Cognitive Radio Nc-ofdm Technology Research

With the rapid development of wireless communication technique, limited spectrum resource becomes the main bottleneck of an increasing number of wireless communication applications, which draws much concern on the conflict between spectrum allocation and utilization. As an effective solution to ever-growing needs of spectrum resource, Cognitive Radio(CR) has become a new hot spot in related literature. Cognitive Non-Contiguous Orthogonal Frequency Division Multiplexing (NC-OFDM) divides the frequency band into several sub-carriers and only uses those unoccupied sub-carriers for data transmission, which guarantees desired flexibility on several aspects such as spectrum cutting, combination, power control, etc., and enables dynamic spectrum access. Therefore, it is considered as a competitive candidate for transmission techniques of adaptive link in Cognitive Radio. NC-OFDM system breaks the limitation of traditional wireless communication systems in which data must be transmitted in a contiguous frequency-band, while it also introduces new challenges on a number of key techniques, among which how to suppress the interference caused by the coexistence of multi-users is an important problem left open.Based on previous research, this thesis discusses the technical mechanism, characteristics and new challenges of NC-OFDM system, provide simulation results of the NC-OFDM system using simulink for analysis of the BER under different spectrum allocations. As our main concentration, the synchronization and interference suppression strategy are explored, under certain assumptions, some feasible ideas and algorithms are proposed and validated by theoretical analysis and simulation experiments. In addition, a NC-OFDM system experiment platform has been set up using hardware devices such as FPGA. The main contribution of our work is listed as follows: Firstly, Normalization ML estimation of time and frequency offset algorithm is investigated comprehensively and applied to NC-OFDM system, the performance analysis based on matlab simulation results and discussion of its utility scope are also provided. In addition, three classical timing synchronization algorithms, i.e. Schmidl&Cox, Minn, and Park, are simulated and compared, from which an adaptive synchronization algorithm base on training sequence for NC-OFDM system are proposed, which yields a single peak on timing metric, and promises satisfactory timing accuracy.Secondly, the problem of mutual interference between users is endeavored, the model describing interference between cognitive users and licensed users in cognitive NC-OFDM wireless is established with detail, and some existing interference suppression methods are also provided. Active interference cancellation(AIC) algorithms are also introduced with detail, of which the windowing and AIC combined algorithm is modified, and the modified algorithm is demonstrated to have a better performance on interference cancellation in simulation. The pre-filtering method is then introduced for interference removal at primary users in supplement of the newly-proposed adaptive synchronization algorithm, whose performance is evaluated by simulation results in matlab.Finally, a NC-OFDM system experiment platform has been set up using hardware devices such as FPGA, along with function and goal description of each module in the NC-OFDM system. The implementation of each module is also described in detail. Test results of some of the modules are also provided.