Intelligent Deep Learning and Softmax Routing for Energy-Efficient Wireless Sensor Networks in Public Space Design

Main Article Content

Mengmeng Chang

Abstract

The increasing usage of several nodes to transfer the massive volume of data to the remotes in wireless sensor networks is a challenging task to reduce the loss. The high volumes of data transmission in wireless sensor networks (WSN) can surpass their capacity, resulting in congestion, latency issues, and packet loss. However, computational intelligence (CI) models can aid in managing and creating intelligent networks in WSN. The WSN congestion issues result in information loss and increased energy usage. CI-based models have been used to resolve this issue, reducing the latency. This paper proposes SoftMax Routing with Deep Neural Network (SRDNN) for efficient routing in WSN. This will route the data packets by choosing the high energy and lower load. It consists of two parts, such as the construction of the routing path, which determines the residual energy of the node. It is analyzed using SoftMax routing to decide whether the node is efficient in energy. The route request and reply established various paths between the source and destination. The path with minimum buffer space and maximum bandwidth is chosen in the optimal routing. The simulation results under the metrics such as energy consumption, data loss rate, throughput, and delay show the proposed model performance.

Article Details

Section
Special Issue - Scalability and Sustainability in Distributed Sensor Networks