Network Optimization of Optical Performance Monitoringwocc2008.aoetek.com/PDF/O21.pdf · 2008-05-05 · Network Optimization of Optical Performance Monitoring ... OPM OPM OPM OPM

WOCC 2008 Centre for Advanced Research in PhotonicsThe Chinese University of Hong Kong 1

Network Optimization of Optical Network Optimization of Optical Performance MonitoringPerformance Monitoring

Lian K. Chen (Lian K. Chen (陳亮光陳亮光))The Chinese University of Hong KongThe Chinese University of Hong Kong

Hong Kong, SAR, ChinaHong Kong, SAR, China


OutlineOutlineMotivationsAdvanced Monitoring Techniques for point-to-point link

OSNR MonitoringChromatic Dispersion MonitoringMulti-impairment Monitoring

Optimization of performance monitoring networksOptimization of Fault-diagnosis with Minimum ProbingAlgebraic Monitoring NetworkOther related works

Conclusions


Optical Performance MonitoringOptical Performance Monitoring

Definition: Physical layer monitoring of the signal quality, for the purpose of determining the health of the signal in the optical domain

OPMOPM

OPMOPM

Applications:Applications: Examples: Examples:

Signal quality characterization

Relating OSNR with BEREarly signal degradation alarm

Fault management Fault detection, localization, and isolationResilience mechanism activation

Active compensation Dynamic CD + PMD monitoring and compensation

Quality of service (QoS) provisioning

SLA fulfillment verificationRef: A. Vukovic, H. Hua, M. Savoie, “Performance Monitoring of Long-haul Networks,” Lightwave Magazine (2002)


Wish List for the Enhancements of OPMWish List for the Enhancements of OPMhigher dynamic range; higher sensitivity;robustness to various effects; multi-impairment monitoring; and lower costlower cost.

But how? ⇒ Optimization of Performance Monitoring System


OSNR monitoring moduleOSNR monitoring module

PM driven by low frequency sinusoidal signalCW & CCW experience periodic phase differenceSignal switched out periodicallyTrack Pmax and Pmin to calculate OSNR

PhaseMod

Power Meter

bandpassfilter isolator

polarization scrambler

polarization controller

Proposed OSNR Monitoring Module

Pmax

PminO

utpu

t pow

er

VPhase Mod.λ

Tran

sfer

func

tion

(dB

)

λsignal

(Ref: Y.C. Ku et al., OFC 2006)

,

,

( / 0.1 )0.1

(1 )[ (1 ) (1 )] 0.1

sig in f

ASE in

f

sig sig

P NEBOSNR dB nm

P nmr NEB

r D D nm

⋅=

⋅

− ⋅=

− − + ⋅

Pmax/Pmin(Pmax-Pmin)/(Pmax+Pmin) when noise is absent


Experimental results: 10Gb/s, back-to-backExperimental results: 10Gb/s, back-to-back

Without PMD:Monitoring error < 0.25 dB40-dB monitoring dynamic range

OSNR by OSA (dB)0 10 20 30 40 50

OSN

R b

y pr

opos

ed s

chem

e (d

B)

0

10

20

30

40

50

Mon

itorin

g Er

ror (

dB)

-4

-2

0

2

4

DGD (ps)0 10 20 30 40 50

Mon

itorin

g er

rror

(dB

)-4

-2

0

2

4

With PMD:Influence of DGD

reference OSNR 25 dB/0.1nmMonitoring error < 0.25 dB


Experimental results: 10Gb/s, 150km, w. ~1.5ps PMDExperimental results: 10Gb/s, 150km, w. ~1.5ps PMD

Influence of chromatic dispersion and transmitted power

An EDFA was added after each 50-km SMF span EDFA output power1 or 10 dBm

Distance (km)0 50 100 150

OSN

R (d

B/0

.1nm

)20

22

24

26

28

30

OSNR by OSA, Pout = 1dBmOSNR by proposed method, Pout = 1dBmOSNR by OSA, Pout = 10dBmOSNR by proposed method, Pout=10dBm


Experimental results: 10Gb/s, back-to-back, w/ polarized noiseExperimental results: 10Gb/s, back-to-back, w/ polarized noise

Influence of partially polarized noiseWorst case: DOP=99.69%

OSNR by OSA (dB)0 10 20 30 40 50

OSN

R b

y pr

opos

ed s

chem

e (d

B)

0

10

20

30

40

50

Mon

itorin

g Er

ror (

dB)

-4

-2

0

2

4

ASE noise // SignalASE noise // SignalASE noise _|_ SignalASE noise _|_ Signal


Chromatic Dispersion (CD) MonitoringChromatic Dispersion (CD) Monitoring


Previous work on chromatic dispersion monitoringPrevious work on chromatic dispersion monitoringClock phase difference monitoring

Pros: No need to modify transmitter, highly sensitiveCons: requires high-speed electronics

Ref.: Q. Yu, et. al., IEEE JLT, 20, 2267-2271 (2002)

We propose to use a birefringent fiber loop (BFL) to realize a simple, polarization insensitive CD monitoring scheme without transmitter side modification, and demonstrate in NRZ system

Ref.: Y.C. Ku, et. al., OFC 2006


Proposed Dispersion Monitoring Module based on BFLProposed Dispersion Monitoring Module based on BFL

Measure RF power at dispersion dependent frequency dip (fRF) produced by birefringent fiber loop (BFL)

PC

Hi-Bi Fiber

3-dB coupler

SMF

IM BPFEDFA

RF Spectrum Analyzer

10Gb/s 1023-1 PRBS

Dispersion Monitoring Module

1550nm

DCF

EDFA

2 22 2sin sin

2 2RFfP D

ϕ ϕ πλΩ −Ω ⎛ ⎞+⎛ ⎞∝ = ⎜ ⎟⎜ ⎟⎝ ⎠ ⎝ ⎠

/ 2 1/ 2RFf π τ= Ω =

W a ve le n g th (n m )

1 5 4 9 .8 1 5 4 9 .9 1 5 5 0 .0 1 5 5 0 .1

Pow

er (d

Bm)

-5 6

-5 4

-5 2

-5 0

-4 8

-4 6

-4 4

-4 2

0.8nm

Ref.: Y.C. Ku, et. al., OFC 2006


Experimental ResultExperimental Result

Measured unambiguous monitoring range: 1500 ps/nm for 10-Gb/s NRZ signalIn principle, monitoring range can reach c/2λ2fRF2 = 2497ps/nm

Dispersion (ps/nm)

-1000 -500 0 500 1000 1500 2000

Pow

er a

t 5G

Hz

(dBm

)

-34

-32

-30

-28

-26

-24

-22

0ps

110ps

450ps


Multi-impairment MonitoringMulti-impairment Monitoring


Multi-impariment monitoring - Delay Tap Asynchronous Waveform SamplingMulti-impariment monitoring - Delay Tap Asynchronous Waveform Sampling

Combines asynchronous sampling with two tap delay lines, so thateach sample point comprises two measurements (x and y), separated by a fixed time corresponding to the delay length.

Sarah D. Dods and Trevor B. Anderson, “Optical Performance Monitoring Technique UsingDelay Tap Asynchronous Waveform Sampling”, OFC OThP5, 2006

Ts Ts

x1

x2

x3y1

y2

y3Δt

x

y

Δt=bit period

Low Speed

A/D ΔtCPU


Application of asynchronous sampling in PMD monitoringApplication of asynchronous sampling in PMD monitoring

The proportion of sample points on the left and bottom edge is used as the effective parameters to evaluate PMD value

0ps 10ps 20ps 30ps

40ps 50ps 60ps 70ps

80ps 90ps

number of sample points falling on the left and the bottom edges total number of sample points on the scatter plot

m =

DGD (ps)

0 20 40 60 80

m: p

ropo

rtion

of s

ampl

e po

ints

falle

don

left

and

botto

m e

dges

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00


Application of asynchronous sampling in misalignment monitoring for RZ-OOKApplication of asynchronous sampling in misalignment monitoring for RZ-OOK

The increasing dispersion and rotation of the sample points on the diagonal can serve as effective parameters to evaluate alignment status

0ps-30ps-40ps-50ps +30ps +40ps +50ps

2 21 sin arctan4

ii i

i i

yd x yn x

π⎛ ⎞= + −⎜ ⎟⎜ ⎟

⎝ ⎠∑

1 arctan ii i

ytn x

= ∑

Ref.: Y.C. Ku, et. al., ECOC 2006

d: deviation from diagonal linet: average angle of sample points


Experimental resultsExperimental results

Sign and amount of misalignment can both be estimated

Timing Misalignment (ps)-50 -40 -30 -20 -10 0 10 20 30 40 50

d: d

evia

tion

from

dia

gona

l (a.

u.)

6.0e-4

8.0e-4

1.0e-3

1.2e-3

1.4e-3

1.6e-3

1.8e-3

t: av

erag

e an

gle

of s

ampl

e po

ints

(rad

)

0.72

0.74

0.76

0.78

0.80

0.82

0.84

0.86


Correlation of Monitored ParametersCorrelation of Monitored ParametersMonitored parameters may be correlated.The monitoring schemes need to be able to differentiate different possible impairments

Question: Can we derive one parameter based on the measurement of the other parameters?How about other dimensions of correlations, e.g. spatial & temporal?

Ref: ITU-T G.697 Optical monitoring for DWDM systems


Optimization of Performance Monitoring SystemOptimization of Performance Monitoring SystemConsider a performance monitoring system SPM that is formed by all the monitoring devices in a network and the corresponding measurements

Optimization of SPM : 1. the minimum number of monitors required2. the optimal locations of the monitors – monitor placement problem3. the minimum number of measurements needed

Issues:static network vs. dynamic (reconfigurable) networkdistributed vs. centralized monitoring

CAPEX

OPEX


Centralized vs. Distributed OPMCentralized vs. Distributed OPM

single monitoring point for the whole network

MonitoringDiagnosis

distributed monitoring/centralized diagnosis

centralized monitoring/centralized diagnosis


Number and Placement of OPM Number and Placement of OPM Optimization of Placement and Amount of OPM needed►One-monitor-per-link/component is not optimum if we consider

♦failure probability of link/component♦scalability with the network size♦amount of redundant alarm

1. Channel-Based: Design Based on Established Lightpaths

1

6

4

2

7

5

3

8

MM

M

Optimal monitor placement : monitorM

1

6

4

2

7

5

3

8

LSP1

LSP2

LSP3

Alarm matrixcomputation

E.g.

Ref: Sava Stanic et al, “Efficient alarm management in optical networks”, in Proc.DARPA Information Survivability Conference and Exposition 2003, pp. 252-260

In-service channel performance based Reduce monitor number considerably For static network only


Number and Placement of OPMNumber and Placement of OPM

Suitable for dynamic networks Reduce monitor number considerably Does not measure in-service channel quality Extra wavelength needed

2. Network-Based: Design Based on Network TopologyE.g. Break down a network into a number of “monitoring cycle” and assign a monitor to each,

using an extra supervisory wavelength to probe the health of the cycle

1

3

4

5

7

6

8

9

2 10

M M

M

M

1

3

4

5

7

6

8

9

2 10

Cycle findingalgorithm

Ref: Hongqing Zeng et al, “Monitoring cycles for faultdetection in meshed all-optical networks”, in Proc. ICPPW’04, pp. 434-439


Scalable Network Diagnosis for All-Optical Networks via Group Testing Over GraphsScalable Network Diagnosis for All-Optical Networks via Group Testing Over Graphs

Group Test: r = T(Xa,Xb,Xe,X1,X8)

X1

X2

X3

X4

X5

X6 X7 X8

X9

Xa Xb

Xc

XdXe

Xf

Xa Xb

XcXf

Xe Xd

OXCIP

OXCIP

OXCIP

OXCIP

OXCIP

OXCIP

Y. G. Wen, V.W. S. Chan and L. Z. Zheng, "Efficient Fault Diagnosis Algorithms for All-Optical WDM Networks with Probabilistic Link Failures (Invited Paper)," Journal of Lightwave. Technology, vol. 23, no. 10, October 2005, pp.3358-3371.

The Minimum Number of Measurements Needed

Analogy: fake coins problem

Objective: to minimize the number of measurements to identify the fake coin(s) that are of less weight

http://web.mit.edu/eewyg/www/JLT-05.pdfhttp://web.mit.edu/eewyg/www/JLT-05.pdfhttp://web.mit.edu/eewyg/www/JLT-05.pdf


Asymptotically Optimal Run-Length Probing Schemes via an Information Theoretic ApproachAsymptotically Optimal Run-Length Probing Schemes via an Information Theoretic Approach

00100000000001000000

F

F

S

S

SF

S

F

F

S

Fault Detection

Fault Localization

Probe Syndrome FFSSSFSFFS 1100010110

Guidelines for Fault Diagnosis Algorithms• The total number of probing is

approximately equal to the entropy of the network state

• Each probe should provide approximately 1 bit of state information

21

36

45

7 8

0

9


Network Diagnosis – an algebraic approachNetwork Diagnosis – an algebraic approachObjective: To find an efficient way to determine the quality of different paths, so that it can be used as a metric for path computation in the network layer for channel setup

For example:Two existing monitoring channels c1 and c2New request to setup c3NO estimation

S.T. Ho, L.K. Chen, C.K. Chan, “On Requirements of Number and Placement of Optical Monitoring Modules in All-Optical Networks,” OECC, paper 7A2-2 (2005)

Network diagnosis that explores the spatial correlation to minimize the number of monitors


Linear expressions of optical impairmentsLinear expressions of optical impairmentsNoise Figure (NF)

Chromatic dispersion (CD)

Polarization-mode-dispersion (PMD)

321

2 3

koverall

c c ck

NF NFNFNF NFG G G

= + + + +L

Overall impairment for all paths can be derived from impairments of individual fiber link and component

NFi : Noise figure of the ith hop

Gci : cumulative gain up to ith hop

Δλ : spectral width

D: dispersion parameter

Li : total length of ith hop

1 2 3overall NNF NF NF NF NF= + + + +L

2 2 2= +∑ ∑overall fiber componentPMD PMD PMD

1 1 2 2( )overall k kCD L D L D L Dλ= Δ ⋅ ⋅ + ⋅ + + ⋅L


A sample networkA sample network

(The monitoring module contain both monitoring source and monitoring detector.)


Problem formulationProblem formulation

For the previous example, total 9 impairment variables to be determined and 2monitoring modules are used 9:2

Assumption: monitoring modules are able to monitor different probing channels from all ports by time division manner or by proper labeling

0 0 0 1 0 0 0 0 00 0 1 0 0 0 0 0 00 1 1 0 0 0 0 1 0

⎡ ⎤⎢ ⎥⎢ ⎥=⎢ ⎥⎢ ⎥⎣ ⎦M

Μ

1 11 1 1

2 2

3 3

1

N

N N NN N

y m m xy xy x

y m m x

⎡ ⎤ ⎡ ⎤ ⎡ ⎤⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥=⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎢ ⎥ ⎢ ⎥ ⎢ ⎥⎣ ⎦ ⎣ ⎦ ⎣ ⎦

L L L

M O M

M O M

M M O M M

L L L

y = Mx -1x = M y

Q: What is the minimum number of monitoring modulesand where should they be placed?

M: Monitoring Matrix

x: individual impairment

y: Accumulated impairment


Bus & Ring networksBus & Ring networks


Two-link-connected networkTwo-link-connected network

= − −( ) ( ) ( )green redImpairment c Impairment c Impairment p x


Summary of Algebraic Network DiagnosisSummary of Algebraic Network DiagnosisProposed a novel algebraic approach of monitoring optical impairments in both un–directional nodes and links

Proved that two monitoring modules are necessary and sufficient to support our monitoring scheme in two–link–connected network

For every node that becomes a leaf node after removal of all leave nodes in the original network, installing one monitoring module at it is necessary and sufficient to support the proposed probing scheme

For general networks with bridges, first transform the network to a tree-like structure and then solve it by “divide-and-conquer” method

With the probing scheme, any single-fault can be located in the whole network


Intelligent Fault Diagnosis for Optical Networks: Detection, Isolation and PrognosisIntelligent Fault Diagnosis for Optical Networks: Detection, Isolation and Prognosis

Data CollectionNetwork

Network

Fault Detection

Fault Isolation

Fault Prognostics

Operations & Maintenance

Correction & Restoration

ModelDifference/ Residues

Sensor Readings

Decision & Classification

Knowledge & Training

Inference & Estimate

Sensor Readings

Decision & Classification

Model-based Approach Data-driven ApproachYG Wen, MIT

Two fault-diagnosis formulation approaches:• model-based• data-driven


ConclusionsConclusionsOPM in next-generation high-speed transparent reconfigurable networks is essential

Enhancements of OPM, including higher dynamic range, higher sensitivity, robustness to various effects, and multi-impairment monitoring, are desirable to make the OPM system more effective.

By exploring the correlation of different spatial or time domaininformation, it is possible to further enhance the efficiency of OPM.

Some works on the Optimization of OPM network to derive theoretical bounds for the min. number of monitoring module or the number ofprobing are presented.


Acknowledgement:Projects supported in part by Hong Kong SAR Government CERG Grant 411006Contributors S.T. Ho, Z.C. S.T. Ho, Z.C. XieXie, Y.C. Ku, C.K Chan, and Y.G. , Y.C. Ku, C.K Chan, and Y.G. WenWen

The End

OutlineOptical Performance MonitoringWish List for the Enhancements of OPMOSNR monitoring moduleExperimental results: 10Gb/s, back-to-backExperimental results: 10Gb/s, 150km, w. ~1.5ps PMDExperimental results: 10Gb/s, back-to-back, �w/ polarized noiseChromatic Dispersion (CD) MonitoringPrevious work on chromatic dispersion monitoringProposed Dispersion Monitoring Module based on BFLExperimental ResultMulti-impairment MonitoringMulti-impariment monitoring - Delay Tap Asynchronous Waveform SamplingApplication of asynchronous sampling in PMD monitoringApplication of asynchronous sampling in misalignment monitoring for RZ-OOKExperimental resultsCorrelation of Monitored ParametersOptimization of Performance Monitoring SystemCentralized vs. Distributed OPMNumber and Placement of OPM Number and Placement of OPMScalable Network Diagnosis for All-Optical Networks via Group Testing Over GraphsAsymptotically Optimal Run-Length Probing Schemes via an Information Theoretic ApproachNetwork Diagnosis – an algebraic approachLinear expressions of optical impairmentsA sample networkProblem formulationBus & Ring networksTwo-link-connected networkSummary of Algebraic Network DiagnosisIntelligent Fault Diagnosis for Optical Networks: Detection, Isolation and PrognosisConclusions

Documents

Network Optimization of Optical Performance Monitoringwocc2008.aoetek.com/PDF/O21.pdf · 2008-05-05 · Network Optimization of Optical Performance Monitoring ... OPM OPM OPM OPM