23
6.829 Computer Networks 1 Compressed Sensing for Loss-Tolerant Audio Transport Clay, Elena, Hui

6.829 Computer Networks1 Compressed Sensing for Loss-Tolerant Audio Transport Clay, Elena, Hui

Embed Size (px)

Citation preview

6.829 Computer Networks 1

Compressed Sensing for Loss-Tolerant Audio Transport

Clay, Elena, Hui

6.829 Computer Networks 2

Introduction to CS

Basic idea:

Given a signal S of length d (large)

S can be recovered from a much smaller measurement vector v ! ( if S is sparse )

Sparse compressedsignal

6.829 Computer Networks 3

Introduction to CS

signal: s= (0, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0, 0, 1)

measurements: projections of s onto some small number of basis vectors

Questions: 1. what basis vectors?2. how many measurements are enough?

6.829 Computer Networks 4

Intro to CS

Sometimes imperfection is OK! We only want to have to transmit enough for a “reasonable” reconstruction.

Reduce the number of bits used to transmit a signal

6.829 Computer Networks 5

Motivation

Direct applicability to low-power sensor networks (data is sparse)

Applications to medical imaging

How does CS apply to audio signals?

6.829 Computer Networks 6

CS and sound reconstruction

Compressed Sensing is:

loss-tolerant

universal

But:

is it practical? Particularly for audio?

how about quality of reconstructed sound?

6.829 Computer Networks 7

Approach/contributions

- Use a modified version of the classical Orthogonal Matching Pursuit

1. optimized the main iterative step2. dealt with MATLAB memory overflow for

matrix storage3. split original large data samples into

smaller frames and combine at the end4. Quantify relationship between quality and

compression parameters m, c.

6.829 Computer Networks 8

Parameters

• m: sparsity level of original data

• d: data space dimension

• N: # of measurements

N= c m ln(d)

6.829 Computer Networks 9

OMP(Orthogonal Matching Pursuit)

• InputΦ: N x d measurement matrixv: N-dimensional data vectorm: data sparsity

• Outputs: estimated signal in Rd

• Procedure

v= Φ * s

6.829 Computer Networks 10

OMP Procedure

Determine which columns of Φ participate in the measurement vector v, in greedy fashion.1. Initialization 2. IterationIn each iteration, choose one column Φ that is most strongly correlated with the remaining part of v. Then we subtract off its contribution to v and iterate on the residual.3. ReconstructionUse the chosen columns of Φ and approximation to reconstruct the signal.

6.829 Computer Networks 11

I-OMP on Audio Signal Recovery

• Original sound signal (Source: s4d.wav)

• Reconstructed by setting m = 256 and 500

6.829 Computer Networks 12

Tests

Test the impact of the parameters m, c on the quality of the reconstruction

Method: MOS (Mean Opinion Score)

6.829 Computer Networks 13

Sparsity and MOS

m as fraction of number of samples

MO

S

scor

e

6.829 Computer Networks 14

Quality of reconstruction

Sum of squared differences between original and reconstructed signal

m = 1233

d = 8821

c

6.829 Computer Networks 15

Piecewise Compression

• Original:

• Recovered:

• MOS = 2.8

6.829 Computer Networks 16

I-OMP on image recovery

Different m = 256, 512, 1024

Source: moon.bmp

6.829 Computer Networks 17

I-OMP on Image Recovery

• Different value of parameter c

• Original, c=2,4,20

6.829 Computer Networks 18

The End

6.829 Computer Networks 19

• 1. Initialization residual r = v; Index set Λ = empty;

• 2. Iteration

• 3. Reconstruction

6.829 Computer Networks 20

OMP Procedure• 1. Initialization• 2. Iteration

For t=0: m-1• Find the index λ that solves• λ= arg max j=1,…,d |<r,φj>|• Λ = Λ U {λ}• Re-compute projection P on φΛ.

A = P* vr = v - A

• 3. Reconstruction

6.829 Computer Networks 21

OMP Procedure

• 1. Initialization

• 2. Iteration

• 3. Reconstruction

The estimate s for the ideal signal has non-zero coefficients sλ at the components li

sted in Λ.

A = Σ λ∈Λ φλ* sλ

6.829 Computer Networks 22

Iterative OMP

• 1. Initializationr = v;s = 0d;

• 2. IterationFor t=0: m-1

• Find the index λ that solves λ= arg max j=1,…,d |<r,φj>|

• sλ = <r, φλ >/ || φλ ||2

• r = r - sλ * φλ

• A= A + sλ * φλ

• 3. Reconstruction

6.829 Computer Networks 23

Iterative OMP -2