Cooperative Spectrum Sensing Algorithm Based On Majority Decision Fusion

In the last few years， The use of frequency spectrum is level off to saturate.Consequently, the requirement of the spectrum resource usage becomes higher than before.The Cognitive Radio (Cognitive Radio, CR) is a study of wireless communication, whichbecomes a hot spot of research. Cognitive radio network including cognitive users (unlicensedusers) and authorized user (licensed users), and the cognitive users can dynamic access to thefree spectrum of authorized users. So as to the cognitive users can realize the share of thespectrum. When cognitive users use the sharing frequency band, each cognitive user can scanthe spectrum of an authorized user, so that perceive the use of free frequency band. In order toavoid cognitive users make the interference on the authorized user of using the spectrum, so itis important for the cognitive user whether can accurate to perceive the authorized user isusing the spectrum. Spectrum sensing research becomes the core of the cognitive radiotechnology. Compare with the cooperation spectrum perception algorithm, in the cognitiveradio system coordination spectrum sensing can provide more reliable detection. Minimizethe authorized users of interference. And the data fusion is the key technology.This paper studies that respectively combine single user energy detection,“and” fuserule,“or” fuse rule and majority decision fusion rule with the double threshold energydetection under the conditions of restricted bandwidth of collaborative spectrum sensingmethod. This article adopts the distributed data fusion theory can reduce the average numberof bits perception of sending information between cognitive users and the fusion center,which save the transmission bandwidth. This paper analyzes the performance of collaborativespectrum sensing under the condition of without existing error rate. And then compare theperformance with single user energy detection. In the analysis of cognitive radio systemperformance, we mainly conduct a mathematical research. The mathematical expression givesa variety of circumstances of spectrum sensing algorithm the detection probability of theclosed type. The MATLAB simulation results shows that the detection performance ofmajority decision fusion rule algorithm is much better than the "and" fusion rule and "or"fusion rule algorithm under the parameter change. the signal-to-noise ratio of the newalgorithm performance improvements10%when it is-10db, SNR performance improvements20%when it is-20db, and when SNR between (10db,-20db), the testingperformance improvements between10%to20%.This paper finally analysis the collaborative spectrum sensing algorithm, it under existerror rate. It is also to say that make the error rate which is produced by the practicalcommunication condition of influence into algorithm. Use error rate to measure theinterference of digital signal transmission. We reanalyzed three kinds of algorithm ofcooperative spectrum sensing in the presence of an error rate. And according to the principleof mathematical deduction get the mathematical expression of cognitive radio system innon-ideal channel. The MATLAB simulation results show that when send1bit digital signaldecision-making information to fusion center, the detection performance of majority decisionfusion rule algorithm is almost make no difference in the non-ideal channel. But theperformance of "and" fusion rule and "or" fusion rule decrease in the presence of an error rate.It follows that majority decision fusion algorithm have good stability in the non-ideal channel.