether blocking artifactsare likely to occur horizontally, or vertically or both, and to filter with varying strengthaccording to these flag classifications. For de-ringing, similar adaptive means are used ascontrolled by two flag values. Special treatment of the classification flags is provided fordetermining new flags after motion compensation. In INTRA pictures, a method is provided todetect and correct “corner outlier” pixel values in which the extreme corner of an INTRA-codedblock may have a value that unusually deviates from the values of its neighbor pixels in otherblocks.

A D-1 demonstration of the filter’s performance was provided. Its performance was compared tothat of H.263+ Annex J with and without the de-ringing filter of Section 10 of TMN10. Onecomment was that the demonstration conditions seemed to often consist of material coded at asufficiently high quality level that filtering was less necessary than would be seen under moreadverse conditions. When compared with Annex J without Section 10 of TMN10, the presentedfilter sometimes had somewhat visibly improved quality (with smooth areas sometimesrepresented more smoothly and edges sometimes retaining greater crispness), and sometimes didnot. When compared with Annex J with Section 10 of TMN10, the presented filter was lessblurry. The reduced blurriness was sometimes judged to be a subjective improvement, butsometimes was not.



MVC - Affine MC of segmented frame with multiple transformIntra Frame Coding Pixel prediction (7 directions) and DC prediction

Prediction error is coded; similar concept of Inter prediction DC transmitted separately Different VLC tables from Inter coding One more 4x4 ECVQ code book for /8 direction Three directional 8x8 KLT Predetermined most probable basis functions

Inter Frame PredictionM.C. Segmentation of frame

Quad-tree like splitting Composed of 8x8 blocks Motion assisted merging for motion field codingAffine Motion Model Motion of all pixels are modeled by affine function Six motion coefficients affine function Subpixel values by cubic spline interpolation (separable 2D filtering with 4x4 window)

M.E. Motion estimation and split decisionGauss-Newton minimization 5 iteration on subsampled imageMotion model adaptation

Residual codingTransform Multiple transform (4x4)

Multishape DCT, directional VQ, extrapolation 7 ECVQ code books, (sizes are 100 to couple of 100) Low/high energy horizontal Low/high energy diagonal Low/high energy texture Flat

Attribute Classification concept Classified by localizing active area Directionality of activity Calculated on prediction frameClassifier Calculate variance for 4x4 pixels in predicted frame 5 different cluster sizes: 8x8, 4x8, 8x4, 3x8, 4x4 4x4 block is coded one of multiple transform methods Other sizes by DCT

Current implementation No long-term memory is usedRemarks on the results Improvement significant in both Inter and Intra

Similar bit savings percentage in low to high bitrate rangeNever worse than referenceSmoother, less blocky, less ringing, very good on sky areas

Table 3. Nokia affine MC with directional VQ, Q15-F-24.

The complexity of the presented filter was described as being a significant improvement relativeto the use of Annex J with Section 10 of TMN10, and this was presented as a key part of why thisfilter should be considered for adoption as a loop filter. It should be noted that a significantmotivation for the adoption of Annex J was reduction in complexity allowed by obviating the needfor an extra frame store in a decoder when using the Annex J technique as a loop filter ratherthan as a post filter. At least two apparent criteria would need to be well measured to justifyadoption of another loop filter annex. The first is the relative quality of the filtered content ascompared to Annex J, and the second is the complexity of the filter (e.g., as compared to eitherAnnex J or a good post-filter design or as compared to overall decoding complexity).



Long-term frame memory with affine MCIntra Frame Coding Identical to the anchorInter Frame Prediction

M.C. Long-term frame memory with affine MC 6 parameters affine transform Reference picture selection is signaled MB by MB

M.E. Searching affine motion parameters Gradient-based approach

Coding Affine parameters for warped version of reference frames Transmitted in picture header

Residual coding Identical to the anchorCurrent implementation Long-term memory of 10 previous coded frames

32 affine warped reference framesRemarks on the results Appeared generally sharper and “cleaner” than reference

Reduced ringingGenerally reported 2-3 QUANT values lower than reference

Table 4. U. Erlangen-Nurem. and 8x8 long-term memory with affine MC, Q15-F-33.

Post-filtering for de-blocking and de-ringing

Q15-F-23 (M. Hong, T. Choe, and C. Yon, LG electronics) presents a post-filtering method using aregularization method with a smoothing constraint applied with four directional smoothnessfunctions. The method provides an iterative filtering approach; results show use of either one andtwo iterations. A single filtering method was applied to obtain both a deblocking and a de-ringingbenefit. A consistent PSNR gain was shown as a result of using this filtering method (which isoften not seen from post-filtering methods), relative to H.263+ Annex J (section 10 of TMN10 wasapparently not used).

A D-1 demonstration of the filtering method’s performance was provided, relative to the use ofAnnex J without Section 10 of TMN10 (although there was some problem with the CB, CR dataformat on the D-1 tape). There appeared to be some relative benefit shown in some cases (relativeto Annex J use only). A slight improvement was seen, with some reduction in ringing, although itwas seen overall as visually about the same quality. There was also a reddish tinge to somesequences for an unknown reason.

The complexity of the presented method was estimated by the proponent as about half thecomplexity of the combination of H.263+ Annex J with Section 10 of TMN10. The filteringmethod was presented as potentially providing an improvement to the test model reference design(an improvement in both complexity and quality), and should be considered more in the future inlight of its performance and estimated complexity - although further testing seems necessarybefore such action could be justified.

SAMPLE-ADAPTIVE PRODUCT-CODE QUANTIZER

Q15-F-21 (D. Kim, N. Shroff, J. Kim, and Y.-M. Park, Purdue University and LG Electronics)gives a method that seeks to improve the coding fidelity of “busy” blocks in the coded content(blocks which have significant coded content). This method differs significantly from “adaptivequantizer” designs seen at previous Q15/16 meetings and designated as an H.263++ KeyTechnical Area. The premise of this design was to use block classification (without sideinformation) to determine which blocks had a certain level of busyness, and to refine thequantized representations of the coefficients in these blocks. The inverse quantization rule waschanged, using an E8 lattice to represent the refining data. When compared with a referenceencoder (TMN8) the technique showed some (relatively minor) PSNR gain in the busy blockareas, although it appeared essentially neutral in overall picture PSNR fidelity. A demonstrationwas provided to show the performance of the method. The benefit was unclear in thisdemonstration—it may have been roughly equivalent in performance to the use of a quantizationparameter one increment smaller, but was not a very visible improvement.



EZW-STRUCTURED DCT CODING

Q15-F-22 (Y.-A. Jeong, C. Yon, and Y.-M. Park, LG Electronics) presents a method thatrestructures the DCT coefficient filtering structure along the lines of embedded zero-tree wavelet(EZW) coding. The resulting design offers enhancements in coding efficiency, rate controlprecision, scalability, and error robustness. Significant gains in coding performance werereported in both INTRA and INTER coding performance. When compared with H.263 with highquantization step sizes, a gain was reported up to 1 dB for INTER coding, and up to 2.5 dB forINTRA coding. With smaller step sizes, the reported gain diminished. A demonstration ofperformance was provided, although not using the anchors generally used by the group (e.g., notusing Annex I Advanced INTRA coding and not using Annex J deblocking). Some scenes (HallMonitor, Foreman, News) showed a quite visible performance advantage in this comparison. Thegroup thought that this method showed promise, and thought it to be potentially relevant toH.26L, where more freedom is available for significant restructuring of the codec design.

DATA PARTITIONING FOR ERROR RESILIENCE

Q15-F-29 presents further data on the use of data partitioning for error resilience enhancement.The scheme presented was basically the same as had been proposed at the Whistler meeting (e.g.,in Q15-E-19 and Q15-E-20). The proposed method was tested using essentially the testconditions recently agreed upon for use within the group (except for not using R-D optimization,which was argued to be not entirely relevant). A significant gain in both subjective and objectiveterms (e.g., 2 dB) was shown, particularly under conditions in which the reference design resultedin lost frames. Less gain was apparent with better channels, partly to due an increase inoverhead and INTRA update information (which is partly simply to be expected, but which mightbe improved upon by better use of the same syntax). The method proposed was tested in someways using a design that resembled slice structured coding, although it was not precisely usingthe slice design in H.263+. Further work appeared to be needed on this and some other aspects(e.g., investigating why so many frames were lost in the referenced anchors). A demonstration ofrelative performance was shown, and it was agreed that the difference was quite visible. Therewas some question why the anchor reference implementation did not INTRA update certaincorrupted areas of the scene - and this appeared to be due to a fault in the design of the test modelreference conditions, resulting in not updating skipped areas of the picture. The method wasdemonstrated both with and without the use of the MUX model (applying errors just to thebitstream rather than using the flags and CRCs available from a mux as in the mux-modelreference which essentially used Q15-F-16), although only test results using the mux model wererequested by the group. Without a mux for providing error indications, the anchor performancesuffered, resulting in a larger relative improvement for the data partitioning method.

The data partitioning method has received a significant level of interest within Q15/16, is adesignated Key Technical Area (KTA) of H.263++ development, and appears to be maturing as apotential tool for an H.263++ annex (as reported in the H.263++ ad hoc report Q15-F-09).However, it has still not been adopted in any official draft, and the group thought it wise tocontinue its current status until further verification of test results are available. Given thestability of the design and the level of interest shown by the group, two proponents, Dr. J.Villasenor (UCLA) and D.-S. Park (Samsung), were jointly appointed to write proposed draft textfor consideration when verification of test results are available. The test conditions agreed uponfor evaluation of such proposals were also refined at this meeting, and these modified conditionsare described below (under Test Model, Software, and Encoding).

ENHANCED REFERENCE PICTURE SELECTION

Q15-F-32 (T. Weigand, N. Färber, and B. Girod, U. Erlangen-Nuremberg and B. Andrews, 8x8Inc.) is a proposed draft for an H.263++ annex on the coding efficiency performance improvementfeature known as enhanced reference picture selection (also known as long-term memory); itdescribes the text which could be adopted for this feature. The proposed syntax is based on thecore experiment description on Enhanced Reference Picture Selection (Q15-E-52). Significantcoding efficiency performance improvement was demonstrated for this feature at the Whistler andTampere meetings (e.g., Q15-E-25 and Q15-E-44), and this feature has also been mentioned aspromising in other proposals (e.g., Q15-E-17 and Q15-F-11).



Further results are also presented for this method in Q15-F-33 (T. Weigand, N. Färber, and B.Girod, U. Erlangen-Nuremberg and B. Andrews, 8x8 Inc.). Additional results with additionalperformance improvements were presented in this contribution using an affine motion model formotion compensation.

Q15-F-39 (M. Gallant, U. British Columbia) reports on initial efforts (no suitable results yet)toward verifying the performance of the long-term memory motion compensated prediction coreexperiment described in Q15-E-52. This contribution also provided some clarification on theintended design. The authors intend to do further work on the topic.

This coding efficiency enhancement technique has also received a significant level of interestwithin Q15/16, is a designated Key Technical Area of H.263++ development, and appears to bematuring as a potential tool for an H.263++ annex (as reported in the H.263++ ad hoc reportQ15-F-09). However, it has still not been adopted in any official draft, and Q15/16 thought it wiseto continue its current status until further verification of test results are available. Despite theongoing investigation of further enhancement of this technique by adding an affine motion model,the basic design has been stable, and the level of interest shown by the group led to theappointment of T. Wiegand (U. Erlangen-Nuremberg) to write a proposed draft text forconsideration when verification of test results are available.

TEST MODEL, SOFTWARE, AND ENCODING

Q15-F-08 (K. Hibi, Sharp) is the Test Model, Software, and Encoding ad hoc report.

Reference Software Availability

The University of British Columbia (UBC), which has made software available for the use of thepublic in experimenting with H.263, provided Q15-F-40 (M. Gallant, G. Côté, S. Wenger, and F.Kossentini, UBC). UBC will continue to allow public use of version 3.2.0 of this software, fromweb site http://www.ee.ubc. ca/spmg/research/motion/h263plus and ftp siteftp://dspftp.ubc.ca/pub/tmn/ver-3.2. They will not provide updates to this software (even bugfixes).

UBC plans to enhance its software, and to include updated Q15/16 test model designs in thatsoftware. However, they have indicated that they can no longer afford to conduct this activitywithout some monetary support. They stated that UBC is not attempting to make a profit fromproviding access to this software for research purposes, but is only seeking a way to share someof the costs of this project. UBC will provide software which includes some very significantenhancements (enhanced GUI, speed-up features, enhanced performance, etc.) to the priorversions of this software, but will seek a cost-sharing fee (e.g., US$1000, plus US$500 for eachupdate) from most users (exceptions may be made for certain organizations such as academicinstitutions).

Video Coding Quality Experiments

Q15-F-36 (S. Wenger, TU Berlin and TELES) gives the results of some experiments usinghigher-quality coding than usual. Various source material was encoded at 768 and 1536kbit/sec, and the results of this test were shown to the group. The basic coding method used wasthat of the test model, although without rate-distortion optimization due to an interaction problembetween rate distortion optimization and rate control. This effort also uncovered a need to find away to operate the test model reference encoder design without allowing frame skipping and otheraspects not generally encountered in the work at lower bit rates (such as a potential need forbetter rate control for B pictures).

This experiment shows the high degree of effectiveness of the H.263+ standard at bit rates thatare not yet typical for conversational services. Q15/16 can benefit from an understanding of theissues involved in video coding at bit rates beyond those found commonly in their experiments.



Error Resilience Quality Experiments

Q15-F-06 (S. Wenger, TELES) reports that the ad hoc work on error resilience simulationconditions progressed well. The ad hoc report also mentions that some relevant error resiliencework is occurring in the Internet Engineering Task Force (IETF) and has resulted in a draftavailable to the public from the IETF site ftp://ftp.ietf.org/internet-drafts/draft-ietf-avt-fec-04.txtby J. Rosenberg and H. Schulzrinne; ftp://ftp.ietf.org/internet-drafts/draft-ietf-avt-reedsolomon-00.txt by J. Rosenberg and H. Schulzrinne may also be of interest.

Packet multiplex simulator for error resilience testingQ15-F-16 (G. Sullivan, PictureTel, et al.) contains software for the simulation of the effects of anerror-prone channel on video data. It simulates the Level 2 mobile multiplex of H.223 Annex Bprocessing video data using Adaptation Layer 3 transmission. It allows an error pattern fileobtained from channel measurements or simulation conditions to be applied to the pseudo-multiplexed bitstream and for the effects of this corruption to be simulated on the multiplexeddata. The program was written using a design developed collaboratively once it becameapparent that no other good packet-oriented H.223 mobile mux simulator was available forQ15/16 experiments.

While this simulation program is not perfect, Q15/16 believes that it represents a reasonablyaccurate model of the effects of a mobile multiplex on video data and provides a good basis forsuch experiments. As a result, Q15/16 plans to use this model, or an improved version of it, infuture experiments.

The contributor indicated that some enhanced version of the program will probably be developed(such as allowing video packets longer than 254 bytes), but that the basic overall degree ofrealism of its behavior is not expected to change significantly as a result.

Mobile error resilience test conditionsQ15-F-38 (S. Wenger, TU Berlin, TELES, and UBC) proposes an alteration of the test conditionsplanned for error resilience testing: it describes a new technique known as rate-distortionoptimized error resilient coding. This technique appears to provide a significant benefit inperformance by adding an awareness in the encoder of the likelihood that data will be lost, and ofthe presumed error concealment to be applied in the event that the data has been lost. However,because Q15/16 did not have time to assess this method fully, it was judged premature to ask forthe error resilience tests to be conducted based on its design. It was also stated that theperformance gain which could be obtained by use of this technique is likely to be applicable tomany other encoder designs (e.g., with data partitioning) in addition to being applicable to the useof the existing H.263+ syntax.

Q15/16 decided to simplify the error resilience test conditions (more along the lines described inAlternative Proposal number 2 of Q15-F-38) because of problems in implementing the prior testmodel design for rate-distortion optimization and because of its questionable relevance errorresilience experiments. Issues concerning the possible anomalous content at the beginning of theRayleigh error pattern files were also discussed, leading to adoption of a plan to jump past 80bytes of these files when starting simulations (but using circular wrapping so that these bytesmay be used later). Several of the error patterns that had previously been decided uponappeared to provide little useful information and were dropped from the experiment plan. Errorpatterns for higher bit rates (e.g., 128 and 384 kbit/s) are desired by the experimenters, and helpis requested in obtaining such data.

The common conditions adopted by the group for error resilience testing were written into theoutput document report, Q15-F-45 (S. Wenger, TU Berlin), and the test model reference encodingand decoding design for use of these test conditions were similarly described in output consensusdocument, Q15-F-46 (S. Wenger, TU Berlin and G. Côté, UBC).

RATE CONTROL

Frame-rate controlQ15-F-14 (H. Song, J. Kim, and C.-C Jay Kuo, USC) describes methods for controlling framerate adaptively during changes in scene motion activity and channel bit rate. The selection of an



appropriate frame rate for use in encoding a video sequence is a fundamental part of the design ofa video coding system, yet it has not previously received significant attention in the test modeldesign efforts: more typically we simply pick a frame rate we think is appropriate forexperiments using a video sequence, without regard for the fact that a real-time encoder wouldneed to decide on its own what frame rate is appropriate, and would need to do so without priorknowledge of the difficulty of the content of the video sequence. The design in Q15-F-14 attemptsto maintain both perceptual visual motion smoothness and adequate picture fidelity whileadapting the frame rate for encoding based on the bit rate available for use and the amount ofmotion in the scene.

Q15/16 agreed that this topic is worthy of study, and hopes to do further work to assess thefeasibility of this method and how it might be improved. It is hoped that some reasonable framerate control scheme can be soon adopted into the test model reference encoder design, and thismethod appears to be a good initial candidate.

Fixed frame-rate codingQ15-F-34 (G. Côté, M. Gallant, and S. Wenger, UBC) points out that no method is currentlyfound in the test model reference encoder design which provides a truly fixed frame rate fortesting in implementations that need this feature. Q15/16 authorized this method for inclusion inthe next test model.

Interference between rate control and rate-distortion optimizationQ15-F-35 (M. Gallant, G. Côté, and S. Wenger, UBC) reports further on the problem that the testmodel designs for rate control and rate-distortion optimization do not work well together. Thiswas also noted in the report from the Whistler meeting (Section 4.7 of Q15-E-53). Q15/16 expertson rate control and rate-distortion optimization were asked to find a solution for this problem. If areasonable solution is found, it is authorized to be included in the next test model.

DECODER POST-PROCESSING ENHANCEMENT

De-blocking and de-ringingQ15-F-13 and Q15-F-23 provide for enhancing the test model reference design for post-processing filtering. The results are described above.

Q15-F-41 (G. Sullivan, PictureTel) seeks clarification of the current description found in the testmodel. Q15/16 agreed on the need for this clarification, and also noted that another aspect of thefilter design which had previously been authorized for inclusion in the test model had beenneglected in the current description. The other aspect was modification of the processing of theedges of blocks in the post-processing when Annex J is not in use. This was based on Q15-B-14 ofthe Sunriver Q15/16 meeting (Sept. 1997), which was intended to have been adopted into the testmodel (as reported in CSR-T Vol. 8.9). Q15/16 agreed to revise its test model to include both theclarification requested in Q15-F-41 and the modification requested in Q15-B-14 as authorized foradoption in Sunriver.

Frame rate interpolationQ15-F-15 (T. Kuo, J. Kim, and C.-C Jay Kuo, USC) presents a technique for a decoder toincrease the output video frame rate by estimating the contents of the depicted scene at timeinstants between those of the transmitted pictures. The method used morphological processingoperations in conjunction with a human visual system model and change detection. The groupexpressed a keen interest in this concept, and discussed some previous uses of it. Thedemonstration shown used only a rather small set of test material (including some relatively easycontent such as “Miss America” and “Suzie”) and did not show real-time display of results. Therewere some comments that objectionable artifacts might occur if this concept was applied todifficult scenes. It was also noted that this scheme increased delay by requiring the decoder todelay the display of the latest decoded picture until after any interpolated pictures had beenshown. The complexity of the method was not well understood by the group, although it wasrecognized that having such a technique which provides enhanced quality for decoders with extracapabilities makes this an intriguing topic. Pending further investigation, Q15/16 decided not toinclude this technique in its test model at this time.



DELIVERY OF VIDEO OVER ATM USING H.323

Q15-F-28 (D. Y. Suh and J. M. Seok, Kyunghee Univ.) describes the delivery aspects of video useover ATM networks using H.323. Various aspects of variable bit-rate, constant bit rate, and“available bit rate” video control are discussed, including in particular the potential for providinga rate-control buffer within the network. Statistical multiplexing can be used in suchenvironments to allow the bit rate of video streams to have high peak bit rates while carryingmore streams than could be supported at the peak bit rate of each stream. The hypotheticalreference decoder design of H.263 Annex B was thought to provide a relatively good ability toadapt H.263 video codec performance to these varying bit-rate conditions (given a sufficientlygood use of the design). The amount of overhead required by the 53-byte cell size of ATMnetworks is also mentioned as a factor affecting the characteristics of video use over suchsystems. Q15/16 appreciated the information about these important networks and encouragesfurther study of the issues involved.

TEST MODEL DEVELOPMENT RESULTS

Since a number of changes were authorized in regard to the test model document (currentlyTMN10, the same as Q15-D-65), the group has decided to produce a new version of the testmodel: see above and input documents Q15-F-38 and Q15-F-41, and the output document Q15-F-46 (S. Wenger, TU Berlin). The editor was also asked to review the last few meeting reports toensure that decisions made on the test model are accurately reflected in the document.

The current editor of the test model, T. Gardos (Intel), noted that he has limited ability to continueas editor, and would prefer the designation of another editor. Q15/16 has therefore jointlyappointed S. Wenger (TELES), F. Kossentini (UCB) and M. Gallant (UBC) as the editors of thenext version.

SIGN LANGUAGE AND LIP READING APPLICATIONS

An updated draft Q15-F-12 was provided on the use of video coding in sign language and lipreading applications. Q15/16 would generally prefer to have its content adopted in some form bythe ITU-T. However, two issues were identified that need to be resolved first:

1. The document refers to a particular video sequence data set which is available on thePictureTel informal ftp site (ftp://standard.pictel.com/video-site/sequences/irene). However, ifthis data set is to be referenced in a formally-approved document such as an appendix to aRecommendation, then it should probably be made available in a more official manner from theITU-T itself. The ITU-T TSB will be consulted regarding the potential difficulty of the ITUmaking this electronic data set available. The issue of the existence of a copyright on this dataset also needs to be addressed in the ITU archiving and publishing process evaluation.

2. Some members thought that it would be best if the document is adopted in a form other than asan appendix to H.263, since its content is relevant when using essentially any particular videocodec design (e.g., H.261, H.262 or H.26L).

WORKPLAN FOR H.263++

As reported in Q15-F-09, the ad hoc report on H.263++ enhancement efforts (G. Sullivan,PictureTel), seven of the eight prior Key Technical Areas (KTAs) of H.263++ development wereaddressed in contributions to this meeting. Two of these KTAs in particular have generated anongoing high degree of interest with relatively stable designs (data partitioning and enhancedreference picture selection). However, the group has still not adopted any draft for the adoptionof any features into H.263 version 3, and this status remained unchanged at this meeting.Members were appointed to create proposals for draft text for potential adoption in these twoKTAs at the next meeting.

Given the lack of a current draft, the workplan has been modified (although without any changein the target dates for Determination and Decision). Actual timing of the project will need to beevaluated as the work progresses, and different features may proceed at different paces. Thecurrent list of Key Technical Areas for the H.263++ project was also reviewed and somewhat



altered. The current workplan is shown below in Table 5 and the new KTA list is provided inTable 6.

Note that the last KTA in Table 6 includes specifying how a decoder would detect and react toerrors and packet losses, so that an encoder may optimize its encoding with an assumption of howthe decoder would react (e.g., see above on error resilient rate distortion optimization).

WORKPLAN FOR H.26L

The H.26L project reached an important milestone. This meeting included the first formalevaluations of responses to the Call for Proposals issued at the Jan/Feb 1998 meeting of SG16;Q15/16 is pleased that several proposals provided the information essentially as requested.Several other technology proposals also appear to have relevance to the H.26L project.

Meeting Approx.Date

Type Milestone

SG16-1 17 March 97 Study GroupQ15-A 24 Jun 97 ExpertsQ15-B 8 Sep 97 Experts Adoption of WorkplanQ15-C 2 Dec 97 Experts Start of Significant EffortSG16-2 26 Jan 98 Study GroupQ15-D 21 Apr 98 ExpertsQ15-E 21 Jul 98 ExpertsSG16-3 14 Sep 98 Study GroupQ15-F 3 Nov 98 ExpertsQ15-G 23 Feb 99 Experts First Formal Draft AdoptionsSG16-4 May 99 Study GroupQ15-H Jul 99 Experts Last Formal Draft AdoptionsQ15-I Nov 99 Experts Final Draft for DeterminationSG16-5 Feb 00 Study Group DeterminationQ15-J Apr 00 Experts Bug-checkingQ15-K Jul 00 Experts Final Draft for DecisionSG16-6 Nov 00 Study Group Decision

Table 5. H.263++ Workplan.

Key Technical Area Example RelevantDocument(s)

Error Resilient Data Partitioning Q15-F-29Enhanced Reference Picture Selection Q15-F-32, Q15-F-33, Q15-F-39Affine Motion Compensation Q15-F-33Adaptive Quantization Q15-E-21, Q15-E-29IDCT Mismatch Reduction Q15-F-11Deblocking and Deringing Filters(normative or informative)

Q15-F-13, Q15-F-20, Q15-F-23,Section 10 of TMN10 Q15-D-65

Error Concealment(normative or informative)

Q15-F-29, Q15-F-38, Q15-F-45,Q15-F-46

Table 6. H.263++ Key Technical Areas.

The following four proposals were evaluated:• Q15-F-11 (G. Bjøntegaard, Telenor)• Q15-F-18 and Q15-F-19 (R. Fryer, U. Strathclyde)• Q15-F-24 (Nokia)• Q15-F-33 (T. Weigand, E. Steinbach, and B. Girod, U. Erlangen-Nuremberg, and B. Andrews,

8x8)



These proposals are summarized in Tables 1-4, above.

The proposal in Q15-F-22, which was evaluated in joint technical evaluation category, wasoriginally intended for H.26L proposal, although it addressed a residual texture coding part only.It was agreed that the content of Q15-F-22 would be treated as a candidate for H.26L byincluding it in the list of action items for the progress of H.26L work.

H.26L SCHEDULE, KTAS, AND FUNCTIONALITY DEVELOPMENT AREAS

Significant progress was made, in particular, in the coding efficiency area, although no particularproposal fully meets all the requirements for H.26L at this time.

According to the discussion on the similarities in technical natures of H.26L proposals, six KeyTechnical Areas of interest for the H.26L project have been identified:

Inter frame prediction coding(1) Variable-size block-based segmentation

-Quad-tree like decomposition of frame-Small block size to 8x8, 4x4

(2) Long-term memory-Increase candidates of reference materials for prediction-Combination with other prediction scheme (e.g., affine warping)

(3) Affine motion model-Adoption of six parameter affine motion model-Coding method and syntax of affine motion parameters-Searching of affine motion parameters

Inter frame residual coding (also applicable to Intra texture coding)(4) 16-sampled waveform coding (4x4)

-Coding with smaller block size than 8x8 DCT(5) VQ, EZW with DCT, multiple waveform coding

-Adoption of (variable-dimension) VQ scheme-Embedded EZW structured coding-Selection from multiple coding modes

Intra frame coding(6) Directional (pixel-prediction) Intra coding

-Improvement of prediction in Intra coding-Consideration of activity direction of picture

Note: Coordination with JPEG-2000 to obtain an information of the state of the art of Intra codingtechnology would be helpful.

Functionality development areas to be covered by H.26L were also discussed. The H.26Lproposals mainly focused on performance improvement with higher coding efficiency, which isconsidered to have a first priority in technology development. However, all functionalities listed inthe H.26L requirement document should be finally supported by the H.26L standard. Thefollowing list was agreed, which shows the functionality area to be included in the H.26Ldevelopment process.

(a) Simplification “back to basics” approach- Adoption of a generally simple, straightforward design using well-known building-blocks- For example, use of one VLC for all parameters to be coded

(b) Low delay (e.g., no B pictures)- Use of more reference pictures for prediction (e.g., long-term memory)- Constant bitrate coding of each frame to reduce buffering delay- To be extendible to different delay environments

(c) Error resilience- Packet loss resilience- Mobile channel corruption resilience

(d) Complexity scalability in encoder and decoder



- Asymmetry of encoder and decoder processing complexity- Scalability between amount of encoder processing and achievable quality

(e) Full specification of decoding (no mismatch)- Resolve mismatch problem (e.g., integer transform, VQ,…)

(f) High quality application- Performance improvement in higher bitrate- Applicability to entertainment-quality applications

(g) Network friendliness- Ease of packetization- Information priority control

Discussion of the H.26L workplan addressed two major items; how to connect goals withprogress, and how to get from here to a draft.

The next step in the H.26L development is to find initial consensus. The members were,therefore, encouraged to come to an integrated proposal by investigating KTAs of H.26L at thenext meeting, when the first algorithm draft (not necessarily a complete draft) is expected to bedefined. The proponents were also requested to provide a detailed description of their proposedalgorithm for cross validation and for the definition of the algorithm draft.

It was emphasized that network friendliness of the video codec, for example, a coded informationstructure suitable for packet transmission, is an important feature supported by H.26L. Q15/16agreed with this opinion, but first may work on a definition of basic coding algorithm, and thenmove to consider such aspects as a definition of the top level syntax suitable for transmittingthrough packet based network.

“Simplification” will be a key word of technology development for H.26L.

The group discussed the test conditions and results of H.26L proposals to be provided at the nextmeeting. The error resilience test results of H.26L proposals were requested, according to theerror resilience test method described in Q15-F-45 and Q15-F-46. Information on the transfermethod of coded information and its implication for the proposed codec over mobile, IP real-time,and IP store-forward environments should also be provided.

Q15/16 agreed that the Test Model for Long term (TML) and relevant core experiments should bedefined in a later stage. One opinion stressed that Q15/16 should be careful with how to definethe draft H.26L standard. Q15/16 discussed a process of defining the draft and agreed thatcross validation of coding elements was indispensable and only validated tools could beincorporated into the Test Model and the draft of H.26L standard.

Q15/16 also recognized that reaction from users and others should be taken into consideration.Therefore, it was decided to learn the expectations of users of the video coding standard, althoughit is a moving target and may be difficult to estimate. As a first attempt, Q11-14/16 was asked inthe liaison (Q15-F-48) for their guidance to this group.

It was agreed that the current time schedule of H.26L project would be kept unchanged. Thecurrent H.26L workplan schedule is in Table 7.

PLANS FOR FUTURE WORK, AND AD-HOC COMMITTEE DESIGNATIONS

The future workplans for H.263++ and H.26L projects and for deployment of prior video codecstandards were briefly reviewed. The following ad hoc committees were established to progressthe work between now and the next meeting:

• H.263+ in H.320 [activity now to be primarily in Q.11] (S. Gupta)• Error Resilient Simulation Conditions and Evaluations (S. Wenger)• Test Model and Software Development (K. Hibi)• H.263++ Development (G. Sullivan)• H.26L Development (K. Hibi)

Gary Sullivan, PictureTel Corporation



Meeting Approx. Date Type MilestoneSG16-1 17 March 97 Study GroupQ15-A 24 Jun 97 ExpertsQ15-B 8 Sep 97 ExpertsQ15-C 2 Dec 97 Experts Modified Workplan AdoptedSG16-2 26 Jan 98 SG16 Issue Call for ProposalsQ15-D 21 Apr 98 ExpertsQ15-E 21 Jul 98 ExpertsSG16-3 14 Sep 98 Study GroupQ15-F Nov 98 Experts First Formal EvaluationsQ15-G Feb 99 Experts First Draft Text and Test ModelSG16-4 Apr 99 Study GroupQ15-H Jul 99 ExpertsQ15-I Nov 99 Experts Final Major Feature AdoptionsSG16-5 Feb 00 Study GroupQ15-J Apr 00 ExpertsQ15-K Jul 00 ExpertsSG16-6 Nov 00 Study GroupQ15-L Apr 01 ExpertsQ15-M Jul 01 ExpertsSG16-7 Aug 01 Study Group DeterminationQ15-N Oct 01 Experts Bug-CheckingQ15-O Jan 02 Experts White Document GenerationSG16-8 May 02 Study Group Decision

Table 7. H.26L Workplan

Communications Standards Reviewformerly Communications Standards Review-Telecommunications

regularly covers the following committee meetings:

TIA TR-29 FacsimileTR-30 ModemsTR-41 User Premises Equipment

ITU-T SG8 Telematic TerminalsSG15 WP1 Network AccessSG16 Multimedia

ETSI ATA Analog Terminal AccessDTA Digital Terminal AccessTC SP Speech ProcessingTC STQ Speech Transmission QualityTIPHON Voice Over InternetTM6 Transmission & Multiplexing

10902



Q15/16 VIDEO CODING EXPERTS GROUP MEETING ROSTER, NOVEMBER 3 - 6, SEOUL,REPUBLIC OF KOREA

Gary Sullivan, PictureTel (US) Q15/16 RapporteurKeiichi Hibi, Sharp Corporation (CIAJ (Japan) Q15/16 Associate Rapporteur

FinlandNokia Research Ctr Ari HourunrantaNokia Research Ctr Marta KarczewiczNokia Research Ctr Jani LainemaGermanyTELES AG / TU Berlin Stephan WengerUniv. of Erlangen-Nur. Thomas WiegandJapanFujitsu Wolfgang NiemSharp Keiichi HibiKoreaHankuk Aviation Univ. Dae-Gwon JeongKAIST Seung-Kyu LeeKAIST Yung-Lyul LeeKAIST Kyuchan RohKJIST Yo-Sung HoKJIST Daehee KimKyunghee Univ. Doug Young SuhLG Electronics Tae-Eun ChoeLG Electronics Min-Cheol HongLG Electronics Soo-Young HuhLG Electronics Yeong-An JeongPusan National Univ. Jae-Ho KimSamsung Elect. Jae-Woo JungSamsung Elect. Sang-Ug KangSamsung Elect. Dong-Seek ParkSamsung Elect. Jeong-Hoon ParkSamsung Elect. Jeong-Wook SuhSamsung Elect. Kook-Yeol YooNorwayTandberg Telecom Tom-Ivar JohansenTelenor Satellite Gisle BjöntegaardSwedenTelia Research Marie DahlqvistTaiwanCCL/ITRI Fang-Chu ChenUKImageCom Ltd. John MasonOrange PCS Richard CraneUniv. of Strathclyde Richard FryerUSA8x8, Inc Barry AndrewsPictureTel Gary SullivanPurdue Univ. Dong-Sik KimPurdue Univ. Ness ShroffReal Networks Gary GreenbaumSorenson Vision, Inc Chien-Min HuangUCLA Matthew FongUCLA Maximilian LuttrellUCLA John VillasenorUSC Jongwon Kim



REPORT OF ETSI PROJECT DIGITAL TERMINAL AND ACCESS (DTA)MEETING #4, NOVEMBER 16 - 18, 1998, BERLIN, GERMANY

Note: All documents referred to in this meeting report are available to ETSI memberselectronically from the EP DTA Permanent (98) area of the ETSI Server at www.etsi.fr.

ETSI/EP DTA(98)029r01 is the Agenda. ETSI/EP DTA(97)025B is the report of DTA#3.ETSI/EP DTA(98)032R1 is the Chairman’s report.

EFFECTS ON EP DTA OF THE R&TTE DIRECTIVE

The Chairman introduced the R&TTE (Radio and Telecommunications Terminal Equipment)Directive, and its effects on EP DTA. Details are in ETSI/EP(98)37 (early draft Workplan),ETSI/EP(98)43 (draft R&TTE Directive), ETSI/EP(98)44 (early draft work plan for R&TTE), andETSI/EP(98)45 (draft Guide to the production of Harmonized Standards for Application underthe R&TTE directive).

M. Sharpe (ETSI Secretariat) gave a further explanation of the new Directive and outlined therole of ETSI OCG (Operational Coordination Group) TG6. TM6 has been asked to develop aconsolidated ETSI standardization program for Harmonized Standards under the proposedR&TTE Directive.

ETSI/EP(98)48, the Liaison from TG6, requests each Technical Body that has responsibility forthe production of Harmonized Standards under the R&TTE Directive to produce a draft workprogram for consideration by TG6 at their meeting on December 3, 1998. ETSI/EP(98)48 alsoincludes a draft list of ENs used for type approval to be examined for possible conversion toHarmonized Standards in the first phase, and a draft list of existing type approval standards tobe evaluated for possible conversion into Harmonized Standards in the second phase. TG6 plansto produce an overall ETSI work plan for agreement by OCG at its January 1999 meeting.

DTA considered various scenarios, including the question of what happens to existing servicesthat are covered by National approval procedures.

It was agreed that the DTA Chair would produce a draft for TG6; he will e-mail this draft to DTAmembers for comments. This draft work plan should consider the various possible scenarios andworking methods. Equipment groupings and new technologies will also be considered wheninformation becomes available. It was agreed that this draft work plan would be tentative due tothe uncertainty about some aspects of the R&TTE Directive. Note: This draft (EP DTA98)37R3)has been sent to EP DTA members for comment.

It was accepted that it is not the job of EP DTA to criticize the new Directive. If members havecomments, they should be addressed to their National delegations to ACTE (ApprovalsCommittee for Terminal Equipment).

DEFINITIONS , ABBREVIATIONS AND SYMBOLS

TR 101 380 (Definitions, abbreviations and symbols) was presented by the Rapporteur, B.Martenson. The TR (Technical Report) was provisionally approved with members being givenseven days from the end of the meeting to submit final comments. The Chairman thanked B.Martenson for his work as Rapporteur.

ISDN (NARROWBAND)

Basic Access Cabling ConfigurationsTSI/EP DTA(98)41 (P. Nooren, I. Keesmaat, KPN Research) contains proposals for an additionalwiring configuration to be specified for the S-bus. In ISDN Basic Access, two types of point-to-multipoint in-house cabling have been specified: short passive bus and extended passive bus.This submission suggests that a star (max. 100 meter long branches) configuration is a desirableoption as well. Members were sympathetic to this suggestion but pointed out that it was theresponsibility of CENELEC (European Electrotechnical Standards Committee) TC215 to specifycustomer premises cabling. P. Nooren will draft a Liaison to TC215 proposing such a change; he



will include the results of tests conducted by his organization. It was agreed that no change toTBR 3 or ETS 300 012 (ISDN Basic User-Network Interface [UNI]) would be required. Thisliaison will be submitted by e-mail for approval by EP DTA prior to submission to CENELEC(and copied to STC TM6, Sub-Technical Committee Transmission and Multiplexing).

TBR 8 Edition 2The approval of the National Vote on TBR 008 Edition 2 (ISDN Telephony 3,1 kHz teleservice;Attachment requirements for handset terminals, was noted and accepted. Comments werereceived from Germany, Italy, Netherlands, Slovenia and Sweden (see ETSI/EP DTA(98)40). EPDTA was also informed that a new ITAAB (ISDN Technical Advisory Ad hoc Board) AdvisoryNote is in preparation to cover some of the comments received.

Future ISDN WorkIt was noted that TBR 003 (ISDN Attachment requirements for terminal equipment to connect toan ISDN using ISDN basic access) and TBR 004 (ISDN Attachment requirements for terminalequipment to connect to an ISDN using ISDN primary rate access) now contain references tobase standards that are out of date. Furthermore, various ITAAB ANs exist and theAmendments have not yet been incorporated into the main text. However, in spite of this, it wasconsidered that the cost and resource required to update these TBRs could not be justified in thelight of the impending R&TTE Directive. EP DTA decided that they would not update TBRs 3and 4 unless a specific request was received from TRAC/ACTE (Technical RecommendationApplication Committee / Approvals Committee for Terminal Equipment) or the CEC (EuropeanCommittee for Standardization). It was thought that any repairs could be covered by ITAAB forthe time being.

FUTURE B-ISDN WORK

ETSI/EP DTA(98)28, Optical interface standardization (I. Marshall, Nortel), proposes aharmonized standard for optical interfaces, with a non-data rate dependent approval regime forthis category of terminal. DTA agreed that it is the appropriate ETSI Technical Body to dealwith such a matter. During the discussion, it was discovered that in those countries in whichsuch a service exists, only safety (and perhaps optical power limiting) requirements are applied.

Considering that a number of Network Operators within Europe are providing, or willing toprovide, Optical Fiber interfaces according to ITU-T Recommendation G.957, and thatcharacteristics of these network interfaces will not differ significantly between member states,ETSI EP DTA agreed that type approval requirements for connection to such interfaces shouldbe harmonized.

After detailed technical discussion regarding the degree of regulation of terminals intended toconnect at the NT1, it was noted that the network is normally self-protecting in respect ofnetwork harm, as defined in article 5(d). Accordingly, ETSI EP DTA decided to advise TRACthat the degree of regulation applied to such terminals should be limited to requirements underarticles 5(a), 5(b), and 5(c) of Directive 98/13/EEC.

ETSI EP DTA also noted that a number of ATM services are being delivered on electricalinterfaces conforming to ITU-T G.703 for which no current harmonized regulation applies.Therefore ETSI EP DTA agreed that any future regulation for such interfaces should be nogreater than that currently applied to ONP (Open Network Provision) leased lines, e.g., CTR24and CTR25.

Terminal Equipment utilizing optical interfaces incorporating Wavelength Division Multiplexing(WDM) technology are excluded from the above policy until further technical discussion has takenplace.

Finally, ETSI EP DTA agreed that the provision of these services be optional for the NetworkOperator.

ETSI EP DTA produced a liaison, ETSI/EP DTA(98)047R2, to TRAC expressing theseconclusions and requesting that TRAC endorse them.



APPROVAL OF EN 301 401

ETSI/EP DTA(98)33, Draft EN 301 401 (V1.2.2, 1998-11), Attachment requirements for DTE toconnect to public networks that have physical and electrical network presentations based uponthe ITU V-series Recommendations, was submitted for approval.

ETSI/EP DTA(98)34 describes 21 changes made to TBR 2, which was the starting point for thedraft document EN 301 401.

Members offered information on the missing V.12 connector arrangements; the Rapporteuragreed to update the connector clause to enable proprietary connectors to be used. It was alsoagreed to update the clause on Signal Element Timing provisions for testing.

The rapporteur agreed to incorporate these changes and submit the revised draft for approval (togo for PE - Public Inquiry) by e-mail. The draft will be available by December 7, 1998 with aview to approval early in January 1999.

SAFETY, PROTECTION AND EMC

The formation of the joint CENELEC/ETSI Safety Group JTC Safety was noted.

ETSI/EP DTA(98)36 (I. Marshall, Nortel) contains a proposal to remove references to ETS 300046 (ISDN Primary rate access - safety and protection) and ETS 300 047 (ISDN Basic access -safety and protection) in standards that predate the EMC directive. While this proposal wassupported by some delegates, other expressed reservations. The cost of revising the TBRs wasmentioned along with the position of ITAAB.

As these clauses in ETS 300 046 and ETS 300 047 relate to EMC (ElectroMagneticCompatibility) phenomena, it was agreed that the EP DTA Chair should write to TC ERM (ETSIcommittee on EMC and Radio spectrum Matters) seeking advice before taking action.

LIAISONS

ETSI/EP DTA(98)38 is a liaison from STC SPS5 which has the responsibility for TBR 33 (ISDN;Attachment requirements for packet mode terminal equipment to connect to an ISDN using ISDNbasic access) and TBR 34 (ISDN; Attachment requirements for packet mode terminal equipmentto connect to an ISDN using ISDN primary rate access). It requests that SPS5 be involved inany protocol work specification required in a Harmonized Standard under the R&TTE Directive.EP DTA agreed with this request; the Chair will reply to SPS5.

EP DTA WORK PROGRAM

ETSI/EP DTA(98)39, the EP DTA Work Sheets, were reviewed; it was agreed that the Chair willupdate them off-line.

ETSI/EP DTA(98)42 was introduced by J.-P. Senkeisen (Orsenna). It is a Work Item that hadoriginated in EP MTA (Multimedia Terminal and Applications) before its closure. The Work Itemcontains a specification (draft DE/TE-01066 Terminal Equipment: File Transfer: B channelaggregation and compression) on channel aggregation and compression techniques produced bythe EUROFILE Working Group. J.-P. Senkeisen requested that EP DTA adopt this Work Item.France Telecom and Datakom Austria supported adoption of the Work Item; Siemens objected.The exact status of the Work Item was questioned; M. Sharpe (ETSI) agreed to check this withthe ETSI database.

The Chair did not expect a decision to be taken at this meeting and asked all members to maketheir views known to him by December 18, 1998 (preferably by e-mail) on whether EP DTAshould adopt the Work Item.

FUTURE OF EP DTA

The Chair introduced discussion on the future of EP DTA. The Chair explained that the choicewas between EP DTA continuing as a separate entity or merging with another terminal-basedTechnical Body such as EP ATA. The Chair noted that he had received two e-mails on the



subject, one supporting a merger and one opposing a merger. ETSI/EP DTA(98)35 (I. Marshall,Nortel) argues for smaller committees (e.g., DTA) as they are less bureaucratic and thereforemore efficient and speedy. There was also concern that ATA as a larger committee wouldsubsume the DTA work.

During the ensuing discussion, the consensus was for EP DTA to remain as a separate entity andthat joint meetings should only be considered in special circumstances. There was a greaterdesire to have joint meetings with ITAAB than with any other Technical Body.

The Chair will report this decision to OCG with copies to EP ATA and ITAAB.

DTA MEETING ROSTER, NOVEMBER 16 - 18, 1998, BERLIN, GERMANY

David Maxey, BT GB DTA ChairFrank Jürgens, Deutsche Telekom AG DE DTA Vice ChairHosts: Deutsche Telekom AG and Siemens AG

Austria Datakom Austria Hans ZankeDenmark Tele Danmark Gunvor TindFrance ETSI Mike Sharpe

France Telecom Jean-Michel LescopFrance Telecom Georges SebekOrsenna Jean-Philippe Senkeisen

Germany Reg TP Thorsten SchildtSiemens Andreas Kamke

Hungary Comm. Authority Sandor SzilagyiNetherlands KPD Research Pieter NoorenSweden Ericsson Bus. Sys. Berndt MartensonSwitzerland Afcon Claude A. Polier

Swisscom AG Christof KaeserUK Babt Hilton Carr

Nortel Ian Marshall

COMMUNICATIONS STANDARDS REVIEW ALSO PUBLISHES:

Communications Standards Summary (ISSN 1075-5721), a quarterly publicationreporting on all active projects and recently completed standards of the TIA’s(Telecommunications Industry Association’s) TR-committees. Authorized by TIA.

For more details visit http://www.csrstds.com. To receive a complimentary issue ofeither of CSR’s technical journals, please contact Elaine Baskin, tel +1 650 856-9018,fax +1 650 856-6591, e-mail: [email protected]



REPORT OF Q11/16 CIRCUIT-SWITCHED NETWORKS (CSN)MULTIMEDIA TERMINAL AND SYSTEMS EXPERTS GROUP

NOVEMBER 17 - 20, 1998, TORINO, ITALYThe Q11/16 Rapporteur is T. Geary (Conexant, USA, previously Rockwell). There were 29contributions, of which 27 were technical (see document list Q11-H-00r2). The meeting report isQ11-H-25d3.

INTELLECTUAL PROPERTY

Members were reminded to make disclosures of known patents to the ITU Secretariat in a timelymanner. ITU Patent Policy and related information is at: http://www.itu.int/ITU-Databases/TSBPatent/.

H.324 AND H.320 SYSTEM, CONTROL AND MUX

The Chair of H.324 and H.320 System ad hoc is M. Nasiri (Siemens, Germany).

MPEG-4 VISUAL IN H.324

Q11-H-17, H.245v5 codepoint revisions for MPEG-4 visual support (Y. Kikuchi, Toshiba), wasaccepted by Q11/16 with the agreement that Q15/16 had until close of this meeting for review.Additional comments were received, documented in Q11-H-17r2. During review of Q11-H-17r2,it was pointed out that the drawingOrder default value needs to be other than “0”. This issue willbe coordinated on the Q11, Q14 and Q15 reflectors and the agreed value will be noted in thewhite document prior to submittal. An additional sentence was added to the last paragraph ofthe contribution. (See the joint meeting report below for further discussion.)

Future work of the H.324 & H.320 systems ad hoc includes:• Prepare H.324 Appendix or Implementor Guide for use of V.250 (ex V.25ter Annex A) and

V.80 with H.324.• Continue work on non-normative H.324 Appendix listing PICS (Product Implementation

Conformance Statement) statement for H.324; LBC-96-244 (C. Hansen, Intel) begins thiswork.

• Continue work toward H.246 Annexes for H.324/H.324M and H.324I interworking with otherH.series terminals.

• Commence the generation of detailed requirements for H.324L terminals.• Study requirements for future IMT-2000 terminals and provide inputs to the full group for

consideration.

H.263+ IN H.320

S. Gupta (VTEL, USA) is the Chair of the H.263+ in H.320 ad hoc. Editorial changes to theDetermined text for H.320 (Narrowband visual telephony), H.230 (Synchronous control) andH.242 (Communications establishment using digital channels up to 2 Mbit/s) submitted to thismeeting.

Q11-H-04 (S. Gupta, VTEL) is draft text for H.320. It was reviewed and the proposal of Q11-H-12 (see below) was added. Q11/16 agreed that the document should be submitted to the TSB as awhite document.

Q11-H-05 (S. Gupta, VTEL) is draft text for H.221 (Frame structure, 64-1920 kbit/s). Q11/16agreed that the document should be submitted to the TSB as a white document.

Q11-H-06r1(S. Gupta, VTEL) is draft text for H.230. Q11/16 agreed that the document shouldbe submitted to the TSB as a white document. The inputs from PictureTel (G. Sullivan) relativeto H.230 (Q11-H-30) were reviewed; it was noted that potential performance issues may surfacein the future (if Group of Blocks [GOB] headers are not sent by an H.320 terminal operating withan H.323 or H.324 terminal); the experts should review this issue in February, 1999. Q11/16agreed that the document should be submitted to the TSB as a white document but noted thatfurther input may be required.



Q11-H-07 (S. Gupta, VTEL) is draft text for H.242. It was noted that the profiles that aredefined in 5.2.4.4 are now being made normative in H.320 terminals but are based on aninformative Appendix of H.263 (1998). The experts will review this situation at the meeting inFebruary 1999. Q11/16 agreed that the document should be submitted to the TSB as a whitedocument.

The revised documents are to be posted to the Q11/16 ftp site at ftp.standards.pictel.com/lbc-site/in the “white” folder by Dec. 1 and submitted to the ITU-T TSB by Dec. 19, 1998.

T.140 in H.224

Q11-H-12 (M. Nasiri, Ericsson) proposes a new codepoint for T.140 (T.chat), in H.224. It wasnoted that in the past addition of enhancements to H.320 were challenged in favor of H.324/I butQ11/16 expressed no objections to accepting this change.

MOBILE

B. Wimmer (Siemens, Germany) is the Chair of the mobile ad hoc.

H.223/ANNEX D

Q11-H-08, Proposed white text for H.223 Annex D (Optional multiplexing protocol for low bitratemultimedia communications over highly error-prone channel), was presented by H. Tanaka(Toshiba). There were minor editorial changes including: D.4.7.1.3, (etarget was emax); D.4.1.7.3first sentence add “/8” which implies byte alignment, D.4.1.7.2, the descriptions of CRCpolynomials were changed to refer to relevant sections of Recs. H.223 for 8 and 16 bit CRCs andV.42 for 32 bit CRC.

INTEROPERABILITY EXPERIMENTS

M. Luomi (Nokia, Finland) is the chair of the Interoperability Testing (Mobile terminals) ad hoc.

Q11-H-21, Interoperability Test Plan version 0.1 (M. Luomi, Nokia), was presented forinformation. Q11/16 noted that the experimentation testing has progressed successfully andminor clarifications are being submitted to the experts for consideration. Members interested inparticipating in this work were encouraged to contact the ad hoc chair.

Future work of the Interoperability Testing ad hoc includes:• Coordinate the activities relating to member agreed interoperability testing.• Target development of a test plan (considering contents that would be applicable to the ITU

recognized PICS statement format).• Provide status and progress to Q11/16 experts as appropriate. (The members are reminded

that such activity is proprietary in nature and participation implies agreement ofconfidentiality as relates to participants’ product development status and especially test and orsimulation results.)

CLARIFICATIONS TO H.324M

Q11-H-22 (M. Luomi, Nokia) suggests two areas of clarification for items identified duringinteroperability experiment testing. An item concerning octet alignment will be covered byadding an Annex C section to the H.324 Implementors Guide (I/G). It was agreed to fill with “0”sto achieve byte alignment. The item concerning the content of the level 2 optional header instuffing sequences will be covered by additions to the H.223 I/G. Draft text for these I/Gs will bereviewed at the February 1999 experts meeting.

MULTIMEDIA SERVICES IN IMT-2000

Q11-H-14 (S. Okubo, TAO) discusses the control items to be considered for IMT-2000 terminals.The group agreed to make a decision on in-band or out-of-band signaling and to generate a liaisonto SG11 at the next Q11 Experts meeting in February 1999.



ENHANCED ERROR RESILIENCY FOR VIDEO IN H.324/M

Q11-H-27 documents the common conditions for use in determination of video qualityimprovements by Q15/16. There will be a new TMN11 generated incorporating the latestchanges.

Q11-H-26, Test Model Definition, is for information only. It was suggested that if mobile systemscould provide real time channel end-to-end Quality of Service (QoS) indication to the video layer,it could offer increased coding gains.

Future work of the mobile ad hoc includes:• Coordination with Q15/16 to continue improvements to video quality in the mobile

environment.• Evaluate future system, signaling and network interface for mobile terminals targeting

multimedia services on IMT-2000.• Evaluate new methods for unequal error protection strategies between systems and video.

NON-CONVERSATIONAL SERVICES

C. Quist (KPN, Netherlands) is the Chair of the Non-Conversational Services ad hoc. Nocontributions were received at this meeting.Future work on non-conversational services includes:• Identify or define a means for file storage applicable to H.324 sessions (considering other

H.series multimedia terminals) or applications such as networked database servers, and videorecording/answering function.

• Identify related areas of coordination with other standards organizations.

JOINT MEETING WITH Q11-14/16

IETF

According to T. Taylor (Nortel), IETF is close to establishing working groups for “decompositionof gateways” and “device control.” The 9 kbyte packet size should change: it will not affect theH.323 systems as it is “optional;” in the targeted high speed segments, the added delay is only acouple of microseconds.

FUTURE TERMINAL REQUIREMENTS

Q11-H-15 (APC-1458) (S. Okubo, TAO Japan) outlines future terminal (H.32L) interworkingissues. AT&T asked if there is anything to be learned from current network centric solutions.BT noted that H.323 has a weakness in that conference control is not distributed and it should beincluded in future requirements.

Q11-H-16 (APC-1459) (S. Obuko, TAO) contrasts the various efficiencies relative to transmissionrate for packet networks. TELES noted that, concerning video, full pictures included in a packetcan be more efficient but could result in more artifacts.

Q11-H-13 (APC-1460) (T. Suzuki, NTT DoCoMo) notes that gateways will convert control as wellas bearer channels. AT&T noted that this paper indicates that there will continue to be arequirement for two optimized for each network. One member stated that inclusion of too manyprotocols in a terminal is not practical, and protocol conversion should be supportable byelements in the network. Ascend felt that there will continue to be needs for two protocols basedon network characteristics and that Q11/16 should state a clear position. The meeting concludedthat multimedia systems for IMT-2000 could not be singled out and probably the currentapproach may be applied to IMT-2000. Rapporteurs S. Okubo and T. Geary will draft a liaisonwith comments to ITU-R TG8/1 (Q11-H-29).

FUTURE AUDIO CODECS

Q11-H-11 (APC-1453) (A. Crossman, PictureTel) discusses the end requirements for layeredaudio coding. The term “layered codec” implies that any network node or receiver may “peel off”bits from the bit stream (or remove layered packets); every frame of audio is effectively split into



layers of bits, each layer may be successively removed; as long as the core layer remains, thelistener still hears audio. Layered coding allows receiving terminals to select the bit rate mostappropriate for their connections and allows service providers to scale audio quality for economicreasons. When Ascend asked about the status of this work, the WP3/16 chair (S. Campos-Neto,Comsat) indicated a target completion date of 2001. BT asked what techniques are applicable.The Q20/16 Rapporteur (R. Drogo De Iacovo, CSELT) provided an overview of concepts andnoted that the technical feasibility suggested in Q11-H-11 are achievable. However, videolayered work has not led to implementation to date, and audio layering may result in morevariations in quality than video layering. Q11-H-10 (Q11/16 Rapporteur) contains draft terms ofreference (Wideband [7 kHz] speech coding algorithm around 16 kbit/s).

MPEG-4 AUDIO CODEC SUPPORT IN H.245

Q11-H-19 (L. Chiariglione, MPEG) is the liaison from ISO/IEC JTC1/SC29/WG11, concerningsupport for MPEG-4 audio in H.324 terminals. Q11-H-20 (H. Harasaki, NEC), supportingH.245 codepoints for MPEG-4/Audio, was also presented. TOA asked about the targetedapplications for this codec; the presenter agreed to provide this information to next SG16 Q11-14experts meeting (Feb. 1999). BT asked what implications there were, from the ITU perspective,given that this codec has not been tested by SG12.

Q11-H-23 (APC-1463) (S. Campos-Neto, Comsat, USA) explains that inclusion of codepoints inH.245 for non-ITU codecs indicates to the marketplace some level of ITU-T endorsement of thesecodecs, and proposes that future normative support of non-ITU vice codecs should be done onlyafter sufficient studies have been satisfactorily conducted in the ITU-T. It was suggested thatnon-ITU-T codecs be included in an Appendix to H.245. TOA noted that video and audio qualitydefinitions are not now consistent, and suggested that audio has wider ranging effects due tonetwork provider quality requirements. AT&T noted that additions to H.323 may also be moreapplicable than H.324, especially with regards to quality. Terminal recommendations defineminimum supports and others are optional and market driven, but adding functionality to onefamily and not others poses greater responsibilities for network providers and the cost of theirgateways. Adding codecs also brings related intellectual property issues. Q11/16 felt thatapplications must be defined and a “compelling” reason defined before a codepoint is added toH.245 (comparable quality at equivalent bitrate is not a compelling reason). A liaison to thiseffect will be sent to ISO/IEC (Q11-H-31r1, see below).

MPEG-4 SYSTEM SUPPORT IN H.245

Q11-H-18 (L. Chiariglione, MPEG) is a liaison from ISO/IEC JTC1/SC29/WG11 concerningcomplete support for MPEG-4 in H.324/H.245. It proposes future inclusion of the completeMPEG-4 (due to be completed in December 1999) in H.324 series terminals. A comment wasmade that addition of codepoints to H.245 enables support by other H.series terminals. TELESsuggested that a response to ISO could include that Q11/16 will consider their input and solicitclose cooperation as DSM-CC Multimedia Integration Framework (DMIF) version 2 is developedso that it can fit efficiently into the future H.32L terminals. The proposal will be classified as“under study,” pending further contribution and discussion, especially in view of the fact thatMPEG-4 version 2 is still in development. A liaison to this effect will be sent to ISO/IEC (Q11-H-31r1, see below).

FUTURE VIDEO CODECS

Q11-H-24 defines the status of work in Q15/16 and notes that with no limit in computationalcapability, (and given that H.263L need not be backward compatible with H.263 [1998]), H.263Lcan demonstrate the same performance at 0.5 the bitrate or twice the quality at the same bitrate.H.26L is considering error-prone environments and seeks response from “customers”, especiallyconcerning the following:

1) What the “next” generation terminal transport will be, bit-by-bit or packet;2) Detailed goals for layering needs;3) “Low” delay (relative to MPEG-4);4) Should “long” delay be considered?



AT&T asked if there is a “quality” effort for video codecs in ITU. The response was yes, VQEG(Video Quality Experts Group), but it is targeting higher bitrates typical of TV broadcast qualityobjectives and working on objective test methods. INMARSAT asked if there was there any workin the H.26L group relating to the IMT-2000 network needs for error resiliency in the assumedinherent error prone environments; Q15/16 is soliciting direction in this area.

This work is soliciting inputs for the joint experts meeting in February 1999.

OUTGOING LIAISONS

The draft liaison to ITU-R TG8/1 discusses which multimedia protocol(s) would be appropriate forIMT-2000. The informative Table (Table 1, below) included in the draft liaison was notsupported. The decision was to include the proposed draft liaison as an Annex A of the meetingreport (Q11-H-25d3) along with explanatory comments provided by Siemens and TELES.Members are asked to review and make contributions responding to the original requests fromthe ITU-R TG8/1 and ITU-T SG11/Q.23.

Terminal Rec. Network Bitrate(kb/sec)

BER(1x10\-n)

Delay(2)(mSec)

Jitter (mSec)

Min(1)

Goal Min Goal Max Goal Max Goal

H.310 ATM networksH.320 N-ISDN 64 9 n/a n/aH.321 ATM networks (H.320

emulation)H.322 QoS guaranteed LANs

(such as ISLAN16T(H.320 emulation)

H.323 Packet basednetworks such asLANs, Internet

6

H.324 GSTN 28.8 5 n/a n/aH.324 Annex DA N-ISDN 56 9 n/a n/aH.324 Annex CD Mobile (wireless) 9.6 4 800 200

Note(1): Minimum net bit rate to achieve reasonable customer acceptanceNote(2): Network delay only, end-to-end

Table 1. Informative Table not included in draft liaison to ITU-R TG8/1

Q11-H-31r1 (APC TD-20), Q11-14 liaison to ISO/IEC JTC1 WG11 SC 29 (MPEG-4, summarizesthe issues discussed. Support of the MPEG-4 system (including video, audio, systems, DMIF) istechnically possible as indicated in the MPEG liaison. However, the current MPEG proposal doesnot reflect the interoperability of the various SG16 multimedia families (e.g., H.324, H.323, H.310).SG16/Q11-14 expects detailed proposals for language in all the relevant recommendations. Inaddition, a clear justification why the MPEG audio coding schemes would be beneficial for H.32xsystem users is desired, bearing in mind that when codepoints are added they become applicablefor all second generation H.32x terminal architectures. If such an advantage is recognized,MPEG-4/audio would need to undergo extensive evaluation of its impact regarding complexity,delay, end-to-end quality and quality assurance, the same as other ITU codecs.

NEW BUSINESS

It was requested that a study item be opened to define what is required of an H.324 Annex D(Supplementary services) terminal to support H.324 Annex C (Error prone channels) endpoints.The group agreed to add this item to the February 1999 experts meeting agenda. No new itemsof business were introduced.



FUTURE WORK PLAN

Table 2 below shows the Q11/16 future work plan.

Recommendation Determ. Decision/Approval

Comment

H.324 Annex F (multilink operation) 9/98 ApprovedH.226 (H.multilink) 9/98 ApprovedH.324 Annex D I/G 9/98 ApprovedH.320 (Revision) 9/98 5/99 White generatedH.221 (Revision) 9/98 5/99 White generatedH.230 (Revision) 9/98 5/99 White generatedH.242 (Revision) 9/98 5/99 White generatedH.223 Annex D (multiplex over error pronechannels)

9/98 5/99 White generated

H.324 I/G update 5/99H.223 I/G update 5/99

Table 2. Q11/16 Future Work Plan.

Visit the CSR Web Pages: http://www.csrstds.comThe Web Pages include an updated Telecom Acronym Definitions list, updated meeting schedules,a list of web sites and ftp sites as listed in all issues of CSR journals, background material ontelecom standards and CSR (the company), data sheets on both CSR technical journals, andmore.



REPORT OF Q12/16, Q13/16, AND Q14/16 RAPPORTEURS MEETINGNOVEMBER 17 - 20, 1998, TORINO, ITALY

Objectives of the meeting were:

• Future vision and liaison response to TG 8/1 and SG11/Q23 on multimedia systems on IMT-2000

• Policy for the audiovisual coding for the H-300 series systems• Reaction to the proposed new Question of SG15• Preparation for Decisions to be addressed at the May 1999 SG16 meeting (H.222.0 Amd. 5,

H.222.0 Amd. 6, H.225.0 Annex G, H.245 v5, H.283, H.323 Annex E, H.323 Annex F, H.341,H.450.4, H.450.5, H.450.6, H.450.7)

• Preparation for Determination to be addressed at the May 1999 SG16 meeting (H.225.0 v3,H.246 Annex B, Annex C?, H.323 v3)

Document from this meetings are available from ftp://standard.pictel.com/avc-site/9811_Tor/

REVIEW OF RELEVANT MEETINGS

SG16 Meeting, September 1998F. Tosco (CSELT) gave some messages from the SG16 management team, including the activestatus of next SG16 meeting planned in Santiago, Chile. S. Okubo (TAO) reported a defect foundin the Decided Amendment 5 to H.222.0 (see Q12/16 of this report).

SG15 meeting, October 1998F. Tosco (CSELT) reported that SG15 WP2 is proposing a New Question on gateway equipmentinterconnecting Public Switched Telephone Network (PSTN) and Internet Protocol (IP) networksand requested action of this group.

Q15/16 Seoul Meeting, November 1998Requirements for H.26L are requested (see both H.26L, below, and the Q15/16 report in thisissue).

OthersT. Taylor (Nortel) noted that IETF had established new work groups on gateway decompositionand device control.

APC-1468 (G. Meyer, Siemens) reports the progress made in the recent ETSI TIPHON meeting(see report in this issue).

HARMONIZATION: FUTURE DIRECTION

APC-1458 (S. Okubo, TAO) and APC-1459 (S. Okubo, TAO) present general requirements forthe future system H.32L and discuss some technical aspects. APC-1460 (T. Suzuki, NTTDoCoMo) discusses multimedia terminals on IMT-2000 networks; it proposes that H.324 andH.323 be used for circuit switched service and packet switched service, respectively.

A joint session among Q11-14/16 addressed the appropriate multimedia systems for IMT-2000 inthe context of currently conceptual future system H.32L. A crucial question for the severalmultimedia communication systems is what type(s) of network service would prevail in the futureincluding those of IMT-2000.

After some discussion, the meeting concluded that the multimedia system (including but notlimited to voice only) for IMT-2000 could not be singled out, rather the approach taken so farshould be applied to IMT-2000 as well, resulting in systems optimized to the given networkcharacteristics such as H.323 and H.324/M.

Based on this discussion, the meeting agreed to return a liaison response to ITU-T TG8/1 andITU-T SG11/Q23, in TD-14. However, due to lack of quantitative requirement description whichis thought essential for sending the liaison, the meeting concluded to attach TD-14 to this reportas a basis for further study at the next meeting. An alternative way of presenting therequirements was suggested by S. Wenger (TELES). Contributions were solicited.



HARMONIZATION: AUDIOVISUAL CODING FOR H.300-SERIES

The joint session among Q11-14/16 addressed audio visual coding for H.300-series.

Layered Audio Coding

APC-1453 (A. Crossman, PictureTel, same as Q11-H-11) discusses a possible audio layeredcoding which is now considered in Q20/16, indicating its need and some high level detailsnecessary for its specifications. During the discussion, the following were pointed out:

• Application areas, such as distributed conference, centralized conference where Multi-pointControl Unit (MCU) distributes audio, should be clarified in which layered coding is effective.

• Packet loss resilience, Variable Bit Rate (VBR) should be considered as possible merits of thelayered coding.

• WP2/16 and WP3/16 should jointly work out the requirements at the next SG16 meeting inMay 1999.

The meeting concluded that APC-1453 is a stimulus contribution and further contributions aresolicited toward the next Rapporteur meeting to clarify the points raised above.

MPEG-4 Audio

Q11-H-19 (ISO/IEC JTC1/SC29/WG11) requests definition of H.245 codepoint for MPEG-4 audio,while Q11-H-20 (NEC) proposes H.245 syntax and semantics corresponding to the MPEGliaison. APC-1462 (same as Q11-H-23, S. Campos-Neto, COMSAT) discusses the criteria toinclude non-ITU coding in the ITU-T system standards. The following views were expressed:

• Assigning a codepoint in H.245 to non-ITU audiovisual coding implies some ITU-Tendorsement to it.

• Looking into applications and quality evaluation are required before the code assignment.• Availability of implementable codecs is the problem.• Approaches to quality are different between voice coding and video coding even in ITU-T.• From the system perspective, it is most essential to guarantee interworking by defining

mandatory coding; use of optional coding can be left to the design of product and serviceproviders.

• Once the H.245 codepoint is defined, application of MPEG-4 audiovisual coding is not limited toH.324; H.323 and other systems can also use it.

After discussion, the meeting concluded that further investigation of compelling necessity to addMPEG-4 audio in the H.300 systems is required before defining H.245 capabilities. Liaisonresponse to MPEG is in Q11-H-20.

Whole MPEG-4 System

Q11-H-18, is a liaison from MPEG requesting consideration of applying full MPEG-4 system toH.324 after the definition of MPEG-4 version 2 to be finalized in December 1999. The followingviews were expressed:

• MPEG-4 does not fit the H.300-series model, philosophy is different; it may be worthconsidering for H.32L.

• For mobile terminals, H.324/M is already complex.• Careful study must be made of its implication from H.300-series system perspective,

interworking with existing systems in particular.• Excluding H.323 from the consideration is not appropriate.• Close collaboration is needed while version 2 is under development. Being given a solid

document at its final stage does not work.

The meeting concluded to continue the study. Liaison response to MPEG is in TD-20.



Requirements for H.26L

APC-1467 (G. Sullivan, PictureTel, Q15/16 Rapporteur) describes the progress made in therecent Q15/16 meeting in Seoul and requests feedback on the H.26L requirements from thesystems group. During the discussion, two candidate items were raised:• Packet network aware syntax• Low delay coding for conversational services with performance comparable to the speech

codingBased on the list in APC-1467, well-thought requirements from the “customers” should beformulated. This work should be completed at the next Rapporteurs meeting in February 1999,using the opportunity of joint sessions with Q15/16. Contributions are solicited.

Q12/16 B-ISDN MULTIMEDIA SYSTEMS AND TERMINALS

AMENDMENT 5 TO H.222.0|ISO/IEC 13818-1 FOR DECISION

APC-1450 (S. Okubo, TAO) is a liaison statement from MPEG addressing the defect inAmendment 5 to H.222.0 found at the MPEG meeting in October 1998 and the way to resolve theproblem. Amd. 5 as written will cause the P-STD to be broken by ISO/IEC 13818-7 streamswhich exceed the maximum bitrate permitted by those streams. ISO/IEC JTC 1/SC29/WG11recommends that the text of Amd. 5 not be published and the corrected text (in APC-1450) besubmitted for a second ballot. APC-1451 (S. Okubo, TAO, Rapporteur Q12) proposes a course ofaction for the ITU-T to also address the identified defect.

Q12/16 supported the proposal and agreed to return a liaison response to MPEG (TD-22, S.Okubo, TAO, Rapporteur Q12/16).

INTEGRATED EDITION OF H.222.0|ISO/IEC 13818-1 AND H.262|ISO/IEC 13818-2

APC-1450 (S. Okubo, TAO) is a liaison statement from MPEG proposing to issue integratededition of the two common text Recommendations|International Standards. APC-1452 (S.Okubo, TAO, Rapporteur Q12) clarifies the Amendments and Corrigenda to be included in theintegrated edition.

Q12/16 supported the plan and agreed to return a liaison response to MPEG (TD-22, S. Okubo,TAO, Rapporteur Q12). It supports the two integrated publishing plans of H.222.0|ISO/IEC13818-1 and -2:

Recommendation DateH.222.0|ISO/IEC 13818-1Original 7/95

Amd.1 11/96Amd. 2 11/96

Cor. 1 3/97Amd. 3 2/98Amd. 4 2/98Amd. 5 to be 5/99Amd. 6 to be 5/99

H.262|ISO/IEC 13818-2Original 7/95

Cor. 1 6/96Cor. 2 6/96

Amd. 1 11/96Amd. 2 11/96Amd. 3 2/97Amd. 4 2/97Amd. 5 to be 5/99Amd. 6 to be 2/00



There was concern with delay caused in the publication of the approved Recommendations. Thepre-publication is thought to solve the problem to a large extent. However, a desire wasexpressed that those pre-publication Recommendations should also be included in the quarterlyupdate of CDs from the ITU. Q12/16 recognized that this point should better be raised to TSAGwith a national position paper.

IMPLEMENTORS GUIDE FOR H.321

APC-1477 (S. Okubo, TAO, Rapporteur Q12) proposes a plan for issuing H.321 ImplementorsGuide noting the replacement of “Notification Indicator IE” with “Broadband Report Type IE” toindicate the clock source. Q12/16 agreed to this plan.

Q13/16, PACKET SWITCHED MULTIMEDIA SYSTEMS AND TERMINALS

COORDINATION/JOINT WORK

APC-1438, Liaison from ETSI Telecommunications and Internet Protocol Harmonization OverNetworks (TIPHON) transmitting the draft Phase 1 and 2 TIPHON architecture (DTS/TIPHON-02002), was considered as part of the various work items. APC-1468, Report about ETSITIPHON meeting of October 26-30, Tel Aviv (G. Meyer, Siemens), presents an overview ofTIPHON work. (See also the TIPHON report in this issue.)

APC-1449 (G. Thom, Delta Information Systems) notes the New Question in SG15 WP2 on PSTNto Internet gateway. A response to SG15 was prepared jointly with Q13/16.

H.323 IMPLEMENTORS GUIDE

Changes to H.323, H.255.0, H.235, and H.450.x recommendations are included in a singleimplementors guide so that implementors need look in only one place for all corrections todocuments related to H.323.

APC-1444, Clarifications on the use of the CRV field in H.323 Annex E (D. Gurle, VocalTec),resulted in a change to the implementors guide. It clarifies that the value of a CRV is valid forpeer-to-peer relationships only - it cannot be used uniquely end-to-end.

RECOMMENDATIONS FOR DECISION

H.323 Annex E “Connection over UDP”

APC 1432, H.283 (ex H.RDC, Remote device Control) and H.323 Annex E harmonization (M.Duckworth, PictureTel), calls for additional addressing information related to H.283 to accountfor multipoint acknowledgments. Objections to this focussed on a mixing of protocols and protocollevels. In the end, it was decided to modify Annex E as requested. The substance of APC 1433,General comments on Annex E (K. Hedayat, PictureTel), will also be incorporated by the editor.

APC 1441, H.323 Annex E: Call Signaling over UDP (M. Duckworth, PictureTel), waspresented and accepted with a few minor changes. Text toward the white paper appears as TD-23. It was agreed that G. Kimchi (ETSI) would continue to develop Annex E as H.R-UDP forpossible use in H.323 Annex G (Inter-Domain Communications) and H.RDC. This may result, byagreement of the editors of H.323 Annex G and H.RDC, in white papers being issued that refer toAnnex E.

APC 1444, Info on CRV Mapping for Annex E (D. Gurle, VocalTec), was accepted for Annex E,and also forwarded for inclusion in H.225.0 V3 implementors guide (to be decided where finalinclusion place will be).

H.323 Annex F “Single Use Terminal”

APC-1479 is Draft H.323 Annex F (J. Ott, TELES). It was noted that APC-1436, Master-slavedetermination procedure in GK routed call model (RADVision), is important to having Annex Fwork: this was to be considered later. The editor presented and reviewed the work. Variouscomments were accepted from the floor. It was stated that, as there is no stateless forwarding, ahop count is not needed. If stateless forwarding is added, the hop count is needed.



Text toward the white paper appears as TD-27. J. Ott (TELES) was charged with looking intointeroperability issues before issuing the white paper.

H.225.0 V3 Annex G (GK to GK Communications)

APC-1473, Draft Annex G (G. Freundlich, Lucent), is the basis for the work. It was agreed toissue the white paper to retain the possibility of Decision in May 1999. However, considerationwill be given to the ideas of APC-1480, Back-End Service Transaction Protocol (D. Clowes,Interline, Australia), which may result in changes of a sufficient magnitude to require re-Determination. Text for the white paper appears as TD-26. It was concluded that the encodingof the templates and messages will be binary using ASN.1 Packed Encoding Rules (PER). Thismay be revisited in favor of Basic Encoding Rules (BER) at the next meeting.

APC-1434 (AT&T) is Intra-Domain/Inter-Zone Gatekeeper-to-Gatekeeper and Inter-DomainBorder Element-to-Border Element Communications. APC-1435 (R. Roy, AT&T) is ViewgraphPresentation of APC-1434. The proposal for a pathValue field was discussed. It was agreed thatavoiding loopback is important, but that several proposals (a path history) were not good ideas.A. Draper (Madge), G. Kimchi, and R. Roy will present text to the editor.

APC-1447, Comments on H.225 Annex G working draft (R. Callaghan, Siemens), seeks to allowthe endpoint or endpoint’s GateKeeper (GK) to get back enhanced information about the protocolssupported by the egress GateWay (GW). This was generally agreed, but some more editorialwork is needed.

APC 1448, Coding and usage details for Annex G (Intel, PictureTel, VocalTec, VideoServer),calls for the usage of a directed Gatekeeper Discovery Request (GRQ) to allow determination ofthe protocol used by the GK; it also clarifies the usage of User Datagram Protocol/TransmissionControl Protocol (UDP/TCP).

Agreements:1) Both UDP/TCP will be used.2) Only well-known ports will be used; leave dynamic FFS.3) Large messages should use TCP; others may use UDP.4) Annex E for reliable UDP will continue to be considered and developed, but there is not a

commitment to this course.5) A TCP and a UDP well-known port will be requested from Internet Assigned Number

Authority (IANA).Other changes were accepted; some with a need for editorial modifications.

APC-1455, Comments on GK to GK communication protocol (N. Frid-Ruppin, RADVision), pointsout that the direct call model may not work properly. No conclusion was drawn, but the issue canbe considered again at future meetings.

It was agreed that the contents of APC-1469, Scalable Security for Annex G (A. Draper, Madge),should be incorporated into Annex G. Note that H.235 tokens will continue to be used for end-to-end authentication.

APC-1475 (Lucent, VocalTec) contains comments on Annex G:Sec 2 - Admission support, proposes adding various fields to the Address Query message. Thisproposal was rejected as being a large change in the protocol.Sec 3 - Timing considerations - values rejected; some text will be added indicating that values areimplementation-dependent.Sec 4 - Response in process - accepted proposal for request in progress.Sec 5 - accepted proposal for non-standard message.Sec 6 - Keep Alive - between domains accepted. Point was added that fail-over and redundancymust be added to make the best use of this feature.Sec 7 - Message Not Understood. If a border element receives a message it cannot understand, itshould be able to indicate this to the message transmitter. Accepted.Sec 8 - Proposal for call information distribution.



APC-1480, Back-End Service Transaction (BEST) Protocol (Interline, Australia) and APC-1481,“Back-End Service Technical Rationale” (Interline, Australia), were presented and discussed.The BEST protocol implements basic functionality for back-end services for an H.323 network.The protocol provides for authentication, authorization, address translation, call processing,pricing exchange and usage recording. It operates between H.323 entities, such as a gateway orgatekeeper, and one or more back-end servers. It supports vendor specific extensions. This workmay result in changes to Annex G. See TD-25 (D. Clowes, Interline, Australia) for a comparisonof APC-1480 and Annex G which describes the characteristics of APC-1480 which are notpresent in APC-1473:

• Relationship Management• Redundancy (alternate service points)• Clean Shutdown• Transaction Identifier• Authorization Token (present in APC-1473 messages but use not specified)• Pricing Exchange (only address resolution)• Access Control• Call Authorization (not specified in APC-1473)• Call Validation at destination• Request In Progress• Non Standard Message and Response• Unknown Message Response• Call Detail Records (Usage Recording, not Billing or Accounting)

In addition, APC-1480 supports an authenticated call model that works in all scenariosdiagrammed in APC-1473. APC-1473 does not explicitly support authenticated calls, and themodel does not appear to be adaptable to Figure 6 of APC-1473 where calls span unrelateddomains, such as A to C.

It was noted that many items listed in TD-25 are no longer true due to changes made during themeeting, and reflected in TD-26, Updated H.225.0 Annex G.

TD-11, Annex G Ten open issues (from APC-1448) (D. Gurle, VocalTec). Q13/16 responded:1) Annex G is a NNI not UNI, but this term will not used as it is not accurate.2) In Annex G, there is no concept of multi-tiers (e.g., clearing house).4) Probing is implicit (i.e., well known port mechanism).5) Agreed to need for Zone Implicit verbs.6) Agreed to usage of H.235 as appropriate.7) Agreed ASN.1 shall be binary codable.8) Proposal to tunnel Registration, Admission, and Status (RAS) needs further development and

justification.

H.450.4 (Call Hold),H.450.5 (Call Park and Call Pickup), and H.450.6 (Call Waiting)

There were no changes; it was agreed the editor (M. Korpi, Siemens) should generate whitepapers.

H.450.7 (Message Waiting)

APC-1454, Draft H.450.7 (D. Walker, Mitel) was presented, including some small changes; itwas agreed the editor should generate a white paper.

ADDITIONAL TOPICS AND FUTURE WORK

H.323 V3

APC-1436, Master-Slave Determination Procedure in GK routed call model (RADVision), will beconsidered further at future meetings; no decision was taken.

APC-1470, Requirements for IVR Control in VoIP (BT), was accepted as the basis of future worktoward H.323 V3. It provides a basic overview of Interactive Voice Response (IVR) control for



Voice over Internet Protocol (VoIP) networks. It includes a contribution on generic IVR controlrequirements for VoIP networks. Although this document assumes an H.323 networkarchitecture (with Gatekeeper routed signaling), the IVR control requirements are applicable forall VoIP signaling protocols. The IVR control examples and scenarios outlined in this documenthave been influenced by a desire to provide services available from existing PSTN networks in aVoIP network. As such, IVR control capabilities from existing IN standards are regarded as abaseline set of requirements. Detailed contributions were solicited. Both call models must besupported.

Existing IN standards include: ETSI, “Intelligent Network (IN); Intelligent Network CapabilitySet 1 (CS1); Core Intelligent Network Application Protocol (INAP); Part 1: Protocol Specification”,ETS 300 374-1 and “Set 2, Intelligent Network Application Protocol (INAP); Part 1: ProtocolSpecification for Capability Set 2”, DEN/SPS 03038-1, November 1997; “InterfaceRecommendation For Intelligent Network CS-1”, ITU-T Recommendation Q1218, March 1993and “CS-2”, ITU-T Recommendation Q1228, September 1997; Bellcore, “ISCP-IP InterfaceSpecification”, SR-3511.

H.323 Interoperability Profiles

APC-1429, H.323 version 2 compliant profile (Intel), presents a possible H.323 interoperabilityprofile. It offers an H.323-compliant profile for PacketCable client to network signaling. Theclient can be a PC, Telephone Adapter, or any other client device that is H.323 compliant. Oneconcern raised was the usage of pre-granted admission (no Automatic Repeat Request [ARQ]) asthis greatly reduces the functionality provided. Q13/16 was asked to consider whether thisshould be pursued as a work item, either in the ITU-T or possibly the International MultimediaTeleconferencing Consortium (IMTC). Input to both profile work and to a framework for profileswere solicited.

H.323 Annex C and ATM Matters

APC-1478, H.323 Media transport over ATM (G. Ratta, ATM Forum), made a variety of points,mainly requests for additional clarifications. It was requested that section 4 (Interworking) beput forward to M. Reid, editor H.323, to become the basis of the Asynchronous Transfer Mode(ATM) interworking section. H.246 will then also refer to the ATM forum document in aninformative way. Section 4 will be updated to refer to H.246 (Interworking of H-Seriesmultimedia terminals with each other and with voice/voiceband terminals on the GSTN andISDN) for interworking details. A request for changes to H.245 (Control) are expected by thenext Q14/16 rapporteur’s meeting to allow existing Annex C (H.323 on ATM) terminals and newAnnex C terminals using Real Time Transport Protocol (RTP) compression to be distinguished.Good progress has been made on this document, and it was noted that the next ATM Forummeetings were Nov. 30-Dec 4, 1998, and Feb 8-12, 1999. Both meetings will occur before thenext Q12/Q13/Q14 rapporteur’s’ meeting.

It was suggested that a request be made to SG11 for codepoints to describe this type of terminal(H.323 Annex C). This had been previously done, but there has been no response from Q11/11since they have not yet met.

There seem to be three ways forward:1) Refer to the ATM Forum document. This may not be procedurally allowed, but views were

expressed that it was allowed.2) The text related to RTP compression could be brought into H.323 Annex C.3) The entire document could be included as H.246 Annex for ATM Interworking.

One of these three courses must be followed to support the inclusion of H.245 codepoints related toRTP compression over ATM. It was decided to make a final decision at the May meeting.

H.225.0 V3

APC-1472, Draft H.225.0 V3 (Editor H.225.0) suggests a few small changes.

APC-1428, RTP Maximum Packet size (R. Evan, Accord, Israel) proposes text for H.225.0 V3with the change: an exception will be added to state that if the smallest independently coded



element of the coding scheme (e.g., a macroblock) is bigger than the Maximum Transfer Unit(MTU) size, it is not required to break up the packet over MTUs.

APC-1443, New SCN Redirect Reason (D. Gurle, VocalTec), is a proposal adopted with thefollowing change, that it should only be used for ingress gateways, and the E.164 number ischanged to the PSTN/SCN number. This also requires that the forwarding address be a partynumber for some cases.

H.332 V2 (H.323 extended for Loosely Coupled Conferences)

APC-1427 (S. Kim and H. Kim, Korea Telecom) concerns required functions of RTP receiver andpacket forwarding procedure for Dynamic Multicast Network. TD-21 is related. It contains aset of open issues related to APC-1427.

There were no objections in principle, but issues raised in TD-21 include:1) How to determine the set of recipients, perhaps via unicast.2) How to organize the tree.3) What entities need to be higher in the tree than others.4) How to create links between entities.5) How to identify who can do the forwarding.6) Messages needed to create the tree and to dynamically manage the tree.

It is expected that this proposal will be developed further and considered at the next meeting.

H.450.x Future Work

APC-1445, Conference out of consultation (K. Klaghofer, Siemens and S. Wiryaman,VideoServer), presents a plan for a new supplementary service similar to a three partyconference or add-on conference in ISDN and PBX environments, respectively. Work willcontinue toward May Determination. A major issue is that the proposal uses a new conferencetype; this will be discussed further at the next meeting. This would be proposed as H.450.8.

Future Work in Q13/16

The schedule for further Q13/16 work is shown in Table 1.

Recommendation Doc. Det. Dec. EditorH.323 Annex E (Call Connection over

UDP)TD-23 9/98 5/99 D. Gurle (VocalTec)

H.323 Annex F (Single Use Terminal) TD-27 9/98 5/99 J. Ott (TELES)H.225.0 Annex G (Communication

between Administrative Domains)TD-26 9/98 5/99 G. Freundlich (Lucent)

H.450.4 (Call Hold) No changes 9/98 5/99 M. Korpi (Siemens)H.450.5 (Call Park/Pickup) No changes 9/98 5/99 M. Korpi (Siemens)H.450.6 (Call Waiting) No changes 9/98 5/99 M. Korpi (Siemens)H.450.7 (Message Waiting) APC-1454 9/98 5/99 D. Walker (Mitel)H.450.8 (Conference out of consultation) APC-1445 5/99 2/00 M. Korpi (Siemens)H.450.9 (Call Completion on Busy) APC-1463 5/99 2/0

Table 1. Q13/16 Work plan from the November 1998 Rapporteurs meeting.

Q14/16 COMMON PROTOCOLS, MCUS AND PROTOCOLS FOR INTERWORKING WITH H.300-SERIES TERMINALS

RECOMMENDATIONS FOR DECISION

H.283 Real-Time Device Control

APC-1430, Draft H.283 (M. Duckworth, PictureTel), was presented. There were no comments.It was agreed to put this Recommendation forward for Decision at the next SG16 meeting. Theeditor will prepare a White Contribution for the May 1999 SG16 meeting.



APC-1431, Changes to draft H.323 related to recommendations H.283 and H.283 (K. Hedayat,PictureTel), was presented and accepted. The changes for H.323 Version 2 will go intoimplementer’s guide.

APC-1432, H.283 and H.323 Annex E harmonization (M. Duckworth, PictureTel), will continueto be developed by the editors and may be included at their discretion.

APC-1456, Tunneling of H.281/H.224 FECC frames (Y. Miranda, RADVision), was acceptedwith limitations. H.281 is a Far End Camera Control protocol for videoconferences using H.224(A real time control protocol for simplex applications using the H.221 (frame structure)LSD/HSD/MLP channels). This will be an optional method for communications between twoH.323/H.320 gateways over the IP network. It shall not be used by an H.323 terminal for FarEnd Camera Control (FECC). This text will be incorporated into Annex A of H.246 (H.323-H.320 interworking). Draft text is contained in TD-13 and will be developed further at the nextmeeting.

APC-1464, New code point for T.140 (Protocol for multimedia application text conversation) inH.224 (M. Nasiri, Ericsson), consists of a change that was approved by Q11/16 and submitted toQ14/16 for approval. The change was approved. M. Nasiri will act as editor and produce achange to H.224 to be Determined in May 1999.

H.341 Multimedia Management Information Base

APC-1437, Update to Draft H.341, and TD-10, H.341 Issues View Graphs, present draft Rec.H.341 and view graphs describing three illustrative cases of management agent complexity. Thefirst uses a single agent for each instance of interface, port, stack, GW.

The second uses a single agent for multiple instances of interface, port, stack, GW complex cases.The third uses multiple agents for each instance of stack, GW. It was agreed that the secondcase was important, and the third case was not. The editor will incorporate changes to supportthe second case in time to prepare the white contribution. It was agreed to move thisRecommendation forward for Decision at the May 1999 SG16 meeting.

H.245 V5 Communications Control Protocol

APC-1440, New H.245 messages for bit-rate management (H. ElGebaly, Intel), was presented,discussed and approved. It may be desirable to make mandatory that the FlowControlIndicationbe sent in response to a FlowControlCommand. This will be decided by the next meeting. Theeditor is instructed to incorporate these changes into H.245 V5.

APC-1457, Proposed changes on H.245 codepoints for MPEG-4 visual support (Y. Kikuchi,Toshiba) was approved by Q11/16 and presented to Q14/16. The proposed changes wereaccepted. The editor will incorporate these changes into H.245V5. The editor should check to seeif specifying unbounded octet strings are a problem. Additional changes are included in TD-16.It was agreed to include this in H.245V5 using the traditional coding scheme, in case that objectoriented coding scheme is not ready for Decision. Coding of the MPEG-4 parameters using theobject oriented method of H.245V5 is in Q11H17R2, subject to review by Q15/16. The decisionas to which to use will be made at the Feb. 1999 Q14/16 rapporteurs meeting.

H.245V5 was approved for Decision at the May 1999 SG16 meeting.

TD-8, An Object Oriented Approach to H.245 Extensions, presents the current status of thegeneric capabilities work. It was asked if the multidimensional complexity is really needed, andthat perhaps the complexity of the proposal can be reduced. Everyone was encouraged to look forproblems with this approach. It was also noted that generic capabilities may be useful fordistributed control which is desirable future work. TD-8A (also TD-24) contains a revised versionof the object oriented coding scheme to be included in H.245V5. This should be reviewed in detailand discussed on the e-mail reflector prior to the December White Contribution deadline. Thetext of TD-24 will be included in the white paper.



GATEWAY PROTOCOLS

APC-1449, New Question of SG15 on PSTN to Internet gateway (G. Thom, Delta InformationSystems), was presented and discussed. A group was selected to draft a response whichdescribes the work being done in Q14/16, and suggests a modification to the definition of Qx/15.The response is in TD-15.

Transport of ISUP Signaling

Contributions included:• APC-1439, Liaison from ETSI TIPHON regarding transport of ISUP in an H.323 network• APC-1446, H.450 tunnel transport of SCN signaling protocols (R. Callaghan, Siemens)• APC-1471, ISUP and FAS/NFAS call signaling for H.323 gateways• APC-1474, SS7 information transport over intermediate H.323 network (B. Hill, VideoServer)• TD-9, Mechanisms for extending H.225.0 without adding to the core protocol

There was a detailed discussion on transporting ISDN User Part (ISUP) and other signalingprotocols within H.323 protocols. Several proposals were discussed for transcoding andtunneling ISUP messages in H.225.0 and/or H.450. A group was charged with converging thetwo approaches but did not converge. Contributions are solicited for the February Rapporteursmeeting. The rapporteur requested that the work continue on the reflector, with an initialemphasis on relating this work to the terms of reference, and on defining the conflictingapproaches under consideration.

H.GCP (Gateway Control Protocol)

H.GCP defines a protocol for use between media gateways and media gateway controllers.Contributions included:• APC-1438, Liaison from ETSI TIPHON transmitting the draft of phase 1 and 2 TIPHON

architecture (DTS/TIPHON-02002)• APC-1466, Requirements for the device control interface (T. Taylor, Nortel)• APC-1461, Media Gateway Control Protocol (MGCP) (H. Liu, Bellcore)• APC-1465, Connection model for device control• APC-1470, Requirements for IVR control in VoIP• APC-1483, Network Architecture for the separation of control and transport planes (G.

Roullet, Alcatel)• APC-1484, Media Gateway Control Protocol (Lucent Technologies, Intel)

The various contributions on Media Gateway Control were presented and discussed. It wasagreed that any architecture coming from this group must address both direct routed andgatekeeper routed call models. It may be desirable to add an interface E to the terms of referencefor signaling between the call services management entity and the signaling control gateway.One unresolved issue is the use of binary versus text string encoding.

In order to move this work forward, it was agreed that T. Taylor (Nortel) would continue todevelop the requirements document and to include requirements for interfaces B, C, and D.Requirements will be developed under five headings: A, B, C, D, and X, where X is concernedwith how A/B/C/D relate. This should be presented at the Feb. 1999 Rapporteurs meeting. Aneditor (B. Hill, VideoServer) has been assigned to begin producing an H.GCP draftrecommendation. It will include a description of the decomposed gateway elements, and willattempt to harmonize the two approaches present at this meeting. This should be completed bythe Feb. 1999 Rapporteurs meeting so that it can be compared to the requirements document andthe terms of reference, with the plan of May Determination. It was noted that significant effort isrequired to complete this work.

The proponents of the two main proposals (APC-1461 and APC-1484) were asked to develop aTD which identifies the similarities and major differences between the proposals. This is in TD-30, which was approved as an input to work in progress. It was noted that the documentsconsidered in TD-28 have limited scope compared to H.GCP, especially in their focus on voiceoperations.



It was agreed that terminology in H.GCP should be consistent with the terminology in H.323,e.g., should not use the same term for different things.

H.235 V2 H.323 SECURITY

APC-1442, H.235 Comments (L. Moscovici, VocalTec), proposes two syntactic correctionsconcerning the security tokens defined in H.235. The proposed solution in section 3 was notaccepted. The proposal in section 4 (Password Based Hashing) was not rejected but needsfurther consideration from the experts and will be reviewed at the next meeting.



REPORT OF ETSI TM6 ACCESS TRANSMISSION SYSTEMS ON METALLICCABLES MEETING, NOVEMBER 24 - 27, 1998,

SOPHIA ANTIPOLIS, FRANCEThe following permanent documents apply to this TM6 meeting:

TM6(98)05 980p05a3.PDF, Living list for DTS/TM-06003-1 (VDSL)TM6(98)06 980p06a2.pdf, Current draft of DTS/TM-06003-2 (VDSL) *NEW*TM6(98)07 980p07a3.PDF, Living list for DTS/TM-06003-2 (VDSL)TM6(98)08 980p08a0.pdf, Current Draft of DTS/TM-06011-1 (SDSL)TM6(98)09 980p09a1.pdf, Living List for DTS/TM-06011-1 (SDSL)

TD-02 of the Sophia meeting gives the list of work items (ADSL, HDSL, VDSL, ISDN-BA, SDSLand B-ISDN) assigned by ETSI to TM6 and their status. For the three documents which showedin the status “at ETSI Secretariat,” the status in TD-02 should now read “published.” Thosethree documents are:• The new version of the ISDN basic rate access (ETR 080)• Issue 5 of the HDSL TS including the ISDN Basic Rate Transport• ADSL over ISDN TS (which was contributed to ITU-T).

H. Frizlen (Ericsson) will continue as Chair of ETSI TM6 in 1999. A. Carrick, (Ascom,Switzerland) is now official secretary of the TM6 Sub-Technical Committee (STC); he previouslywas acting secretary. All other rapporteurs will continue in their roles.

ITU-T SG15 LIAISON

TM6 Chair (P. Wery, Nortel) introduced TD-16, Work on Very high speed ADSL (VDSL) in ITU-TSG15 for information. During the Plenary meeting of SG15 in October, the SG15 managementteam discussed their coming work with VDSL. The position of SG15/Q4 is reflected in theexcerpt from SG15 TD-5(PLEN), which is copied in TD-16 for the information of TM6. It states,“the management team recommends to continue work on VDSL in SG15 in a cooperative waywith the ongoing activities of ETSI and ANSI Committee T1”.

In the discussion that followed, the TM6 Chair insisted on the necessity of having liaisons in bothdirections between ITU-T and ETSI; while he cannot nominate liaison from ITU-T to ETSI, heproposed to appoint S. Schmoll (Alcatel) as ETSI TM6 liaison to ITU-T SG15. The TM6 meetingagreed. (Technical Editor’s note: D. Pagnotta [Marconi] has offered to be liaison from SG15/Q4 toTM6 but this has not yet been approved.) Another discussion opened on the topic of what toprovide ITU-T with on VDSL and on SDSL. Delegates feel that ETSI TM6 should providemature documents only and not unstable drafts; the delegates also feel that clear work plansshould be provided to ITU-T indicating when documents will become available and what they willcontain (ETSI work items and milestones).

TD-08, Relationship between TM6 and ITU-T SG15 (A. Carrick, Ascom), was introduced fordecision. Ascom wishes to encourage TM6 to continue to produce high-quality xDSL standards ina timely fashion, and to make the results of this work available to ITU-T SG15 at the earliestpossible opportunity for inclusion/merging in international recommendations. To this end, Ascomproposes that:

• A liaison officer to ITU-T Q4/15 be appointed.• The work plan for SDSL and VDSL be liaised immediately to ITU-T Q4/15.• TS 101 270-1 v1.1.1 (VDSL part 1: Functional Requirements) be liaised immediately to ITU-T

Q4/15.• Input from ETSI members to ITU-T Q4/15 be coordinated by TM6.

TD-09, Proposed revision to work plan for DTS/TM-06011-1 (SDSL) (A. Carrick, Ascom),proposes to also change the work plan on SDSL so as to fit into the ITU-T plans. In order tocoordinate the work on SDSL with the work on G.shdsl, it was proposed that the milestones berevised to better fit the needs of ITU-T Q4/15, and that these milestones then be liaised to Q4/15.Key milestones proposed were:



• Milestone 2 - Table of Contents and Scope, November 27, 1998• Milestone 6 - First complete draft, May 7, 1999

A work item should be created, and work started, on part 2 once Milestone 6 of part 1 isachieved. The TM6 Chair postponed the approval of this new plan to the discussion on SDSL (seebelow).

ADSL FORUM LIAISON

Three liaison statements were received from ADSL Forum meeting just completed in LosAngeles; they were introduced by B. Dugerdill (Motorola), who reported on the main events ofthat forum in WD-05.

WD-02 expresses the concern of ADSL Forum on the HomePNA (Home Phoneline NetworkingAlliance) Group use of the home wiring and its potential negative effects on other Telecom wiringand systems including ADSL and VDSL.

WD-03 informs TM6 that the ADSL Forum is creating an SDSL watch group and preparing awhite book on SDSL (part of the VDSL sub-group).

WD-04 sends the latest text of ATM over ADSL to TM6 and asks for inclusion of that text inappropriate parts of TM6 documents (this concerns mostly the former TM3 WP2 group).

ATA (ANALOG TERMINAL ACCESS) LIAISON

TD-52 (ATA) replies to TM6 on concerns expressed by TM6 delegates at the last (Vienna)meeting on limits to spectrum compatibility of CTR21 approved non-voice terminals and ADSLsystems, and possible interference between such terminals and other xDSL equipment. TD-52includes an attachment which discusses the history and rationale for TBR21 frequency limits,noting the difficulty of reproducible measurements above 200 kHz and the small percentage ofterminal equipment affected by DSL.

TD-52 notes also that the new Radio equipment and Telecommunications Terminal Directive(R&TTE) is in an advanced phase and foresees “the end of the type approval regime” forTelecommunications Terminal Equipment. ETSI OCG-TG6 and 4 other Groups created by theEuropean Commission are actively working on the implementation of the new regime.

VDSL

The work plan of the VDSL session is contained in WD-06; the session was lead by B. Waring(Lucent Technologies). The session split into two parts: part 1 related topics (FunctionalRequirements), and part 2 related topics (line code and related topics). Reporter’s comment:Progress on VDSL at this meeting seemed very slow, possibly due to decreasing market pressureand increasing technical difficulities.

TD-26, Noise model of xDSL systems (M. Meninger SPT Telecom), assigned to the VDSL session,was withdrawn.

TD-04, Update of Cable Reference Models (T. Kessler, DTAG), discusses the cable referencemodel of cables in the network of Deutsche Telekom, covered in the current version of thepermanent document TM6(97)02 “Cable reference models for simulating metallic accessnetworks.” This model results in impulse responses with on-causal precursors. A new phasemodel for the phase of the transmit function leads to significant improvements. It was proposedto introduce this new model into the permanent document. This was agreed without presentationor discussion.

CONTRIBUTIONS FOR VDSL PART 2

TD-07, VDSL Alliance DMT PMD Proposal (K. Jacobsen, Texas Instruments), was suggested asa replacement for the text presently in the Part 2 Technical Report. A few questions fromdelegates were triggered by the presentation of this document: Why is there need for several FECcodes? How Zipper (duplex



scheme for DMT VDSL) and non-Zipper systems start up and communicate?

TD-20, Line Coding, Modulation and Duplexing Method Based on Single-Carrier (V. Oaxman, forthe VDSL Coalition), is a part of the current working draft specification for VDSL systems whichproposes line coding, modulation, and duplexing methods using single-carrier technologies; itspecifies the Physical Media Dependent (PMD) sub-layer for both Network Termination (NT) andLine Termination (LT) which complies with the VDSL systems requirements as specified in theDTS/TM-06003-1 document.

TD-21, Physical Media Specific TC Sub-layer for a Single Carrier VDSL Transceiver (V.Oaxman, for the VDSL Coalition), specifies a general PMS-TC reference model, interfaces,transmission frame format, randomization, forward error correction and interleaving. Thetransmission frame header bears frame alignment, NTR (8 kHz Network Timing Marker) andcontrol information. The proposed PMS-TC contains optionally a Fast virtual channel and a Slowvirtual channel to provide both single and dual latency transmission.

TD-22, Activation/Deactivation of a VDSL Link Based on Single Carrier (V. Oaxman, for theVDSL Coalition), proposes activation and deactivation procedures both for the VDSL transceiverand the VDSL link, controlled over a VDSL Operation Channel. The document is based on VDSLlink State and Timing diagram which complies with both ANSI T1E1.4/98-046 and ETSIDTS/TM-06003-1 V.007 system requirement documents.

WD-01 (N. Quimby, Texas Instruments) contains the first draft of part 2 as resulting from inputsto the Vienna meeting.

WD-07 gives the views of the rapporteur for VDSL on each of the running study points.

CONTRIBUTIONS FOR VDSL PART 1

The first set of contributions deal with study points 1-1 (low frequency cut-off) and 1-2 (PSDboost).

TD-13 Progression on VDSL SP1-1 and SP1-2 (R. Kirby, BT), indicates that these study points(SP 1-1 and SP 1-2) have attracted no debate on the TM6_ALL server since Vienna. The contentof this document concerns the points that were not agreed in Vienna and is simply a restatementfrom a previous BT contribution.

TD-15, Interpretation of PSD for PSD masks (R. Kirby, BT, for FSAN), states that the concept ofa PSD for a random signal is something that is often given very little thought. This can give riseto a raft of issues concerning interpretation of requirements and specifications if these are notwell defined. The measurement of Power Spectral Density (PSD) inherently involves an averageover both time and frequency. Measurement resolution in one domain can be exchanged formeasurement resolution in the other. The real world problems of spectral incompatibilitysuggests sensible values for the averaging bandwidth (~2.4 kHz) and averaging time (~250 µsec)for PSD requirements, standards and measurement methods.

WD-08 is a presentation on the FSAN working group done by R. Kirby (BT). FSAN, Full ServiceAccess Networks, is a telco-only trade association focusing on spectral compatibility issues. As ofthis meeting, the only US-based telco member is US West.

As a result of those TDs, an ad hoc group met and produced WD-11, which presents two PSDmasks for both the LT and NT ends of the link in FTTEx and FTTCab scenarios. This wasprovisionally agreed by TM6.

The following contributions address the VDSL noise model.

TD-14, Revised noise models for VDSL (R. Kirby, BT, on behalf of FSAN), gives more informationabout the derivation of the FSAN VDSL noise models, including all the network scenarios onwhich they are based. It also presents a revision of the noise models themselves, taking intoaccount many developments since the models were first proposed in late 1997. The set of noisemodels has also been extended to include models in which it is assumed that ADSL does not use afull 1.1 mHz bandwidth. This noise model contains information also valuable for SDSL.



TD-37, Methods to defend (V)DSL against non-stationary cross-talk (P. Reusens, Alcatel), statesthat any power control technique will create a non-stationary cross-talk. The worst cross-talk isa dynamic power saving state, which has the most rapid level variation. A VDSL system mustinclude a method to defend the noise margin dynamically in the presence of such non-stationarysignals. Potential techniques are discussed.

Time variation of noise must also be included in impairment generators for performance testing.

In TD-39, Discussion of the impairment generator (P. Reusens, Alcatel), the noise models in theimpairment generators for the performance test of VDSL contain a white noise source G4. Thisgenerator is thought to model the thermal noise of the DSL receiver. This is not practical: it isproposed to limit the use of the G4 model to computer simulations, e.g., when comparisons aremade between different line-codes and modulation methods. In that circumstance, G4 models thesensitivity of the receiver, to force a more objective, or fair, comparison.

TD-41, VDSL Noise Models - Integrating ADSL FEXT (D. Manor, Orckit), indicates that one ofthe most dominant noise sources in the VDSL noise environment is ADSL cross-talk appearing asNear End Cross Talk (NEXT) and Far End Cross Talk (FEXT). It is shown that the currentnoise model masks do not represent this noise component accurately because its loop dependencyis not taken into account and the noise model is fixed over all loops. Two methods to remedy thisfault are suggested.

TD-27, Noise tests of xDSL systems M. Meninger, SPT Telecom), presents views of VDSL tests.

TD-23, Proposal for cross-talk combination method (L. Magnone, CSELT, for FSAN), notes thatthe FSAN Noise Combination Method has been accepted for combining cross-talk contributionswithin the TM6 VDSL project; the FSAN VDSL working group continues to use and support thismethod in ANSI T1E1.4. This contribution provides a detailed comparison between this methodand the Mean PSD method and shows that while neither is ideal, the FSAN method is a bettercompromise in the absence of any other alternative. It is suggested that a spectral compatibilityproject be started and that this document be taken as a basis. TM6 decided that this TD will beincorporated into the Cable Reference model for Europe (permanent document).

TD-50, Time domain requirements on noise in performance tests (R. van den Brink, KPN),presents a literal text proposal for the VDSL performance tests in part 1. It solves the problemof inadequate time domain requirements on the impairment noise. The solution is to specify amask on the amplitude distribution. This mask specifies how close the impairment noise shall beto true Gaussian distribution and what its crest factor will be. A match is proposed within 10%and a crest factor of 5, but those values are open for discussion.

The following contributions address the installation procedure and the power back-off.

TD-36, Some thoughts on the VDSL installation test procedure (P. Reusens, Alcatel), contains alist of issues related to the VDSL installation, noting that this has not been defined in the currentfunctional requirements. Based on these issues, TD-36 compiles a set of requirements of theVDSL signals and procedures at installation. TD-36 notes the need for an initial handshakeprocedure.

TD-05, “Equalized FEXT” Upstream Power Cutback Method (K. Jacobsen, TI), describes yetanother method of upstream power cutback to mitigate near-far FEXT in distributed cabletopologies. The method involves shaping upstream transmit PSDs such that all VTU-O (VDSLTransceiver Unit at the Optical network unit) receivers see the same FEXT PSD.

TD-24, Power Back-off strategies for VDSL, TDD versus FDD performance (P. Reusens, Alcatel),discusses power back-off strategies for upstream VDSL, Time Division Duplexing (TDD) versusFrequency Division Duplexing (FDD) performance. TD-24 proposes a new power back-offstrategy, the reference noise power back-off. It requires frequency dependent adjustments of thetransmit level. DMT modulation offers the flexibility to shape the transmit PSD as required bythis power back-off. Both duplexing methods can be tuned to have optimal performance, but TDDis easier to manage than FDD.



TD-25, Power Back-off for multiple target bit-rates (D. Bengtsson, Telia), describes a new efficientmethod to perform power back-off in the upstream of a VDSL system. The method leads to higherdata rates than the earlier proposals, i.e., the constant back-off method and the reference lengthmethod.

TD-06, TDD CAN Support Multiple Data Rates in the Same Binder (K. Jacobsen, TI), showsthat, contrary to the assertions of some FDD proponents, TDD VDSL solutions can indeedsupport a wide range of data rates in the distributed cable topology. As a result, the requiredupstream power cutback at higher frequencies is not as high as previous contributions haveassumed. In fact, a wide range of data rates can be supported in the same binder.

TD-48, Involvement from management concerning power back-off (T. Kessler, DTAG), gives andexplains the requirement that the management system must be able to control the power back-offof each VDSL-NT. Thereby, a question asked by Savan/Siemens in TD-19 (Vienna meeting) isanswered.

TD-29, Application mixing in VDSL (Telia, Sweden), presents more views on mixing of differentapplications in VDSL.

SDSL

The plan of the SDSL session under the rapporteur S. Schmoll (Alcatel) is in WD-09. In order tospeed up the process within ETSI and be able to provide functional requirement inputs, TD-09,modified SDSL workplan, was introduced by A. Carrick (Ascom) (see above).

In TD-11, Comments on Table of contents of DTS/TM-06011-1, A. Carrick (Ascom) proposes anew table of contents for part 1 of the SDSL Technical Specification to deal with functionalrequirements and to be in sync with the ITU-T SG15 work.

It was proposed in TD-09 that Milestone 2 of this work item, Table of Contents and ScopeCompleted should be achieved by the end of the present meeting. In order to facilitate this,improvements to the existing draft were proposed for discussion. This document was accepted byTM6 delegates as the new table of contents.

TD-10, Text proposal for clause 1 Scope of DTS/TM-06011-1 (A. Carrick, Ascom), proposes textas a basis for discussion and agreement. This text was agreed with minor editorial revisions.

TELECOM OPERATOR REQUIREMENTS FOR SDSL.

TD-51, The position of the Dutch operator on SDSL (R. van den Brink, KPN), states that SDSLbecomes attractive for KPN if it will transport broadband services to the SOHO and consumermarkets. Simultaneous transport of telephony is therefore a key requirement; telephony must betransported in-band to outperform ADSL under equivalent noise conditions. Features such as n x64 kbits/s data transport only, G.703 interfacing, low jitter, and wander are no more than “niceto have” features. Lifeline service is not considered essential. This TD was accepted as input toSP 1-1 of the SDSL living list.

TD-28, Further items for SDSL (T. Stefanson, Telia Research), presents views regardingapplications, line code, and remote power for the emerging SDSL standard.

TRANSPORT OF NARROW-BAND SERVICES

In TD-43 (A. Kliger, Metalink), several operators express their requirement to transport narrow-band services, such as ISDN or Plain Old Telephone Service (POTS) lines, over SDSL.Multiplexing the narrow-band service digitally with the wide-band data transport isrecommended. This allows more efficient use of bandwidth, flexibility, and any number of POTSchannels, and avoids the complexity of splitter issues. It also proposes to use the multiplexingmethod that is provisionally agreed for the transport of ISDN over HDSL.

TD-46, Multi-service Application for SDSL (Marconi Communications Ltd.), outlines a potentialmulti-service network application for SDSL transmission between user and access networkequipment which can be supported by new ETSI signaling proposals for V5.2 controlled Internet



Access. TM6 delegates recommend sending this document to the ANAI (Access NetworkArchitecture and Interfaces) group, where expertise on higher layers is more widespread. ANAIis meeting in Bonn and has recently become part of TM6.

TD-12, Text proposal for clause 4 Reference Model of DTS/TM-06011-1 (A. Carrick, Ascom),proposes text for clause 4 of the Reference Model, as it is felt that the reference model is anintegral part of the scope. L. Humphrey became the owner of the Reference Model for SDSLstudy point 1-12 of the SDSL living list. A discussion took place around the requirement totransport the SDH TU12 payload (~2.3Mbit/s) with or without additional ISDN/POTS services.TM6 delegates agreed that when transporting TU12 payload, there would be no requirement foradditional ISDN/POTS service. Note that the line codes for SDSL will be tested at 2.048 Mbit/s.Therefore maximum reach will not be met for the SDH TU12 payload.

Some delegates indicated that the reference model is critical to the progress of SDSL since it isnot sufficient to have an efficient physical layer, but one must know what to do with it. Delegateswere invited to comment on the content of TD-12; the reference model should be finalized by thenext meeting in February. As part of this reference model discussion, TM6 delegatesprovisionally agreed that both ISDN and POTS are to be carried in-band and that the splittersolutions will be discarded.

FRAME FORMAT

A third set of contributions address study point 1-10 of the SDSL living list and deal with frameformat.

TD-18, Frame Structure for SDSL Subscriber Loop (J. Hausner, Siemens), proposes a flexiblebandwidth efficient SDSL frame structure. Payload bit rates can be varied between 256 kbit/sand 2304 kbit/s in steps of 8 kbit/s. Therefore, the payload can be composed of combinations ofISDN B-and D-channels or D/A-converted POTS. ISDN M-bits can be mapped onto the SDSLoverhead. Thus maintenance as proven over several years in ISDN systems which have beendeployed in high volume can also be used for SDSL applications. This frame format assumes thatthe network clock is looped back at the network termination (NT) side; some delegates feel thatthis is a very restrictive condition in a deregulated or leased line environment. J. Besseyre(Consultelecom) asked if the transport of an existing HDSL frame was studied and the answerwas no; a question also came up about the transport of 2.304 SDH TU12 payload and the answerwas not quite clear.

TD-19, Embedded Protocols for SDSL Subscriber Loop (A. Tanhauser, Siemens), notes thatpayload blocks of the SDSL frames are grouped together to comply with the requirements of thesupported services. The data transmission over the SDSL line will be ensured using reliableservice specific protocols to be embedded within the dedicated channels of the SDSL frame.These protocols, described in this contribution, will satisfy all necessary premises.

Part of TD-43 deals also with frame structure; A. Kliger (Metalink) was appointed as owner ofStudy Point 1-12 Frame format; TD-18, TD-19, and TD-43 are inputs to that study.

POWER FEEDING

TD-03, Power feeding issues on SDSL (M. Peschke, TELES AG), includes power calculations fordelivering lifeline POTS/ISDN services within the lower frequency range of SDSL for remotepower regeneration. Assuming 70% efficiency of DC to DC converters, 1.56 W is the net poweravailable. Assuming POTS services with no power averaging at the remote (for ring generation),this leaves 0.97 W available power after supporting POTS service, or 0.81 W (assuming 0.33 Wfor ISDN U to S transceiver) after supporting ISDN service. This document is based on WD-19from the Vienna meeting.

TD-42, Remote power feeding for ISDN/POTS lifeline services (J. Benndorf, Alcatel), is afeasibility study on the delivery of ISDN/BA and POTS lifeline services for SDSL systemsfeaturing remotely powered regenerators or NTs. It concludes that a separate transceiver fornarrowband service is not necessary. This paper gives good insights into the relation between



line code, line driver actual implementations and required power. J. Besseyre suggested that thepower estimation contained in TD-42 should be compared with those contained in TD-35.

TD-35, The Effect of Spectral Shaping on Line Driver Power Consumption (M. Kimpe,ADTRAN/Siemens), describes how spectral shaping is used in ANSI HDSL2 to tailor thetransmit spectrum to a particular set of cross-talk interferers. By deviating from the two-pairHDSL spectrum, the ANSI HDSL2 peak-to-rms ratio is increased; this contribution estimates thepower dissipation due to the analog line driver for various line code transmit PSDs. Theconclusions are that 2-pair HDSL PSD allows for considerably lower power than ANSI HDSL2and that care should be taken to minimize the peak-to-rms ratio. This excellent paper led to adiscussion; the numbers given are theoretical numbers not taking into account the actual linedriver performances and efficiency. P. Brackett (Adtran) became the owner of the power studypoint. J. Besseyre asked the Telco operators for three figures: what is available in terms ofDirect Current (DC) power at the end of the worst case line, what is needed for POTS, and forISDN. S. Schmoll, rapporteur, recalled that those numbers are in WD-19 of the Vienna meeting,even though some operators challenged the numbers contained in that WD.

NOISE MODEL

TD-14, Revised noise models for VDSL (R. Kirby, British Telecom, on behalf of FSAN), givesmore information about the derivation of the FSAN VDSL noise models, including all the networkscenarios on which they are based. It also presents a revision of the noise models themselves,taking into account many developments since the models were first proposed in late 1997. Theset of noise models has also been extended to include models in which it is assumed that ADSLdoes not use a full 1.1 mHz bandwidth. This noise model also contains information useful forSDSL.

TD-49 (T. Kessler, DTAG) proposes use of the FSAN VDSL noise models “LT FTTEX” and “NTFTTEX” for the urgent SDSL performance calculations until a complete ETSI model is agreed.

R. van den Brink (KPN Research) intervened to highlight that the VDSL noise model does notinclude the distance which is assumed relatively small for VDSL compared to SDSL; this VDSLnoise model does not include the SDSL spectrum and the length dependent FEXT fromasymmetrical system such as ADSL. The model cannot be picked up as is: it needs to beadapted to the model contained in TD-27 of the Vienna meeting. The whole topic of relationbetween FSAN and ETSI led to an animated discussion. An ad hoc group met to discuss theSDSL noise model; its conclusions, in WD-15, are that for line code performance simulations, twonoise models need to be used until further notice, the noise model contained in TD-27 of the Viennameeting and the FSAN B noise model which can be found on the FSAN server(http://btlabs1.labs.bt.com/profsoc/access/fsan_vdsl.htm). A liaison contained in WD-10 was sentto FSAN; R. van den Brink was asked to participate at the next FSAN meeting in January to tryto harmonize the noise models. Note that some operators, such as the Dutch PT&T KPN, are notpart of FSAN.

TD-30, Performance of TS 101 135 2-pair HDSL under various noise models (M. Kimpe,Adtran/Siemens), examines the performance of the current Pulse Amplitude Modulation (PAM)and Carrier-less Amplitude Modulation Phase Modulation (CAP) HDSL 2-pair systems, asdefined in TS 101 135, under different noise masks. Based on the FSAN scenarios, a tighternoise upper bound was created. Under the new mask, termed “Revised FSANB,” HDSL two-pairsystems have a reach comparable to TS 101 135. This “Revised FSANB” noise model is proposedfor use in the evaluation of SDSL line codes.

According to TD-30, the existing HDSL 2-pair systems would neither satisfy the stringentrequirements of the model of the TD-27 of the Vienna meeting nor those of the FSAN B model interms of reach. Some manufacturers expressed their concerns that both noise models are still toopessimistic. TD-30 was also sent to FSAN as part of the liaison from TM6 to FSAN.

TD-38, ISDN a non-stationary NEXT source to SDSL (P. Reusens, Alcatel), states that mostISDN-Basic-Access U-interfaces are deactivated unless a call is ongoing. The non-stationarybehavior of the cross-talk can cause problems in modern xDSL systems. Indeed, such systemscan optimize their noise margin around 6 dB, to reduce cross-talk and save power. If SDSL



optimizes its noise margin, it must include methods to survive any non-stationary ISDN-BANEXT. Moreover, a non-stationary ISDN-BA NEXT should be included in the relevant noisesources of the respective xDSL requirements.

TD-47, OPTIS (Overlapped PAM Transmission with Interlocking Spectra) Performance underthe loop and noise conditions of TD-27 (September, 1998), was presented by J. Benndorf, KEKommunikations Electronik GmbH, (for Level One). In Lulea, Sweden (June, 1998), Level Onepresented TD-24 which proposed a derivative of the ANSI OPTIS PSD template that wasmodified to operate within the noise conditions of Europe. The contribution demonstrated howshaped PAM modulation may be employed to improve spectral compatibility. This contribution isa continuation of TD-24 (June, 1998) and demonstrates OPTIS performance under the noiseconditions of TD-27, September, 1998.

PERFORMANCES OF DIFFERENT LINE CODES UNDER DIFFERENT NOISE CONDITIONS

TD-44, Performance projections of SDSL at 2320 kbit/s (A. Kliger, Metalink), presentsperformance evaluations of 2320 kbit/s SDSL system in conjunction with several noise models. Itis clear that the noise environment of the new SDSL system should be defined before the systemline code and spectral shaping can be determined.

TD-31, Performance of E1 rate SDSL systems: Symmetrical Single Carrier (M. Kimpe,Adtran/Siemens), compares the performance of various symmetric single carriers modulationtypes on various 0.4mm loop lengths. Three noise masks are considered: the one in TD-27, theFSANB noise model, and the revised FSANB noise model introduced in TD-30. The bestperforming modulation type from the PAM family is generally Coded 16-PAM on the LT side andcoded 8-PAM on the NT side. The best performing modulation of the CAP/QAM family is theCoded 64-CAP. The best-performing PAM outperforms the best-performing CAP modulation.These results are based on infinite length, ideal DFE margins and 5.0 dB of coding gain. Inaddition, it is assumed that the spectra extends to 0 Hz.

TD-32, Performance of E1 rate SDSL systems: Asymmetrical PAM (M. Kimpe, Adtran/Siemens),examines the performance of asymmetrical PAM approaches to E1 rate SDSL. Such a solutiontakes advantage of the asymmetry of the noise masks at the LT and NT while reducing selfNEXT so that it is not the limiting interferer. A rather comprehensive study of possiblecombinations of symbol rate, bandwidth, and power level was conducted. With 16.3 dBmtransmit power at the NT and 13.3 dBm at the LT, and a manageable spectral shape, margins ofthe order of 6 dB (using a revised FSAN model as in TD-30) can be achieved at a reach of 2.5 kmover 0.4mm wire. This is similar to the reach of current HDSL 2-pair systems under the sameconditions. If only self-NEXT is considered, a much greater reach can be achieved. Spectralshaping is briefly discussed.

TD-33, Performance of sub-E1 rate SDSL systems: Symmetrical PAM (M. Kimpe,Adtran/Siemens), examines the performance of symmetric PAM on various 0.4mm loop lengths inorder to deliver sub-E1 rates. Three noise masks are considered: the one in TD-27, the FSANBnoise model and the revised FSANB model introduced in TD-30. The target reach of 4500m on0.4mm at 384 kbit/s can be reached with any of the Coded systems simulated, when using therevised FSANB model. Therefore, performance at the sub-E1 rates will not be the designbottleneck.

TD-34, Potential Common Elements Between ANSI HDSL2 and ETSI SDSL (M. Kimpe,Adtran/Siemens), reviews the various elements of the ANSI HDSL2 draft standard physicallayer, and discusses ways for those elements to be reused in a device which contains both theANSI HDSL2 standard and an ETSI SDSL standard. The conclusion is that most of thephysical blocks used for HDSL2 can be reused for SDSL.

TD-17, Low frequency cut-off for SDSL spectrum (A. Carrick, Ascom AG), notes that the low-frequency cut-off of existing HDSL systems is about 10 kHz. Previous SDSL contributions haveproposed that this cut-off frequency be either increased to above the POTS band, or decreased to4 kHz to increase the data capacity. This contribution outlines some potential considerations inthe choice of low frequency cut-off for the SDSL spectrum and notes that the capacity gains with



different low frequency cut-offs are relatively small. It was proposed that a new study point beopened to address this issue.

J. Besseyre noted that most of the results presented in the three TDs above could apply to CAPwith a variation of 0.2 dB on overall performances; S. Schmoll encouraged contributions in thatdirection. The TM6 Chair created an ad hoc group for line coding under M. Kimpe (AdtranSiemens). Some delegates objected to starting work on part 2 of SDSL when part 1 is notcompleted; in fact, it is not clear what the schedule for completion of part 1 is. A review of “identified” (identified replaces agreed) part 1 requirements was conducted. The critical pointconcerns the payload; all simulations of performances and tests are to be conducted at the bitrate of 2.048 Mbps; the payload may extend to 2.320 Mbit/s and maybe even higher for somedelegates to the extent of the reach; in the case where TU12 SDH payload is carried, there is norequirement to also carry a POTS/ISDN narrow band service.

TD-45, Performance projections of SDSL at variable data rates (A. Kliger, Metalink), presentsperformance evaluations of an SDSL system in conjunction with several noise models at variabledata rates.

TD-40, The use of Discrete Multi-tone Carrier (DMT) for a flexible SDSL system (P. Reusens,Alcatel), proposes the use of DMT as line code for SDSL and explains the advantages in terms offlexibility and re-use of a number of ADSL solutions with the elimination of the POTS/ISDNsplitter.

LIVING LIST

The study points in Table 1 are active for SDSL part 1.

Jacques Besseyre, Consultelecom

Number Title OwnerSP 1-1 Functional Requirements Lucent MicroelectronicsSP 1-2 Investigation of an optimum line code Martin Kimpe (Adtran/Siemens)SP 1-3 Rate Adaptation George Eisler (Rockwell)SP 1-4 Asymmetric Transmission Method Rob van den Brink (KPN)SP 1-5 Noise models Rob van den Brink (KPN)SP 1-6 Multiplexing functions of sub-channelsSP 1-7 Investigation of a MIB Les Humphrey (Nortel)SP 1-8 Management of bit rate in case of failureSP 1-9 Power budget for remote power feeding Peter Brackett (Adtran/Siemens)SP 1-10 Transport of narrow-band services Siegfried SchmollSP 1-11 Impact of wiring in the customer premises Massimo Sorbara (Globespan)SP 1-12 Frame format Avi Kliger (Metalink)

Table 1. SDSL Study Points.



TM6 MEETING ROSTER, NOVEMBER 24 - 27, 1998, SOPHIA ANTIPOLIS, FRANCE

Hans-Jörg Frizlen, ETSI TM6 Chair

Austria Telekom Austria AG Manfred GindelBelgium Telindus Jan Marien

Alcatel Thierry PolletAlcatel Peter Reusens

Czech Republic SPT Telecom a.s Milan MeningerFinland Nokia Telecommunication Jari Lindholm

FINNET Group Pasi PoikolainenTellabs Oy Janne Väänänen

France France Telecom Belgacem DjebariRockwell International George EislerETSI Hans-Jörg FrizlenLevel One Com. Europe Simon HollandTexas Instruments SA Krista JacobsenRockwell International Ragnar JonssonST Microelectronics Simone MazzoniTexas Instruments SA Neil QuarmbySAGEM Group Roger SamyFrance Telecom Marc Vautier

Germany KE Kommunikations Elektronik GmbH Jens BenndorfGlobespan Semiconductor Jacques BesseyreSiemens AG Peter BrackettSiemens AG Josef HausnerDeutsche Telekom AG Werner HenkelLucent Technologies Chris Jollyeutsche Telekom AG Thomas KesslerSiemens AG Marc KimpeSiemens AG Shimon PelegAlcatel Siegfried SchmollLucent Technologies Mark SimkinsSiemens AG Klaus StarnbergerSiemens AG Armin TannhäuserSiemens AG Josef Waldinger

Israel IAEI Avi KligerIAEI Daniel Manor

Italy Telecom Italia SPA Flavio CucchiettiTelecom Italia SPA Lorenzo MagnoneMarconi Communications Davide PagnottaMinisterio Communicazioni ISCTI-VFF G. Pierri

Norway Norwegian Post&Telecomms. Authority Jarle KiilSweden Telia AB Daniel Bengtsson

Ericsson Telecom AB Jan BoströmTelia AB Tomas Stefansson

Switzerland BAKOM / OFCOM Kurt BartschiAscom AG Angus CarrickPro Telecom Christof Heidelberger

The Netherlands KPN Research Rob van den BrinkLucent Technologies EMEA B.V: Vladimir OksmanLucent Technologies EMEA B.V: Brian Waring

UK Fujitsu Europe Graham ArnoldMotorola Ltd. Bernard DugerdilBT plc Kevin FosterNortel Networks Les HumphreyBT plc Rob KirkbyConsultronics Europe Steve MlikanMarconi Communications Ltd. Graham Phillpott

USA Cisco Systems Inc Justus Osude



REPORT OF ITU-T Q4/16 AND Q23/16 RAPPORTEUR MEETINGSDECEMBER 7 - 9, 1998, SAN DIEGO, CA

Q23/16, PCM MODEMS

The primary purpose of this Q23/16 PCM Modem Rapporteur meeting was to complete the text ofV.91 for submission as a White document to the next SG16 meeting. V.91 (ex V.adm) defines theoperation of duplex all digital mode of V.90. The issues remaining from the September 1998SG16 meeting at which V.91 was Determined were:

• Finalization of the provisional agreement on the structure of Phase 1 of the startup procedure,which depends on the definition of robust DCPc and DCPa signals

• Further consideration of test loops (sec. 9)• Clarifications to the scope and title of the Recommendation pertaining to the definition of

switched digital network, and type(s) of network over which V.91 is intended to operate• Work on an Appendix describing typical network configurations for V.91

L. Brown (Motorola) is the Rapporteur. There were no incoming liaison statements.

No contributions were received on V.90 Issue 2, i.e., PCM in the upstream direction.

REVIEW OF V.91 DRAFT TEXT

PCM98-035 (Richard Williams, 3Com, editor) is the latest draft text of V.91. It includesproposed wording changes to the title and scope to address some of the concerns from theSeptember SG16 meeting. The use of the term “4-wire chain,” taken from RecommendationG.101 (Transmission Plan, General Definitions), was proposed as a generic term covering digital4-wire connections.

PCM98-031 is the updated issue list. It includes the agreements from the September SG16meeting, and the two options identified for the Phase 1 start-up. PCM98-031 was approved withsome clarifications.

PRESENTATION OF TECHNICAL PAPERS

PCM98-032 (G. Hoefer, Siemens) proposes a DCP (digital connectivity probing) signal which issquare wave with maximum allowable amplitude and frequency not greater than 100 Hz. It isintended to:• Differentiate between a digital channel using ADPCM and a digital channel free of ADPCM;• Distinguish between a digital and an analog channel.The contribution was supported by PCM98-036 (G. Hoefer, Siemens) which includes test resultsof the proposed DCP signal on channels with combinations of ADPCM, robbed bit signaling, anddigital pads. PCM98-036 concludes that, with minimum complexity, the square waveaccomplishes the objectives for the DCP signal.

PCM98-033R1, revision of TR-30.1/98-11-028 from 3Com & Lucent (3Com, GDC, Lucent,MultiTech), proposes the use of V.8 for the start-up rather than the DCP sequences, i.e., thefallback position from the September SG16 meeting. It requires the specification of a V.91 bit inV.8. It is proposed on the basis of:• Optimizing start-up times• Avoidance of interoperability and reliability problems• Avoidance of need for definition of new procedures for fallback to V.8• Existing V.8 implementations can be reused

PCM98-034R1 (T. Liau, Rockwell, same as TR-30.1/98-11-32) proposes a DCP signalrepresenting a constant magnitude signal with a scrambled and differentially encoded sign bit.For DCPC the signal has an energy level less than -48 dBm0 to avoid compatibility issues. TheDCPA signal uses a different scrambling polynomial for the sign bit and a higher magnitudesignal to ensure that network echo cancellers are disabled. PCM98-034R1 discusses thedetection procedures and concludes that this signal will allow the detection of whether the



connection is all digital or not, and for the former, whether it includes ADPCM. Test results forthe DCPA signal are included.

PCM98-037 (G. Hoefer, Siemens) is a critique of the Rockwell-proposed scrambled sign DCPsignal in PCM98-034. Because the analysis was based on PCM98-34, an earlier contribution,rather than PCM98-34R1, it was not reviewed in detail.

DISCUSSION AND AGREEMENTS

The main issue was the resolution of Phase 1 of the start-up. In discussion it became clear that acompromise could be reached through the use of V.8 procedures, along with a new signal as apreamble to Phase 2, which could be used for connection analysis prior to the INFO exchange. Itwas thus agreed that V.91 would use V.8, followed by 75 ms of silence and then a new signal EZ

which is defined as 24T of binary zero mapped as the sign bit of Ucode 66, just prior to the INFOsignal.

The use of V.8bis procedures is optional as for V.90. Both V.8 and V.8bis will require thedefinition of V.91 codepoints. (See the report of the Q4/16 meeting, below.)

Other agreements included:• The wording of the title and scope was clarified, but the term “4-wire chain” was not adopted

as it was not thought to be universally understood.• The use of the same scrambling polynomial in both directions was agreed, i.e., that used by a

V.34 call modem.• The start-up and retrain procedures were reviewed and time-outs added as appropriate.• Clarification added on the procedures for escape to V.34 mode - new section 8.3.• New section added on transparent mode operation - new section 8.7.• Clarifications made to the informative Appendix.• Other editorial corrections and clarifications.

These agreements were incorporated in an updated draft of the Recommendation in PCM98-038,A Digital Modem Operating at Data Signaling Rates of up to 64000 bit/s for Use on a 4-WireCircuit Switched Connection and on Leased Point-to-Point 4-Wire Digital Circuits, which wasapproved with some further amendments. The final output is PCM98-038R2©, which has beensent to the ITU-T as the White document version of V.91 for approval in May 1999.

INTELLECTUAL PROPERTY

Companies were reminded that Patent statements in relation to Recommendation V.91 must besubmitted to the ITU TSB prior to the next SG16 meeting in May 1999.

Q4/16, MODEMS

At the September 1998 SG16 meeting, it was agreed to focus all modulation work under Question23/16. Q4/16 now handles the remaining work on modem issues; all work of the previousAssociate Rapporteur Groups, e.g., on Recommendations V.8 and V.8bis, and the new work oncoordination of modem managed objects, is now brought together under the new Rapporteur B.Adams (Hayes). Technical editor’s note: Hayes recently ceased operations. However, B. Adamsnoted on e-mail to the reflector that he plans to continue as Rapporteur.

WP1/16 intends to revise the text of Question 4/16 at the next SG16 meeting. Suggestions forrevisions were requested by the WP1/16 Chair, J. Magill, [email protected].

Q4-16_98-00 is the document list, which was amended during the meeting. The agenda is in Q4-16_98-09.

Q4-16_98-11 was distributed for information and is an extract of the WP1/16 report from theSeptember SG16 meeting covering the Q4/16 work.

There were no incoming liaison statements.



RECOMMENDATIONS V.8 AND V.8BIS

Q4-16_98-04 (TIA, USA) was introduced by L. Brown (Motorola); he informed the RapporteurGroup that TIA is not now planning to use the information category octet assigned to it in V.8,and proposed that this assignment in Table 2/V.8 be removed. This was agreed, and V.8 will beamended accordingly at the next opportunity.

L. Brown, the Q23/16 Rapporteur, informed the meeting that the Q23/16 meeting had agreed touse V.8 in the Phase 1 start-up of V.91. Accordingly a V.91 bit was required in V.8, and also inV.8bis. The meeting agreed to the following allocation:• In Recommendation V.8 : “V.pcm availability octet” - bit 7 assigned to Recommendation V.91;• In Recommendation V.8bis : “Data {NPar(2)} Octet 4” - bit 1 assigned to Recommendation

V.91.

It was also agreed that the V.pcm category in V.8 would be renamed to “PCM ModemAvailability” to more accurately reflect its meaning and recognize the existence of both V.90 andV.91.

MANAGED OBJECTS FOR MODEM DIAGNOSTICS

The Q4/16 Rapporteur proposed that the work on V.mmo (modem managed objects) make use ofan issue list as had been used in the development of V.90 and V.91. He introduced a first draft ofsuch an issue list in Q4-16_98-07. It was agreed that such a list would be used.

Q4-16_98-01© (Managed Objects for Modem Diagnostics, Rockwell) and Q4-16_98-02© (Proposalfor Defining Managed Objects for Diagnostic Information of Voice-Band Modems, USA) areD.198© and D.210© respectively, from the September SG16 meeting. They were introduced byK. Chu (Rockwell) primarily as background for the work. D.210© (F. Lucas, USA) is the USAproposal to initiate this work, and D.198© (K. Chu, Rockwell) is the first attempt at definingmanaged objects for modem diagnostics.

Q4-16_98-03© (K. Chu, Rockwell, editor) is a proposed template for a Recommendation “ManagedObjects for Diagnostic Information of V-series DCEs.”

Q4-16_98-05 (B. Adams, Hayes) proposes a mapping of #UD commands into the proposedmanaged object structure of D.198©.

Q4-16_98-08©, Unimodem Diagnostics Command Reference Specification Version 1.0.1(Microsoft), was introduced by Fred Lucas (3Com), who is also Chair of TIA TR-30.2 andRapporteur for Q7/16. It was noted that Version 1.0.2 now exists and is believed to be availableon the Microsoft web site.

Introduction of these contributions stimulated general discussion on how the work shouldprogress. Issues raised included:• Should this work be a revision to Recommendation V.58 (Management information model for

V-Series DCEs) or a new Recommendation?• If the latter, how are overlaps with V.58 handled?• Should the work be strictly limited to diagnostic information only, rather than the more

general content of V.58?• Both machine and human readable information output is required• Should the work be in two parts - definition of the objects, and definition of mechanisms to

access them?

The main issues were left open for further consideration and discussion again at the next meeting.It was felt useful however, to continue with the definition of the required managed objects suchthat these could then be reviewed against the existing content of V.58.

It was agreed to categorize objects such that they could be put together in a tree structure, i.e.,• “Gross” objects - those encompassing many sub-objects• Mid-level objects• Lowest level objects



Mid and lowest level objects can appear in more than one place in the tree, e.g., parametersapplying to more than one modulation mode.

Based on the contributions and the ensuing discussions, an initial set of gross objects and sub-categories was agreed. These will be included into a working text based on Q4-16_98-03, withthe intention of progressing the work by correspondence prior to the next SG16 meeting in May1999.

RECOMMENDATION V.250 IMPLICATIONS

Q4-16_98-10 (J. Ellwood, Hayes) proposes some specific extensions to the V.250 command set forthe retrieval of managed object information. It was noted that this document is for information,as the responsibility for Recommendation V.250 lies in Q7/16.

OTHER ISSUES

K. Chu (Rockwell) informed the meeting that Conexant (née Rockwell) intends to submitcontributions to future meetings on the subject of V.dispatch, i.e., a method of changing operatingmode while maintaining the modulation, e.g., transitioning from H.324 to V.34 fax.

It was noted that SG8 has Determined their Recommendation T.66 (T.codes) which defines theuse of V.8 and V.8bis for fax applications. It is expected to be approved at the next SG8 meetingin March/April 1999.

No further meeting of the Q4/16 Rapporteur Group is planned before the next meeting of SG16 inMay 1999.

Q4/16 AND Q23/16 RAPPORTEURS ROSTER, DECEMBER 7 - 9, 1998, SAN DIEGO, CA

Bruce Adams, Hayes Q4/16 RapporteurLes Brown, Motorola Q23/16 RapporteurHost: 3Com

3Com Fred Lucas [email protected] Richard Williams [email protected] Devices Primo Haim [email protected] Agarwal Sandeep [email protected] Systems Kevin Riley [email protected] Technology Jordan Cookman [email protected] Telecom Georges Sebek [email protected] Bruce Adams [email protected] Ali Sadri [email protected] UK John Magill [email protected] Les Brown [email protected] Telephone Juha Kauppi [email protected] Keith Chu [email protected] Thomas Liau [email protected] Glen Griffith [email protected] Gerald Hoefer [email protected] Neal King [email protected] Peter Plets [email protected] Technologis Victor Demjanenko [email protected]



REPORT OF THE ANALOG TERMINALS AND ACCESS (ATA) MEETING,DECEMBER 7 - 11, 1998, CAPE TOWN, SOUTH AFRICA

The Chair of ETSI ATA is N. Encarnação (OFCOM/BAKOM). Forty-five delegates attended thisfirst ETSI ATA meeting held outside of Europe.

TD-02 is the report of the previous ATA meeting.

TD-03, the ATA meeting action list, includes:• A report from the Chair of possible reorganization in the Terminals area• A report on a network terminating point from P. Dofnas (Ericsson Business Networks)• Getting information on ATA documents from the ATA web site (TD-27)• The technical report on a terminal support interface for harmonized analog Public Switched

Telephone Network (PSTN) terminals, published as TR 101 185; the Guide is out for members’vote as EG 201 185.

• TD-20, Disposition of comments received during EG 201 185 voting. Alcatel Telecom,Deutsche Telekom, France Telecom, and Norwegian PTA voted no with comments. However,the document reached the required level of support for approval.

ORGANIZATIONAL

The Chair gave a report of activities since the last meeting (June, 1998). There were no workinggroup meetings of ATAc (conventional technologies). TR 101 182 on definitions, abbreviationsand symbols was published, as was EG 201 184 on multiline terminal equipment and EG 201121 on the application of TBR 21. EN 301 437 (PSTN attachment requirements for voice TE inwhich network addressing, if provided, is by DTMF) was out for public inquiry. ATAe (enhancedsignaling systems) had two working group meetings; ES 201 071, Protocol over the local loop fordisplay services; Server Display and Script Services (SDSS), has been published. This standardis a “delta” document, setting out the differences in the European adaptation of Bellcore SR-INS-002461 and TR-NWT-001273.

It was noted that CTR 21 (PSTN access requirements for non voice terminals) and CTR 38(requirements for handset telephony) have both been adopted by the European Union (EU). TheChair wishes to see the speech requirements of TBR 38 input to SG12 of the ITU; ETSI TC STQ(Technical committee for Speech processing, Transmission and Quality) had not put an input intothe last SG12 meeting.

TD-36 (D. Barry, Technical Recommendation Application Committee [TRAC] Chair) describesthat, following a clearer definition of the justified case for voice telephony, an I-CTR 37 (proposalfor a committee decision on PSTN voice access) had been put to the Approvals Committee forTerminal Equipment (ACTE) by the Commission; TD-35 (D. Barry, TRAC Chair) gives the textof the proposal. The proposal is similar to EN 301 437 except that there is no requirement foraudible ringing and no requirement to limit the current to 60 mA. France had objected to ETSItechnical work being rejected by the Commission. J. Long (DTI) noted that the Commission’saction was a useful initiative to make the market move and should be welcomed bymanufacturers.

The new edition of TBR 10 on Digital European Cordless Telephone (DECT) telephony that wasgoing for vote has adopted TBR 38 test methods and test impedances although not itsharmonized loudness ratings. ETS 300 001 (Attachment to the PSTN; General technicalrequirements for equipment connected to the analog subscriber interface in the PSTN) is now atthe fifth edition. Approximately half of the document is in the new ETSI style. ETR 075 hasproposed a significant restructuring and review of the standard. This matter needs to bereviewed soon. J. Montenot (France Telecom) queried the continuing usefulness of the standard.

TM-ATA-1 (TM6 Chair) is a liaison from TM6 asking for changes to TBR 21 to take account of allthe different Digital Subscriber Line (xDSL) terminals. The ATA Chair had provided a reply(ATA-to-TM6) giving preliminary information. He suggested to ATA that changes could not bemade without a new work item, but that the matter could possibly be dealt with as a comment atthe public inquiry on TBR 37 (see ATAc below for additional information).



TD-12 (E. Vidal, Telefonica) is a liaison proposing teleconferencing to reduce the number ofmeetings. TD-58 (ATA Chair) is a reply noting the low frequency of ATA meetings (2-3/year) andrejecting the proposal.

In response to suggestions from OCG (TD-08) that the terminals area should be reorganized, theChair introduced TD-50, a liaison to OCG noting the wish of DTA to remain a separate body andrecommending no change in the present organization of ATA.

TD-53 (W. Mellors, ETSI User Group) is a brief report of the October User Group meeting.Concerns were raised over the R&TTE Directive and its effect on QoS and the needs of disabledusers. Billing is currently the “hot” subject.

P. Dofnas (Ericsson Business Networks) spoke about ATA(98)020 which set out recent activitiesin the field of EMC. He also reported in TD-19 on the completion of the Simpler Legislation forthe Single Market (SLIM) initiatives in the field. Proposals had been made for revision of theEMC Directive to clarify its use and prevent further national measures from being brought intoforce.

R. Trevland reported in ATA(98)021 on safety activities since the last ATA meeting, particularlynoting the new edition of EN 41003 (based on draft IEC 60950 Edition 3) and potential safetyproblems (raised by Alcatel, Belgium) with using POTS/ADSL and ISDN/ADSL over the samecables.

F. Howett (Nortel) reported that STQ has taken on the cordless handset work but no progresshas been made; the author has been asked to revise the document to present it as more generalstandard rather than its present form as a description of a proprietary product.

TD-64, a liaison statement to TC PTS (Pay Terminals and Systems), provides comments to TD-29, PTS’ draft TS 100 xxx, PTS Data Over Voice for analog pay terminals. The reply discussesthe use of data over voice ASK or FSK (carrier/center freq. of 12, 16 or 32 kHz) modulation andrelated protocols for pay phones. J. Long (DTI) raised concerns that such systems upset POTS,that the work item was based on one manufacturer’s specific product and was not restricted topay phones. TD-64 was approved.

The meeting approved TD-65r1, a rather non-responsive reply to TD-23 from M. Bull of MSBdesign (a UK manufacturer of telecoms test equipment) which sets out some simple technicalconcerns on TBR 21 (now CTR 21).

S. Wüst (Deutsche Telekom) input a useful document (TD-66) setting out 10 differences betweenTBR 21 and EN 301 437 (TBR37).

TD-67 is the first draft of TR 101 292 (PSTN; Protocol over the local loop for display and relatedservices; proposed enhancements and maintenance of existing standards). It will be the mainissue for the next ATAe meeting.

R&TTE DIRECTIVE

The Chair spoke on the impact of the new R&TTE Directive and the work necessary in ETSI totake account of its new reduced essential requirements. Additional information on type approvalis available at http://www.ispo.cec.be/infosoc/telecompolicy/typeappr/. A number of ad hoc groupshad been set up covering market surveillance and One Stop Notification (OSN), InterfaceNotification and Equipment Classes, Interface Publication, Essential Requirements & UserInformation, and EG 201 399, a guide on the work. Details are given in TD-04 (ATA Chair); acollection of files of documents were put on the ETSI server.

There was a significant discussion on the implications of the new R&TTE Directive on the work ofATA. TD-07 (N. Encarnação, ETSI, ATA Chair) suggests a structure for harmonized standardsderived from the new essential requirements; TD-09 (D. Barry, TG6 Rapporteur) is a draft of aguide on their production. TD-10 is a liaison from the Operational Coordination Group (OCG)setting out a draft program of work. TD-11 (not distributed) is a collection of files on the serveron various aspects of the work.



J. Long (DTI) explained that the ad hoc groups should be seen as pre-TCAM organizations.TCAM (the Telecommunication Conformity Assessment and Market Surveillance Committee)would replace ACTE and TRAC under the new Directive. ETSI has no mandate to deal witharticles 3.1 and 3.2 of the Directive (on safety, protection, harm to the network, and EMC),pointing out that the essential requirement group was only dealing with article 3.3 (oninterworking, privacy, fraud, emergency services and disabled users). He warned that themembers of the ad hoc group on surveillance were very pro-surveillance and suggested that more“anti” members were needed.

It was noted that there was as yet no ad hoc group on harm to the network. J. Long said that adecision on article 3.1(c) was expected soon. The Chair explained that the ad hoc groups workedon a top-down basis whereas the work of ATA was bottom-up.

In addition to the five new work items created during this meeting concerning the R&TTEDirective, further work items were notified to permit publication of the DTMF signalingdocuments as TSs, with a further work item to allow for the production of a general part. Theparts will now consist of a general part, parts for DTMF transmitters and receivers in exchangesand another part for DTMF receivers in terminals. It is hoped that approval of the TS and theES to go for voting can be completed by correspondence in February.

A liaison statement to DTA (TD-56) on the impact of the R&TTE Directive on the present andfuture standards in the TE area was agreed, informing DTA of the five new work items andinviting them to join in the work. A similar liaison statement (TD-57) to OCG and TG6 wasagreed.

ATAC (CONVENTIONAL TECHNOLOGIES)

A discussion considered the liaisons with TM6 on the revision of TBR 21 to take account ofDigital Subscriber Line (DSL) performance requirements by modifying the send limits above 4.3kHz. TD-44 (B. van Leerdam, KPN Research) provides useful information on POTS/ADSLsplitter requirements. It was agreed that the TM6 input was an oversimplification and thatother out-of-band and regulatory aspects need to be considered, as explained in TD-21 (F. Howett,Nortel) and TD-30 (M. Harris, OFTEL). TD-54 proposes a new work item to report on thetechnical issues on the mutual compatibility of current and future POTS terminal equipment withxDSL systems operating at frequencies up to 30 mHz. It was supported by BT, OFTEL, Nortel,Ericsson LM, TDK Electronics, OFCOM and NPT, and approved. F. Howett (Nortel) will be therapporteur. A reply explaining the situation was drafted to TM6 in TD-55; it was approved atthe final plenary. It notes the use by “Home Network” equipment of frequencies up to 10 mHz onspare wire pairs and the necessity of new systems to coexist with existing TE designs.

The meeting was unable to make progress on the draft standard for Dual Tone Multi Frequency(DTMF) Transmitters and Receivers (ES 201 235) which was derived from CEPTRecommendations T/CS 46-02 and T/STI46-04 (now ETR 206 and ETR 207). The files on therapporteur’s (C. Harley, BT) portable had become corrupted and could not be reconstructed. Anew version will be put on the server in January.

TD-14 (ETSI WG TM4) is the final version of Access networks - UNI; Requirements for analogUNI dedicated to access networks terminated by a stand-alone unit at the customer’s premises(fixed wireless access).

TS 101 187 / ES 201 187

Draft documents TS 101 187 / ES 201 187 (loop disconnect specific dialing requirements) werereviewed together with TD-22 (predominately editorial comments by F. Howett, Nortel) and TD-46 (Z. Gütter, TESTCOM). The rapporteur (P. Erasmus, Telkom SA) noted that TD-22 would betaken into account in a new version of the document. TD-46 proposes reducing the break currentfrom 1mA to 0.5mA. This led to a long discussion on the definition of break time, reopeningarguments thought to have been settled long ago. Finally a small discussion group was set up tosolve the problem and a final version was produced for the plenary.



Draft documents TS 101 187 / ES 201 187 were approved in plenary for publication as a TS(Technical Specification) and submission to the membership approval process as an ES (ETSIStandard). A table of coordinates will be added for figure 2 (pulse shape).

TR 101 188 / EG 201 188

Draft documents TR 101 188 / EG 201 188 (harmonized analog presented network terminationpoint, NTP) were reviewed by P. Dofnas, (Ericsson Business Networks), the rapporteur. Thefollowing comments in were dealt with in clause order:TD-13, PSTN: NTP analogue interface (T. Salaun, TRT Lucent)TD-26, Comments to Draft EG 201 188 (R. Ceruti, CSELT/Telecom Italia)TD-32, Comments to Draft EG 201 188 (W. Schellhaas, Deutsche Telecom)TD-43, Comments to Draft EG 201 188 (France Telecom)TD-49, Comments to Draft EG 201 188 (Siemens AG)

There were many comments on the document. It was agreed that maintenance tests proposed inTD-49 were outside the scope of the document, being a quality matter which was up to thepurchaser of the equipment. It was noted that, as the specification would call for minimalrequirements, there might be some difficulty under European legislation for a purchaser whowished to specify additional requirements in order to purchase a better quality product. Thediagram proposed in TD-13 (showing a short access line and a long access line) was rejected butit was decided to insert Figure 1 of TBR 21. It was agreed that the document should contain aclause on out-of-band signals and immunity to them but it was decided not to deal with unbalance.

In clause 6 the dc feed resistance was changed to 500 ohms with a note to be added referring toinstallation resistance. France wanted a high maximum current to be set to ensure thatregulation in terminals was working. In reply it was pointed out that the guide was intended forfuture networks. After much discussion, France settled for a recommended range of currentsfrom 25 to 40 mA with 55mA maximum and 18mA minimum.

France Telecom in TD-51 noted the position it would be taking at the next ACTE meeting tooppose proposals (removing the 60mA current limit) creating differences between I-CTR 37 andthe previous consensus in ATA.

The requirement for polarity reversal was made optional and clause 7.x specifying a connectsignal was deleted. J. Long (DTI) requested a bandwidth wider than 300-3400 Hz so that theequipment was suitable for high speed modems.

There were other minor editorial changes. The rapporteur will put an updated version of theguide on the server in February. The updated version of Draft document EG 201 188 (analogpresented NTP) will be submitted for approval by consensus during February 1999, with theMarch ATA meeting available for discussion of any comments. It will be published first as TR101 188 after receiving ATA approval.

R&TTE DIRECTIVE FOR TE STANDARDS

J. Long (DTI) was unable to offer any input on prTR 101 389-1 on the general aspects of theimplications of the R&TTE Directive essential requirements on standards in the TE area.

Document prTR 101 389-2 (R. Trevland, Norway P&T) on the prevention of harm to the networkwas introduced. This document will now need significant change as the meeting was informed ofthe new decision that “the prevention of harm to the network” was no longer an essentialrequirement under Article 3.1(c) of the Directive but has been moved to Article 3.3 (with a slightrewording adding misuse of resources). As it is in Article 3.3, the technical requirements will bedecided in TCAM. Ad-hoc group D will have an input to TCAM but the Commission will makethe decisions.

A French delegate pointed out that the TBRs will be required until replaced by the requirementsof the new Directive. J. Long (DTI) confirmed that the CTRs continue for the transitional period.

Document prTR 101 389-3 (P. O’Keefe, Telecom Eirann) deals with the obligation of networkoperators to publish details of their access interfaces. The main proposal is for ETSI to publish



templates for interface publication that calls for information up to layer 4. Ad hoc group B islooking for volunteers to write templates.

It was pointed out that one of the jobs for TCAM would be to decide on the equivalence ofinterfaces so as to prevent unnecessary proliferation of specifications. It was hoped that theexistence of templates might assist in this work. Equipment classes for connection to the variousinterfaces also need to be considered.

P. Dofnas (Ericsson Business Networks) did not yet have available a draft of prTR 101 389-4dealing with service aspects in terminals.

prTR 101 389-5 (M. Harris, OFTEL) deals with new technologies. Attention was drawn to adocument in TD-11 (not distributed), which, in clause 3, set out a decision process which was usedto decide whether a requirement is essential under Article 3.3.

1. What is the problem?2. Does the problem in principle fall under the scope of Art. 3.3? If yes,3. Do we have substantial, documented and well-grounded reasons that there is a problem? If

yes,4. Is it necessary to make a decision or may the problem be solved through market forces? If

necessary,5. Do we have substantial, documented and well-grounded reasons for this assessment? If yes,6. Should the problem be solved through other Community provisions, e.g. horizontal or

telecommunications specific? If not,7. Does the objective that we want to protect through the decision prevail over the possible

negative influence on the market? If yes,8. A decision shall be taken that an equipment needs to comply with particular essential

requirements according to art. 3.3, 6.2 and 14.

J. Long (DTI) opined that the decision process contains worrying issues. He did not enlarge onhis worries.

It was noted that when the Directive comes into force, new harmonized standards will replaceCTRs for regulatory purposes but that TBRs will remain in existence and could be used on avoluntary basis. Their use may give a presumption of conformity.

The point was made that it is important not to lose standards that contain harmonized testmethods. Further work on the 389 series of documents will be by e-mail. It was stated that allfive parts of TR 101 389 will be placed on the server shortly and will be reviewed in detail bycorrespondence, with the intention of obtaining approval jointly with DTA in March 1999.

ATAE (ENHANCED SIGNALING SYSTEMS)

ATAe focuses on enhanced features, e.g., CLASS.

Liaison SPS-ATA-1 (Working Group SPS5) gives a contact name, T. Egeland (Telenor AS), forwork on analog display services. TD-61 (U. Jansson, ATAe Chair, Ericsson Telecom)a liaison statement to SPS5 (and copied to NA2), acknowledges the contact for display services.

Liaison SPS-ATA-2 (Working Group SPS5) notes that a new work item has been defined toenhance DSS1 protocol for IN/ISDN interworking.

Liaison NA-ATA-1 (partial) (B. Reed, BT) is a reply from NA2 rejecting an earlier proposal for ajoint meeting on PSTN display services. A liaison to NA2 (TD-68) expressed regret that theywere unable to attend a joint meeting.

Liaison NA-ATA-1 (partial) (P. Veentra, KPN) from NA2 gives details of a new work item toenhance ISDN and PSTN with a generic display/prompting transfer capability for services suchas advice of charge. TD-60 (P. Lottin, France Telecom, ETSI EP ATA-e) requests comments fromNA2 on the ATAe draft TR 102 088 describing the advice of charge during (AOC-D) the call andadvice of charge at the end (AOC-E) of the call services. A second liaison to SPS5, be copied to



NA2, on generic display/prompting transfer capability was also approved. This liaison, TD-62,was not submitted to the meeting.

Liaison from NA2 NA-ATA-1 (partial) (B. Reed, BT) points out a number of difficulties inimplementing Anonymous Call Rejection (ACR) as required by the EC Directive on Privacy(3626/97). An ACR service introduced currently would reject far more calls than only ones fromparties who did not wish to identify themselves. The liaison informs ATA that a new work itemhas been started on enhancing the existing CLI (calling line identity) supplementary services tolimit the number of situations in which calls would be rejected due to ACR reasons.

The work on ETS 300 659-1 (PSTN; Subscriber line protocol over the local loop for display (andrelated) services; Part 1: On hook data transmission) has now been transferred from SPS toATA. TD-15 (T. Salaun, TRT-Lucent) proposes the addition of a note referring to pair gainsystems. TD-24 (France Telecom) proposes an amendment to Annex A adding a networkoperator parameter. TD-40 (U. Jansson, Ericsson) proposes an extension of the call typeparameter. TD-38 (U. Jansson, Ericsson) and TD-47 (U. Jansson, Ericsson) deal with NetworkProvider Identification (NPID). TD-48 (U. Jansson, Ericsson) proposes how the protocol shouldsupport advertising. TD-37 (G. Engström, Telia AB) proposes new message parameters in ETS300 659-1 and ETS 300 778-1 (PSTN; Protocol over the local loop for display and relatedservices; Terminal Equipment requirements; Part 2: On-line data transmission) to deal with theShort Message Service (SMS). TD-39-r01 (G. Engström, Telia AB) proposes changes in the samedocuments to deal with the message origin parameter.

TD-41 provides an initial input to the report on Advice of Charge display services (TR 102 088).TD-33 (W. Schellhaas, Deutsche Telekom) proposes a number of amendments to deal particularlywith AOC-E; TD-34 incorporates these amendments into the draft. TD-25 (France Telecom) notesthe changes necessary to deal with charge units as well as currency. TD-42 (F. Provenzale,CSELT) proposes a number of mainly editorial changes. A second revision of TD-41r2 (TR 102088) was approved at the plenary for publication. TD-60 (P. Lottin, France Telecom, ETSI EPATA-e) is a liaison to NA2 asking for their comment.

TD-16 (TR 102 089) is Selection of specific terminal or terminal function. TD-17 (F. Provenzale,CSELT) suggests the use of the called line identity parameter for multiple subscriber numberprotocol implementation; TD-52 (W. Schellhaas, Deutsche Telekom) gives a caveat on its use. TD-18 (P. Plaitano, Ericsson) proposes sub-addressing protocol implementation to enable theselection of one or more terminals on the same access. Arising from a TC HF (human factors)suggestion that the title “basic service” might be misleading, TD-28 (Z. Gütter, TESTCOM)suggests changing the name to “preferred service” together with a number of other minorimprovements. TD-31 (W. Schellhaas, Deutsche Telekom) suggests an alternative name of“Calling type” for “basic service,” in the end, “Connection Type 2” was chosen. TD-16-r2, reviseddraft of TR102 089, was approved at the plenary for publication. F. Howett (Nortel) pointed outthat TD-16 introduces features that Bellcore systems do not support and asked whether theywere being kept informed. He was told that this would be done. TD-59 is a liaison to NA2 (ccSPS1/SPS5) asking for their comment and informing them of the publication of TR 102 089; TD-63 (F. Provenzale, ATAe) is a liaison to TC/HF requesting comments on the selection of a specificterminal or terminal function.

TD-67 is a first draft of TR 101 292, which sets out the proposed enhancements of the existingstandards for protocols over the local loop and embodies the services from TR 102 088 and TR102 089.

Walter J. G. Mellors, WM Services



ETSI ATA MEETING ROSTER, DECEMBER 7 - 11, 1998, CAPE TOWN, SOUTH AFRICA

Nuno Encarnação, OFCOM/BAKOM ATA ChairHost: Telekom SA

Austria Austrian Fed. Ministry Walter MarxtTelekom Austria AG Gustav MagdoinTelekom Austria AG Josef Halbertschlager

Czech Rep. Testcom Prague Zdenek GutterDenmark Teledanmark A/S Finn RavensholtFinland Finnet Group Kaj Andersson

Finnish Telecommunication Antero SaarinenFrance Alcatel Jean-Marie Notter

ETSI Secretariat Claire d’EsclercsFrance Telecom Jerome DescosFrance Telecom Philippe LottinFrance Telecom Jean MontenotTRT Lucent Technologies Thierry Salaun

Germany Deutsche Telekom Werner SchellhaasDeutsche Telekom Stefan WüstREG TP Christina GandyraSiemens AG Josef-Peter ZuckTDK Electronics Europe Peter de Wit

Ireland Telecom Eireann Pat O’KeeffeItaly CSELT Rodolfo Ceruti

CSELT Fabrizio ProvenzaleTelecom Italia Sonia LalliTelecom Italia Paolo Vitiello

Netherlands KPN Bert van LeerdamKPN Jo Salemans

Norway Norwegian Post and Telecom Eskil ElnessNorwegian Post and Telecom Mikael KarlssonNorwegian Post and Telecom Jan NergardNorwegian Post and Telecom Rasmus TrevlandTelenor R&D Finn Stafsnes

S. Africa Telkom SA Ltd Kimon AposporisTelkom SA Ltd Pieter ErasmusSiemens Telecommunications Paul Alberts

Spain Alcatel Espana SA Jose Maria GarciaSweden Ericsson Business Networks Per Döfnäs

Ericsson Business Networks Berndt MartensonEricsson LM Urban JanssonTelia AB Göran EngströmTelia AB Christer Karlsson

Switzerland OFCOM/BAKOM Nuno EncarnaçãoUK British Telecommunications Chris Harley

DTI John LongETSI User Group rep. Walter MellorsNortel Fred HowettOFTEL Martin Harris



REPORT OF ETSI TIPHON MEETING #11JANUARY 11 - 15, 1999, SOPHIA ANTIPOLIS, FRANCE

H. Schink (Siemens) is the chair of TIPHON (Telecommunications and Internet ProtocolHarmonization Over Networks). This meeting was a paperless meeting. 11TD-02 is the list of135 documents submitted. The meeting agenda is in 11TD-01r1. 11TD-05 is the report fromTIPHON meeting #10, October 26 - 30, 1998 in Tel Aviv (see also report in CSR 10.1).

The chair reminded all participants of ETSI’s IPR policy and invited the attendees to announceIPRs of relevance to TIPHON. No announcements were received.

The following were the TIPHON 11 goals at the opening of the meeting (see 11TD-01r1) :

GoalsDTR-01002 to be approved by TIPHONDTS-02002 to be approved by TIPHONDTR-02004 stable draftDTS-03003 first textDTS-03005 first textDTS-03007 first textDTS-04002 stable draftDTR-05002 first textDTR-06002 first textDTS-06004 to be approved by WG6Approval of revised ToR

H. Schink (Chair) informed the attendees that S. Petrack (WG3 Chairman) and B. Castle(TIPHON vice Chairman & WG5 Chairman) resigned from their chairmanship positions. A newjob position in 3Com prevents B. Castle from carrying on his chairmanship commitment; howeverhe will remain active in TIPHON and in the marketing activities of TIPHON. S. Petrack leftVocalTec to join Metatel as Vice President and Chief Technical Officer.

The election of the new WG chairs will be held at the next meeting; candidates are invited tosubmit their nomination to the chair.

MARKETING ACTIVITIES

The following Press Releases were approved:

• Joint release by ETSI and the IMTC, January 8, 1999: “ETSI and IMTC announcecooperation agreement” (see http://www.etsi.org/press/press.htm or 11TD-129)

• Joint release by ETSI and the IMTC, January 15, 1999: “First TIPHON/IMTCInteroperability Testing Event” (see http://www.etsi.org/press/press.htm or 11TD-131)

• ETSI press release, January 14, 1999: “ETSI Project TIPHON launches preparations forTIPIA” (see http://www.etsi.org/press/press.htm or 11TD-130). The main objective of TIPIAwill be to set up a TIPHON-compliant global IP-telephony service by the year 2000. The firstTIPIA meeting will be held in Vienna, Austria February 8-10, 1999.

H. Schink will give a presentation at the PBX2000 conference in Washington in February. B.Castle and H. Schink will give a joint presentation at the INET99 (the annual meeting of theInternet Society), June 22-25, 1999 in San Jose, CA.

IETF

During the meeting of the IETF in Orlando, Dec. 7-11, 1998, the TIPHON chair, assisted by theSTF and S. Bradner (AD Transport and VP for standards from ISOC), discussed cooperationbetween IETF and TIPHON. The work of TIPHON is well known in the IETF due to severalpresentations and Internet Drafts which were submitted by the STF (which were highly



appreciated). IETF and ISOC support further enhancing this cooperation and bringing it to amore formal state. Two main actions were agreed:

• The TIPHON chair will participate in the announce list maintained by IETF where all newwork items of the IETF are announced prior to their approval by the IESG. TIPHON will alsosend its new work items to this list to inform IETF about its new work.

• TIPHON and IETF will develop a document describing the detailed areas and procedures ofcooperation between the two bodies. This will be based on a similar document between ITU-Tand IETF (RFC 2436). The STF will prepare a draft for the next TIPHON PMC meeting onFebruary 28.

ICANN

B. Cosgrave, ETSI Deputy Director General, presented ETSI’s Internet policy regarding theInternet Corporation for Assigned Names and Numbers (ICANN). It is considered essential forETSI to gain representation in the various decision-making bodies underlying ICANN. On behalfof E. Dyson, interim chairman of the Board, B. Cosgrave invited ETSI members to nominateparticipants in a number of bodies to support the ICANN decision-making process. 11TD-115,Collective Letter 1914 on ETSI Internet Policy, contains B. Cosgrave’s presentation.

IMTC

The IMTC, in its Fall Forum meeting in Miami, Florida on November 9-11, 1999, achievedworking group approval of two extensions to its Voice over IP Forum InteroperabilityAgreements. Those extensions include a basic security profile and conventions for useridentification that permit easy integration with existing RADIUS-based AAA services. Inaddition, the IMTC announced an expansion of its existing interoperability testing activities. Theexpansion is intended to address the needs of end users by giving them assurance that purchasedequipment will successfully interoperate.

MSAF

MSAF’s (Multimedia Services Affiliate Forum) Internet Services Study Group (ISSG) and ETSI’sTIPHON project formed a joint cooperation agreement in November, 1998. Since then, thecooperation agreement has extended to cover the Ten Telecom TIPHON, TTT-Net project. Theterms of the joint cooperation include:

• To jointly develop an Implementation Agreement for IP-Telephony applications based uponTIPHON technical specifications.

• To jointly prepare General Service level Agreements based upon TIPHON technicalrequirements to expedite interoperability for Service Providers and Equipment Vendors.

• MSAF and TTT-Net will have a co-located meeting in February 1999.

• MSAF members will participate in TTT-Net trials.

TTT-Net is best positioned to demonstrate IP-Telephony interoperability between members andthe industry based on TIPHON-developed technical specifications. MSAF’s core competency is toenable service providers to adopt and implement these technical specifications via itsmembership base. By combining these strengths, a complimentary, integrated approach can beachieved for the global deployment of IP telephony services.

11TD-084 (NSAF/TTT-Net) is a liaison to TIPHON WG5 on QoS Requirements for Inter-domainScenarios, noting the desire to work with TIPHON to resolve QoS issues and expedite thedeployment of TIPIA.

ITU WORKSHOP ON IP RELATED MATTERS AND GII

Last September, the ITU-T TSAG appointed SG13 as the lead study group for IP related issues.As a result, ITU-T TSAG and IETF have developed a cooperation agreement.



A workshop was held in Orlando December 14-18, following the IETF meeting. TIPHON,represented by its chair and the chair of WG2, gave a key-note presentation in the openingplenary and a presentation in the session on quality of service. TIPHON participated in thepanel on multimedia services. Experts from TIPHON gave additional presentations on differenttopics.

ITU-T has significantly increased its activity in the IP related area. The work in the followingITU-T areas is of relevance for TIPHON:• SG2 (Numbering, Naming, Service Description)• SG11 (Signaling Extensions)• SG12 (End-To-End QoS)• SG13 (Architecture, Network Performance)• SG15 (Transport Aspects of Gateways)• SG16 (Multimedia, H.3xx series)

The following actions have been requested:

• TIPHON to input its architecture to SG16 and SG13• TIPHON to input its requirements for the reference point N to SG16• SG16, in cooperation with IETF, to deliver a protocol for reference point N according to

TIPHON requirements in a timely manner• TIPHON to continue its input to SG2 on a global resource• Ongoing exchange of results and requirements with the other SGs

OCG AD HOC GROUP ON VHE

The ETSI Organization Coordination Group (OCG) set up an ad hoc group (OCG-VHE) to adviseon how to deal with VHE (Virtual Home Environment) in ETSI, given that there are severalactivities related to VHE in ETSI. H. Schink and L. Klostermann represented TIPHON. L.Klostermann (Ericsson) gave a short presentation on the work ongoing in TIPHON on mobilityand VHE, and on the discussions on wireless access.

There were further presentations of

• SMG-UMTS• NA6 UMTS• 3GPP (not in ETSI, but strongly related to ETSI)

Given these activities, the task of the OCG-VHE is to advise the OCG on how to organize thiswork within ETSI, and how to communicate with the outside world.

Although EP UMTS (ETSI project Universal Mobile Telecommunications System) is about to bestarted, and the terms of reference have not yet been agreed, the preliminary conclusion of themeeting was that EP UMTS is probably the best positioned to coordinate the VHE activitieswithin ETSI for future phases of UMTS, taking into account and basing itself on the work of3GPP, defining UMTS phase 1.

TIPHON has just started to look into VHE. The ad hoc group on VHE will study how muchharmonization with the TIPHON work is appropriate.

Within the ITU, the activity of GII (Global Information Infrastructure) under the leadership ofSG13 has the responsibility to coordinate all relevant activities within ITU. SG11 (IMT2000)was identified as the most active body working on VHE, producing standards.

OCG-VHE will meet again to finalize its work.

BOF REPORTS

11TD-127 is the BOF session on Wireless aspects of IP Telephony meeting report. D. Gurle(VocalTec) chaired this BOF Session. The group agree that the wireless and mobilitymanagement issues should be addressed within TIPHON. The Terms of Reference and



appropriate work items need to be refined. The majority were in favor of the creation of aworking group to address the wireless and mobility management issues.

11TD-132 is the BOF session on Implementation of a Global IP Telephony Service meeting report.H. Schink chaired this BOF Session. The group agreed to support the establishment of aseparate group, TIPIA (TIPHON IP Telephony Implementation Agreement) to provide input toTIPHON in the area of requirements.

WORKING GROUP 1, REQUIREMENTS

The WG1 Chair is L. Klostermann (Ericsson, [email protected]).

INBAND INFORMATION IN DTS-03005

11TD-69 and 11TD-70 (J. Holm, Ericsson Telecom) propose solutions for the problems ofreceiving and responding manually to inband tones, announcements and alerting signals for SCN(Switched Circuit Network) initiated and IP-network initiated calls in DTS-03005 (Signaling forbasic calls between an H.323 terminal and an SCN terminal). During the TIPHON plenary, itwas requested to check whether the requirements were already covered. It was discovered thatthis is the case (Requirements 11 and 12).

SECURITY IN DTR-01002

11TD-73 (S. Moore, Siemens) identifies the issue of the optionality of security profiles, andidentifies inconsistencies between current WG3 and WG1 documents. It was agreed at theprevious meeting that security requirements only apply if the business role requires it, whichmeans that either:

1) Zero security profile is restored, or2) Use of security profiles is optional, depending on deployment.

It was agreed to modify the text in DTR-01002 according to the second option. DTR-01002V2.2.1 thus needs to be approved.

11TD-61 (C. Perry, Motorola) raises the issue of firewalls. It was identified that security isrelated to business domains; business roles define the border between them. After somediscussion it was agreed to change requirement 1 from “The use of firewalls shall be possible in aTIPHON compliant system” to “Between business roles within the IP domain, the use of a borderelement within a TIPHON compliant environment shall be possible. The functions of this borderelement can among other things be security, address translation and flow control.”

11TD-58 (H. Daanen, HP Labs) requests that WG1 specify the requirements for support ofFirewalls in TIPHON systems.

VALUE ADDED SERVICES

During last meeting, it was agreed not to accept H.450 supplementary services en-block for themoment, but to focus on trying to achieve service differentiation. The general requirement for thesupport of the Virtual Home Environment (VHE) has been agreed. 11TD-45, 11TD-46 and11TD-47 (all from J. Korpi, M. Korpi, and K. Klaghofer, Siemens) provide the PICS (ProtocolImplementation Conformance Statement) from SG16 on H.450.1 (generic functional protocol),H.450.2 (call transfer), and H.450.3 (call diversion), respectively. 11TD-79 (L. Klostermann,Ericsson) provides a high level description on how service differentiation and VHE could berealized. 11TD-79 was not discussed in great detail, since it contained no further detailedrequirements on supported services, but is more of an architecture to provide VHE. There was ageneral discussion on services, the relation between standardized supplementary services, andthe kind of service framework proposed by 11TD-79. It was clarified that in case of 11TD-79there is more freedom to provide services based on flexible format data, e.g., user profiles forpersonal secretary service. It was also clarified that many services can be provided bymanipulating H.323 signaling and bearers, but that some extra signaling support might beneeded for some service features.



More detailed contributions on VHE, value added services and on H.450 and their interworkingwith other types of supplementary services are invited. Once there is agreement on the degreeTIPHON shall support supplementary services, the PICS provided by SG16 will be used toexpress the degree of H.450 conformance.

IDENTITIES

11TD-71 (J. Talbot, Nortel Networks) distinguishes more clearly between user identity, locationidentity and presentation identity, and provides examples on how these are used in case of MCID(multi-line caller identification), emergency call and certain call center calls. Presently CallingLine Identity is used, which can be used to indicate both the user identity or a location identity,and maps on SCN signaling. After some discussion it was agreed that the distinction betweenthe three identities is indeed useful, and are also being standardized in the SCN (ISUP2000). Theissue of providing the identities in the IP domain, and mapping them towards especially existing(pre-ISUP2000) SCN signaling for TIPHON scenarios 1 and 2 are issues which still remain. Thecontributor agreed to provide some more clarification, proposed text for DTR-01003, and toclarify the exact meaning of the locationID in the Voice over IP context.

11TD-77 (J. Horrocks, DTI), Access to emergency services, addresses the same issue, and alsopoints to SPS1 ISUP2000 activities for guidance.

The following plan was agreed:1. Focus on service requirements (MCID, CLIP, emergency),2. Identify required identities,3. Create text on requirements on the interface between IP-SCN.

On the interface, the issue was raised that in certain cases (CTM, UPT) the IN dialing number isa relevant parameter. Contributions on the interface issues were requested.

CLEARING HOUSE

11TD-55 (H. Wermescher, Telekom Austria) proposes a clearing house architecture. It proposestwo new requirements, but makes a distinction between Back-end-services within a domain, andBES (Back End Services) in the Clearing House domain. However, it proposes the use of DTS-03004 (Inter-domain pricing, authorization and usage exchange) within a domain and possiblybetween domains, while it has been discussed only as used between domains.

11TD-60, Positioning the BEST Protocol with H.225 Annex G (Communications betweenadministrative domains) and RAS (D. Clowes, OzEmail Interline), was presented, but no explicitrequirements were proposed for inclusion in DTR-01003 (Service requirements forinteroperability and technical aspects of accounting and security, Phase 3) for the moment. Itwas clarified that the requirements related to DTS-03004 apply to the protocol proposed in 11TD-57 and 11TD-59 (Back End Service Transaction [BEST] protocol description and technicalrationale, respectively, D. Clowes, OzEmail Interline) and 11TD-60.

MOBILITY

11TD-75 (W. Legrand, Matra Nortel Communications) notes the difference between portable(service location can change) and mobile (service location changes while service is provided) andnotes the requirement of wireless access for private networks. Wireless access for privatenetworks was considered to be within the TIPHON work scope already, thus no addition wasrequired to DTR-01003.

11TD-78 (L. Klostermann, Ericsson) is a proposal for the scope of TIPHON activities related towireless access. At the TIPHON 9 meeting, the proposal (09TD-45) was made to initiate aworking group on wireless access. 11TD-78 elaborates on mobility and wireless aspects, andgives a proposal on what work can be done in TIPHON, and what work can be done in otherbodies.



MSAF

11TD-48r1 (A. Dilber, AT&T, MSAF) is a statement of cooperation between MSAF-ISSG andETSI-TIPHON. 11TD-49 (A. Dilber, AT&T, MSAF) outlines a table of contents for the TIPHONTechnical Service Description as one of the deliverables from the Statement of Cooperation.

H.450

There was reluctance to include the H.450 supplementary signaling in the TIPHON requirementswithout further study. One of the main issues is harmonizing the IETF’s MEGACO (MEdiaGAteway COntrol) work chaired by T. Taylor (Nortel) with the ITU and TIPHON work on VoIP.In MEGACO, downloading scripts to Gateways, including residential gateways, is supported,which is a different approach from the one in H.450.

The conclusion was that WG6 can work on PICS for interworking testing, and could possiblyinclude this material in an annex to the working group’s test deliverable.

RETAILER CONCEPT

11TD-62 (F. Panken and P. Sijben, Lucent) introduces the retailer concept to TIPHON. Theretailer concept describes a role played by service providers. The activities of the retailerencompass the access (registration), user profile management and user mobility management.The retailer (e.g., hotmail.com) would be the primary and only point of access for a particularuser for a particular set of services. Within the current TIPHON architecture, most of theretailer functions would be encompassed within the current definition of the gatekeeper.

WORKING GROUP 2, ARCHITECTURE

WG2 Chair is J. Vandenameele ([email protected]). The agenda is in11TD-07.

DTS-02002

11TD-72 (S. Moore, Siemens) provides technical as well as editorial comments to DTS-02002V0.3.1, Network Architecture and Reference Configurations; Phase II: Scenario 1 + Scenario 2.The editor will include all editorial changes. Detailed discussions of the technical commentsresulted in a number of changes as well.

Sections 7 to 10 (those wrongly numbered to 6 to 9 in the current draft) were reviewed asplanned. The meeting agreed not to apply any changes except for those already agreed from11TD-72.

J. Bouwen (Alcatel Bell), the ad hoc interim editor, undertook the final drafting with support of S.Moore (Siemens). The result is DTS-02002 V 0.4.1., which was approved by WG2.

INITIAL WORK ON DTS-02003

J. Bouwen (Alcatel Bell) agreed to act as the interim editor during this meeting.

11TD-076 (J. Bouwen, B. Van Doorselaer, Alcatel) provides a proposal for the architecture forTIPHON Scenario 3. It introduces the concept of Call Agents as well as the new reference pointsM and B*. M covers the reflexive relationships for the Call Agents at the edges of the IP basednetwork, while B* has to cover this relationship for the Media Gateways.

The proposal was discussed in detail. No consensus was achieved to include this material even asa straw proposal in the first draft of DTS-02003 (Reference configuration, network architectureand interfaces, Phase 3). The proposal for the new M reference point turned out to be verycontroversial. Worries were expressed on the upward compatibility between DTS-02002 and thenew deliverable that has to cover additional scenarios. As a result no progress was made on thenew deliverable.



The other contributions could not be dealt with because the architecture has to be clarified first,but also because lack of time. It was agreed to continue the discussion on the TIPHON_WG2 e-mail list in order to be prepared for a decision at the next TIPHON WG2 meeting.

In TIPHON 10, the J-reference point was left for further study. 11TD-056, Requirements for theJ interface (the interface between the Signaling Gateway and the Media controller) (C. Groves,Ericsson), was also not discussed. 1TD-056 builds on the requirements in 10TD-140 andintroduces additional requirements for this reference point.

Contributions are solicited on:• Changes to Reference architecture to cover Scenarios 3 and 4• Requirements for reference point J• Need for and nature of additional reference points / interfaces as proposed in 11TD-076

especially D, M and B*• Are Network Access Servers (NAS) sufficiently covered by the reference architecture• Problem of ringing tones• Attention for back-end services (list of possible back-end services), reference points to BES,

existing protocols

DTR-02004

The responsibility for DTR-02004 has been moved to the new Working Group on Mobility andWireless Access. Hence no action was undertaken by Working Group 2.

REQUIREMENTS FOR A PROTOCOL AT REFERENCE POINT N (BETWEEN MEDIA GATEWAY AND THE MEDIAGATEWAY CONTROLLER)

When discussing DTS-02002 it was agreed to start a new specific deliverable dealing with thedetailed requirements for reference point N. It was also agreed to use the requirements in thecurrent draft as a starting point. Hence WG2 proposes to start a new Work Item DTS-02005.

Two drafting sessions chaired by the interim editor J. Bouwen (Alcatel) created the first draft.The resulting draft is available as DTS-02005 Version 0.1.1.

LIAISON ACTIVITIES

11TD-118 is the report of the joint meeting between WG2 and SPS5. The purpose of the meetingwas to clarify the required architectures to support network access servers, and to identify ifthere should be a DSS1-based protocol providing a short term protocol solution to support theconcept of network access servers. WG2 will take into account the conclusions of the meetingwhen drafting DTS-02003.

11TD-025 from SPS1 requests a liaison with TIPHON regarding activities related to thetransport of ISUP information over IP. This relates to a liaison sent by WG3 to the ITU-T SG16Q13/14 experts meeting after the last TIPHON meeting. It was agreed that a liaison should besent to SPS1 attaching 10TD-138r1 (Liaison to SG16 Q13&14 on SS7 Information TransportOver Intermediate H.323 Network) from the TIPHON 10 meeting in October 1998.

11TD-026 addresses the management of CLI, and is in response to a liaison from TIPHON. SPS1will keep TIPHON informed on the status of this work.

It was agreed to submit a revised draft using the new deliverable on requirements for a protocolon the N reference point. R. Scholl (STF) will prepare the proposal and get final comments usingthe TIPHON_WG2 e-mail list. Since DTS-03005 has been approved, no changes to the technicalcontent will be accepted.

The STF will send a liaison to the next meeting of Q13/16 and Q14/16 in Monterey, CA, February15-19, 1999, including both DTS-02002 and DTS-02005.

WORKING GROUP 3, CALL CONTROL

D. Gurle (VocalTec) acted as WG3 chairman.



WG level agreement of DTS-03005 stable text was achieved. DTS-03005, Signaling for basiccalls between an H.323 Terminal and a Terminal in a Switched-Circuit Network (SCN), will beuploaded to the docbox server by the editor.

DTS-03005

11TD-63, Comments on DTS-03002 v1.6.4 (K. Mossel, Lucent), requests that AutomaticGatekeeper Discovery, Registration KeepAlive, and FastConnect fallback be incorporated andspecified in DTS-03005. A question was raised regarding the relevance if IETF IPTEL GWLOC(gateway location protocol, see http://www.ietf.org/internet-drafts/draft-ietf-iptel-gwloc-framework-01.txt) relates to the GK discovery; the author said no. A request to limit ThreeLetter Acronyms (TLAs) in TIPHON documents was noted.

11TD-89 (P. Sibjen, Lucent) provides additional text for DTS-03005 based on 11TD-63. Section4.1 was accepted with changes. Sections 4.2 and 4.3 were not accepted. Accepted text will beadded to the next revision.

11TD-69, In-Band Information During Call-Setup and Release (J. Holm, Ericsson), proposes amodification to DTS-03005. The use of Fast connect procedures will make it possible for a callinguser to listen to announcements. A question was asked whether this relates to charging; theanswer was no, it is more general. It was decided to add the text to DTS-03005 (current version),and to send a liaison to SG16 WP2 Q13 asking for comments (see 11TD-100, below).

11TD-70, Alerting tone towards calling party (ring back signal) connected to SCN (J. Holm,Ericsson), proposes two solutions: One, gateway starts the tone when alerting signal is receivedfrom the IP network; Two, anyone may start the tone. A comment was made that the secondsolution may be problematic as the signal message is not supported in the European DSS1protocol. It was clarified that this is not relevant, as this scenario never happen towards theSCN. It was agreed that the contribution will be enhanced and re-submitted at the next TIPHONmeeting.

11TD-100 is a liaison to Q13/16 based on 11TD-69 and 11TD-70, asking for their view on thesolutions.

11TD-73, Implications of the Removal of the “null” Security Profile (S. Moore, Siemens), requestsa change in section 5.2 of 03005; it was accepted with minor editorial changes.

GATEWAY DECOMPOSITION

11TD-29r1, MDCP - Media Device Control Protocol (P. Sijben, Lucent), defines a protocol for usebetween media devices (commonly known as media gateways) and their controllers. A questionwas asked if the hard-delay limit that resulted in not using DNS in MDCP is required for allsystems. The author responded that in many systems, such as IVR, where real-time control isrequired, DNS lookup just takes too long. It was mentioned that using DNS makes buildingscalable systems simpler.

Regarding scripting, the question was asked: Is MDCP standardizing the scripting language andmodel or just creating an architecture? The author mentioned that this is a work-in-progress.

Regarding the element responsible for Resource Allocation: the author clarified that MG itselfshould control its resources, but the question whether where it is done is related to the controlmodel is a open issue. It was mentioned that Element-Management is out of scope for MDCP.

The H.245 channel is terminated either in the MGC or the MG.

QoS gets provisioned as follows: EndPoints have properties describing QoS capabilities andcommunicate this to the MGC. RSVP support was presented, and other QoS methods arepossible.

Binary encoding may imply light-weight implementations, but this is an open issue that need tobe studied, in terms of performance (both size and speed).



11TD-67, MGCP (H. Liu, Bellcore), describes the media gateway control protocol (MGCP)resulting from the convergence of SGCP and IPDC (same as APC-1461). The requirements andarchitecture are not yet final, hence it is difficult to converge or select one protocol.

11TD-43, MDCP Call Flows (L. Spergel, Lucent), was not presented. It contains sample callflows for the MDCP protocol (11TD-29r1). 11TD-68, MGCP Call Flows (H. Liu, Bellcore), wasalso not presented. It provides examples of MGCP (Media Gateway Control Protocol) usage witha variety of call flows.

It was noted that WG2 is in the midst of adding the architecture behind both Gatewaydecomposition protocols, and that the effort should be placed there, and not in WG3. It wasdecided that WG2 and WG3 are the most appropriate groups to discuss and propose a completelist of requirements as an input to the appropriate standard bodies, such as IETF Working Groupand ITU-T Q14/16.

As an action item, it was agreed to review and comment on the requirements produced by theWorking Group in order to help the progress on the establishment of the media gatewaycontroller to media gateway controller protocol. This effort will be concentrated in WG2.

It was decided to revisit the selection of the protocol for reference point N at the appropriate time,after the WG2 architecture work is completed and at least one protocol is standardized in Q14/16and the Working Group.

FIREWALL ISSUES

11TD-33, Firewall Considerations (K. Sambor, Telekom Austria), considers both firewalls and“H.323 proxies” as important components for TIPHON compliant systems. A question was askedif this is a product availability issue or a protocol interoperability issue; it seems to be more of thelatter.

The chair called for contributions that specify any required changes to support TIPHONoperation via a NAT (Network Address Translation). The decision as to whether thespecifications of NATs/Firewalls are required in WG3 should be requirements driven, i.e., WG1,then WG2 should determine the requirements; only then should WG3 solve any presumed protocolissues.

INTER-DOMAIN AND INTER-GATEKEEPER

11TD-41, H.323 Annex G, Inter-Domain and Inter-Gatekeeper Communications (G. Kimchi,STF114), requests that WG3 verify that H.323 Annex G is on track as originally specified in theTIPHON requirements submitted to ITU-T SG16. No liaison to SG16 was deemed necessary.

WORKING GROUP 4, NAMING, ADDRESSING

The WG4 chair is L. Spergel (Lucent, [email protected]). The agenda is 11TD-09.

PROGRESS ON WG4 DELIVERABLES

11TD-50 (A. Dilber AT&T, R. Stastny, Telecom Austria) and 11TD-74 (J. Horrocks, DTI) bothrecommend changes to DTS-04002 (Naming, Numbering, and Addressing; Interworking fromE.164 Terminals (PSTN/ISDN/GSM) to IP Terminals, Phase 2). After reviewing thesecontributions, it was decided to restructure the deliverables. It was concluded that a smallnumber of the requirements contained in DTS-04002 are necessary, and that the remainder ofthe contents belong in a technical report. As a result, WG4 moved the requirements to approvedTIPHON document TS 101 324 as a revision.

The text in this new document is considered stable as no technical details were added. It isrecognized that the integration of the document may lead to some editorial changes before WGapproval at the March meeting.

There were no contributions towards DTS-04003 (specifications for scenarios 1,2,3 & 4).



New material will be added to DTS-04004 (E.164/IP address resolution) that was removed fromDTS-04002, as described below.

STUDY GROUP 2 STATUS

R. Stastny (Telecom Austria) reported on the SG2 and NA2 (develop service description forGlobal IP-Telephony Subscriber Number, GISN) activities. The GISN, in combination with the to-be-assigned country code, is an E.164 number. At the ITU SG2 plenary in November, 1998, itwas agreed that the country code assignee of the Global IP-Telephony resource, should a countrycode be assigned, would be the ITU on behalf of the international telecommunications sector. R.Stastny presented draft contributions 11TD-53 and 11TD-54 to support those actions that heplans to submit to NA2 and SG2. 11TD-53 is a draft of the Administration of the Global IPTelephony Resource; 11TD-54 is a draft of the Service Description for the Global IP TelephonySubscriber Number - Phase 1a. Comments and suggestions towards these contributions wererequested.

It was noted that the current plan is to request a global code for services. Therefore, it isnecessary to write a service description and to choose an appropriate name for the service.

11TD-27 is a liaison from NA2 NAR, requesting that WG4 continue to send drafts to NA2. It wasdecided to send a liaison to NA2 containing a copy of the revised draft of TS 101 324. Theproposed liaison is in 11TD-126.

OTHER HIGHLIGHTS

J. Ott (TELES) presented the IETF TENUM BOF status. It was noted that the planned workcould be applicable to the global code. WG4 will initiate contact with the leaders of this activityvia e-mail to ensure this applicability.

11TD-77, Access to Emergency Services (J. Horrocks, DTI), discusses general requirements forservices and numbering issues. It was noted that this issue had been identified in SG2; this issueis similar to legal interception, on which there will be a presentation at TIPHON 12. There wasagreement to add a section to TR-04004; J. Horrocks will serve as editor of this section. 11TD-71, Definition of Identification Services (J. Talbot, Nortel Networks), was noted as relevant aswell.

11TD-94, Issues and Objections Surrounding the Formation of ICANN (Internet Corporation forAssigned Numbers and Names) (M. Shore, Nokia), was presented for information. It was notedthat 11TD-115, Collective letter 1914 on ETSI Internet Policy (B. Cosgrave, ETSI), is alsorelevant. NA2 is already involved with ICANN. No action is necessary.

11TD-116 (R. Scholl, STF114) describes the draft IETF Handle System. The Handle System is aglobal naming service that is more scalable, extendable security-wise and more reliable thanwhat is currently available on the Internet. The internet draft is available athttp://search.ietf.org/internet-drafts/draft-sun-handle-system-01.txt. There may be a speaker onthis at the next meeting.

WORKING GROUP 5, QUALITY OF SERVICE

M. Buckley (Lucent, [email protected]) was the acting chair of WG5.

APPROVAL OF DELIVERABLE TR 101 329 V 1.3.0

11TD-099r1 is a new work item RTR-05006 for revisions to TR 101 329 v 1.2.5, General Aspectsof QoS. The changes agreed at TIPHON 10 and those detailed in 10TD-148 were incorporated;the revised deliverable was agreed.

FURTHER WORK ON TR 101 329

A new work item RTR-05007 as described in 11TD-111 for revisions to TR 101 329 v 1.3.0 wasagreed.



11TD-33 (K. Sambor, Telekom Austria) indicates possible QoS implications of the use of firewallsin TIPHON networks. It was agreed that this issue should be included in future work.

11TD-39 is a liaison from ITU-T SG12 on acceptable levels of delay in telephone connections.

11TD-42 (M. Perkins, AT&T) provides a draft of the new ITU-T Recommendation G.177,Transmission Planning for Voiceband Services over Hybrid Internet/PSTN Connections. G.177is based on the work of ANSI/T1A1 and ETSI/TIPHON TR 101 329.

11TD-52 (E. Diedrich, H. Gierlich, Deutsche Telekom) is a proposal for terminal testing inTIPHON systems. It was agreed that this work should form the basis of an Annex to be added toTR 101 329; H. Gierlich agreed to lead an editing group to produce draft text.

11TD-64 presents the result of simulation work of TIPHON systems carried out by DeutscheTelekom (T. Scheerbarth et al). Some of the packet loss test conditions (packet loss greater than10%) outlined in TR 101 329 resulted in severely degraded speech performance and as such wereprobably too severe. It was agreed to keep these test figures under review and to maintain theresults of 11TD-64 for future reference in the form of a permanent TIPHON document (see below).

11TD-85 (M. Perkins, AT&T) is a liaison from ITU-T SG12 Qs14 & 20 in response to the TIPHONliaison on packet loss, frame loss, and E-model impairment factors. Appendix I ofRecommendation G.113 contains information on the effects of packet loss and other impairmentsin IP networks. It was agreed that this useful information should be included in the permanentdocument.

11TD-86 (P. Coverdale, Nortel Networks) is a liaison from ITU-T SG12 Qs16 & 20 in response toa request from TIPHON to include wideband codecs into the E-model. This is not possible, as theapplicability of the E-model is restricted to voiceband systems.

11TD-87 (M. Perkins, AT&T) is a liaison from ITU-T SG12 Q1 notifying TIPHON (and others) ofthe creation of a new study question Q23, Transmission performance considerations for voicebandservices carried on networks that use Internet Protocol.

11TD-88 (J. Erreygers, Raychem) is a liaison from ITU-T SG12 Q16 notifying WG5 of theDetermination of Recommendation G.109, Definition of categories of speech transmission quality.It includes the text of this new recommendation.

11TD-108r1 (K. Adler, H. Klaus, J. Pomy, ETSI TC STQ) contains text for a new Annex to TR101 329 on transmission planning for TIPHON networks using E-Model calculations. This wasderived from 10TD-47. It was agreed to incorporate this text into the new draft with some minormodifications.

11TD-112 (R. Reynolds, BT) describes the effect of packet delay variation (PDV) on speechquality and proposes the need for more accurate modeling of these effects. A corporate networkwas analyzed using IP packet analysis and BT’s Perceptual Analysis/Measurement System(PAMS). The IP analysis showed that corporate network introduced all the PDV. This effect wastraced to a firewall. The PAMS analysis showed that large PDV significantly impacts speechquality. It was agreed to include this in the work program on TR 101 329.

DTR-05002

10TD-34, 10TD-44 and 10TD-64 from the TIPHON10 meeting were presented. They contain textof two IETF Internet drafts on “An expedited forwarding PHB” and “A framework for Use ofRSVP with DiffServ Networks” and a draft new ITU Recommendation I.380 IP “Internet protocoldata communication service - IP packet transfer and availability performance parameters.”These contributions were noted as relevant to DTR-05002.

11TD-51 (G. Manor, VocalTec) notes the IETF DiffServ work in RFCs 2474 and 2475. It wasagreed that DiffServ would provide a suitable mechanism of QoS signaling and that TIPHONshould make input to the IETF work in the near future. It was agreed to ask the STF to providea diagram showing the interrelationship between the QoS application under study in DTR-05002and the DiffServ mechanism.



11TD-84 (M. Blaschitz, TTTNet, A. Dilber, MSAF) is a liaison from MSAF/TTTNet on QoSRequirements for Inter-Domain scenarios. MSAF/TTTNet requests help from TIPHON WG5 toidentify engineering methods to provide inter-domain QoS, to provide guaranteed levels ofreliability and availability as defined in 11TD-24r1 (Draft Agreement on the Implementation of aTIPHON Compliant Global IP Telephony Service by 2000), and in specifying QoS budgets forinterdomain configurations. It was agreed that the scope of DTR-05002 covered these topics andthat WG5 would work closely with MSAF/TTTNet on these issues. A return liaison will be sent.

11TD-123 (M. Buckley, Lucent) summarizes the discussions at the two earlier BOF sessions onthe QoS end-to-end signaling application.

CREATION OF PERMANENT DOCUMENT CONTAINING TEST & SIMULATION RESULTS

It was agreed to seek approval at the TIPHON plenary to form the permanent document referredto above (from 11TD-64 and 11TD-85). Further relevant test and simulation results will beadded to this document from time to time. The document will also contain a list of work items andinformation relevant to their progress. 11TD-125 is draft text for this document; it was agreed.

WORKING GROUP 6, VERIFICATION

The WG6 Chair is M. Blaschitz (Infonova, [email protected]).

WORK ON DTS-06004 AND DTS-06005, TEST SPECIFICATIONS FOR TIPHON SYSTEMS PHASES 1 AND 2

11TD-091, Mailing LRQ an ARQ (G. Koerbler, STF), was presented for use in DTR-06004.H.225 defines both a location request (LRQ) and an admission request (ARQ) with littleexplanation on how and when to use each. This document tries to make the distinction clearer. Itwas accepted. 11TD-96 is a liaison to WG3 requesting additional clarification on duplication offunctionality.

DTS-06004 was reviewed in detail. A number of modifications were made. The revised DTS-06004 was approved by WG6 and is ready for Project approval at TIPHON 12.

11TD-031, High Level Test Plan for TIPHON Interoperability Tests Framework forConsideration of the SCN Side (M. Bien, Robert Bosch GmbH), will form the basis of DTS-06005,High Level Test Plan for TIPHON Interoperability Tests. DTS-06005 needs further work.

11TD-092 (G. Koerbler, STF) is e-mail correspondence concerning the test specification. Itproposes some additional test specifications for DTS-06004. The purpose of the test has beenconsidered valid but the base standards are considered to be not stable enough to approve themat this meeting. Consequently, the test is identified in DTS-06004 as for further study. Thedetailed test description, however, was put into the baseline document (DTS-06005).

DTS-06005 was developed as a first draft. The draft will serve as a baseline document for thenext revision of DTS-06004.

11TD-110 (J. Korpi, Siemens) proposes a new Work Item for PICS on Supplementary Services.Four ETSI Members have declared support for the Work Item: Siemens, PTA, Finnet, andMotorola. It was decided to coordinate the launch of the Work Item with WG1.

INTEROPERABILITY EVENTS

The Interop event was a great success and the chair thanked the ETSI secretariat and theSTF114 for the excellent organization. 32 companies participated with more than 100participants.

There was a general feeling that a fixed virtual testbed should be established for ad hoc testingbetween the interop events. The network should be based on a secure ISDN architecture with acentral, managed ISDN router through which different network configurations can bedynamically established.



It was decided that STF shall draft a proposal for implementers net architecture for the nextmeeting and contact the participants of the interop to evaluate their commitment.

The current plan is to have the next joint interop event around April in the North-East of theUSA. Another interop event is planned around June in Europe. The “Super-Op” is thenscheduled for fall, most likely in Hawaii.

ETSI TIPHON MEETING #11 ROSTER, JANUARY 11 - 15, 1999, SOPHIA ANTIPOLIS, FRANCE

Helmut Schink, Siemens TIPHON Chair

Australia Ericsson Ian RytinaOzEmail Interline Pty Ltd Douglas Clowes

Austria ETSI Secretariat Guenther KoerblerInfonova GmbH Herwart WermescherÖFEG Richard StastnyTelekom Austria AG Michael BlaschitzTelekom Austria AG Klaus SamborTelekom Austria AG Andreas Steiner

Belgium Alcatel Bell Jan BouwenAlcatel Bell Jozef VandenameeleDialogic Telecom Europe Steven Magnell

Canada Ericsson Michel HoudeEricsson Sylvain LabontéVienna Systems Corporation Donald Fraser

China Shanghai Bell Shen QunShanghai Bell Yan Xueqiang

Denmark Motorola A/S Chuck PerryTele Danmark A/S Niels KnudsenTele Danmark R&D Christian Thystrup

Finland Finnet Group Markku LaasonenFinnet Group Jari PohtioNokia Corporation Pekka PessiNokia Corporation Senthil SengodanNokia Corporation Melinda ShoreNokia Corporation Marko SuoknuutiSonera Limited Sampsamatti Tanner

France Alcatel France Jean-Pierre AlbinetAscend Communications Matt HoldregeCegetel Hubert CastellanCegetel Philippe LucasCompaq Bertrand FarabetCompaq Laurent GaumeCompaq Isabelle Valet-HarperENST Bretagne Hossam AfifiENST Bretagne Laurent ToutainETSI Secretariat Laurence FerrandETSI Secretariat Laurent VreckFrance Telecom Francis GaucheHewlett-Packard France John O’connellIBM France Yves CoutardIBM La Gaude Aline FichouMatra Nortel Communications Walter LegrandMatra Nortel Communications Didier RocchiaQUALCOMM Europe S.A.R.L. Peter JacksonSecrétariat d’ Etat Industrie Didier ChauveauTexas Instruments France Patrick Le Poultier

Germany Alcatel Sel AG Klemens AdlerAlcatel Sel AG Rainer MuenchBMWi Theo MetzgerBMWi Reinhard WalterDeutsche Telekom AG Hans Wilhelm GierlichDeutsche Telekom AG Harald Klaus



Deutsche Telekom AG Ingmar KlicheDeutsche Telekom AG Thomas ScheerbarthEricsson Eurolab Frank EcksteinETSI Secretariat Reinhard SchollEURESCOM GmbH Magnus KrampellITK Jan ElligerNEC Electronics (Europe) GmbH Stefan GesslerRobert Bosch GmbH Matthias BienRobert Bosch GmbH Joachim PomySiemens AG Keith DrageSiemens AG Juha KorpiSiemens AG Gerald MeyerSiemens AG Steve MooreSiemens AG Klaus NimphiusSiemens AG Mourad Oulid-AissaSiemens AG Helmut SchinkSiemens AG Sergio VerduciSiemens AG Wilhelm WimmreuterTELES AG Jöerg OttViag Interkom GmbH & Co Axel Stossno

Israel IAEI Aharon SegevVocalTec Ltd Elad Sion

Italy Italtel S.p.A. Patricia MacrinaNetherlands Ericsson Telecommunicatie BV Lucas Klostermann

Lucent Technologies Krijn MosselLucent Technologies EMEA B.V. Paul SijbenLucent Technologies EMEA B.V. Hans Van Der VeerPhilips Communication Systems Frank DerksTerena Eric MannieTerena Vincent Van Lier

Norway Ericsson AS Lill KristiansenPT Jarle KiilPT Arne Steimoeggen

Spain Telefonica de España S.A. Martinez Santiago InigoTelefonica de España S.A. Mario Munoz

Sweden Ericsson L.M. Jan HolmEricsson L.M. Ulf ThuneTelia AB Rune BomanTelia AB Ake HedevarnTelia AB Christer NilssonTelia AB Markus Stenman

Switzerland Ascom Raymond WuOFCOM Rene KochSwisscom John Riordan

UK AT&T Communications (UK) Ltd Aysé DilberAT&T Communications (UK) Ltd Romeo ZwartBT Morgan PotterBT Laboratories Steve CondieBT Laboratories Richards ReynoldsBT Networks And Systems Graham TraversCable & Wireless Communication Tony AnvonerDatabeam Europe Ltd Paul JonesETSI Secretariat Gur A KimchiFujitsu Europe Telecom R & D C John HopkinsFujitsu Europe Telecom R & D C Fiona KeaneHewlett-Packard Ltd Sam BauerHewlett-Packard Ltd Hans DaanenIntel Corporation (UK) Ltd James TogaLucent Technologies James M. PrattLucent Technologies N. S. UK Mike BuckleyLucent Technologies N. S. UK Peter HollandLucent Technologies N. S. UK Emiliano Antonio MastromartinoMarconi Communications Philip MartMotorola Paul Guram



Motorola Edward HatalaMotorola Willy VerbestelMotorola Ltd Andy MullanMotorola Ltd William SchmidtNEC Technologies (UK) Ltd Neil LaceyNortel Networks (Europe) Ali JamasebiNortel Networks (Europe) John TalbotNortel Networks (Europe) Jerome VoglimacciNortel Networks (Europe) John WarneOffice Of Telecommunications Alan PenneySamsung Electronics John B. FennVocalTec Communications Ltd David GurleVocalTec Communications Ltd Gadi Manor

US AT&T Dewayne SennettBellcore Christopher CelibertiBellcore Hong LiuBellcore Michael RamalhoCisco Systems Inc. Massimo LucchinaECI Telecom Matthew DragiffGRIC Communications Inc. Richard H. BrennanGTE USA Jay HiltonINTELSAT Fawad AbbasLucent Technologies Inc Louise SpergelMediaOne Labs James DahlMotorola Malik KhanMotorola Edgar MartinezMSAF Ronald HirstNuera Communications Inc David WangPulver.Com. Inc. Carl S. FordTransnexus LLC Stephen Thomas



REPORT OF Q4/15 RAPPORTEURS MEETINGJANUARY 18 - 22,1999, SUNRIVER, OR

The primary goal of this meeting was to prepare final text for the six xDSL Recommendationswhich had been Determined by SG15 at the previous SG plenary. This goal was met. At the endof each editor’s session, the proposed text for each Recommendation received a detailed reviewbefore approval.

LIAISONS

PO-023R1 (TI, Nortel, Ameritech, Lucent, SBC, ADI, 3Com, Cisco, GTE, AMD) provides theT1E1.4 VDSL system requirements document (T1E1.4/98-043R6) as a basis for USA VDSLrequirements.

PO-032 is a copy of a liaison to the IETF ADSL Working Group from the ADSL Forum; itprovides recommended updates to the IETF MIB for ADSL on power, fault and performancemanagement.

PO-042 (ADSL Forum) notes potential interference issues between in home data distributionsystems (e.g., Home Phone-line Network Alliance) and xDSL systems, especially VDSL; itrequests that Q4/15 evaluate and if appropriate, consider these issues.

PO-069 (T. Starr, Ameritech, ADSL Forum) provides a set of detailed changes to G.ploam.

PO-063 (H. J. Frizlen, ETSI TM6) expresses a willingness to work cooperatively on SDSL,HDSL and VDSL. TM6 is studying two different deployments scenarios for VDSL: Fiber to theexchange (FTTEx) and Fiber to the cabinet (FTTCab) which will likely require two different PSDmasks. They are working on the noise model which may result in slight changes to the values forimpairment noise. They also plan to provide a functional specification for a European SDSLaccess system to the Q4/15 Melbourne meeting in March.

IPR

D. Stuart (3Com, Q4/15 Rapporteur) encouraged all members who have previously indicated thatthey had IPR associated with this work to send a formal notification to the ITU (TSB Circular156) prior to the approval meeting.

G.972.2 (G.LITE)

PO-008© (C. Hansen, Intel) is the updated version of G.972.2; it incorporates a number ofeditorial changes. PO-018 (C. Hansen, Intel) is the Issues List. There was a short discussionabout the issues associated with the bis versions; it was agreed that a new list would bedeveloped essentially starting from scratch.

PO-024 (3COM, Alcatel, AMD, Aware, Centillium, Intel, ITeX, PairGain, TI, NationalSemiconductor) identifies and proposes a solution for a problem in the current fast retrainprocedure, which does not allow the ATU-R to indicate “unknown profile” after the ATU-C profileselection has been sent. PO-031 (R-R Damm, Deutsche Telekom) proposes a series of editorialand style corrections to the latest version of G.972.2. These proposals were accepted.

PO-048 (R. Verbin, MOC Israel) provides for a cutback in the ATU-R power when a phone is offhook.

PO-049 (R. Verbin, MOC Israel) proposes to dedicate one bit in R_MSGS1 for on/off hookindication. This proposal was modified in PO-049r1 to add 2 bits.

PO-053 (D. Van Bruyssel, Alcatel) proposes a set of 36 changes to G.972.2. It was decided toaddress these proposals in an ad hoc session. Essentially all of the proposals were accepted. Onthe optionally of R16: this contribution proposes that it be mandatory in both the receiver andtransmitter. PO-060 (J. Mueller, 3Com) proposes that it only be optional in the receiver, butmandatory in the transmitter. AMD offered a third possibility i.e., that R16 be mandatory in the



ATU-C transmitter and optional in both the transmitter and receiver in the ATU-R. In the end,none of the proposals were accepted.

PO-054 (F. Van der Putten, Alcatel) proposes using the new concepts of “Reference PSD level”and “Politeness Power Cutback” in Section 11. PO-087 (T. Cole, AMD) is the report of the ad hocon the power cutback in fast retrain. The recommendation was to use the “politeness” cutbackapplied during a previous initialization sequence. If this information is not available, then anestimate would be made (by the ATU-C) from the R-RECOV signal level which should have beensent at a nominal level. This was basically in alignment with the proposals made in PO-054.

PO-055 (F. Van der Putten, Alcatel) proposes the application of a Politeness Power Cutback to C-REVERB-FR1. Specifically, it proposes the use of 5 bits of C-MSG-N to convey the final absolutedownstream PSD level to aid the ATU-R in its calculation of the “Estimated average loopattenuation” for R-MSG-RA, R-MSG2 and the “Line Attenuation” primitive.

PO-056 (F. Van der Putten, Alcatel) proposes that the REVERB signal be sufficiently widebandin order to probe the complete transmit filter to obtain correct Equalizer Training Performance.

PO-057 (D. Van Bruyssel, Alcatel) proposes to modify the position in Fast Retrain where thetransmitter to receiver frame alignment takes place. Specifically it proposes that R-QUIET-FR1may be shortened by any number of samples (at a sampling rate of 276 kHz).

PO-058r1 (Siemens, AMD, Aware) proposes to include the use of low spectrum in the bis revisionof G.992.2. Specifically, it identifies six application areas for the use of low spectrum:• Increase bandwidth for ADSL-only loops for applications where long loops reduce throughput• Increase bandwidth for data during POTS on-hook periods for both splitterless and distributed

splitter CPE configurations• Transport on or more POTS service instances on a single loop using the digital bandwidth and

compression techniques• Provision of advanced services such as VoIP and/or hi-fi telephony• Reduced rate ADSL for data only service on very long (rural) loops• More symmetric service offerings

PO-079 (S. Abbas, Centillium) proposes that all performance enhancements of G.992.2 should beclearly stated as one term of reference for the work on G.litebis.

PO-029R2 (B. O’Mahony, Intel, plus Fujitsu, GTE, and MOC Israel) presents results of a fieldtrial conducted in the fall and winter of 1998. One CO in Hillsboro, OR and 47 homes wereinvolved. The results were encouraging in that they indicated splitterless operation was possiblein about 90% of the locations (80% of the homes required at least 1 filter connected in series withlocal telephone equipment) with top data rates (512/1.536 kbit/s). 95% of the locations supported384/384 kbit/s data rates.

G.977.1 (G.PLOAM)

PO-009© and PO-009R1© (A. Johansson, Ericsson) are drafts of G.977.1, G.ploam. PO-019 isthe issues list.

PO-027 (P. Silverman, 3Com) proposes that the SNMP-based OAM interfaces presently definedin G.997 should not be made optional unless changes are made to the body of G.997 and eitherG.994 or G.992.1 (and G.992.2) which would support this change in a consistent andinteroperable way. PO-075 (P. Silverman, 3Com) proposes new and changed text for G.997 tomake the SNMP interface optional while still maintaining a standard OAM interface.

PO-028 (D. Allan, Nortel Networks) recommends revised text for G.997.1, which clarifies theinstrumentation of the ADSL link MIB in the ATU-C referring to section 9 of G.992.1 for currentOAM practices. It also makes optional the implementation of the data link layer defined insection 7.1 of G.997.1 and those MIB elements, which are considered to be more properly a partof ATM layer performance monitoring.

PO-043 (F. Van der Putten, Alcatel) raises issues with some of the present text of G.997.1 for theidentified issues. It also proposes resolutions.



PO-072 (A. Johansson, Ericsson) proposes clarifications and modifications of draftRecommendation 997.1. Specifically, it proposes that:• Appendix II be replaced with a pointer to IETF• In Section 8.2.3.2 TR1s shall occur 10 seconds after the 15-minute threshold• The number 16 be used as the default for ES, SES & UAS.It also proposes that an open point be added to the Issues list for G.997bis for transfer of clock tothe ATU-R.

The breakout group recommended that only those features which are mandatory in T1.413 Issue2 be mandatory in G.ploam. This proposal was agreed by the main meeting.

G.966.1 (G.TEST)

PO-010 (M. Tzannes, Aware) is Draft Recommendation G.966.1; however, it was not available atthis meeting. PO-020 (M. Tzannes, Aware) is the issues list.

PO-040 (R.K. Smith, G. Bremer, Paradyne) summarizes the results of a survey of NorthAmerican builders associations, electrical contractors associations, electrical inspectorassociations and standards organizations plus a number of Florida architects, builders, andelectricians. It notes that most homes are wired with “bell wire,” more properly termed non-paired cable station wire. The results of this survey form the basis for two premises wiringmodels (a new model and a modified version of the Antwerp model) to be used to evaluate alltechnologies where the ADSL device is connected to existing wiring without the use of a splitter.PO-084 (R.K. Smith, Paradyne) provides reference diagrams for the additional North Americanresidential wiring models (daisy chain and star models) proposed at this meeting.

PO-040 was discussed in detail. R.K. Smith drew a new daisy-chain model (Figure 1) and amodified star model (Figure 2). The group decided to retain the original Star model and add thetwo proposed by Paradyne.

Figure 1. Model A (the boxes with the line indicate a phone is connected) 4 conductor -24 AWG non-paired cable station wire.

PO-041 (F. Van der Putten, Alcatel) proposes two modifications to align the procedures withthose specified in T1.413 Issue 2.

PO-066 (F. Van der Putten, Alcatel) identifies a problem with the G.lite initialization procedureduring the performance tests for North America Annex D, test case 12, because of the present



definition of upstream induced noise based on G.lite PSD-mask. It proposes to align the modelingof G.lite, upstream induced crosstalk with the G.dmt filter-based approach.

PO-067 (K. Kikushima, NTT) proposes additions and modifications to the sections that addressthe TCM-ISDN environment.

PO-083 (R.K. Smith, Paradyne) reports on the proposed wire type and wiring configuration forNorth American residential wiring Model A. It further proposes that both pairs be daisy chainedto all jacks.

Figure 2. Model B (the boxes with the line indicate a phone is connected) Category 3cable.

G.992.1 (G.DMT)

PO-007© (S. Palm, Matsushita) is draft Recommendation G.992.1. It includes the changes madeat the editors’ meeting in December. PO-044 (C. Hansen, Intel) is the report of that meeting.PO-017 (S. Palm, Matsushita) is the G.dmtbis Issues List.

PO-022 (Alcatel, Cisco Systems, Texas Instruments) proposes that G.dmtbis and G.litebis bemerged into a single Recommendation.

PO-033 (K. Kikushima, NTT) proposes editorial changes in Annex E.4 Type 4 of G.dmt, theJapanese POTS splitter.

PO-034 (F. Van der Putten, P. Reusens, Alcatel) proposes a small adjustment in the set of sub-carriers (Section B.3.1) used for measuring the downstream power cutback.

PO-035 (F. Van der Putten, Alcatel) proposes a set of changes to G.972.1.

PO-038 (F. Van der Putten, P. Reusens, R. Cassiers, Alcatel) proposes a new quiescent modehaving the following characteristics:



• Free run operation: all clocks (symbol, frame, superframe, ATM cell) keep running• Qmode transmit signal consists of a negotiated pilot tone only• Periodic return towards Data Mode (L0), Transition to/from Qmode occurs through a

progressive auto-reset of the transmission chain• Transition to/from Qmode under transmitter control and can occur at the beginning of any

symbol

PO-050 (I. Sharfer, MOC Israel) proposes to amend the sections in the text specifying constraintson the fine gains by removing the constraint imposed by the sum of logs formula. An aggregatepower constraint should be specified in terms of the number of sub-carriers used.

PO-052 (W. Matsumoto, Mitsubishi) provides a tutorial on rate-converter operation and itsperformance of delay and bitrate in G.992.1 Annex C.

PO-071 (J. A. Torres, V. Demjanenko, VoCAL Technologies) proposes using a ConcatenatedConvolutional Code technique for G.992.1 and G.992.2.

PO-073 (P. Silverman, 3Com) proposes minor word changes to the text defining the clear EOC insection 9.1 of G.992.1 to correct the errors which appear to make this interface optional. Duringthe meeting this proposal was changed and replacement text for the entirety of sections 8.5 and9.1 was provided (PO-073r2).

PO-082 (Y. Goldstein, PC Tel) proposes to remove the requirement that the shortening of thenumber of samples of the final symbol of R-QUIET3 must be in integer multiples of four.

G.994.1 (G.HS)

PO-006© is draft Rec. G.994.1, G.hs. PO-016 (L. Brown, Motorola, G.hs editor) is the updatedG.hs issues list. The editor detailed all the changes to this list.

PO-025 (S. Blackwell, Adtran) proposes to replace the words “shall alternately attempt” with“should alternately attempt” in line 4, Annex A (the fallback mechanism).

PO-026 (S. Blackwell, Adtran, and J. Joyce, Cisco) proposes that a requirement for a minimumdetection interval of 250 ms and minimum duration of 300 ms be added for signal R-Tones-Req.It also proposes that a requirement for a minimum detection interval of 150 ms be added forsignal C-Tones to provide a placeholder for a C-Busy signal. This proposal was accepted.

PO-046R1 (S. Palm, Matsushita) proposes several editorial suggestions for G.994.1. It alsoprovides suggestions for completing missing sections.

PO-047 (S. Palm, Matsushita) describes a half-duplex procedure that requires minimalmodification to the current duplex procedures in G.994.1. Although this was presented for“information only,” it became the basis of an agreement on half-duplex timing.

PO-059 (S. Blackwell, Adtran) proposes that the 4K family be defined as half-duplex. It wasagreed that G.994.1 would support both duplex and half-duplex operation.

PO-070 (Alcatel, Aware, Ericsson, Telia Research, Texas Instruments) requests that ahandshaking procedure for VDSL be added to the issues list.

PO-074 (P. Silverman, 3Com) proposes a handshake to indicate the use of the G.997.1 SNMPinterface.

PO-076 (S. Abbas, Centillium) proposes editorial changes to add clarity to the text in Section 9(information field coding format).

PO-080 (B. O’Mahony, Intel) supports a previous proposal (AB-144) for the definition of C-Busy.

PO-081© (B. O’Mahony, Intel) is a copy of Draft G.994.1 marked up with proposed editorialcorrections.



PO-036 (F. Van der Putten, Alcatel) identifies several issues and proposes solutions. PO-062 (J.Mueller, 3Com) identifies areas of the text in the draft Recommendation that need clarification.

PO-036 and PO-062 both raise an issue about Message Segmentation; it was decided to notaccept either.

It was decided to add an octet to NSF (Non Standard Facilities) to indicate number on NS blocksand remove the bit indicating more than one segment.

G.995.1 (G.REF)

PO-004© (S. Abbas, Centillium) is draft Rec. G.995.1. PO-014 (S. Abbas, Centillium) is theissues list.

PO-078 (S. Abbas, Centillium) proposes that statements be added to G.ref to identify which partsof the Recommendations are needed to ensure a compliant transceiver implementation. It furtherproposes that these statements be added to the appropriate Recommendations.

Dick Brandt, dB Consulting

Q4/15 RAPPORTEURS GROUP ROSTER, JANUARY 18 - 22,1999, SUNRIVER, OR

Dick Stuart, 3Com Q4/15 RapporteurHost: Intel

3Com Joe Mueller [email protected] Peter Silverman [email protected] Richard Stuart [email protected] Steve Blackwell [email protected] Peter P.F. Reusens [email protected] Arthur J. Carlson [email protected] Terry Cole [email protected] Tom Starr [email protected] Marcos Tzannes [email protected] Networks Aleksandar Purkovic [email protected] Nigel Billington [email protected] Technology Syed Abbas [email protected] Technology Guozhu Long [email protected] Systems John Joyce [email protected] Rabah Hamdi [email protected] Systems Mickey Rushing [email protected] Systems Xuming Zhang [email protected] Telekom Ralf-Rainer Damm [email protected] Albin Johansson [email protected] Technology Yuanjie Chen [email protected] Technology Ping Dong [email protected] Technology Peter Kan [email protected] Telecom Georges Sebek [email protected] Digital Technology Yutaka Awata [email protected] Instrument/Next Level Comm. Sabit Say [email protected] Datacomm, Inc Yuri Goldstein [email protected] Semiconductor Inc Massimo Sorbara [email protected] Gary McAninch [email protected] Telephone Co. Juha Kauppi [email protected] Telecom Express Wen-Juh Kang [email protected] Barry O’Mahony [email protected] Chris Hansen [email protected] One Juan Garza [email protected] John Magill [email protected] Harry Mildonian [email protected]



Lucent Carl Posthuma [email protected] Communications Davide Pagnotta [email protected] Graphic Comm.System Co.. Stephen Palm [email protected] Electric Corp. Wataru Matsumoto [email protected] Richard Brandt [email protected] ISG Les Brown [email protected] Sayfe Kiaei [email protected] Semiconductor Frank Meye-Zu-Drewer [email protected] - USA Alan Weissberger [email protected] Lippo Rantanen [email protected] (Canada) Christian Bourget [email protected] Networks (Europe) Les Humphrey [email protected] Networks Alexandar Purkovic [email protected], Access Network Systems Koji Kikushima [email protected] Electric Industry Akihiro Takahashi [email protected] Ilan Sharfer [email protected] Rami Verbin [email protected] E. James Infusino [email protected] George Zimmerman [email protected] Bill Betts [email protected] R.K. Smith [email protected] Tel Inc. Enrico Cadorin [email protected] Tel CSD Yuri GoldsteinPulsecom Dennis Rauschmayer [email protected] Communications Richard Simeon [email protected] Communications Boaz Porat [email protected] Technology John Roquet [email protected] Rudi Frenzel [email protected] Bob Jenness [email protected] Neal King [email protected] Dirk Schmucking [email protected] Electric Kazuya Matsumoto [email protected] Instruments Shoji Kobayashi [email protected] Instruments Xiaohui Li [email protected] Instruments Satoru Yamauchi [email protected] Takanao Amatsubo [email protected] Technologies Aberto Torres [email protected] Comm. for MOC Israel Ken Krechmer [email protected]

THE CSR LIBRARY

Subscribers may order copies of documents shown in boldface type from CommunicationsStandards Review, where not controlled. Documents listed with © are controlleddocuments. These documents are not for sale, but we can provide you with the author’scontact information. -e means we have electronic copy only; +e means we have electronicand hard copy available. We have a large library of standards work in process and canhelp you locate other information you may need.



Acronym DefinitionsAAL ATM Adaptive Layer3GPP Third Generation Partnership ProjectAAA Authentication, Authorization and

AccountingACTE Approvals Committee for Terminal

EquipmentADPCM Adaptive Differential PCMADSL Asymmetric Digital Subscriber Loop

(Line)AN Advisory NoteAOC-E Advice Of Charge at the End of a callARQ Automatic Repeat RequestATM Asynchronous Transfer ModeATU-C ADSL Transceiver Unit - Central Office

EndATU-R ADSL Transceiver Unit - Remote

Terminal EndB-ISDN Broadband ISDNBA Basic Access (ISDN)BCM Back Channel MessageBES Back-End ServicesBEST Back-End Service TransactionBOF Birds of a FeatherCAP Carrier-less Amplitude Modulation

Phase ModulationCLI Calling Line IdentificationCLIP Calling Line ID ParameterCO Central OfficeCPE Customer Premise EquipmentCRC Cyclic Redundancy CodeCTM Cordless Terminal MobilityCTR Common Technical RequirementsDC Directionality ClassificationDCE Data Circuit Terminating EquipmentDCP Digital Connectivity ProbingDCT Discrete Cosine TransformDFE Decision Feedback EqualizerDMIF DSM-CC Multimedia Integration

FrameworkDMT Discrete Multi-tone CarrierDNS Domain Name ServerDSL Digital Subscriber LineDSS1 Digital Subscriber Signaling 1DTA Digital Terminal and Access ProjectDTMF Dual Tone Multi FrequencyDTR Draft Technical ReportDTS Draft Technical Standard (ETSI)ECVQ Entropy Constrained Vector

QuantisationEMC ElectroMagnetic CompatibilityEN European NormEOB End of BlockEOC Embedded Operations ChannelEP ETSI ProjectES Errored SecondsES ETSI StandardETR ETSI Technical ReportsETS European Telecommunications Standard

(ETSI)EZW Embedded Zero-tree Wavelet (coding)FAS Frame Alignment SignalFDD Frequency Division Duplexing



FEC Forward Error ControlFECC Far End Camera ControlFEXT Far End Cross TalkFSAN Full Service Access NetworksFTTCab VDSL to the cabinetFTTEx VDSL to the exchangeGII Global Information InfrastructureGK GateKeeperGSM Global System for Mobile

Communications (formerly GroupeSpeciale Mobile)

GSTN General Switched Telephone Network(e.g., PSTN)

GUI Graphical User InterfaceGW GateWayHDSL High-rate Digital Subscriber LoopHomePNA Home Phoneline Networking AllianceIDCT Inverted Discrete Cosign TransformIEC International Electrotechnical

CommitteeIESG Internet Engineering Steering GroupIETF Internet Engineering Task ForceIMTC International Multimedia

Teleconferencing ConsortiumIN Intelligent NetworkingIP Internet ProtocolIPDC IP Device ControlIPR Intellectual Property Rightsiptel Internet Protocol Telephony (IETF

Working Group)ISDN Integrated Services Digital NetworkISDN-BA ISDN basic accessISO International Organization for

StandardizationISOC Internet SocietyISSG Internet Services Study Group (MSAF)ISUP ISDN User PartITAAB ISDN Technical Advisory Ad hoc BoardITU International Telecommunication UnionITU-T ITU Telecommunications SectorIVR Interactive Voice ResponseJPEG Joint Photographics Expert GroupJTC Joint Technical CommitteeKTA Key Technical AreasLRQ Location RequestLT Line TerminationMB Macro BlockMC Message CenterMDCP Media Device Control ProtocolMG Media GatewayMGC Media Gateway ControllerME Mobile EquipmentMIB Management Information BaseMPEG Motion Picture Experts GroupMSAF Multimedia Services Affiliate ForumMUX MultiplexerMV Motion VectorNEXT Near End Cross TalkNNI Network to Network InterfaceNT Network TerminationNTP Network Termination PointNTR Network Timing Reference markerOAM Operations, Administration, and

Maintenance



OCG Operational Coordination Group (ETSI)PAM Pulse Amplitude ModulationPC Personal ComputerPHB per-hop forwarding behaviorPICS Product Implementation Conformance

StatementPICS Protocol Implementation Conformance

StatementPMC Packet Mode ChannelPMD Physical Media Dependent LayerPMS Physical Media SpecificPOTS Plain Old Telephone ServicePSD Power Spectral DensityPSNR Peak Signal to Noise RatioPSTN Public Switched Telephone NetworkQAM Quadrature Amplitude ModulationQoS Quality of ServiceQUANT Quantization parameterR&TTE Radio and TTE (Telecommunications

Terminal Equipment) DirectiveRAS Registration, Admission, and StatusRFC Designation for an Internet StandardRSVP Resource Reservation Setup ProtocolRTP Real Time Transport ProtocolSCN Switched Circuit NetworkSDH Synchronous Digital HierarchySDSL Symmetrical high bit rate Digital

Subscriber LineSDSS Server Display And Script ServicesSES Severely Errored SecondSGCP Simple Gateway Control ProtocolSMG Special Mobile GroupSNMP Simple Network Management ProtocolSOHO Small Office Home OfficeSP Study PointSPS Signaling Protocol And System Technical

Committee (ETSI)SS7 Signaling System 7TBR Technical Basis for RegistrationTBR Technical Basis for Regulation (ETSI

standard)TBRL Terminal balance return lossTC Technical CommitteeTC Transmission Convergence LayerTCAM Telecommunication Conformity

Assessment and Market SurveillanceCommittee (ETSI)

TCM Time Compression MultiplexTDD Time Division DuplexingTE ISDN TerminalTE Terminal Equipment (ETSI Committee)TIPHON Telecommunications and Internet

Protocol Harmonization Over Networks(ETSI Project)

TIPIA TIPHON IP telephony ImplementationAgreement

TM Transmission and Multiplexing (ETSICommittee)

TMN Test Model NumberTRAC Technical Recommendation Application

CommitteeTSAG Telecommunication Standardization

Advisory Group



TSB Telecommunications StandardizationBoard (ITU)

TTT Trans European Network Telecom(TIPHON)

UAS Unavailable Seconds (Rec. G.997.1)UMTS Universal Mobile Telecommunications

SystemUNI User-Network InterfaceUPT Universal Personal Telecommunications

(CCITT)VDSL Very high speed ADSLVHE Virtual Home EnvironmentVLC Variable Length CodewordVoIP Voice Over Internet ProtocolVQ Vector QuantizationWG Working GroupxDSL all the different Digital Subscriber Lines

The next issue of Communications Standards Review - Telecommunications (Vol. 10 #3) isscheduled for March-April 1999.



1999 STANDARDS COMMITTEE MEETING SCHEDULESAS OF JANUARY 28, 1999

Subject to Change without Notice

Committee Date(s) LocationT1E1.4 Feb 1 - 5 Orlando, FLTR-29 Feb 8 - 10 Dana Point, CATR-30 Feb 8 - 10 Dana Point, CATR-41 and TR-42 Feb 15 - 19 Sarasota, FLQ12 - 14/16 Rapp. Feb 15 - 19 Monterey, CAQ15/16 Rapp. Feb 16 - 19 Monterey, CAETSI ATAc res. Feb 16 - 17 S. Antipolis, FRQ11/16 Rapp. Feb 17 - 19 Monterey, CAETSI TM6 Feb 22 - 26 Tel Aviv, IsraelETSI TIPHON Mar 1 - 5 Philadelphia, PAETSI DTA Mar 8 - 12 S. Antipolis, FRT1E1 Mar 8 - 12 Tustin, CAITU-T SG8 Mar24-Apr 1 GenevaQ4/15 Rapp. Mar29-Apr 2 Melbourne, AUTR-30 Apr 12 - 15 ---T1E1.4 Apr 19 - 23 Washington, DCETSI TIPHON May 3 - 7 ---ETSI TM6 May 3 - 7 Annecy, FranceTR-29 May 10 -12 East Coast USQ4/15 Rapp. May 10 - 14 Boston, MAT1E1.4 Interim May 17 - 18 Boston, MATR-41 and TR-42 May 17 - 21 Portland, ORITU-T SG 16 May 18 - 28 Santiago, ChileT1E1 Jun 7 - 11 Ottawa, Ont.TR-30 Jun 14 - 17 CaliforniaITU-T SG15 Jun 21-Jul 2 GenevaETSI ATAc WG Jun 29 - 30 S. Antipolis, FRQ15/16 Rapp. July ---ETSI TIPHON July 19 -23 AmsterdamTR-30 Aug 16 - 18 Quebec City, CATR-41 and TR-42 Aug 16 - 20 Ottawa, OntT1E1 Aug 23 - 27 Portland, ORETSI TM6 Sep 20 - 24 Edinburgh, UKETSI TIPHON October 4 - 8 ---TR-30 Oct 11 - 14 Baltimore, MDETSI ATA Oct 11 - 15 ---Q15/16 Rapp. NovemberTR-41 and TR-42 Nov 8 - 12 Reno, NVTR-30 Nov29-Dec 2 Clearwater, FLETSI TM6 Nov29-Dec 3 NetherlandsT1E1 Dec 6 - 10 ---

Communications Standards Review -Telecommunications (ISSN 1081-4655) is published8 - 9 times per year, within days after the latest, related standards meetings. Editor: Elaine J.Baskin, Ph.D. Technical Editor: Ken Krechmer. Copyright © 1999, Communications StandardsReview. All rights reserved. Copying of individual articles for distribution within a subscriberorganization is permitted. Subscriptions: $695.00 per year worldwide, $795.00 in electronicformat. Corporate Intranet subscriptions (site license for multiple copies) are available. Submitarticles for consideration to: Communications Standards Review, 757 Greer Road, Palo Alto, CA94303-3024 U.S.A. Tel: +1-650-856-9018. Fax: +1-650-856-6591. e-mail: [email protected]: http://www.csrstds.com 10902.

CSR Vol. 10 #2

FAX BOUNCE-BACKThe fastest and easiest way to get the standards information you need!

FAX this form to Communications Standards Review (or mail or phone) andthe standards contributions ordered will be shipped the next business day.

FAX: +1-650-856-6591

COMMUNICATIONS STANDARDS REVIEW757 Greer Road

Palo Alto, CA 94303-3024 USATel: +1-650-856-9018

e-mail: [email protected]: http://www.csrstds.com

BILL TO: SHIP TO: (Include Telephone)_________________________ _____________________________________________________ _____________________________________________________ _____________________________________________________ ____________________________

CSR provides to CSR subscribers copies of communications standardswork in process. Once standards are completed and available,

they should be acquired from the appropriate, authorized source.

Documents Ordered:

______________ _________________ _______________________________ _________________ _______________________________ _________________ _______________________________ _________________ _______________________________ _________________ _________________

Documents in addition to those listed in boldface in this issue of CSR mayalso be ordered. Please ask if we have what you are looking for.

Subscribers, Per Order North America Outside North America1st Document $50.00 $50.002nd Document $40.00 $40.003rd + Documents $25.00 $25.00Shipping 2nd day air included Invoiced at cost

Shipping Instructions: _________________________________________

Invoices for documents and any applicable shipping charges or sales tax areincluded with the order.

Payment may be made through MasterCard or Visa.

_________________________ _____________ Authorizing Signature Date PO (if Appl.)

Documents Cited in Communications Standards Review Vol 10 #2, February, 1999

All documents listed in boldface in the text are for saleto CSR subscribers. They are copies of work in progress.

These documents are not approved or published standards.

Documents listed with © are controlled documents.These documents are not for sale, but we can provide you with

the author’s contact information.

Documents in addition to those mentioned in this issue of CSR-Tmay also be ordered. Please ask if we have what you are looking for.

TIA Standards,SP- documents,

and ITU Recommendationsmay be ordered from

Global Engineering DocumentsTel: 800 854-7179, +1 303 792-2181

Fax : +1 303 397-7935.http://global.ihs.com