Task ID 1836063 Prof Brian L Evans Wireless Networking and Communications Group Cockrell School of Engineering The University of Texas at Austin bevanseceutexasedu httpwwweceutexasedubevansprojectsplc ID: 531805
Download Presentation The PPT/PDF document "POWERLINE COMMUNICATIONS FOR ENABLING SM..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
POWERLINE COMMUNICATIONS FOR ENABLING SMART GRID APPLICATIONS
Task ID: 1836.063
Prof. Brian L. Evans
Wireless Networking and Communications Group
Cockrell School of Engineering
The University of Texas at Austin
bevans@ece.utexas.edu
http://www.ece.utexas.edu/~bevans/projects/plc
May 3, 2013Slide2
1
Task Description:
Improve
powerline communication (PLC) bit rates for monitoring/controlling applications for residential and commercial energy usesAnticipated Results: Adaptive methods and real-time prototypes to increase bit rates in PLC networksPrincipal Investigator:Prof. Brian L. Evans, The University of Texas at AustinCurrent Students (with expected graduation dates):Ms. Jing Lin Ph.D. (May 2014) Summer 2013 intern at TIMr. Yousof Mortazavi Ph.D. (Dec. 2013)Mr. Marcel Nassar Ph.D. (Aug. 2013) Defended PhD April 15, 2013Mr. Karl Nieman Ph.D. (May 2015) Summer 2013 intern at FreescaleIndustrial Liaisons:Dr. Anuj Batra (TI), Dr. Anand Dabak (TI), Mr. Leo Dehner (Freescale),Mr. Michael Dow (Freescale), Dr. Il Han Kim (TI), Mr. Frank Liu (IBM),Dr. Tarkesh Pande (TI) and Dr. Khurram Waheed (Freescale)Starting Date: August 2010
Task Summary
| Background |
Noise
Modeling and Mitigation |
Testbeds
| ConclusionSlide3
Task Deliverables
2
Date
TasksDec 2010Uncoordinated interference in narrowband PLC: measurements, modeling, and mitigationMay 2011Testbed #1 based on TI PLC modems to investigate receiver improvementsDec 2011Narrowband PLC channel/noise: measurements/modeling
May 2012
Standard-compliant receiver methods (3x bit rate increase)
Dec 2012
Testbed #2 based on Freescale PLC modems to investigate transmitter improvements (2x bit rate increase) On-goingTestbed #3 based on NI equipment to map noise mitigation algorithms onto FPGAsTestbed #4 for two-transmitter two-receiver (2x2) systems based on TI PLC modems to investigate scalability
Task Summary
| Background |
Noise
Modeling and Mitigation |
Testbeds
| ConclusionSlide4
Recent Project HighlightsPaper in Smart Grid Special Issue (Sep. 2012)IEEE Signal Processing Magazine (impact factor 4.066)
Paper on channel impairments, noise, and standards
Co-authored with Dr.
Anand Dabak (TI) and Dr. Il Han Kim (TI)Channel Model Adopted (Oct. 2012)Reference model for IEEE 1901.2 Standard for Low Frequency Narrow Band Power Line Communications for Smart Grid App.Mr. Marcel Nassar, Dr. Anand Dabak (TI), Dr. Il Han Kim (TI), et al.SRC Technical Transfer Talk (Dec. 2012)Best Paper Award (Mar. 2013)2013 IEEE Int. Symp. On Power Line Comm. and Its ApplicationsCo-authored with Dr. Khurram Waheed (Freescale)3Task Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide5
Central power plant
Wind farm
Houses
OfficesHV-MV TransformerIndustrial plantUtility control centerIntegrating distributed energy resourcesSmart metersAutomated control for smart appliancesGrid status monitoringDevice-specific billing4High Voltage (HV)33 kV – 765 kVMedium Voltage (MV)1 kV – 33 kVSmart GridTask Summary | Background
|
Noise
Modeling and Mitigation |
Testbeds | ConclusionSlide6
Smart Grid GoalsAccommodate all generation types
Renewable energy sources
Energy storage options
Improve operating efficienciesScale voltage with energy demandReduce peak demandAnalyze customer load profiles andsystem load snapshotsImprove system reliabilityPower quality monitoringRemote disconnect/reconnectOutage/restoration event notificationInform customer5Source: Jerry Melcher, IEEE Smart Grid Short Course, 22 Oct. 2011, Austin TX USA Enabled bysmart meter communications ISTOCKPHOTO.COM/© SIGAL SUHLER MORANTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide7
Local utility
MV-LV transformer
Smart meters
DataconcentratorHome area data networksconnect appliances, EV charger and smart meter via powerline or wireless linksSmart meter communicationsbetween smart meters and data concentrator via powerline or wireless linksCommunication backhaulcarries traffic between concentrator and utilityon wired or wireless links6Low voltage (LV)under 1 kVSmart Meter CommunicationsTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide8
Use
o
rthogonal frequency division multiplexing (OFDM)
Communication challengesChannel distortionNon-Gaussian noisePowerline Communications (PLC)CategoriesBandBit RatesCoverageEnablesStandardsNarrowband
3-500 kHz
~500 kbps
Multi-kilometer
Smart meter communication (ITU) PRIME, G3 ITU-T G.hnem IEEE P1901.2Broadband1.8-250 MHz~200 Mbps<1500 mHome area data networks
HomePlug
ITU-T G.hn
IEEE P1901
7
Task Summary |
Background
|
Noise
Modeling and Mitigation |
Testbeds
| ConclusionSlide9
FFT in receiver spreads impulsive energy over all tones
Signal-to-noise ratio (SNR) in each
subchannel
decreasesNarrowband PLC systems operate -5 dB to 5 dB in SNRData subchannels carry same number of bits (1-4) in current standardsEach 3 dB increase in SNR on data subchannels could give extra bitOFDM Systems in Impulsive Noise8Task Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide10
Narrowband PLC Systems
Problem:
Non-Gaussian impulsive noise is #1 limitation to communication performance yet traditional communication system design assumes additive noise is Gaussian
Goal: Improve comm. performance in impulsive noiseApproach: Statistical modeling of impulsive noiseSolution #1: Receiver design (standard compliant)Solution #2: Joint transmitter-receiver design9Parametric MethodsNonparametric MethodsListen to environmentNo training necessaryFind model parametersLearn statistical model from communication signal structureUse model to mitigate noise
Exploit
sparsity
to mitigate noise
Task Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide11
Narrowband PLC Impulsive Noise
Cyclostationary Noise
Asynchronous Noise
Example: rectified power suppliesExample: uncoordinated interference10Task Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionIncreases with widespread deploymentDominant in outdoor PLC
Rx ReceiverSlide12
Non-Parametric Mitigation MethodsExploit sparsity
of impulsive noise in time domain
Build statistical model each OFDM symbol
using sparse Bayesian learning (SBL)At receiver, null tones contain only Gaussian + impulsive noiseSNR gain vs. conventional OFDM systems at symbol error rate 10-4Complex, 128-point FFT, QPSK, data tones 33-104, rate ½ conv. codeAsynchronous Gaussian mixture model and Middleton Class A noise11SystemNoiseSBL w/ null tonesSBL w/ all tonesSBL w/ decision feedbackUncodedGMM
8 dB
10 dB
-
MCA6 dB7 dB-CodedGMM
2 dB
7 dB
9 dB
MCA
1.75 dB
6.75 dB
8.75 dB
time
Task Summary | Background |
Noise
Modeling and Mitigation
|
Testbeds
| ConclusionSlide13
Time Domain Interleaving12
Bursts span consecutive OFDM symbols
Bursts spread over many OFDM symbols
InterleaveCoded performance in cyclostationary noiseComplex OFDM, 128-point FFT, QPSK, data tones 33-104, rate ½ conv. codeTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide14
Time-Domain Interleaving13
Burst duty cycle 10%
Burst duty cycle 30%
Time-domain interleaving over an AC cycleCoded performance in cyclostationary noiseTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionCurrent PLC standards use frequency-domain interleaving (FDI)Slide15
Testbed #1
Hardware
Software
NI x86 controllers stream dataNI cards generates/receives analog signalsTI front end couples to power lineTransceiver algorithms in C on x86Desktop LabVIEW configures system and visualizes results14
1x1 Testbed
Task Summary | Background |
Noise
Modeling and Mitigation | Testbeds | ConclusionAdaptive signal processing algorithms for bit loading and interference mitigationSlide16
G3 link using two Freescale PLC modemsFreescale software tools allow frame-by-frame analysisTest setup allows synchronous noise injection into power line
Testbed
#2: Noise Playback/Analysis
Freescale PLC G3-OFDM ModemFreescale PLC TestbedOne modem to sample powerline noise in fieldCollected 16k 16-bit 400 kS/s at each locationTask Summary | Background | Noise Modeling and Mitigation | Testbeds | Conclusion15Slide17
Analyzed cyclic properties of PLC noise measurementsDeveloped cyclic bit loading method for transmitter
Receiver measures noise power over half AC cycle
Feedback modulation map to transmitter
Allocate more bits in higher SNR subchannelsTestbed #2: Cyclic Power Line Noise2x increase in bit rateWon Best Paper Award at ISPLCTask Summary | Background | Noise Modeling and Mitigation | Testbeds | Conclusion16Slide18
Testbed
#3: FPGA Implementation
Built NI/
LabVIEW testbed with real-time link (G3 settings)Redesigned parametric impulsive noise mitigation algorithmConverted matrix operations to distributed calculations on scalarsBased on approximate message passing (AMP) frameworkMapped transceiver to fixed-point data/math using MatlabSynthesis: LabVIEW DSP Diagram to Xilinx Vertex 5 FPGAsReceived QPSK constellation at equalizer outputconventional receiverwith AMPTask Summary | Background | Noise Modeling and Mitigation | Testbeds | Conclusion17UtilizationTrans.Rec.AMP+EqFPGA123total slices32.6%64.0%94.2%slice reg.15.8%39.3%59.0%slice LUTs17.6%42.4%71.4%DSP48s2.0%7.3%27.3%blockRAMs7.8%18.4%29.1%Slide19
Testbed #4: 2x2 PLC (On-Going)Goal: Improve communication performance by another 2xOne phase, neutral, ground for 2x2 differential signalingCrosstalk between two channels due to energy coupling
18
Frequency response of a direct channel
Crosstalk highly correlated with direct channel responseTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide20
ConclusionPLC systems are interference limitedStatistical models for interferenceCyclostationary model for impulsive noise synchronous to AC cycle
Gaussian mixture model for asynchronous impulsive noise
Interference management
Cyclic bit loading to double bit rates in cyclostationary noiseTime-domain interleaving to mitigate cyclostationary noise followed by receiver impulsive noise mitigationMapping impulsive noise mitigation algorithms to FPGAsPoor: Non-parametric sparse Bayesian learning algorithmsGood: Parametric distributed approximate message algorithms19http://users.ece.utexas.edu/~bevans/projects/plc/index.htmlTask Summary | Background | Noise Modeling and Mitigation | Testbeds | ConclusionSlide21
Our Publications
Tutorial/Survey Article
M.
Nassar, J. Lin, Y. Mortazavi, A. Dabak, I. H. Kim and B. L. Evans, “Local Utility Powerline Communications in the 3-500 kHz Band: Channel Impairments, Noise, and Standards”, IEEE Signal Processing Magazine, Special Issue on Signal Processing Techniques for the Smart Grid, Sep. 2012. Impact Factor 4.066.Journal PaperJ. Lin, M. Nassar and B. L. Evans, “Impulsive Noise Mitigation in Powerline Communications using Sparse Bayesian Learning”, IEEE Journal on Selected Areas in Communications, Special Issue on Smart Grid Communications, Jul. 2013. Impact Factor 3.413.Conference Publications (more on next slide)J. Lin and B. L. Evans, “Non-parametric Mitigation of Periodic Impulsive Noise in Narrowband Powerline Communications”, Proc. IEEE Global Communications Conference, Dec. 2013, Atlanta, GA USA, submitted.20Slide22
Our Publications
Conference Publications
(more on next slide)
M. Nassar, P. Schniter and B. L. Evans, “Message-Passing OFDM Receivers for Impulsive Noise Channels”, Proc. Asilomar Conf. on Signals, Systems, and Computers, Nov. 2013, Pacific Grove, CA, submitted. K. F. Nieman, M. Nassar, J. Lin and B. L. Evans, “FPGA Implementation of a Message-Passing OFDM Receiver for Impulsive Noise Channels”, Proc. Asilomar Conf. on Signals, Systems, and Computers, Nov. 2013, Pacific Grove, CA, submitted. K. Nieman, J. Lin, M. Nassar, K. Waheed, and B. L. Evans, “Cyclic Spectral Analysis of Power Line Noise in the 3-200 kHz Band”, Proc. IEEE Int. Sym. on Power Line Comm. and Its App., Mar. 2012, Johannesburg, South Africa. Best Paper Award. J. Lin and B. L. Evans, “Cyclostationary Noise Mitigation in Narrowband Powerline Communications”, Proc. APSIPA Annual Summit and Conf., invited paper, Dec. 2012, Hollywood, CA USA.M. Nassar, A. Dabak, I. H. Kim, T. Pande and B. L. Evans, “Cyclostationary Noise Modeling In Narrowband Powerline Communication For Smart Grid Applications”, Proc. IEEE Int. Conf. on Acoustics, Speech, and Signal Proc., Mar. 2012, Kyoto, Japan21Slide23
Our Publications
Conference Publications
(more on next slide)
M. Nassar, K. Gulati, Y. Mortazavi, and B. L. Evans, “Statistical Modeling of Asynchronous Impulsive Noise in Powerline Communication Networks”, Proc. IEEE Int. Global Communications Conf., Dec. 2011, Houston, TX USA. J. Lin, M. Nassar and B. L. Evans, “Non-Parametric Impulsive Noise Mitigation in OFDM Systems Using Sparse Bayesian Learning”, Proc. IEEE Int. Global Communications Conf., Dec. 2011, Houston, TX USA. Standards ContributionA. Dabak, B. Varadrajan, I. H. Kim, M. Nassar, and G. Gregg, “Appendix for noise channel modeling for IEEE P1901.2”, IEEE P1901.2 Std., June 2011, doc: 2wg-11-0134-05-PHM5. Adopted as reference noise model in Oct. 2012 ballot.22Slide24
Thank you for your attention…Questions?
23Slide25
Backup Slides
24Slide26
Today’s Power Grids in USA7 large-scale power grids each managed by a regional utility company700 GW generation capacity in total for long-haul high-voltage power transmission
Synchronized independently, and exchange power via DC transfer
130+ medium-scale power grids each managed by a local utility
Local power distribution to residential, commercial and industrial customersHeavy penalties in US for blackouts (2003 legislation)Utilities generate expected energy demand plus 12%Energy demand correlated with time of dayEffect of plug-in electric vehicles (EVs) on energy demand uncertainGeneration cost 30x higher during peak times vs. normal loadTraditional ways to increase capacity to meet peak demand increaseBuild new large-scale power generation plant at cost of $1-10B if permit issuedBuild new transmission line at $0.6M/km which will take 5-10 years to complete25Source: Jerry Melcher, IEEE Smart Grid Short Course, 22 Oct. 2011, Austin TX USA Slide27
Comparison of Wireless & PLC Systems26
Wireless Communications
Narrowband
PLC (3-500 kHz)Time selectivityTime-selective fading and Doppler shift (cellular)Periodic with period of half AC main freq. plus lognormal time-selective fadingPower loss vs. distance dd –n/2 where n is propagation constante – a(f) d plus additional attenuation when passing through transformersPropagationDynamically changingDeterminism from fixed grid topologySynchronizationVariesAC main power frequencyAdditive noise/ interferenceAssumed stationaryand GaussianGaussian plus non-Gaussian noise dominated by cyclostationary componentAsynchronous interferenceUncoordinated users in Wi-Fi bands;Frequency reuse in cellularDue to power electronics and uncoordinated users using other standardsMIMOStandardized forWi-Fi and cellularNumber of wires minus 1;G.9964 standard for broadband PLCSlide28
PLC Noise Scenarios
27
Background Noise
Cyclostationary NoiseAsynchronousImpulsive NoiseSpectrally shaped noise
Decreases with frequency
Superposition of lower-intensity sources
Includes narrowband interference
Cylostationary in time and frequencySynchronous and asynchronous to AC main frequencyComes from rectified and switched power supplies (synchronous), and electrical motors (asynchronous)Dominant in narrowband PLCImpulse duration from micro to millisecondRandom inter-arrival time50dB above background noiseCaused by switching transients and uncoordinated interferencePresent in narrowband and broadband PLC
timeSlide29
Cyclostationary Noise
28
Noise Sources
Noise TraceSlide30
Uncoordinated Interference Results
29
General PLC Network
Homogeneous PLC NetworkSlide31
Cyclostationary Noise Modeling
30
Measurement data from UT/TI field trial
Cyclostationary Gaussian Model [Katayama06]Proposed model uses three filters [Nassar12]Adopted by IEEE P1901.2 narrowband PLC standardPeriod is one halfof an AC cycleDemuxs[k] is zero-mean Gaussian noiseSlide32
Asynchronous Noise Modeling
31
Ex. Rural areas, industrial areas w/ heavy machinery
Dominant Interference SourceMiddleton Class ADistribution [Nassar11]Homogeneous PLC NetworkEx. Semi-urban areas, apartment complexesGeneral PLC NetworkEx. Dense urban and commercial settingsGaussian MixtureModel [Nassar11]Middleton Class ADistribution [Nassar11]Middleton Class A is a special case of the Gaussian Mixture Model.Impulse rate lImpulse duration mli = l, mi = m, g(di) = g0
l
i
,
mi, g(di) = gi Slide33
Parametric vs. Nonparametric Methods
Parametric
Nonparametric
Must build a statistical model of the noiseYesNoRequires training data to compute model parametersYes
No
Degrades in performance due to model mismatch
Yes
NoHas high complexity when receiving message dataNoYes
32Slide34
Sparse in time domain
Learn statistical model
Use sparse Bayesian learning (SBL)
Exploit sparsity in time domain [Lin11]SNR gain of 6-10 dBIncreases 2-3 bits per tone for same error rate - OR -Decreases bit error rate by 10-100x for same SNR Asynchronous Noise33~10dB~6dBtimeTransmission places 0-3 bits at each tone (frequency). At receiver, null tone carries 0 bits and only contains impulsive noise.Slide35
Performance w/o Error CorrectionCS+LS: [Caire08] MMSE: [Haring02] SBL: [Lin11]
NSI
Gaussian mixture
model noise34ProposedNon-parametric methods in blueParametric methods in redSlide36
Performance w/ Error CorrectionNSI
Gaussian mixture
model noise
35NSIProposedNon-parametric methods in blueParametric methods in redSlide37
Power Line Noise at Residential Site
complex spectrum
f
= 30-120 kHznarrowbandf = 140 kHzfrequency sweepf = 170 kHz36Slide38
Analysis of Residential Noise
though spectrally complex, many components have strong
stationarity
at 120 Hz37Slide39
Testbed #1
Quantify application performance vs. complexity tradeoffs
Extend our real-time DSL
testbed (deployed in field) Integrate ideas from multiple narrowband PLC standards Provide suite of user-configurable algorithms and system settings Display statistics of communication performanceInvestigate Adaptive signal processing algorithms Improved communication performance 2-3x38Slide40
Message-Passing OFDM Receiver
RT controller
LabVIEW RT
data symbol generationFlexRIO FPGA Module 1 (G3TX)LabVIEW DSP Design Moduledata and reference symbol interleavereference symbol LUT43.2 kSps
8.6 kSps
zero
padding
(null tones)generatecomplex conjugate pair103.6 kSps256 IFFT w/ 22 CP insertion368.3 kSpsNI 5781
16-bit DAC
1
0 MSps
RT controller
LabVIEW RT
BER/SNR calculation w/
and w/o AMP
FlexRIO
FPGA Module 2 (G3RX)
LabVIEW DSP Design Module
NI 5781
14-bit ADC
sample
rate conversion
1
0 MSps
400
kSps
time and frequency offset correction
400
kSps
256 FFT
w/ 22 CP removal
368.3
kSps
FlexRIO
FPGA Module 3 (AMPEQ)
LabVIEW DSP Design Module
n
ull tone
and active tone separation
184.2 kSps
51.8 kSps
ZF channel estimation/
equalization
AMP noise estimate
Subtract noise estimate from active tones
data and reference symbol de- interleave
51.8 kSps
8.6 kSps
Host Computer
LabVIEW
43.1 kSps
43.1 kSps
sample
rate conversion
400
kSps
51.8 kSps
256 FFT, tone select
51.8 kSps
368.3 kSps
testbench
control/data visualization
differential MCX pair
Example Input Noise
Resource Utilization
39