Extreme learning machine optimized by artificial cell swarm optimization for the data fusion modal in WSNs

Wireless Sensor Networks (WSNs) have gained substantial prominence in various applications, like healthcare, and surveillance due to its data gathering ability. Effective data fusion in WSNs plays a pivotal role in removal of redundant information transmission and reliability of the collected information. A data fusion model utilizing extreme learning machine optimized by Artificial Cell Swarm Optimization for mobile heterogeneous wireless sensor networks is proposed for decreasing the redundant data transmission in the network. To drastically decrease the count of network data transmit to the sink node, the neural network of the extreme learning machine (ELM) gathers sensory data via mobile heterogeneous WSN and integrates the received sensor data to the clustering route. The output of ELM is uneven, that affect the efficacy of data fusion. Therefore, Artificial Cell Swarm Optimization is proposed to optimize ELM. The proposed model implemented on MATLAB. The efficiency of the proposed technique is analyzed to the existing techniques. The experimental result proves that the proposed method produce 92.8% in evaluation of energy consumption, 93.1% in evaluation of surviving node, 79.3% in evaluation of network in cluster head, 86.5% in evaluation of energy consumption in cluster head, 85.8% in evaluation of network node balance, and 84.3% in evaluation of data fusion in network compared with existing methods, such as data fusion method under ELM optimized by bat algorithm for mobile heterogeneous WSN (ELM-BA-WSN), Bug algorithm-based data fusion utilizing mobile element for WSN (ELM-EMESP-BA-WSN) and proficient data aggregation and node clustering with ELM for WSN (ELM-CN-WSN) respectively.