A Novel Chaotic Shark Smell Optimization With LSTM for Spatio-Temporal Analytics in Clustered WSN

Abstract

Wireless sensor networks (WSN) include massive deployment of sensor nodes to observe the physical environment. At the same time, the spatial and temporal correlations that come with the collaborative characteristics of the WSN pose considerable benefits in the design of effective communication protocols designed for the WSN environment. At the same time, energy efficiency is considered a vital challenge in WSN and can be solved by the use of clustering techniques. In this aspect, this study presents a new chaotic shark smell optimization (CSSO) with long short-term memory (LSTM), called the CSSO-LSTM technique for spatiotemporal analytics in clustered WSN. Primarily, a chaotic shark smell optimization (CSSO) based clustering technique is derived, which is based on the spatial correlation that exists among the sensor nodes. The CSSO algorithm derives an objective function involving different input parameters to select cluster heads (CHs) and construct clusters.