An Artificial Neural Network Based Energy Efficient Wireless Detection System to Extend the Lifetime of the Network

Abstract

The eventual growth, as well as successful convergence of various technologies such as wireless technology, nano-electro-mechanical devices (NEMD), and digital electronics, have aided in the realization of detector systems with low-cost, low-powered, tiny detector units for multi-hop, and quick-range communications are considered. The primary problem in Cellular detector Systems (CDS) is the optimum battery energy extraction to prolong the network lifetime. Energy expenses are further increased by the routine input signals, and operating costs sent among the access point and the detector network. Other significant energy challenges are packet costs and successive retransmissions while minimizing overhead fee for control signals. To address this issue, we suggest combining a random sampling technique and an accurate routing method to enable reliable power conservation and rapid transfer of data in the network while trying to minimize control signal overhead expenses. By employing a neural network-based back propagation network technique to anticipate energy-efficient and close-by units in the information-promoting route toward the end point, our study improves the efficiency of packet transmission. Through simulation findings, we were able to show how battery energy consumption could be improved without sacrificing data connection speed, and this was done at a low overhead cost.