利用CUDA加速连续波雷达测速算法

A New Approach to Accelerate Velocity Measurement Algorithm of Continuous Wave Radar Based on CUDA

针对连续波雷达中事后分析系统的运算数据量大,运算时间长的特点,提出了一种利用CUDA平台的GPU进行加速运算的方法,它利用GPU的并行运算功能来完成FFT算法。实验结果表明,相比以往用CPU进行事后分析,利用CUDA平台进行事后分析的速度有明显提升。

The process of velocity measurement algorithm based on CPU takes much of time because of the complexity of computing,a velocity measurement algorithm based on CUDA is introduced.The experimental results show that the proposed method is more efficient than CPU for accelerating velocity measurement algorithm of continuous wave（CW） radar.