Performance evaluation of spectrum Sensing techniques under Noise

Noise has always been the most important constraint upon the sensing of Primary User in the frequency spectrum. Evaluation of various spectrum sensing schemes under noise gives us a fair idea about their robustness. Energy detection is fast but other Sensing techniques are better when it comes to the performing under Noise. Simulations are done for the three techniques at different noise levels. Detection result at different SNR levels are plotted. Response of the Eigen detection is the least effected with change in SNR(20 dB) and a change of 13.74% in case of matched filter detection is observed and it has been found that Cyclostationary is less affected by noise as compare to Matched Filter detection. Eigen value detection it remained almost constant.