Ternary Hypothesis Testing Scheme for Rapidly Opportunistic Spectrum Access

This letter presents a partially observable Markov decision process framework for the opportunistic spectrum access model, where a ternary hypothesis testing scheme is proposed to perform sensing with the goodness of fit testing. Except for idle and busy states, the proposed ternary hypothesis testing also considers the transition state, which reflects the real state of a licensed channel more precisely. Simulation results indicate that the proposed ternary hypothesis testing scheme achieves better performances in channel utilization and collision rate compared with those of the conventional binary hypothesis testing scheme, especially for the rapid variations of the primary user activities.