Data Compression for Wireless Sensor Networks: A Comparative Analysis

Signal Compression very important aspect in the field of Internet of Things (IoT). The existing compression algorithms present overhead to be implemented on low range devices used in Wireless Sensor Networks (WSNs). To overcome this issue some of the compressed sensing schemes have been developed. Few of the schemes include Basis Pursuit (BP), Orthogonal Least Squares (OLS), and Orthogonal Matching Pursuit (OMP) algorithms. This paper tries to provide a comparison of such algorithms by studying their characteristics and their performance with varying sparsity levels of data. The sparse signals are generated using a sparse-plex library, and the experiments are executed using MATLAB for some of the sparsity levels. The experiments show us that as the sparsity level in the data increases, the recovery error increases, and hence the performance of the schemes degrades.