Soft Computing Based Energy Efficient Routing Protocol For Hierarchical Wireless Sensor Networks

Abstract

The importance of Wireless Sensor Networks (WSNs) is growing daily. Wide-ranging uses newlineof WSNs are available in all domains, ranging from personal vehicles to medical fields, and from newlinehome applications to defense surveillance monitoring. WSNs are widely applicable and have newlineimpacted several areas. Numerous design goals, such as small node size, low node cost, low power newlineconsumption, self-configuration, scalability, application-specific, fault-tolerant, reliable, secure, newlinechannel utilization, and quality of service support, must be met by WSNs. The two main goals of newlinethis thesis are to establish an energy-efficient discovery procedure and to maximize WSN lifetime newlineand to reduce energy consumption. The approach seeks to provide sustainable settings that contribute newlineto an extended network lifetime. To preserve the multi-hop and hierarchical network structure, newlineclustering is recommended. The analysis uses the Random Forest technique and LSTM to anticipate newlinehow much energy will be needed for routing. Node mobility or the addition of new nodes are newlinepermitted by a reinforcement-based learning strategy without causing the network structure to newlinecollapse. Additionally, in order to minimize control, data, and memory overheads, the effort aims at newlineminimal routing entry. The efficacy of the proposed approach might be assessed by comparing it to newlineother affordable approaches that are presently being used. newlineThe initial investigation suggests the introduction of WSN, the internal architecture of a newlineWSN node, clustering in WSN, the phases of clustering, the application and characteristics of WSN newlineand the difficulties encountered in WSN. An in-depth literature review on the research area explores newlinethe various WSN architectures, communication technology developments that are relevant to these newlinenetworks, and routing protocol nuances. newlineThe research work offers a Deep Q-Learning (DQL)-based protocol for routing in WSNs, newlinebased on Reinforcement Learning (RL)-based node clustering. This routing technique provides newlineclustering and routing that is energy-balanc