LAUSR

With the knowledge of channel occupancy rate (COR), cognitive radios can significantly improve their performance in exploring and exploiting spectrum holes. However, most existing COR estimators suffer from overestimation or underestimation even at high signal-to-noise ratios (SNRs). The iterative threshold-setting algorithm (ITA) is promising to address this issue. In this work, we revisit ITA and provide a thorough theoretical analysis of ITA. First, we prove that ITA converges to the true COR with a sufficiently large number of traffic samples. Then, we investigate its convergence when the number of traffic samples is small, and show that ITA deviates from the true COR especially at low SNRs. To address this issue, we analyze the upper bound of the number of traffic samples required to achieve a certain estimation error, and further propose an improved ITA (iITA). The proposed iITA enables us to achieve a prespecified estimation accuracy by adaptively adjusting the number of traffic samples. Extensive simulation results are provided, which validate our analyses and demonstrate the superior performance of ITA and iITA compared to state-of-the-art COR estimators.