Hybrid Deep Learning Approach for Improved Network Connectivity in Wireless Sensor Networks

Wireless sensor networks occupy a prominent role in industrial as well as scientific applications. Lifetime enhancement and coverage are the major factors considered while designing the network. Various research models are evolved by considering the scheduling and routing process to solve the network lifetime issues. However, coverage and connectivity is another important factor that affects the lifetime of the remaining nodes. When a large number of sensors are deployed randomly, scheduling is preferred to enhance the network lifetime, but it leads to coverage issues. Other than scheduling, node damage, battery exhaustion, software and hardware failures might lead to coverage issues. Preserving the network connectivity while maximizing the network coverage is a crucial task in wireless sensor networks. To preserve the network connectivity and improve the wireless sensor networks coverage this research work presents a hybrid deep learning approach using a deep neural network and reinforcement learning algorithm. The Proposed model is experimentally verified and compared with conventional deep neural network and reinforcement learning algorithms to demonstrate the better balancing characteristics between network coverage and lifetime.