Adaptive fuzzy multiple attribute decision routing in VANETs

Vehicular ad hoc networks (VANETs) have attracted much research attentions in the recent years. Although there are many outstanding solutions of geographic routing designed for the VANETs, only one or two features of the VANETs have been specifically utilized in each of the proposed solutions. In this paper, we propose a novel adaptive fuzzy multiple attribute decision routing (AFMADR) scheme, by which a packet carrier is supposed to be a decision maker in the selection of a target for the next hop transmission with more factors to be considered. The AFMADR scheme consists of four steps. Firstly, the candidate vehicles are characterized by four attributes including the distance, the direction, the road density, and the location to obtain a fuzzy attribute score of each attribute. Secondly, according to the real-time conditions of the candidates' attributes, weights of attributes will be calculated by the proposed adaptive weight algorithm, which makes the AFMADR scheme more scalable and robust in different VANET scenarios. Then, a fuzzy performance score will be produced for each candidate based on the fuzzy attribute score and their real-time weights. Lastly, a route decision is made by choosing a candidate with the highest performance score as the target for the next hop transmission. We carry out simulation experiments to compare the performance of the AFMADR scheme with the existing well-known geographic routing protocols for the VANETs. The simulation results demonstrate that the proposed AFMADR scheme can achieve the best performance in terms of the highest delivery ratio and the lowest delivery delay with a reasonable number of hops among all protocols in all the simulation scenarios. Copyright (C) 2015 John Wiley & Sons, Ltd.