Power-Efficient Wideband Spectrum Sensing for Cognitive Radio Systems

This paper proposes a wideband spectrum sensing architecture by utilizing a one-bit quantization at the cognitive radio receiver. A window-based autocorrelation is utilized to provide the power spectral density of the quantized signal. Closed-form expressions are derived for the one-bit quantized correlation, where the critical impact of the ultralow resolution is quantified. It is shown that the introduced method can still provide enough information about the sparse spectrum even at low signal-to-noise ratios (SNR). A suboptimal detection algorithm is presented to sense whether individual subbands are occupied or vacant. To evaluate the detection algorithm, the probability of false alarm and probability of detection are simulated for various system parameters. The sensing performance and the accuracy of the offered expressions are justified through comparisons with respective results from computer simulations. When compared to other methods, in addition to the significant saving in power, results indicate that, with the proper parameter selection and SNR down to -12 dBs, the proposed method provides low complexity, low sensing period, and better performance for less sparsity levels even though an aggressive quantization has been applied.