Gradient Enhanced Regressive Multivariate Artificial Fish Swarm Optimized Data Collection for IoT-Enabled WSN in Smart Environments

The emerging Internet of Things (IoT)-based Wireless Sensor Networks (WSN) consist of small size of sensor nodes for monitoring and collecting data from environmental conditions and it transmits to other sensors through the internet. The major issues in WSN are energy constraints that degrade the efficient functioning and lifetime of WSN. Therefore, a novel technique called Gradient Enhanced Broken-stick Regressive Multivariate Artificial Fish Swarm Optimized Data Collection (GEBRMAFSODC) is introduced. The main objective of the GEBRMAFSODC technique for performing energy-efficient data collection with lesser delay , data loss. Smart cities improve effectiveness of different applications including public transport services. By applying this method, the resource efficient optimal path and the population of artificial fishes (i.e. sensor nodes) is randomly initialized in the search space. For each node, fitness is measured depend on multivariate function namely energy, bandwidth, and distance. The Gradient Enhanced Broken-stick Regression is applied to fitness estimation for analyzing the resources and finding the optimal results. Efficient neighboring nodes are selected to transmit the collected data to sink node via best path. Sink node perform as a data collector with better resource sensor nodes through lesser delay. Simulation is conducted in NS2 simulator using Warrigal Dataset and the performance is analyzed by various parameters namely energy consumption, data collection delay, throughput, and data loss rate based on number of data. The observed result shows the superior performance of the proposed GEBRMAFSODC technique with a higher delivery ratio, throughput by 10%, 48% and lesser loss, delay, and energy consumption by 53%, 37%, and 27% as compared to other related methods respectively.