Context-Aware Handover Analysis in Heterogenous Wireless Network Using Machine Learning

Abstract

The speedy development of wireless access mechanisms offers the mobility management and better interoperability methods to accomplish the necessities of users. Nowadays, heterogeneous wireless networks construct a variety of networks of diverse types namely WiFi, WIMAX to offer the users the required signals. These networks are independent, and they differ comprehensively with respect to the service constraints namely, accessing delay, the area of coverage, throughput, etc. the experimental analysis of various modeling system for handover performance is demonstrated. Here, the performance of NN-based vertical handover model has compared models like Levenberg–Marquardt-NN (LM-NN), Fire Fly-NN (FF-NN), Particle Swarm Optimization-NN (PSO-NN), and Grey Wolf Optimization-NN (GWO-NN). Finally, the performance achievement of the presented WOA-NN technique is distinguished with the algorithms with respect to throughput, Mean Absolute Error (MAE), and predicted RSS. From the entire analysis, the predicted RSS of the presented WOA-NN model appears almost closer to the real model, attaining effectual handoff.