Research On Spectrum Sensing Technology For AD Goodness Of Fit Test Based On Exponential Distribution

With the development of wireless communication, the spectrum resource becomes more and more important. It is undeniable that the spectrum resources are not fully utilized, its utilization rate has been a very important indicator. As an effective way to improve the spectrum utilization, cognitive radio has attracted more and more attentions in the academic field. Spectrum sensing technology can detect whether the channel is occupied by the primary user, which is one of the key technology of cognitive radio. In many spectrum sensing schemes, energy detection and goodness of fit test has been widely used because of its almost no need to know a priori information of the signal. However,energy detection and goodness of fit test also have their disadvantages,spectrum sensing performance of energy detection effect is not good, disadvantage of goodness of fit(hereinafter referred to as "routine test of goodness of fit") is that its sensitivity to signal types. In addition, the compensation is very complex. This paper proposes a new spectrum sensing scheme based on the conventional goodness of fit test, which is based on the exponential distribution of AD goodness of fit test spectrum sensing technology(hereinafter referred to as IP_AD). In the IP_AD scheme, the sum of the power of the two adjacent sampling points is used as the test statistic, and the test statistic satisfies the exponential distribution when the signal does not exist.The thesis will exploit Monte Carlo simulation technology to analyze the spectrum sensing performance of IP_AD plan from multiple aspects and multiple perspectives,and conventional ad fit goodness test spectrum perception and energy detection for spectrum sensing scheme has made a deep comparison. First in the AWGN channel,We discuss the IP_AD spectrum detection technology of signal type sensitivity(its signal types affect the spectrum sensing performance), followed by analysis of the sample points,false alarm probability and signal to noise ratio(SNR) IP_AD spectrum detection performance influence. In addition, the spectrum sensing performance of IP_AD and conventional goodness of fit test is compared under different conditions. In particular, wediscussed the fading channel IP_AD plan performance, and its and energy detection and test of goodness of fitting the conventional spectrum sensing performance were compared(from signal types, sample points, false alarm probability and signal to noise ratio are considerable).In the practical wireless communication system, the noise power uncertainty always exists, and the energy detection and the goodness of fit test are all sensitive to it. In the last part, the spectrum sensing performance of IP_AD under noise uncertainty is compared with that of energy detection and conventional AD goodness of fit test.Multiple angle and many aspects are considered, and a lot of extensive simulations are carried out. The simulation results show that the IP_AD spectrum sensing scheme is not sensitive to the signal type, and its spectrum sensing performance is better than that of energy detection and conventional AD fit test method in most cases.