Improved cooperative sensing strategies using wavelet denoising

Cognitive radio technology allows the unlicensed users to utilize spectrum by sensing it reliably. The main obstacles in this sensing are fading and shadowing effects which are overcome using cooperative strategies as: amplify-and-relay (AR) and detect-and-relay (DR). In these strategies, noise received at relay is also amplified and relayed along with the signal, which limits their detection performance at the detector or the end user. In this paper, wavelet denoising (WD) is used to denoise the received signal before energy detection as well as before relaying the signal, and therefore, improving the detection performance. A Chi-square approximation using the cumulants is used to compute the probabilities, and an inverse mapping pseudo-code is utilized to calculate the threshold values at a given false alarm rate. The agreement of analytical and simulation results show that AR and DR (with WD) strategies perform better than plain AR and DR (without WD) in terms of detection probability.