Experimental evaluation of a cooperative kernel-based approach for robust spectrum sensing

Abstract

The spectrum sensing accuracy has been improved by the introduction of cooperative spectrum sensing (CSS) strategies where the spatial diversity is exploited among nonlegacy users. However, these CSS strategies also bring new impairments, such as the interference from other sources that severely degrade the sensing performance. In this paper, we evaluate experimentally our recent proposal for CSS based on kernel canonical correlation analysis (KCCA), where the effect of an interferer is also modeled. The experiments are conducted on a cognitive radio platform composed of several Universal Radio Peripheral (USRP) nodes, and the measurements show that our scheme is able of implicitly learning the surrounding environment by only exploiting the non-linear correlation among the receiver signals of each SU. Eventually, we provide comparative results where a considerable gain over a conventional energy detector is obtained in spite of the impairments provoked by external interferers.

Spectrum Sensing Using Software Defined Radio for Cognitive Radio Networks: A Survey

“…In addition, the problem of softdecision schemes perfectly matches the matching learning (ML) paradigm [65]. ML has been employed to mitigate the presence of narrowband external interference, and it has been validated by means of experimental evaluations [63].…”

“…In this regard, a triggering signal is implemented using PPS or GPS signals, as well as network time protocols. Quite often, these approaches are reported for indoor environments where a CSS scenario emulates a real world scenario with a limited of nodes [63], [67], as it is depicted in Fig. 14, while outdoor measurements require the deployment of several nodes monitored using GPS signals and network time protocols for the synchronization of the experiments [90].…”