A Simplified Spectrum Sensing Scheme under Multiple Primary Transmit Power

In this paper, we consider the spectrum sensing for cognitive radio (CR) when the primary user (PU) has a set of discrete transmit powers. Different from the traditional sensing approach where the binary hypothesis testing was used, we here apply the the multiple hypotheses testing in order to not only check the ``on/off´´ status of PU but also to identify its power level. To better illustrate our idea, we set the price of erroneous sensing in Bayes risk to be unit, such that maximum a posterior (MAP) detection is optimal in terms of minimizing the total error rate. The closed expression of the decision regions as well as the performance metrics are derived. Simulation results are provided to corroborate the proposed studies.