Compressed Sensing with Incremental Sparse Measurements

This paper proposes a verification-based decoding approach for reconstruction of a sparse signal with incremental sparse measurements. In its first step, the verification-based decoding algorithm is employed to reconstruct the signal with a fixed number of sparse measurements. Often, it may fail as the number of sparse measurements may be not enough, possibly due to an underestimate of the signal sparsity. However, we observe that even if this first recovery fails, many component samples of the sparse signal have been identified. Hence, it is natural to further employ incremental measurements tuned to the unidentified samples with known locations. This approach has been proven very efficiently by extensive simulations.