Study On Spectrum Sensing Technology Of Cognitive Radio

The development of the wireless business results in increasing needs of the frequency. Wireless communication is facing a challenge. Radio resource is distributed by the government, which generally rises inequality and waste of Radio resource. So, there is a contradiction between the waste and the shortage of Radio resource. The cognitive radio, a technology of intelligent spectrum sharing, effectively bridges the contradiction. Based on the perception of time domain, frequency domain and airspace, automatic discovery and reasonably use of free spectrum among distributed spectrum, the cognitive radio improves the utilization ratio in the existing frequency. Spectrum sensing is one of the sky technologies in the cognitive radio.Based on the introduction of Cognitive Radio(CR) and the discussion of applications of CR in WRAN, UWB and WLAN, the key technologies of achieving CR were researched in theory and the factors controlling cognitive network and the interferences raised by cognitive network in normal working of authorized user were analyzed. The related method and technology of signal detection was theoretically derived in the severe environment of weak signal and low signal to noise ratio that is must facing to achieve CR. With the algorithm analysis of digital filter, the selection principle of windows function was proposed. Three spectrum detection methods, namely matching filter detection way, energy detection way and cycle property detection way, were investigation in great detail. For testing and verifying the correction of above results, the detailed modeling and simulation of energy detection and cycle property detection were completed with Matlab and the performances of that two detections were also compared. The results show that the energy detection way , in the low SNR , is low and the property of that is very poorer than the cycle property detection. And this paper suggested that improvement of algorithm of window function can promote the performance of cycle detection algorithm. Further, the digital filter on distribution calculating was designed and simulated by using ISE 8.2i in FPGA belonged to the Xilinx company and Modelsim belonged Mentor Graphics. To simple the design of the filter, this paper, in the design progress, also studied the process of fast getting parameters of the filter by Matlab and the FPGA method of achieving module of the digital filter with calling of IP core of Xilinx. Finally, the simulation was verified in Modelsim. The consistence of test waveform and design demand confirmed the correction. This study is useful to explore CR in theory and application. The results came from this paper is very significant for further research.