Vehicular networking and channel modeling: A new Markovian approach

Vehicular networking is rapidly growing in the last years, due to its ability of enhancing users everyday-life, in terms of comfort, security and QoS. New services are being implemented and the research activity on the topic is embracing different OSI layers, like the network and application ones. Many efforts have been also given at the PHY layer. In this paper we focus our attention on a new approach for modeling the wireless channel of VANETs communications. In the literature many channel models have been proposed, but most of them work at the bit level or they investigate only some aspects of channel interaction. In this work we provide a high level IEEE802.11p channel model based on Discrete Time Markov Chain (DTMC) modeling, useful in every simulation context. The proposed model is called Markovian Vehicular Channel Model (MVCM) and it is based on the concept of Packet Error Rate (PER) discretization: we do not consider the performance analysis at the bit level, but we derive the proposed model by computing the PER, associated to different observation times. The effectiveness of the proposed idea has been evaluated through a deep campaign of simulations.