Enhanced Energy Detection using Matched Filter for Spectrum Sensing in Cognitive Radio Networks

The world of Cognitive Radio (CR) technology comprises of intelligent wireless communication systems, designed using transceivers that are apt to automatically detect and access available communication channels in Radio Frequency (RF) bandwidths, aiming to maximize the employment of the RF spectrum and abate the interference between users. In primary transmitter detection i.e. non-cooperative spectrum sensing techniques, the licensed primary users (PUs) of the network are perceived based on the signals that the unlicensed secondary users (SUs) receive. This paper provides an insight into one such method, namely, the energy detection technique, which has a low computational complexity and can be implemented with ease, due to the generic nature of the concept that it follows. However, the detection of weak PU signals can turn out to be a challenging endeavor, especially across noisy channels, and therefore calls for a more sophisticated approach to sensing spectrum. A matched filter can be used to obtain supplementary information regarding the channel activity, help individuate the transmitted pulses from the noise, and reduce the detrimental effects of unlicensed signal interference. The proposed algorithm attains results from a matched filter and implements it within the energy detector, and analyzes the signals over a channel of Additive White Gaussian Noise (AWGN) for a range of Signal-to-noise ratios (SNRs), which are then assessed via Receiver Operating Characteristic (ROC) curves with probability of detection (Pd) and probability of false alarm (Pf) as performance metrics.