Compressive sensing based maximum-minimum subband energy detection for cognitive radios

To satisfy the growing spectrum demands of emerging wireless applications, cognitive radios have been considered as a viable option. It enables dynamic spectrum access opportunistically using wideband spectrum sensing (WSS) methods to discover the temporarily free frequency bands. WSS requires a high-speed analog-to-digital converter (ADC), which has high power consumption and hardware complexity. Improving the power consumption and hardware complexity of the ADC is one of the existing challenges in energy-constrained applications. To alleviate this problem, we propose compressive sensing (CS) in maximum-minimum subband energy detection method to sense the wideband spectrum by utilizing the sparse nature of spectrum occupancy with the minimal possible number of measurements. The CS method uses Fourier Transform and chaotic sequence in designing the measurement matrix to achieve both determinacy and randomness. The Bayesian method is used to reconstruct the signal from the available measurements. From the reconstructed signal, the maximum-minimum subband energy detection (ED) method is used to decide whether the primary user (PU) is absent or present in a particular frequency band. The simulation results show that the proposed CS-based maximum-minimum subband energy detection approach improves the probability of detection by 7.5% compared to the conventional maximum-minimum subband energy detection method of spectrum sensing. The proposed spectrum sensing method is simple and robust to noise uncertainty and signal strength variations.