Modeling and performance analysis of the IEEE 802.11p EDCA mechanism for VANET under saturation traffic conditions and error-prone channel

Abstract

The IEEE 802.11p standard introduces new specifications to improve communications in vehicular ad hoc networks (VANETs). The MAC sub-layer of the IEEE 802.11p standard uses Enhanced Distributed Channel Access (EDCA) to introduce service differentiation when accessing the channel, by using four access categories (ACs). In this paper we propose an analytical model, based on two-dimensional Markov chain, for the IEEE 802.11p EDCA mechanism under saturation traffic conditions and error-prone channel. The major key aspects of the IEEE 802.11p EDCA mechanism are captured by the model as defined in the standard. These include the AIFS waiting procedure, backoff counter freezing, virtual collisions between the ACs of the same station, external collisions with other stations in the network, post-collision period, and retransmission limit. Moreover, the probability of error in the channel is explicitly considered. This model allows us to analyze the performance of VANET. Normalized throughput and average access delay are considered as quality of service (QoS) metrics. In addition, the impact of Bit Error Rate (BER) and frame size on network performance is also investigated. The obtained results confirm that the probability of error is a parameter that has a significant impact on network performance accuracy. Discarding this parameter in the IEEE 802.11p EDCA model limits the capabilities of the analytical model, because the model will not be able to correctly predict the performance of the network in an error-prone channel. They also confirm that the increase in the density of vehicles and the BER result in less efficient use of the network by increasing the collision rate; and that the use of small frames is one of the necessary aspects for applications with strict constraints of delay.