A multiple-input Nyquist folding receiver architecture for low SNR wideband spectrum sensing

Abstract

The ever increasing demand of communication makes the spectrum more crowded, and thus the Cognitive Radio (CR) emerged as a potential effective scheme to increase the utilization of spectrum. Sub-Nyquist sampling based CR can greatly reduce the sampling rate than those Nyquist sampling based schemes for wideband spectrum sensing. But existing single node sub-Nyquist sampling schmes seldom use spatial diversity and show poor recovery and detection performance in low Signal to Noise Ratio (SNR) condition. NYquist Folding Receiver (NYFR) as one of effective sub-Nyquist sampling schemes exhibits a lower system working frequency than most other schemes, but it is greatly effectted by noise. In this paper, we first investigate the property that the Additive White Gaussian Noise (AWGN) after sub-Nyquist sampling by NYFR is still the AWGN, then, we propose a Multiple Input NYquist Folding Receiver (MI-NYFR) scheme by using the noise white-keeping property and spatial diversity. The proposed scheme is tested respectively in denoising and detection respects in AWGN channel. Simulation results show that the proposed MI-NYFR exhibits good noise robustness and high detection probability very well in low SNR condition.