Reducing the Sensing Errors by Adopting the Effective Matched Filter Threshold Estimation in Lower SNR Conditions

Abstract

Cognitive-Radio Network (CRN) is an enabling and promising technology to enhance the spectrum utilization and the key functionality of cognitive radio (CR) systems is spectrum sensing in order to function in an unoccupied spectral bands. To safe-guard the presence of licensed users from any interference in a spectrum, CR should be capable of detecting reigning channels even in the scenario of low signal to noise ratio (SNR). Here, in this article, various spectrum-sensing parameters such as detection probability, bit error probability (BEP), miss-detection probability, probability of false alarm are amended in the case of lower SNR by employing demand and need based genetic-algorithm(GA) by considering geographical disparities of spectrum-holes. The outcomes convey that the GA can provide a finer real-time solution for the sensing capability of a cognitive-radio network.