Fuzzy Adaptive-Sampling Block Compressed Sensing for Wireless Multimedia Sensor Networks

. 2020 Oct 31;20(21):6217. doi: 10.3390/s20216217

Algorithm 2 BCS-SPL Reconstruction Algorithm [38]

Input: The measurement matrix for the

i

-th block:

φ_{i}

, a constant used to control the convergence speed:

λ

, the measurement: y, signal:

x^{(r)}

, 2D transform:

ψ

Output: signal:

x^{(r + 1)}

{\hat{x}}^{(r)} = W i e n e r (x^{(r)})

2: for each block

i

{\hat{\hat{x}}}_{i}^{(r)} = {\hat{x}}_{i}^{(r)} + φ_{i}^{T} (y_{i} - φ_{i} {\hat{x}}_{i}^{(r)})

4: end for
5:

{\overset{ˇ}{\overset{ˇ}{x}}}^{(r)} = ψ {\hat{\hat{x}}}^{(r)}

{\overset{ˇ}{x}}^{(r)} = T h r e s h o l d ({\overset{ˇ}{\overset{ˇ}{x}}}^{(r)}, λ)

{\bar{x}}^{(r)} = ψ^{- 1} {\overset{ˇ}{x}}^{(r)}

8: for each block

i

x_{i}^{(r + 1)} = {\bar{x}}_{i}^{(r)} + φ_{i}^{T} (y_{i} - φ_{i} {\bar{x}}_{i}^{(r)})

10: end for