WLAN covert timing channel detection

Wireless LANs have been widely used to carry out a system to access Internet. WLAN security becomes mission one, especially a new type of attacks called covert channel based attack surfaced over the past few years. This attack uses different data rates provided in WLAN to transmit a secret message. Detecting this covert channel could be difficult due to existence of rate diversity in 802.11 WLAN. Multiple transmission data rates are supported to exploit the trade-off between obtaining the highest possible data rate and trying to minimize the number of communication errors. In this paper, a feature modal is proposed to form possible hypotheses and then statistic hypothesis testing is applied. Simulation results on publicly available WLAN traffic show that our proposed approach could achieve 100% detection rate.