On The Design And Performance Analysis For Spectrum Sensing And Communication Systems

Cognitive radio (CR) technology serves to solve the problems of spectrum shortage and spectrum efficiency enhancing in the legacy primary user (PU) links. By sensing the legacy channels, CR users can identify whether the bands they are in being utilized by the PU. They use the spectrum holes smartly by using efficient transmission strategies, transmitting their own information with interference strengths within acceptable levels to protect the PU service. CR technology substantially promotes the efficient use of PU link, which represents the direction where the future wireless communications go. Consequently, it attracts enormous attentions all over the world.Supported by the fundamental research funds for the central university (No. 017125), this thesis investigates spectrum sensing, data frame structure and the design of cooperative transmission schemes in depth in the aspects of physical layer and data link layer, derives schemes and analyzes performance. Detailed contents are listed as:Investigations on the performance analysis of spectrum sensing:Performance of energy detection in the generalized fading channels is presented. This thesis focuses on the scenarios with high signal to noise ratio (SNR) and low SNR in sensing channels. When SNR is high, the approximate expression of the missing detection probability is derived based on which sensing gain is proposed. The values of sensing gain are evaluated in various fading conditions and reception schemes. When the SNR in PU-CR link is low, the approximate relationship between the false alarm probability and the detection probability is derived. Then, this relationship is extended to the case where cooperative spectrum sensing using counting rule is utilized. Finally, the impact of error happened in control channels on sensing performance is investigated. The relationships between the bounds of overall false alarm probability and overall detection probability and the bit error probability (BEP) are derived. The upper bound of the error probability is derived given that the BEPs are equal in all the control channels.Investigations on the design for cooperative spectrum sensing:The optimization on cooperative spectrum sensing is considered in the link with burst nature PU. The sensing error probability is derived and named after average error probability (AEP). In order to maximize the utilization of PU link, the optimization is modeled using AEP as the objective function and MAP-based missing detection proability as the constraint function. The optimal fusion rule is given in Algorithmâ… and the optimal solutions of the threshold in the energy detector and the fusion rule are derived in Algorithmâ…¡. Numerical results show that the AEP hits its lowest point(obtained by exhausted search) with our proposed Algorithms, which means the utilization of link reaches its highest point.Investigations on the design of CR data frame:This thesis proposes the optimization model with the data rate received correctly as the objective function. The sensing-throughput tradeoff problem is reconsidered under our proposed model and a high efficient algorithm (Algorithmâ…¢) is given to calculate the length of slot allocated for spectrum sensing. The optimization is also considered under the framework of cooperative spectrum sensing. In order to maximize the throughput of CR users, the optimal length of slot for sensing and the optimal fusion rule are derived and the solution is presented in Algorithmâ…£. At last, comparisons are given between the obtained throughput achieved by our proposed algorithms and the ones achieved by schemes with fixed sensing length and fusion rule.Investigations on the design of cooperative transmission:Research is made on the design of cooperative transmission in the scenario of burst PU traffic. The transmission schemes "CR-AF" and "R-CR-AF" are proposed and the outage probability, diversity order and the transmission length are derived under both schemes. Finally, comparisons are given and it shows that the R-CR-AF scheme outperforms the CR-AF scheme in term of reliability (outage probability and diversity order) while it works inferiorly in the sense of efficiency (transmission length).In summary, investigations are made in depth in the physical layer and data link layer of CR networks. The research on the physical layer focuses on the performance analysis of energy detection whereas the research on the data link layer formats new optimization model and derives solutions with proposed algorithms. Broad prospect of our proposed models and algorithms can be uncovered for various applications, e.g. wireless local area networks, Ad-Hoc networks, sensor networks composed by CR users.