Design And Implement Of An Analysis And Evaluation System For Sdr Receiver Front-end

Software-defined radio as a new communication architecture has been widely concerned and is increasing popular because of its standardized modular structure, strong flexibility and openness. But wide-band digital receiving front-end usually face the problem that the desired weak signals are probably submerged by the nonlinear distortions generated by large signals or strong interferes, which restrict the performance of digital receivers. In this case, the nonlinear signal processing technology which is used to improve the performance of the receivers has been developed vigorously. All kinds of nonlinear signal processing techniques have been applied to the receiver front-end, but due to the difference of their effectiveness and stability, the receiver front-ends differ greatly. In order to measure the performance of receiver front-end, it is becoming a hot spot to find the appropriate standard test method and system to evaluate its performance and technology status.Based on the hardware platform of the high-speed and real-time receiver front-end, this paper designs and implements a special system to analyze, evaluate and test the performance of receiver front-end. Firstly, the paper introduced the receiver front-end platform, proposed the system requirements based on the analysis of methods for Performance Evaluation, and used the unified modeling language(UML) to carry on the demand analysis modeling. Secondly, according to the MVC model, the paper designed the overall scheme of the system, and the system will be designed in modules, software flow, Multi-thread and error processing. Thirdly, the paper described the detailed design and reality of system, especially for data processing such as symbol synchronization, frequency offset estimation, phase offset estimation, signal-to-noise ratio calculation and EVM calculation, and we used the modeling method of class diagram to determine the relationship between each class. Finally, we tested the system designed and implemented in this paper to verify whether it is in compliance with requirements.In this paper, we use Python and C++ mixed programming, it is a bold and successful attempt compared to the existing signal measurement and analysis software. In addition, the frequency offset estimation algorithm for signal processing is improved, which greatly improves the performance of the system. At last, this paper introduces the method of subtraction clustering into EVM parameters for the first time, which brings great convenience and flexibility for the realization of the project. The application of the above three key technologies makes the software has the advantages of beautiful interface, smooth operation, strong operability and short development cycle. The results show that the system is reasonably designed. It can satisfy all the requirement of the task.