Performance analysis of energy and AIC detection for spectrum sensing in cognitive radio networks

Fast growth of wireless communication demands more spectrum but it is a limited resource. Survey conducted by Federal Communications Commission (FCC) shows that spectrum allocated to the primary user (PU) is underutilization. Cognitive Radio (CR) is proposed to improve the utilization of underutilized spectrum. In CR, spectrum sensing is a key technology for opportunistic spectrum access. To reduce interference to the PU, it is necessary to sense the spectrum accurately at very low signal to noise ratio (SNR) region. The main drawback of Energy detection (ED) is that it can't discriminate between PU and secondary user (SU) at low SNR region. To overcome described problem, Akaike's Information Criteria (AIC) method is discussed in this paper. It is based on the Information Theoretic Criteria (ITC). In this paper, the comparison is shown between the ED and AIC method in noise uncertainty. Here, AIC method is considered in frequency domain to avoid the requirement of any prior information related to the PU. Simulation results show that the AIC method is better than the ED for spectrum sensing at low SNR region.