Spectrum Sensing of NOMA Signals Using Particle Swarm Optimization Based Channel Estimation With a GMM Model

Abstract

Spectrum sensing of non-orthogonal multiple access (NOMA) signals is a challenging issue, as not only the presence/absence of primary users (PUs) but also the number of PUs needs to be identified. In this letter, we formulate NOMA spectrum sensing as a multiple hypotheses testing problem, where the estimated channel gains from PUs to a secondary user are used as test-statistics. Then, to reduce the computational complexity, we establish a Gaussian mixture model (GMM) for received signals, and propose a channel coefficient estimator using particle swarm optimization with the GMM. In addition, a new miss-detection probability is defined when some active PUs are correctly detected for the multiple hypotheses testing problem. Simulation results are provided to demonstrate the superior performance of the proposed detector, compared to state-of-the-art NOMA detectors.