Fuzzy-AHP model for Intelligent network selection in VANET for data offloading

Increasing data downloading demands by vehicular users creates the overburden problem for cellular 4G/5G base stations (BS). Thereby, cellular BS delivers poor services to vehicular users. Offloading the cellular data from another network might be the solution to maintain the QoS for vehicular user requests. In vehicular ad-hoc networks (VANET), roadside unit (RSUs) plays a significant role in offloading the cellular data and improving the QoS provided by cellular networks. Urban areas or busy streets consist of several RSUs; to maintain the vehicle user’s constant connection to the best network regarding QoS parameters, an intelligent network (Cellular/LTE BS or one of the RSUs) selection technique is required. This work proposes the Fuzzy-AHP based intelligent network selection method in heterogeneous vehicular networks. Fuzzy-AHP uses four parameters (RSSI, Delay, Jitter, and Cost) to select the best network for two different types of applications (Audio and Video). Fuzzy comparison matrices are used to calculate the parameter’s weight. Finally, the ranking of available networks (BS and RSUs) is exclusively done by two decision-making methods: Simple Additive Weighting (SAW) and MEW (Multiplicative Exponential Weighting). Simulation results show that Fuzzy-AHP followed by SAW and MEW methods, provides better QoS regarding reduced packet loss and delay.