SNR Wall for Cooperative Spectrum Sensing Using Generalized Energy Detector

TL;DR

This paper investigates the impact of noise uncertainty on cooperative spectrum sensing using a generalized energy detector, deriving the SNR wall for both hard and soft decision combining, validated through simulations.

Contribution

It introduces the analysis of the SNR wall for CSS with GED under noise uncertainty, providing new theoretical expressions validated by simulations.

Findings

Derived SNR wall expressions for CSS with GED

Validated theoretical results with Monte Carlo simulations

Showed performance degradation due to noise uncertainty

Abstract

Cognitive radio (CR) is a promising scheme to improve the spectrum utilization. Spectrum sensing (SS) is one of the main tasks of CR. Cooperative spectrum sensing (CSS) is used in CR to improve detection capability. Due to its simplicity and low complexity, sensing based on energy detection known as conventional energy detection (CED) is widely adopted. CED can be generalized by changing the squaring operation of the amplitude of received samples by an arbitrary positive power p which is referred to as the generalized energy detector (GED). The performance of GED degrades when there exists noise uncertainty (NU). In this paper, we investigate the performance of CSS by considering the noise NU when all the secondary users (SUs) employ GED. We derive the signal to noise ratio (SNR) wall for CSS for both hard and soft decision combining. All the derived expressions are validated using Monte Carlo (MC) simulations.