Two stage spectrum sensing for cognitive radio networks using ED and AIC under noise uncertainty

Spectrum Sensing is elementary function in Cognitive Radio Networks (CRN) to identify the white spaces in spectrum for opportunistic communication. In this paper, we proposed a novel two stage spectrum sensing under the environment as noise uncertainty. The robustness of uncertainty of noise power is one of the main challenges in spectrum sensing method. Since detection of primary users (PU) in the presence of noise power uncertainty, performance of spectrum sensing method consequently decreases. The proposed detection technique combine two well-known different detection method are Energy Detection (ED) and Akaike's information criteria (AIC) to perform spectrum sensing. At first stage, ED technique is use to find power average of received signal and second stage is AIC detection technique based on the information theoretic criteria (ITC). Study of spectrum sensing method, the ED technique performance is better, reliable and taking short time at high signal to noise ratio (SNR) and worst in low SNR, whereas AIC based method perform better at low SNR, but implementation complexity is high. Under the uncertain noise combination of these two techniques give more reliable detection. Designee two stage threshold parameter, for maximize the probability of detection and mean detection time is improve for the given limitation on the probability of false alarm.