A Multi‐Objective Energy‐Aware Routing in Ad Hoc Networks Using Improved Heuristic Optimization Based on Energy Prediction via Dense Conv‐LSTM With Coordinate Attention

ABSTRACT Energy efficiency is crucial for ad hoc networks since they are decentralized systems with changing topologies and constrained energy resources. Thus, this research proposed an energy prediction‐based routing approach for maintaining energy‐efficient communication within the network. Initially, the dense convolutional long short‐term memory with coordinate attention (DCLSTM‐CA) is used for the energy prediction within the ad hoc networks. Then, the optimal routing is performed via the structure oriented builder optimization (SOBO), which considers the objective functions such as energy consumption prediction, residual energy, link stability, congestion, routing overhead, scalability, and reliability. The effectiveness of the developed model compared to the other approaches is shown by the performance validation.