A Compressive Sensing Approach to Analyze the Performance of wideband Cognitive Radio Networks

Abstract

In the field of wireless communication systems, Cognitive Radio (CR) technology is the talk of the time for the best utilization of spread spectrum frequency. In the Wideband range, traditional Narrowband Sensing methods are not suitable to apply for performing Spectrum Sensing, as of making a single binary decision (Primary User present or absent) in the entire Wideband signal, thus cannot locate individual spectral opportunities that rely within the Wideband Spectrum. The Compressive Sensing (CS) can recover sparse signals at Sub-Nyquist rates and it depends on this principle of sparsity, so that a brief representation of the signal is possible when expressed in a suitable form. This research work has proposed a model of CR receiver sensing module which can be able to estimate a significant part (which is highly sparse among the segments of the spectrum) of the entire Wideband Spectrum with lower computational complexity. This propose work aims to analyze the compression ratio, i.e. M/N, with different number of Primary User (PUs) present in the wideband frequency from the Receiver Operating Characteristics (ROC) curves. Additionally, it is also analyzed that how the compression ratio, M/N characteristics varies with signal-t-noise ratio (SNR) in the wideband frequency from the ROC curves. This work investigates the probability of detection, Pd versus SNR for a fixed M/N and the throughput of a CR network against sensing period for a fixed frame length and vice versa. This proposed work also aims to find the requirement of less computational complexity and physical memory. Eventually, from the wide investigations through our proposed research work, it is found that the proposed method provides better throughput for fixed frame length as well as fixed sensing slot duration as well as frame duration the throughput is greater for shorter sensing time period. It can hopefully state in the final point that the proposed method proves its significance in CR system.