Idle Channel Time Estimation in Multi-hop Wireless Networks

This paper presents a theoretical estimation for idle channel time in a multi-hop environment. Idle channel time is the time proportion of a node during which the channel state is idle. Thus, it can be used to evaluate the available bandwidth. Major related work considers a cross-layer model, where the idle channel time measure is available at the MAC layer, but it is rarely implemented. Furthermore, this measure is not very flexible: it cannot differentiate the traffic's priorities and it works under the hypothesis that flows are strictly policed. Estimating instead of measuring the idle channel time prevents these drawbacks. This estimation is computed in three steps by (1) calculating the idle channel time bounds, (2) evaluating the probability of a given idle channel time value, (3) computing the expected value of this distribution and deducing the average idle channel time. We show by simulation that our estimation is accurate.