Research On Spectrum Sensing Based On Mobility Of The Cognitive Radio Users

Cognitive radio techniques have been demonstrated to improve the spectrum efficiency by "secondary utilization" of the spectrum. Spectrum sensing is a key technique to explore spectrum opportunities in order to ensure the cognitive radio (CR) users' dynamic spectrum access. Most current studies assume that the CR users which used for the spectrum sensing are static or quasi static. With the development of wireless communications, the users' mobility becomes the main feature of future wireless communications. In spectrum sensing, the CR users' mobility leads to different relations between the user's location and primary users'signal coverage, which results in problems of the spectrum sensing. The spectrum sensing methods of static or quasi static scenarios are used directly to the mobile scenarios, which will degrade the detection performance of spectrum sensing. Therefore, the spectrum sensing based on the users' mobility is an important issue to improve the detection performance of the mobile cognitive radio networks.Based on the National Science Funding project: "Cognitive radio technology research and validation in TD-LTE system"(Grant No:2012ZX03003006) and Special Funding for Beijing Common Construction Project: "Cognitive radio based bus network application technology development",the paper studies the independence spectrum sensing algorithms,cooperative spectrum sensing algorithm, spectrum sensing algorithm of dense scenarios and spectrum sensing mechanism based on cognitive radio users' mobility of cognitive radios respectively.The main research work and achievements of this thesis are as follows:1. Survey on the spectrum sensing based on the mobility of the cognitive radio users and their related techniques. Firstly, the spectrum opportunistic, the users' mobility and the spectrum sensing of cognitive radios are summarized. Secondly, the research is mainly focused on the spectrum sensing based on the users' mobility. Thirdly, the open questions are summarized of spectrum sensing based on the mobility of cognitive radio users, which is submitted for study to the target of research.2. To the problem of degrading the spectrum sensing performance of the users' mobility leading to the time-varying channel, the spectrum sensing algorithm is studied in the time-varying channel. With the Doppler frequency shift impacting on the cycle frequency, the largest cyclic autocorrelation function is estimated, and then the statistic value is proposed by the energy of the function. The simulation and analysis show that the proposed algorithm can weaken the negative impact on spectrum sensing of the time-varying channel, reduce the complexity and improve the spectrum sensing performance.3. Aiming at the unreliability, even wrong sensing resulting problem of the spectrum sensing by which the users step out of the boundary of the primary signal coverage, the algorithm of cooperative spectrum sensing based on the mobile CR users' selection is proposed. The algorithm takes into account the factors of CR users such as mobile CR users' stability,correlation and detection capability. Ensuring the reliability of the spectrum sensing, and utilizing fully the spatial diversity of the users, the cooperative spectrum sensing algorithm improves the detection performance of spectrum sensing.4. Aiming at the problem of too high data dimension and strong signal correlation of the traditional spectrum sensing algorithm based on the users'mobility under the dense scenarios in mobile cognitive radio, the PCA(principal component analysis) based spectrum sensing algorithm is proposed. According to the lots of the CR users numbers and their random behaviors, the received signal powers of the CR users are small and the signal correlations are great, and then the spectrum sensing model based on high dimension matrix is built. With the signal correlation, the high dimension matrix is analyzed by the PCA, then the principal components acts as the test statistic to complete the spectrum sensing. The simulation and analysis show that the proposed algorithm based on PCA of spectrum sensing can solve the problem of too high data dimension with the signal correlation, and sense effectively the spectrum in multiple users' mobility under the dense mobile scenarios.5. In view of the problem that the spectrum opportunistic is underutilized due to the mobility of the cognitive radio users, a non-periodic spectrum sensing mechanism based on the channel state monitoring is proposed. The mechanism triggers the sensing spectrum based on the channel state transition. It can reduce the spectrum resources to be wasted and authorized user to be interfered, and overcome the problem of spectrum opportunity being underutilized. Importantly, it can guarantee the smooth transition of the time spectrum opportunistic and spatial spectrum opportunistic to improve spectrum efficiency.