Vertical Handover Using Mutated-Salp Swarm Optimization Algorithm Based on Radial Basis Function Neural Network in Heterogeneous Networks

Abstract

The heterogeneous network for a 4th generation, not constantly supports enhanced mobility and communication amid the Wireless Access Network. Therefore, the vertical handoff is extremely required. This work sets up a vertical handover that is context-aware via WiFi and WiMax in a heterogeneous environment. For the handover points, thriving handover shows enhanced purposes. Hence, an RBFNN-based network method to appreciate the network distinctiveness is initially elaborated. In the experimented environment, the RSS of the heterogeneous network is experimental to model the training library. In a heterogeneous network, to solve the handover points the trained network predicts RSS. To make sure the exact learning of the NN regarding the RSS network characteristics, a mutated Salp Swarm Algorithm (mutated-SSA) is proposed. The developed technique performance is validated using the conventional FF-NN, LM-NN, GWO-NN, and PSO-NN via handover, throughput, Mean Absolute Error, and predicted RSS analyses. For the developed mutated-SSA-RBFNN-based network model, the predicted RSS appears almost nearer to the actual model, obtaining effectual handoff.