Grey Wolf based compressive sensing scheme for data gathering in IoT based heterogeneous WSNs

Sensor node energy constraint is considered as an impediment in the further development of the Internet of Things (IoT) technology. One of the most efficient solution is to combine between compressive sensing (CS) and routing techniques. However, this combination faces many challenges that makes it an attractive point for research. This paper proposes an Efficient Multi-hop Cluster-based Aggregation scheme using Hybrid CS (EMCA-CS) for IoT based heterogeneous wireless sensor networks (WSNs). EMCA-CS efficiently combines between CS and routing protocols to extend the network lifetime and reduces the reconstruction error. EMCA-CS includes the following: a new algorithm to partition the field into various hexagonal cells (clusters) and based on multiple criteria, selects a node from each cluster as cluster head (CH). Each CH will then compress its cluster data using hybrid CS method. Also, a new Grey Wolf based algorithm to create optimal path for CHs to deliver the compressed data to base station (BS) and a CSMO-GWO algorithm to optimize the CS matrix construction process is introduced. Moreover, a new Grey Wolf and reversible Greedy based Reconstruction Algorithm is proposed to recover the actual data. The simulation results indicate that the performance of the proposed work exceeds the existing baseline techniques in terms of prolonging WSN lifetime, reducing the power consumption and reducing normalized mean square error.