Cyclostationary Approaches to Signal Detection and Classification in Cognitive Radio

Spectrum awareness is currently one of the most challenging problems in cognitive radio (CR) design. Detection and classification of very low SNR signals with relaxed information on the signal parameters being detected is critical for proper CR functionality as it enables the CR to react and adapt to the changes in its radio environment. In this work, the cycle frequency domain profile (CDP) is used for signal detection and preprocessing for signal classification. Signal features are extracted from CDP using a threshold-test method. For classification, a Hidden Markov Model (HMM) has been used to process extracted signal features due to its robust pattern-matching capability. We also investigate the effects of varied observation length on signal detection and classification. It is found that the CDP-based detector and the HMM-based classifier can detect and classify incoming signals at a range of low SNRs.