Quadratic Form-Based Spectrum Sensing Algorithm for 6G Cognitive Radio Networks

Spectrum Sensing will play a pivotal role in 6G Cognitive Radio Networks, in which unlicensed users could dynamically use the spectrum and perform opportunistic transmission. In this paper, we focus on a multiple-antenna data-aided Spectrum Sensing algorithm and we compute the exact Moment Generating Function and probability density function under the null hypothesis of such test statistic, by resorting to tools from finite-dimensional random matrix theory. The test at hand is formulated in terms of a quadratic form with unit-norm random vector and random kernel matrix with determinantal joint probability distribution of its eigenvalues. The algorithm exploits the availability of a set of auxiliary data, collected under the assumption of primary signal presence. Results, presented in terms of Receiver Operating Characteristic curve and detection probability vs. the Signal-to-Noise Ratio, show that the presented test outperforms some well-established sensing schemes, such as the Generalized Likelihood Ratio Test and the Roy's Largest Root Test. Furthermore, we compare the outcomes of our analytical derivation with corresponding Monte Carlo simulated data in scenarios with both spatio-temporally uncorrelated and correlated samples.