10805

9960

كارشناسي ارشد

مخابرات شبكه

اصفهان: دانشگاه صنعتي اصفهان، دانشكده برق و كامپيوتر

1394

[شش]، 64ص.: مصور

مهدي مهدوي

1394/09/28

علي محمد دوست حسيني

1396/10/06

كتابنامه

برق و كامپيوتر

مهندسي برق و كامپيوتر

ID9960

Study of the effect of the quantizing reports of the honest users in the performance improvement of the cooperative spectrum sensing in the cognitive radio networks with the presence of malicious users Fatemeh savarsofla f savarsofla@ec iut ac ir Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan Iran Language ParsiDegree M SC Supervisor Dr Mehdi MahdaviAbstractWith recent rapid growth of wireless technologies in and the limitation of the available bandwidth the staticfrequency allocation schemes cannot support many emerging wireless services Cognitive radio technologyis the approach to enhance spectral efficiency In this technology users without spectrum licenses can useunused licensed frequency band spectrum holes In this way the spectrum sensing as a key parameter ofthe CR systems can be affected by the factors such as multipath fading and shadowing This leadsinterferences with PUs and failure detection of them by CR users In order to alleviate these issues cooperative spectrum sensing has been proposed One of the main issues which can severely threaten theperformance of cooperative scheme is to send false sensing data to fusion center by some CR users who arecalled malicious users The purpose of sending false data by malicious users is to disturb the running of data fusion algorithm and make false decision by fusion center There are methods which have beenproposed to detect and eliminate the negative effects of malicious users on the cooperative spectrumsensing efficiency where a division of them are based on the users reports histories In the method proposed in this thesis each CR user senses the spectrum using energy detection and sendsits report to the fusion center The fusion center calculates the truth value and consistency value parametersfor each CR user based on its report histories where such parameters indicate the malicious level of eachCR user When such parameters of a user are lower than some certain thresholds the corresponding user isconsidered as malicious and its report is removed from the decision process Next the fusion centerquantizes the rest of the users sensing reports and after applying its fusion rule it makes its decision aboutthe presence of primary user in frequency band The effect of the reports quantization in the fusion center has been investigated and showed that theincreases of the quantized levels improve the performance of the primary user detection in the frequencyband Keywords Cognitive radio networks Cooperative spectrum sensing Malicious user detection in cooperativespectrum sensing

مهدي مهدوي

علي محمد دوست حسيني