Comparative analysis of OMP and ROMP for sparse recovery of temporally correlated data in WSNs

摘要

Greedy algorithms have the advantages of low computational complexity and ease of implementation over the linear programming optimization techniques. A few greedy algorithms work on the principle of selecting a single column whereas, in others, multiple columns are selected in each iteration. In this paper, a comparison is performed on the experimental data which is collected using NI WSN test bed, to find out a more efficient algorithm for sparse signal recovery. Algorithms are compared on the basis of peak signal to noise ratio, root mean square error and speed of recovery. Based on the simulation results, ROMP is preferred over OMP in case of sparse signals with enough sparsity.