Joint Transmission-Sensing Framework Using Changepoint Detection

Abstract

A cognitive radio (CR) network is reliable by the ability of the secondary user (SU) to maximize the spectrum usage while minimizing the interference on the primary user (PU). This depends on how sensing and transmission are scheduled by the SU. In this paper, we investigate a new joint transmission-sensing framework for a CR system where there is multiple PU appearances and disappearances. Unlike the existing works that sense the spectrum independently frame by frame, we suggest to perform the spectrum sensing sequentially along several frames, using a changepoint detection method. To compute the optimal sensing and transmission times, we introduce a new metric, called the global cost, that gives a trade-off between interference and spectrum waste. The optimal parameters values are studied in detail, with respect to the signal to noise ratio (SNR) and the primary users statistics.