Spectrum Sensing And Modulation Recognition Based On Cyclotationary Spectrum In Cognitive Radio

With the rapid development of wireless communication technology, spectrum resources become more and more rare. There are three paradoxes in spectrum management. The usage of spectrum is dynamic, but the spectrum allocation is fixed. Spectrum resource is rare, but the using efficiency of the spectrum is low. There are a large number of idle, but the spectrum that can be allocated is little. A large amount of spectrum is required with the rapid development of the wireless communication business and new technologies. In order to solve these problems, cognitive radio can maximize the utilization of available spectrum.It is the key and prerequisite to detect the primary users and cognitive radio spectrum hole fast and reliably. The modulation recognition has important significance to intelligent cognitive radio. This paper is divided into two parts of signal detection and modulation identification. We focus on cyclostationary spectrum detection in the part of signal detection. In order to improve its performance at low SNR, we proposed a detection algorithm with Chair-Varshney applied into cyclostationary spectrum, and using multiple frequencies of the joint detection. Simulation results show that improved algorithm makes full use of each single cyclic frequency. It has better performance at low SNR. Due to the complexity of the wireless environment, this paper also studied the detection of the signal in the multi-path channel. Multi-user co-operation can effectively reduce the multi-path. In the study of modulation recognition, we constructed two statistical parameters by cyclostationary spectrum. The parameters have good stability and scarcely effected by SNR. The algorithm has a high probability of detection in recognition of MPSK.