Skip to main content

Table 26 Performance analysis of the proposed Battle3 wavelet based sensing matrix

From: Application of 1-D discrete wavelet transform based compressed sensing matrices for speech compression

Length of signal (N)

Number of measurements (m)

Compression ratio (CR = m/N)

Sparsity level = (k/N) × 100 (%)

No. of non-zeros (k)

No. of iterations required

Signal reconstruction time (s)

RMSE

Relative error

SNR (db)

Construction time for sensing matrix (s)

2048

205

0.1

50

1024

21

2.229166

0.0776

1.2960

2.2520

4.184386

2048

410

0.2

50

1024

17

3.000747

0.0619

1.0350

0.2991

4.037021

2048

512

0.25

50

1024

17

6.762783

0.0619

1.0350

0.2984

4.214151

2048

614

0.3

50

1024

17

5.495381

0.0576

0.9626

0.3313

3.779477

2048

849

0.4

50

1024

15

7.920600

0.0538

0.8989

0.9261

3.706719

2048

1024

0.5

50

1024

14

11.331212

0.0538

0.8995

0.9199

3.945231

2048

1229

0.6

50

1024

13

13.909190

0.0485

0.8106

1.8235

3.764331

2048

1434

0.7

50

1024

12

19.831917

0.0468

0.7823

2.1321

3.777555

2048

1536

0.75

50

1024

12

34.010388

0.0468

0.7816

2.1404

4.464107

2048

1638

0.8

50

1024

12

30.235744

0.0468

0.7816

2.1406

3.855705

2048

1843

0.9

50

1024

11

40.190816

0.0468

0.7816

2.1406

4.003610

2048

2048

1.0

50

1024

9

45.789195

0.0468

0.7815

2.1409

3.726738