The Optimized Design Of OFDMA-based Cognitive Radio System Design

Cognitive radio (CR) is an innovative and intelligent radio design technique motivated to increase spectrum utilization by the way of exploiting unused or low utilization spectrum resources which are already authorized to primary systems in dynamically changing environments. CR cognizes its environment, analyzes the channel state information and then makes decisions on transmission parameters, thus it has the ability of acting according to its environment. Orthogonal frequency division multiple access (OFDMA) is a potential technique for CR networks' air interface. With theoretical analysis and algorithm simulation, this paper is mainly focused on OFDMA-based CR system design including system platform design, transmission mechanism, resource allocation and power control.Firstly, a brief introduction to the background, current status of cognitive radio and problem confronted of OFDMA-based CR system is presented in Chapter 1.In Chapter 2, a framework of OFDMA-based CR transceiver is built. The feasiblity of using OFDMA in CR system is analyzed, and the cognitive cycle of CR device is illustrated later. The flow of transceiver contains spectrum sensing, data modulation, adaptive control. The initial process of link setup is further discussed, and a timing synchronization and frequency domain compensation algorithm is presented with simulation analysis.The software and hardware design of the whole platform is an emphasis of this thesis. The platform is designed based on the concept of Software Defined Radio with the objective to provide a hardware testbench for verifying key techniques in CR system. This platform is composed of embedded system, FPGA, DSP and AD/DA. Each of these parts is an individual system with distinguish developing environment, while these parts should work cooperatively as an integrated system. In consequence, the difficulties of platform design lie in customized system porting, software development and interface connecting.In Chapter 4, a novel power control method is proposed. According to the mutual interference between the primary users and cognitive users in spectrum sharing CR networks, a power spectrum profile generation algorithm is presented. A progressive access power control algorithm is further discussed under power constraint. Simulation results demonstrate the larger capacity of cognitive users under restrictions on transmit power. In Chapter 5, the problem of joint power and spectrum allocation is studied. An optimatiation model is formulated under the constraint of interference, with the objective of maximizing the spectrum utilization. Then a solution based on genetic algorithm is proposed, and it has been shown that the proposed genetic algorithm can obtain a better performance than the heuristic algorithm through simulations.Although both the effectiveness and practicality are taken into account in this thesis, there is still much work to do in the future e.g. adaptive coding control, distribute rate/power/spectrum control, etc.