Learning Automata Based Contention Aware Data Forwarding Scheme for Safety Applications in Vehicular Ad Hoc Networks

Abstract

With an exponential growth of demands of the users to access various resources during mobility lead to the popularity of Vehicular Ad Hoc Networks (VANETs). Users may access various resources from cloud which consists of many resources for the ease of users. VANETs have been used in wide range of applications such as Intelligent Transport Systems (ITS), Safety alarms on roads/in community, online resource access using Internet connectivity etc. Among these applications, safety applications are most important and various proposals exist in literature for the same. But most of the existing proposals have used unicast sender based data forwarding which results an overall performance degradation with respect to the metrics such as packet delivery ratio, end-to-end delay and reliable data transmission. Keeping in view of the above, in this paper, we propose new Learning Automata based Contention Aware Data forwarding scheme for VANETs using cloud infrastructure. Learning Automata (LAs) are assumed to be located in the vehicles which share the information (such as vehicles density, directions of the vehicles or vehicles velocity etc) with the other LAs for taking the adaptive decisions about data forwarding. Based upon these values, automaton performs its action. Corresponding to each action performed by the automaton, its action may be rewarded or penalized by some constant values from the environment where it is working. Based upon the inputs from the environment, each automaton updates its action probability values for the next rounds. An adaptive Learning Automata based Contention Aware Data Forwarding (LACADF) algorithm is also proposed. The proposed scheme is evaluated with respect to different network parameters such as message overhead, throughput, delay etc. with varying density and mobility of the vehicles. The results obtained show that the proposed scheme is better than the other conventional schemes with respect to the above metrics.