Title: Multiscale Geometric Analysis
1A Single-letter Characterization of Optimal Noisy
Compressed Sensing
Dongning Guo Dror Baron Shlomo Shamai
2Setting
- Replace samples by more general measurements
based on a few linear projections (inner products)
sparsesignal
measurements
non-zeros
3Signal Model
- Signal entry Xn BnUn
- iid Bn Bernoulli(?) ? sparse
- iid Un PU
PX
Bernoulli(?)
Multiplier
PU
4Measurement Noise
- Measurement process is typically analog
- Analog systems add noise, non-linearities, etc.
- Assume Gaussian noise for ease of analysis
- Can be generalized to non-Gaussian noise
5Noise Model
- Noiseless measurements denoted y0
- Noise
- Noisy measurements
- Unit-norm columns ? SNR?
noiseless
SNR
6Allerton 2006 Sarvotham, Baron, Baraniuk
- Model process as measurement channel
- Measurements provide information!
7Single-Letter Bounds
- Theorem Sarvotham, Baron, Baraniuk 2006
- For sparse signal with rate-distortion function
R(D), lower bound on measurement rate s.t.
SNR ? and distortion D - Numerous single-letter bounds
- Aeron, Zhao, Saligrama
- Akcakaya and Tarokh
- Rangan, Fletcher, Goyal
- Gastpar Reeves
- Wang, Wainwright, Ramchandran
- Tune, Bhaskaran, Hanly
8Goal Precise Single-letter Characterization of
Optimal CS
9What Single-letter Characterization?
?
,?
channel
posterior
- Ultimately what can one say about Xn given Y?
- (sufficient
statistic) - Very complicated
- Want a simple characterization of its quality
- Large-system limit
10Main Result Single-letter Characterization
- Result1 Conditioned on Xnxn, the observations
(Y,?) are statistically equivalent to - ? easy to compute
- Estimation quality from (Y,?) just as good as
noisier scalar observation
?
,?
channel
posterior
degradation
11Details
- ?2(0,1) is fixed point of
- Take-home point degraded scalar channel
- Non-rigorous owing to replica method w/ symmetry
assumption - used in CDMA detection Tanaka 2002, Guo Verdu
2005 - Related analysis Rangan, Fletcher, Goyal 2009
- MMSE estimate (not posterior) using Guo Verdu
2005 - extended to several CS algorithms particularly
LASSO
12Decoupling
13Decoupling Result
- Result2 Large system limit any arbitrary
(constant) L input elements decouple - Take-home point interference from each
individual signal entry vanishes
14Sparse Measurement Matrices
15Sparse Measurement Matrices Baron, Sarvotham,
Baraniuk
- LDPC measurement matrix (sparse)
- Mostly zeros in ? nonzeros P?
- Each row contains ¼Nq randomly placed nonzeros
- Fast matrix-vector multiplication
- fast encoding / decoding
sparse matrix
16CS Decoding Using BP Baron, Sarvotham,
Baraniuk
- Measurement matrix represented by graph
- Estimate input iteratively
- Implemented via nonparametric BP
Bickson,Sommer,
signal x
measurements y
17Identical Single-letter Characterization w/BP
- Result3 Conditioned on Xnxn, the observations
(Y,?) are statistically equivalent to - Sparse matrices just as good
- BP is asymptotically optimal!
identical degradation
18Decoupling Between Two Input Entries (N500,
M250, ?0.1, ?10)
density
19CS-BP vs Other CS Methods (N1000, ?0.1, q0.02)
MMSE
CS-BP
M?
20Conclusion
- Single-letter characterization of CS
- Decoupling
- Sparse matrices just as good
- Asymptotically optimal CS-BP algorithm