SECOA: Serial Exponential Coati Optimization Algorithm for MANET routing with link lifetime prediction

Mobile Ad-hoc Network (MANET) is a wireless network that operates without a fixed infrastructure and is highly adaptable to changes in speed and connectivity. The source mobile node can transfer the data to any other destination node; however, it has restrictions on energy utilization and lifetime of battery. In order to overcome this, in the literature several optimization-enabled routing algorithms are developed in MANET. In this paper, an algorithm, named Serial Exponential Coati Optimization Algorithm (SECOA) is proposed for MANET routing. Here, the link lifetime (LLT) is predicted using Recurrent Neural Networks (RNN) to ensure reliable and continuous communication. Once LLT prediction is done, nodes with the maximum LLT values are chosen for the routing purpose. To enhance the routing effectiveness, several objective parameters, like energy, distance, trust, and LLT are employed to devise a multi-objective function. Also, it leads to an optimal path using the proposed SECOA approach. In addition, this model is used to extend LLT by choosing best cluster heads of the conventional clusters. Moreover, trust is computed to improve security and enhance cooperation between nodes, which is employed to accelerate the recognition of misbehaving nodes. Finally, the model attained enhanced performance with a maximum energy of 0.895, maximum LLT of 0.758, maximum PDR of 0.889, maximum throughput of 0.895, as well as maximum trust of 0.778.