On the performance of channel occupancy detectors for vehicular ad-hoc networks

Vehicular networking is becoming an actual disruptive technology which is going to improve, first of all, the overall safety as well as the the driving experience. The standardization efforts are almost completed and the first implementation issues are becoming real. Despite the fact that the physical layer is very similar to the one of the Wi-fi standards, while the classical Wi-fi standards which use crowded 2.4 GHz ISM band, vehicular networking uses the 5.9 GHz band. This advantage however entails a series of issues to reach the best throughput performances and, at the same time, maximize benefits from the vehicular mobility safety point of view. One of the major issue is how to detect and map channels occupancy to take then the best decisions when choosing in which channel transmitting the different messages. In this paper we simulate the most common cognitive channel detectors in vehicular scenarios; considering several channel models, we evaluate the detectors performances in terms of probability of misdetection and detection delay. The obtained results show that in case of short messages, the fastest and most reliable detector is also the simplest one i.e. the energy sensing detector.