Study And Implementation Of High-Speed Spectrum Analysis Algorithms Based On Software Defined Radio

The rapid development of wireless communication has made a great difference, which can be shown by our feelings about the considerate service and convenient communications. Thus, it forces the requirements for the compatibilities among systems of different standards; moreover, platforms for communication should be highly flexible and extendable. Consequently, Software Defined Radio (SDR) inevitably becomes the best possible solution in that it has an open structure with standardization and modularity. The fully programmable and digitalized features, as well as the hardware capabilities of providing real-time processing, make it spread in wide use, like bridging the mobile communication between the 3G to 4G, auto-connections among military radios and so on. Otherwise, many algorithms for Power Spectral Analysis(PSA) are good in resolution, variance and tracing real-time signals, which make PSA seen in many areas such as communication, radar, biomedicine and so the forth. It is obvious that PSA has remarkably enhanced the related equipment and performance. However, PSA algorithms distinguish themselves among various applications in performance. Therefore, a proper combination of PSA with SDR will significantly make PSA more flexible, general and effective.Our paper aims at building up one SDR platform with high performance, on which, total 10 classical and modern PSA algorithms are implemented, detail efficiency and performance analyses are given and finally the conclusion is made accordingly. It mainly consists of several parts: introductions to the PSA theory, hardware selection and system buildup of SDR, realization of PSA with optimization, the designs of high-speed data transfer module and GUI, performance analysis and statistics of PSA on SDR.According to our simulink results and given the input noise mixed with multiple complex sinusoidal signals, the optimized high-speed PSA algorithms were implemented on our SDR platform. Additionally, PSA performance curves under different algorithms, effectiveness and limitation analyses were given as well. Finally, conclusive suggestions concerning real-time and performance requirements were made, which paved the way for our further research.