Efficacy of compressive sensing for dynamic spectrum access

Compressive sensing (CS) relies on the fact that CS sampled signals are much closer to their information rate rather than the signal bandwidth. This attribute helps to provide the much needed benefits of reduced storage or transmission bandwidth for the next-generation broadband wireless communications and to overcome the hardware limitations for wideband spectrum sensing in dynamic spectrum access. In order to opportunistically reuse holes in the spectrum, it is essential to have a spectral detection and estimation technique that is capable of sensing and identifying available frequency bands. Conventional methods of detection are saddled with the high sampling rate requirement of Nyquist rate, however timing requirements limits the number of samples that can be taken from the signals. In a situation whereby the signal spectrum in open-access networks is sparse in nature, this work develops a detection mechanism for identifying spectrum holes using compressive sensing based algorithm technique. Different compressive sensing reconstruction algorithms are investigated and FFT spectrogram with an edge detection algorithm is used to identify the holes in the spectrum. A quick wideband spectrum sensing can be achieved using the compressive sensing technique and a more refined sensing can be used by any of the other available methods such as energy detection. The proposed model is evaluated in different fading propagation environments, taking into account of both additive and multiplicative noise.