Energy Harvesting Technique for Improved WSN Performance: Application of IOT and Deep Learning
Main Article Content
Abstract
Due to technological advances, real-time automation systems using IoT-based sensor networks increasingly need faultless data transmission over wireless sensor networks (WSN). Nowadays, Wireless Multimedia Sensor Networks (WMSNs) have taken precedence over conventional scalar wireless sensors that can extract scalar sensor data together with video, audio, and image data. Such networks are expected to eventually span the whole planet in the future. These advancements with many benefits have introduced significant obstacles too, like high energy consumption, that require the deployment of optimized energy-efficient techniques. Machine learning (ML) approaches that can handle dynamic scenarios with a successful learning process have been utilized recently in WSNs as a solution to this challenge. WSNs specifically employ Deep Learning (DL) techniques, a subclass of machine learning techniques distinguished by the application of deep neural networks, to extract higher-level features from unprocessed sensor data, hence reducing computational complexity and increasing energy efficiency, etc. This paper has reviewed all such techniques based on deep learning for energy harvesting in WSN. Professionals working in this field will find this paper's review analysis extremely fascinating, as it will assist them in selecting the best methods for their particular applications.