Sensing Algorithm for Cognitive Radio Networks Based on Random Data Matrix
Signal detection is a fundamental problem in Cognitive radio. In this paper a new statistical test is proposed based on Random Data Matrix (RDM) for detecting the signals in noise, as opposed to the eigen-value based tests. Among the many spectrum sensing methods, the RDM method detects the primary users without any prior information. The performance of the test is compared with Energy Detection (ED), Covariance Absolute Value (CAV) and eigen-value based algorithms through simulation analysis. This sensing algorithm can be used for very low SNR signal detection without requiring the knowledge of signal, channel and noise. Simulations are based on wireless microphone and Identically and Independently Distributed (IID) signals.