ballard blair [email protected] phd candidate mit/whoi joint program advisor: j im presisig
DESCRIPTION
Comparison and Analysis of Equalization Techniques for the Time-Varying Underwater Acoustic Channel. Ballard Blair [email protected] PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig. 158 t h Meeting of the Acoustical Society of America San Antonio. Channel Model. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/1.jpg)
1
Comparison and Analysis of Equalization Techniques for the
Time-Varying Underwater Acoustic Channel
Ballard [email protected] Candidate
MIT/WHOI Joint ProgramAdvisor: Jim Presisig
10/28/2009
158th Meeting of the Acoustical Society of AmericaSan Antonio
![Page 2: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/2.jpg)
2
Channel Model
10/28/2009
Transmitted Data
Time-varying, linear baseband channel
Baseband noise
Baseband Received Data
+
Matrix-vector Form: Split Channel Convolution Matrix
![Page 3: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/3.jpg)
• Problem Setup:• Estimate using RX data and TX data estimates
• DFE Eq:
• Two Parts:– (Linear) feed-forward filter (of RX data)– (Linear) feedback filter (of data estimates)
Decision Feedback Equalizer (DFE)
10/28/2009 3
Solution to Weiner-Hopf Eq.
MMSE Sol. Using Channel Model
![Page 4: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/4.jpg)
4
DFE Strategies
10/28/2009
Direct Adaptation:
Channel Estimate Based (MMSE):
Equalizer Tap Solution:
![Page 5: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/5.jpg)
5
Central Question:
Why is the performance of a channel estimate based equalizer different than a direct adaptation equalizer?
10/28/2009
![Page 6: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/6.jpg)
6
Comparison between DA and CEB
10/28/2009
• In the past, CEB methods empirically shown to have lower mean squared error at high SNR
• Reasons for difference varied:– Condition number of correlation matrix– Num. of samples required to get good estimate
3dB performance difference between CEB and DA at high SNR
Performance gap due to channel estimation
Similar performance at low SNR
![Page 7: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/7.jpg)
7
Comparison between DA and CEB
• Our analysis shows the answer is: Longer corr. time for channel coefficients than MMSE equalizer coefficients at high SNR
• Will examine low SNR and high SNR regimes– Use simulation to show transition of correlation
time for the equalizer coefficients from low to high is smooth
10/28/2009
![Page 8: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/8.jpg)
8
Low SNR Regime
Update eqn. for feed-forward equalizer coefficients (AR model assumed):
10/28/2009
Approximation:Has same correlation structure
as channel coefficients
![Page 9: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/9.jpg)
9
High SNR
10/28/2009
Reduces to single tap:
Reduced Channel Convolution Matrix: Matrix Product:
Approximation:
![Page 10: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/10.jpg)
10
Correlation over SNR – 1-tap
10/28/2009
AR(1)model
Gaussian model
Channel and Equalizer Coeff. Correlation the Same at low SNR
Equalizer Coeff. Correlation reduces as SNR increases
![Page 11: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/11.jpg)
11
Amplitude of MMSE Eq. Coeff.
10/28/2009
20 dB difference
Tap #
First feed-forward equalizer coefficient has much larger amplitude than others
![Page 12: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/12.jpg)
12
Multi-tap correlation
10/28/2009
Multi-tapAR(1)model
SNR
Strong linear correlation between inverse of first channel tap and first MMSE Eq. tap
![Page 13: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/13.jpg)
13
Take-home Message
• Channel impulse-response taps have longer correlation time than MMSE equalizer taps– DA has greater MSE than CEB
• For time-invariant statistics, CEB and DA algorithms have similar performance– Low-SNR regime (assuming stationary noise)– Underwater channel operates in low SNR regime
(<35dB)
10/28/2009
![Page 14: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/14.jpg)
14
Questions?
10/28/2009
![Page 15: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/15.jpg)
15
Backup Slides
10/28/2009
![Page 16: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/16.jpg)
16
Assumptions
• Unit variance, white transmit data
• TX data and obs. noise are uncorrelated
– Obs. Noise variance:
• Perfect data estimation (for feedback)
• Equalizer Length = Estimated Channel Length Na + Nc = La + Lc
• MMSE Equalizer Coefficients have form:
10/28/2009
![Page 17: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/17.jpg)
17
WSSUS AR channel model
• Simple channel model to analyze• Similar to encountered situations
10/28/2009
![Page 18: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/18.jpg)
18
Direct Path and Bottom Bounce, Time Invariant
Time Varying Impulse Response
10/28/2009
Wavefronts II Experiment from San Diego, CA
Wave interacting paths, highly
![Page 19: Ballard Blair bjblair@mit.edu PhD Candidate MIT/WHOI Joint Program Advisor: J im Presisig](https://reader035.vdocument.in/reader035/viewer/2022070422/568165b1550346895dd8a6f0/html5/thumbnails/19.jpg)
19
Future Work
• Reduce time needed to update channel model– Sparsity– Physical Constraints– Optimization Techniques
• Combine DA and CEB equalizers– Better performance at lower complexity
10/28/2009