Novel extended modified twin test based sensing for cooperative communication under noise uncertainty

With the evolution of 5G wireless communication systems, we have witnessed an increase in the demand for wireless broadband applications and services. However, fixed allocation of the frequency spectrum has led to an under-utilization of the spectral resources, making it hard to find unoccupied bands to deploy new services. To address the spectrum scarcity problem, a new and promising technology has emerged, namely cognitive radio. In particular, centralized cooperative spectrum sensing (CSS) is becoming an effective strategy to discover unused frequency bands, since it allows to overcome issues related to shadowing and noise uncertainty. Here, we propose an extension of the modified twin test based on a double test that takes into account correlated observations in a real communication scenario, considering different noise uncertainty values. In particular, our test employs two fusion rules together (i.e. OR and Majority), salvaging those detection cases that would otherwise go undetected due to the noise uncertainty. The obtained results show that the proposed method outperforms the conventional CSS and modified twin test, highlighting its robustness in the presence of noise uncertainty.