Methods of Signal Detection and Recognition in Cognitive Radio Networks

Abstract

The paper employs techniques of signal detection and recognition to perform spectrum sensing in cognitive radio networks. The main focus is on extensive use of signal recognition techniques, which is not the case yet in the context of cognitive radio networks. In particular, the use of unconventional methods of signal recognition allows us to not only divide received signals into classes of those which belong to primary or secondary users but, additionally, register appearance of new, uncategorized, sources of radiation. As a consequence, all considered in the paper procedures may be employed by the local frequency control authorities to monitor radio frequency resource utilization. Performance analysis of presented in the paper algorithms for signal selection and recognition is performed by the method of Monte-Carlo.