基于GPU的扩频测控信号快速捕获算法研究  

Fast Acquisition Algorithm for Spread Spectrum Signals Based on GPU Acceleration

作　　者：刘飞[1] 张纪成[1] 肖银[1] 张迅[1] LIUFei ZHANG Ji-cheng XIAO Yin ZHANG Xun(Unit 75707 of PLA, Nanning 530000, Chin)

机构地区：[1]中国人民解放军75707部队

出　　处：《军事通信技术》2016年第4期14-18,共5页Journal of Military Communications Technology

摘　　要：文章主要研究了两种基于GPU的扩频测控信号快速捕获的算法,即基于GPU的频域并行捕获算法和基于GPU的时域并行捕获算法,采用并行算法,使用CUDA、VC和Matlab进行了仿真和比较。仿真结果表明,基于GPU的时域并行捕获算法并行化程度高,较基于GPU的频域并行捕获算法速度快约10倍,较基于CPU的时域并行捕获算法快约50倍。Two kinds of acquisition method of spread spectrum signals based on GPU acceler- ation were developed in this paper, namely, the frequency domain parallel acquisition algorithm and the time domain parallel acquisition algorithm. The parallel algorithm was implemented first- ly. Then, CUDA, VC and Matlab were used for simulation and comparison. The result shows that the time domain parallel acquisition algorithm based on GPU acceleration has a better per- formance than the frequency domain parallel acquisition algorithm based on GPU acceleration. The former is about 10 times faster than the latter. In addition, it is about 50 times faster than the acquisition method based on CPU.

关 键 词：图像处理器 并行算法 扩频测控 捕获 

分 类 号：TN919.81[电子电信—通信与信息系统]