Secured energy optimization of wireless sensor nodes on edge computing platform using hybrid data aggregation scheme and Q-based reinforcement learning technique

Wireless Sensor Network (WSN) security and energy consumption is a potential issue. WSN plays an important role in networking technologies to handle edge devices on a heterogeneous edge computing platform. For faster processing of sensor nodes on an Industrial Internet of Everything (IIOE), an efficient computing technique for an emerging networking technology is being explored. As a result, the proposed study provides a chaotic mud ring-based elliptic curve cryptographic (CMR_ECC)-based encryption solution for WSN security. In the proposed WSN environment, various sensor nodes are deployed to collect data. To enhance the network lifetime, the nodes are combined into clusters, and the selection of cluster heads is performed with a fuzzy logic-based osprey algorithm (FL_OA). After the encryption process, the most optimal key selection process is performed with a hybrid chaotic mud ring algorithm, and the encrypted data are optimally routed to varied edge servers with a hybrid Chebyshev Gannet Optimization (CGO) approach. The data aggregation is performed with a Q-reinforcement learning approach. The proposed work is implemented with MATLAB. For 500, 750, and 1000 WSN sensor nodes, the proposed technique resulted in energy consumption values of 0.28780005 mJ, 0.31141 mJ, and 0.339419 mJ, respectively.