Goodness‐of‐fit‐based non‐parametric spectrum sensing under Middleton noise for cognitive radio

Abstract

A novel goodness‐of‐fit‐based non‐parametric spectrum sensing scheme in a non‐Gaussian noise environment, modelled by Middleton class A distribution, is proposed. The sampling distribution of the proposed test statistic is derived and the detection performance is shown using Monte Carlo simulations. Results are presented and it is concluded that the performance is degraded if the Gaussian component in the Middleton noise is higher than the non‐Gaussian component.