Cooperative Spectrum Sensing Based On Competitive Neural Networks In Cognitive Radio

Facing the present situation of the growing demand for wireless services and the scarcity of spectrum resources, Cognitive Radio (CR) is recognized as a viable solution. The key and the priority issue of cognitive radio technology is spectrum sensing, that is, secondary users must detect the related frequency band to determine whether there is a primary user. The system commands that the secondary users have the ability to occupancy the available local spectrum immediately while avoiding interference to the communications of the primary user. For multi-path fading, shadow effects and hidden terminal problems the single-user detection may encounter, cooperative spectrum sensing has been widely studied.Meanwhile, as a new means of artificial intelligence technology, the artificial neural network technology is the rapid development of the international forefront of research. It has been widely used in pattern recognition, control optimization, intelligent information processing and other aspects.Firstly, this paper summarizes the existing data fusion algorithms, and these algorithms are simulated and summarized; then propose the data fusion algorithm based on artificial neural network, which is applied to varied environments and can give the existence location of primary user while has a high detection probability and zero false alarm probability; finally, put forward a kind of learning mechanism based on the proposed data fusion algorithm, which learns from the human learning and memory mechanism, and makes the system adapt to the dynamic environment.