Mode S Installation and Siting Criteria Sandholm_1982_ATC-99a_WW-15318

8/6/2019 Mode S Installation and Siting Criteria Sandholm_1982_ATC-99a_WW-15318

1/33

Lincoln LaboratoryMASSACHUSETTS INSTITUTE OF TECHNOLOGY

LEXINGTON,MASSACHUSETTS

DOT/FAA/RD-82/58

Project ReportATC-99Rev. A

Mode S Installation and Siting Criteria

R.G. Sandholm

13 September 1982

Prepared for the Federal Aviation Administration,

Washington, D.C. 20591

This document is available to the public through

the National Technical Information Service,

Springfield, VA 22161


2/33

This document is disseminated under the sponsorship of the Department

of Transportation in the interest of information exchange. The United

States Government assumes no liability for its contents or use thereof.


3/33

Technical Report Documentation PaI. Report No. 2. Gonr.mant Accession No. 3. Recipient's Catalog No.

DOT/F AA/RD-82/584. Titla and Subtitle 5. Report Data

13 September 1982Mode S Installation an d Siting Criteria 6. Perlarming Organizatian eade

7. Author(s) 8. Perlarming Orgenization Repart No.Ronald G. Sand holm ATC-99 Revision A

DOT-FA72-WAI261

Project Report

10. Wark Unit No. (TRAIS)Project No. 052241-04

11. Contract Dr Grant Na.

14. Spansaring Agency Code

9. Perlarming Orgenizatian Name end AddressMassachusetts Institute of TechnologyLincoln LaboratoryP.O. Box 73Lexington, MA 021730073

12. Spansoring Agancy Nama and AddressDepartment of TransportationFederal Aviation AdministrationSystems Research and Development ServiceWashington, D.C. 20591

13. Type 01 Repart and P"iad Ca. .red1_- - - - - - -__--------------------1

15. Supplementary Notes

The work reported i n t hi s document was performed at Lincoln Laboratory, a center for research operatedby Massachusetts Institute of Technology, under Air Force Contract FI962880C-0002.

t 6. Abstract

This paper provides information on site-associated phenomena that affect th e proper operation of aMode S sensor and therefore warrant serious consideration when siting a sensor. The Mode S related discussion is intended to be a supplement to th e ATCRBS si ting cr i ter ia presented in th e FAA Primary/Secondary Terminal Radar Siting Handbook. The paper discusses sitin/!: criteria as they relate to t heMode S sensor antenna system, as opposed to th e ATCRBS hogtrough antenna, and importantly, addresses those character ist ics of the surrounding environment t ha t a re crucial to proper Mode S surveillance.

17. Key Words 18. Oistributian StatementSuneillance radar sitingVertical lobingShadowingDiffraction

Document is available to th e public throughth e National Technical Information Serviee,Springfield, Virginia 22151

19. Security Classil. (al this repart) 20. Security Classi!. (01 this page) 21. No. 01 Pages 22. PriceUnclassified Unclassified 32

Form DOT F 1700.7 (8-72) Reproduction of comple ted page authorized


4/33

CONTENTS

1.0 INTRODUCTION 12.0 OVERVIEW 23.0 SHADOWING AND DIFFRACTION 33.1 Obstructions 33.2 Signal Fades Due to Man-Made Obstructions 33.3 ,Azimuth Error Due to Man-Made Obstructions 8

3.4 The Effect of Natu ra l Terrain on Signal Fade and, Azimuth Error 184.0 VERTICAL LOBING 235.0 FALSE TARGET REFLECTIONS 256.0 SUMMARY AND SITING RECOMMENDATIONS 26REFERENCES 28

iii


5/33

la

lb

2

3

4

LIST OF ILLUSTRATIONSHighrise Buildings (Part of Boston Skyline As SeenFrom Logan Airport).Highrise Buildings (A Second View of Part of BostonSkyline As Seen From Logan Airport).Fade Cast by th e Shadow of the Prudential BuildingAs Seen From Logan Airport. Obstacle Width and RangeAre 220-Feet and 22000-Feet Respectively.Envelope of Deepest Fade Nulls Caused by an ObstacleAs a Function of Obstacle Range.Azimuth Estimation Error vs. Obstacle Position.Obstacle, Which Corresponds to Prudential BuildingAs Seen From Logan Airport, is 220-Feet Wide and at a22000-Foot Range.

4

5

6

7

95 Azimuth Estimation Error vs. Aircraft PositionRelative to Obstacle Midpoint. Obstacle, Which

Corresponds to Hanscom AFB Smokestack, is 10-Feet Wideand at a 1500 Foot Range. 10

6 Plot of an Encounter Showing the Effect of DiffractionFrom a Vertical Obstacle on Aircraft Surveillance. 127 Maximum Azimuth Error (Deg) As a Function of ObstacleWidth (f t ) and Range ( f t ) . 148 Angle at Off-the-Center of Obstacle Where Peak AzimuthError Occurs vs. Obstacle Width ( f t . ) . 159 Error Azimuth Extent vs. Obstacle Parameters: Rangeand Width. 171011

Monopulse Error vs. Azimuth: Clementon, NJ.Monopulse Error VS. Azimuth: Philadelphia Airport.

1920

12 Signal Fade Due to a Series of Four Hills as Measuredfrom an Aircraft at 4300-Foot Altitude. Optical Line-of-Sight is at 0 Degrees Elevation. 22

13 Envelopes of 5-Foot Open Array and Hogtrough AntennaLobing Nulls for Flat Grass and Water Surfaces ofInf inite Extent. 24

iv


6/33

MODE S INSTALLATION AND SITING CRITERIA

1 .0 INTRODUCTION

This paper provides i ns ig ht i nt o s i t e - a s s o c i a t e d phenomena t h a t a f f e c t theoperation of a Mode S sensor and which warrant ser i ous consi der at i on i n any ModeS s i t i n g exercise. The Mode S - r e l a t e d di scussi on i s intended to be a supplementto th e ATRCBS s i t i n g criteria presented i n th e FAA Primary/Secondary TerminalRadar Siting Handbook l The paper di scusses s i t i n g cri ter ia as r e l a t e d to th eMode S s en so r a nt en na system, as opposed to the ATCRBS hogtrough antenna, andmost impor,tantly i t descr i bes features of the surrounding environment t h a t arec r u c i a l to proper Mode S surveillance and what ca n be done in s i t i n g th e Mode Ssensor tQ mitigate them.

1


7/33

2.0 OVERVIEWThe most common s i te characterist ics affecting Mode S performance andwhich will be discussed in greater detai l are:a. Signal shadowing and diffraction induced azimuth errorsfrom man-made obstructions and natural terrain.b. In-beam vertical lobing fades caused by specularreflections f rom smooth f la t terrain surrounding thesensor.c. Man-made and natural reflective surfaces that causegeneration of false targets .In general most of the sit ing criteria discussed in the Siting Handbook forATCRBS apply equally well to Mode S with few exceptions. The vert ical pattern

lower edge rol l-off of the Mode S open array antenna will minimize thesensi t ivi ty of the vertical pattern to ground reflections. This in turn willprovide ModeS with improved coverage capability and will afford greater freedomin si te selection as far as in-beam multipath is concerned.The fact that Mode S has the capability of flagging f als e t ar ge t reportstogether with i ts lower edge cutoff will also lessen th e importance of location

with respect to man-made reflecting surfaces.One area of considerable importance to Mode S which is no t covered in theSiting Handbook, is the impact o f obstructions (towers, buildings, smokestacks,etc.) on Mode S surveillance accuracy. The inherently grea te r reso lut ion of theMode S azimuth position estimator w ill re su lt in noticeable cross range and

cross track velocity errors due to the di ff rac tion e f fec ts of shadowingobstruction. Depending on size and distance of th e obstruction, the azimutherror may greatly exceed the surveillance accuracy requirement specified inparagraph 3.3.2.8 of th e Mode S sensor specification, FAA-E-2716.

2


8/33

3.0 SHADOWING AND DIFFRACTION3 .1 Obs tru ction sMan-made and natural obstacles surrounding the Mode S s i te can cause both

a serious fade in the l ink signal strength resulting in noise-induced errorsand a sizeable azimuth error in the position estimate of aircraf t flyingbehind th e obstruction. Signal blockage on both the uplink and downlink canresult in either marginal or no coverage fo r se ve ra l scans and seriouslyimpair th e capability of the Mode S sensor. In addition to shadowing, theobstruction will cause diffraction of the downlink signal wavefront fromaircraf t w.hose l ine-of-sight is in close proximity to i t . Serious diffractioncan cause a sizeable error in the Mode S azimuth position and cross trackvelocity f tst imat e o f the aircraf t .

3.2 Signal Fades Due to Man-Made ObstructionsThe primary cause of blockage in a terminal enviroment are man-made

obstructions such as towers, buildings and smokestacks. In heavily populatedurban areas the proximity of these structures to an airport-located sensorcould provide destructive interference to Mode S surveillance of aircraf t upto a few degrees elevation. As an example Figs. la and lb i l lustrates theBoston skyline (typical of many terminal locations) as seen from the ASR atLogan Airport. Most of the buildings in one particular 11 degree azimuthsector exceed 1 degree elevation and some extend to 2.5 degrees. Fig. 2typifies th e character of signal fading caused by an obstruction, in this caseth e Prudential building in Boston, in which the aircraf t is below th e top ofand at considerably greater range than the structure2 The building is 220feet wide, 22000 feet from the sensor and extends to 2 degrees elevation. Thevariation of the fade pattern as a function of aircraf t offset from themidpoint of th e obstruction as i l lustrated in Fig. 2 ( i . e . , a midpoint lobesurrounded by deep nulls) is characteristic of a l l isolated and geometricallysimple structures except that the width of the structure will determine th efrequency and number of fade nul ls . Fig. 3 is a plot of the approximaterelationship between t he deepes t null value and the obstacle range fordifferent obstacle widths. Generally the fade at midpoint is one-half thevalue of the deepest fade.

The following general comments relative to th e Prudential example can bemade concerning the relationship of signal fade to the obstacle dimension andto the obstacle and th e aircraf t range.a. An increase in obstacle width for a given range willresul t in deeper fades and will increase the number and

frequency of fade nulls.b. A decrease in obstacle range for a g iven wid th willincrease the amount of fade.

3


9/33

11 Degree Azimuth Sector

', ' .: :: '",. . . .A . ', .. ..... ' ...... ...

''' 11 -',' ' ".....,........

J'l.:... ....

2 Degree8 Elevation

Fig. 1a. Hlghrl8e Bu ll d ln g8 (Part of B08ton Skyline A8 Seen From Logan Ai rport ).

4


10/33

2 Degrees Elevation

Prudential Buildingj

Fig. 1b. Hl gh rlse Buildings (A Se co nd View of Part of Boston Sk.yllneAs Se en F ro m Logan Airport).

5


11/33

-4 _-------,..--------r-------...,..-----......--"1

-8

......

CD"0...,w -1 200(U...J0(ZS!(f )

-16

- 20-0 .4

1---0.2

TARGET EXTENT

o o. 2

DEEPEST FADE

..I0.4

Aircraft Azimuth With Respect To Center of Building (Degrees)

Fig. 2. Fade Cast by the Shadow of the Prudential Building As Seen From Logan Airport.Obstacle Width and Range Are 220 Feet and 22000 Feet Respectively

6


12/33

aWD........U.IWA.WWD

IOBSTACLEWIDTH200 FEEt

mm II OBSTACLEWIDTHP 10 0 FEETmw m fl:ttl:lIBSTACLE mmWIDTH !II0 FEETI iI+mI i iI II ANGE OF OBSTACLE FROM SENSOR eNM)Fig. 3. Envelope of De.peet Fade Nulle Caueed by anObetacle Ae a Function of Obetacle Range.


13/33

c. A decrease in aircraf t range will result in deeper fades.As an example an aircraf t at twice th e obstacle rangew ill re su lt in fades 4 dB larger than that due to anaircraf t at much greater range.

In a typical metropolitan environment the ideal location of a Mode Ssensor, in order to eliminate low alti tude shadowing, would be one that ise ither su ff ic ien tly removed from the obstruction (at least 5 om for a 200 footwidth) or at a height comparable to the obstruction. Unfortunately an ideallocation is not always possible in an urban area and therefore considerationshould be given to minimizing shadowing effects for a majority of aircraf t inp r e d o m i n a ~ t l y used airspace. A simple and approximate criteria for th ed istan ce to an obstruction of given wid th in order to maintain an "acceptable"level of ,fade i s given in Reference 3. Assuming that th e midpoint fade valueis a good representation of th e l ikely fade encountered over th e azimuthextent of the obstacle then in order to maintain this value to -6 dB or lessth e range to the obstacle should be at least the square of i t s width.

3.3 Azimuth Error Due to Man-Made ObstructionsDiffraction of the wavefront from an aircraf t whose l ine-of-sight iseither through th e obstacle or in close proximity to i t can cause anappreciable error in the estimate of the aircraf t azimuth. The diffractedsignal will have approximately the same effect on the azimuth positionestimate regardless of whether the estimate is generated by a Mode S monopulseprocessor or by the beam spli t t ing technique currently employed in ATCRBSsensors.Current ATCRBS sensors use a sliding window detection p r o c e s ~ that has aninternal quantization error comparable in magnitude to th e errors' generated bytypical obstructions surrounding an airport. This fact has tended to mask the

effect of diffraction errors in present ATCRBS sensors. Mode S, on the otherhand, employs a surveillance p roce ssor o f inherently grea ter reso lu t ion .The magnitude of th e error and the azimuthal extent or wedge over which aposition estimate is ser iously af fec ted depends on the dimensions of theobstacle as well as on the range of both the obstacle and the ai rcraf t . Figs.4 and 5 i l lustrate the typical nature of the azimuth estimation error as afunction of aircraf t position relative to the obstacle midpoint. The errorvalues in the plots are based on a single Mode S interrogation of an aircraf tthat is ei ther at or very close to the antenna boresight. An interrogation atthe lea ding ( tra i l ing) edge of a clockwise rotating beam would produce asmaller (larger) error i f the target azimuth preceeded th e obstacle azimuth.The resu lt an t e rror associated with a large number of interrogations per dwellsuch as for ATCRBS would tend to average out to a value equ ivalent t o theerror from a single interrogation at boresight.

8


14/33

I I

/if,

, ~ ~ I d l n ~widt... L ...., . . . . a.t J .A..& .. La ...&. .... &.. .... ,.. ~ I J V I V' 'J ..... ..

--- V II

!

Obstacle azimuth- target azimuth (deg)62

0.5

o

1.0

II

Q)-.i -0.5

Fig. 4. Azimuth Eatlm8lton Error VI . Obatacle Poaltlon. Obatacle. Which Correaponda to PrudentialBuilding Aa Seen From Logan Airport. 18 220 Feet Wide and at a 22000 Foot Range.

9


15/33

4.0.0

- - - . - - - - - - T - - - - r - - - - - - - I - - - - - - - I - - . - - - - - T - - - ~ - - - ,AIRCRAFT AT LOW ELEV4TION l-AND LONG RANGE. I

0.5ILl00: 0.250:0:ILlILl2 'fI )ILl:J: I-0,25 r

-Q.5L_L __-'-- L . .L . ~ ...._ __ .L--__ """-4.0 -3.0 -2.0 -1,0 0,0 1.0 2. 0

TARGET/OBSTACLE SEPARATION (DEG)

Fig. 5. Azimuth Elt lmat lon Error VI . Aircraft PoIltlon Relative to ObstacleMidpoint. Obltacle. Which CorrelPondl to Hanscom AFB Smokeltack I I 10 Feet Wide and at a 1600 Foot Range.

10


16/33

Fig. 4 i s th e computed perturbation caused by th e Prudential building inBoston as viewed from th e Logan ATCRBS s i te . The Prudential building is 222feet wide and at 22000 feet range. Fig. 5 is the Hanscom AFB smokestack asviewed by th e Mode S Experimental Facility (DABSEF) in Lexington, MA andi l lustrates th e comparison between th e azimuth error function as computed fromtheory and one derived from actual measurement using a controlled aircraft .The smokestack is 10 feet wide and at a range of 1500 feet . In both cases theaircraf t is well below th e top of the obstruction and at a much greater range.The azimuth error is zero for an aircraft position directly behind th e centerof the obstacle and varies in an oscil latory manner as a resul t ofcon-structive and dest ruc tive in ter ference between the direct and diffractedsignals as, th e aircraf t moves away from th e obstacle. The importantcharacterist ics of this error function, in terms of Mode S performance, arethe m a x i ~ u m peak error value, i t s location relative to the obstacle midpointand th e azimuth wedge over which succeeding peak values are large enough toaffect the position estimate. The plots indicate that a single scan azimuthsurveillance report on an aircraf t which happens to be located in one of th epeak error regions can be in excess of surveillance requirements. Additionallythis error could persist for many scans depending on the aircraf t f l ight pathand then change abruptly due to a maneuver.

The severity of diffraction induced errors are dramatica lly i l lus t ra tedduring flight test ing at the Lincoln Laboratory Mode S Experimental Facil i ty(MODSEF) in August 1975. Fig. 6 is an X-Y plot showing the track history oftwo test a i rcraf t flying a planned near-miss encounter behind and below thetop of the Hanscom AFB smokestack. The target r epor ts a re shown as asterisks.The beginning and end points of the l ine segment associated with each reportrepresents respectively the current smoothed position and a predicted 4-secondadvanced position based on monopulse inputs. The actual aircraft f l ight pathsare shown by dashed l ines and th e optical shadowing extent of th e smokestacki s i l lustrated by th e cross-hatched area. The actual azimuth position ofaircraf t 1 and 2 with respect to the smokestack midpoint varied from +0.8degrees to +1.7 degrees and from +2.4 degrees to +1.7 degrees respectively.

The oscillatory nature of the position estimate with respect to the trueposition is seen to ref lect the same kind of azimuth error behavior observedin F i g ~ 5 in th e region of +0.8 to +2.4 degrees offset . Diffraction in thisinstance was severe enough to seriously degrade th e azimuth estimate.In an analysis of the impact of diffraction on azimuth estimationReference 4 provides s everal bas ic criteria relating obstacle dimension and

range to th e size and extent of the azimuth error. A completely accurateprediction of the effect of a complex grouping of structures, such as found ina typical metropolitan skyline, would involve a lengthy process, particularly

11


17/33

23.0 " . . . . -.....--.....- - . , r - - - " " ' T " - - ~ - ~ - - - r - - - nACTUAL TRACK- AIRCRAFT NO. 1

II. 21.5wcoQ02II. 21.0l -ll :0Zj &Q5z

5 " - - ~ - - " ' - - " " ' - - . . I __ - : - -- 4 U -41.7 -42.2 -41.7 -41.2 -40.7 -40.2 -39.7 -39.2

NML WEST OF MODSEFFig. 8. Plot of an Encounter Exercise Showing the Effect of DiffractionFrom a Vertical Obstacle on Aircraft Surveillance.

12


18/33

i f a large number of si te locations were evaluated. However, Reference 4offers some generalized observations based on relatively simple and isolatedgeometric shapes that should be sufficient in providing a fair ly accurate andpractical guideline for siting Mode S to minimize diffraction induced error .Fig. 7 i l lus t ra tes the general relationship between obstacle width asobserved by the sensor and obstacle range for three different values of

maximum peak azimuth error. The aircraf t is assumed to be well below the topof the obstacle and at much greater range. The plots show that the maximumpeak error is reduced th e narrower th e obstacle and the further away i t isfrom the sensor. The location of the maximum peak with respect to themidpoint ~ the obstacle depends primarily on the width of the obstacle (seeFigure 8) and is represented by the following approximation:

Azimuth of Maximum Peak Error (DEG.) 20obstacle width (feet)

An important consideration to Mode S is not only th e maximum peak errorand i ts position but also the to ta l angular region of destructive error abouta given obstacle. As seen in Fig. 5 additional azimuth error peaks occur inan oscil latory fashion as the aircraf t l ine-of-sight moves away from thecenter of the obstacle. The total azimuth extent of corruptible errors (errorwedge) is a more complicated function of obstacle range and particularly widththan is the value and position of the maximum peak error. Generally, for anygiven obstacle width, the azimuth extent of corruptible errors decreases asth e obstacle range is increased unt i l i t become non-existent, i . e . , zero errorcontribution. Table 1 l i s t s th e approximate maximum range for a number ofobstacle widths at which the azimuth error wedge becomes zero and the obstacleis no longer a corrupting influence on Mode S. Also shown is the range atwhich the maximum peak error does not exceed 0 .25 degrees. Note that thenarrower the obstacle is the shorter the range at which the error wedgebecomes insignificant. As the range is decreased both th e azimuth wedge andthe maximum peak azimuth error (Fig. 7) increase in value.

TABLE 1

Values of Obs tac le Range for Zero Azimuth Error Extent (Zero Error)and for 0.25 Degree Peak Error. Aircraft Well BelowTop of Obstacle and at Much Greater RangeObstacleWidth(Feet) Approximate Obstacle Range (Feet) ForZero Wedge-Zero Error 0.25 Degree Error100402010

320001600080004000

13

20000700030001300


19/33

4 0 0 ~ - - - 4 - - - - - - t - - - - - + - - - - - - i - - - - - 7 f " ~ - - - l

.....200

I.L.....:::I:0iw 100..J(,)4(fI )

40

50,000000 10,000 20,000oaSTACLE RANGE (FT)

200020L-_.....JI.:..- ......._ - . . 01 : - - _ - - " . . . .L._ __'_ _

1000

FIg. 7. Maxlnun Azimuth Error (Deg) Aa a Funct ion of Obatacle Width (ft) and Range (ft).

14


20/33

2 "....CJ !\.w ~ X i m u m azimuth e"a,,,, 2 0 / a b s ~ a C l e width..a:0a:a:w:I:I - 0.5:; ) ' \N

Documents

Mode S Installation and Siting Criteria Sandholm_1982_ATC-99a_WW-15318