The Design And Implement Of Radar Signal Processor Based On CUDA

Continuous wave radar system has been widely used, as it has the advantage ofcompact structure, no fuzzy when measure speed, high accuracy, distinguishing movingtarget easily and easy for digital implementation, With the development of moderninformation electronic war, there are the increasing demanded functionality andperformance indicators of continuous wave radar, the data rate and the amount ofcomputation have been greatly improved. Due to the huge advantages of CUDA parallelcomputing platform, making acceleration solutions based on CUDA parallel computingplatform become a new way.Based on the function and performance requirements of continuous wave radar signalprocessor, this subject will be deeply study the signal processing algorithm and how thealgorithm implementation on CUDA platform. A method is presented to acceleratealgorithm with CUDA platform, designed and implemented continuous wave radar signalprocessor on CUDA platform. Test results show that the computing speed with CUDA hasimproved significantly compared to using of FPGA or DSP for signal processing, and theplatform also has advantages of low development cost and transplantable. Firstly this paperintroduces the background, significance and the domestic and foreign research status anddevelopment trends of CUDA-based platform for continuous wave radar signal processing.Then the radar signal processing algorithms is introduced. The GPU computing model,focusing on NVIDIA’s CUDA programming model, CUDA cores, thread-level, hardwaremapping, software architecture and Memory Organization are introduced in detail. Thenhow to implement the radar signal processing algorithms on CUDA platform, and theoptimization strategies of CUDA programs are presented. Finally, the paper introduces thedesign of radar signal processor software and testing of system performance. On thesmall-scale testing platform, the system performance parameters and the reliability andstability are tested. Focusing on the poor instability of the system when work long time,in-depth analysis and test the operating system and GPU Accelerator, and the solvingmethods are presented.