A Dictionary-Enhanced Clustering Compressive Sensing Routing Protocol for Large-Scale WSNs

Abstract

Designing an efficient energy-saving routing protocol to optimize network lifespan is a pivotal challenge in large-scale wireless sensor networks (WSNs). In this article, a dictionary-enhanced clustering compressive sensing routing (DEC2R) protocol is designed to conserve energy and provide network load balancing. In DEC2R, the optimal number of clusters for each round is accurately calculated based on the analysis of the optimal cluster size. Through learning the sparse dictionary method, a low-coherence sensing matrix is constructed for data transmission and compression. On this basis, the optimal cluster heads (CHs) are selected based on a cost function (including remaining energy and distance). Non-CHs determine whether to join a cluster based on energy and distance, ultimately completing the clustering formation. In each cluster, data nodes multiply the data by measurement coefficients and transmit it to the CH via the shortest path. Between clusters, each CH forwards the data packet to the next CH along the transmission path. In the end, the sink node receives the entire compressed packets. The simulation results demonstrate the effectiveness of DEC2R. Compared with LEACH, PEGASIS, CDG, and EIREC protocols, dictionary-enhanced clustering compressive sensing routing (DEC2R) significantly extends the lifetime of the network and improves energy efficiency.