Cluster-Tree Routing Based Entropy Scheme for Data Gathering in Wireless Sensor Networks

Abstract

Wireless sensor networks (WSNs) have captivated substantial attention from both industrial and academic research since last few years. The major factor behind the research efforts in the field of WSNs is their vast range of applications, such as surveillance systems, military operations, health care, environment event monitoring, and human safety. However, sensor nodes are low potential and energy constraint devices; therefore, energy efficient routing protocol is the foremost concern. In this paper, a new Cluster-Tree routing scheme for gathering data (CTRS-DG) is proposed that composed of two layers: routing and aggregation and reconstruction. In aggregation and reconstruction layer, a dynamic and a self-organizing entropy based clustering algorithm for cluster head (CH) selection and cluster formation is proposed. Data is aggregated and compressed at CHs based on compressive sensing technique. In routing layer, a new proposed algorithm to form the routing tree as backbone of the network is proposed. The routing tree is used to forward the compressed data by CHs to the base station (BS). Finally, as a phase of aggregation and reconstruction layer, an effective CS reconstruction algorithm called Bee based signal reconstruction (BEBR) is proposed to improve the recovery process at the BS. BEBR utilizes the advantages of the greedy algorithm and Bees algorithm to find the optimal solution of reconstruction process. Simulation results reveal that the proposed scheme outperforms existing baseline algorithms in terms of stability period, network lifetime, and average normalized mean squared error for compressive sensing data reconstruction.

Keywords

• Average normalized mean squared error
• clustering-tree based
• compressive sensing
• entropy coefficient
• bees algorithm
• network lifetime