Energy Prediction for Mobile Sink Placement by Deep Maxout Network in WSN

Abstract

In a Wireless Sensor Network (WSN), Numerous cost-effective and energy-constrained sensor nodes are typically used. In a typical Wireless Sensor Network, a single Base Station (BS) gathers information from the whole network, which contributes to concerns including latency, network failure, and congestion. The overwhelming proportion of energy consumption, as well as the energy hole limitation, significantly degrades the overall system performance and network lifetime, which is owing to the sensor nodes that are near the BS consuming more energy. To tackle this problem, it’s essential to determine the perfect spot for mobile sink nodes, which minimizes the power consumed and so increases the network's lifespan. In this work, an effective strategy is designed and developed to detect the location of a mobile sink considering factors such as distance, estimated energy, and fairness, using Deep learning-based energy prediction with an adjacency cell score model. In addition, the predicted energy is determined by employing the Deep Maxout Network (DMN). However, a Minimum distance of 137.364, maximal residual energy of 30.903, maximum standardized fairness of 64.426, maximum network duration of 60, and maximum standardized throughput of 60.613 was obtained using the proposed adjacency-based cell score + Deep Maxout Network.

Combined Osprey-Chimp Optimization for Cluster Based Routing in Wireless Sensor Networks: Improved DeepMaxout for Node Energy Prediction

“…The primary component of DMN's multi-layered structure makes it a form of adaptable activation factor. In this section, positive and negative terms are specified as a nonzero slope through an effective activation function known as Maxout [21]. In general, Maxout supports measures to tackle the optimization challenge by partially defending the hidden elements from transitioning to an abnormal mode.…”