一种基于高性能GPU的实时脉内分析实现方案  

Design of real-time intra-pulse analysis of high performance GPU

作　　者：庄跃迁[1] ZHUANG Yue-qian(Southwest China Institute of Electronic Technology,Chengdu 610036,China)

机构地区：[1]西南电子技术研究所

出　　处：《电子设计工程》2019年第19期100-104,109,共6页Electronic Design Engineering

摘　　要：传统的雷达信号分选方法,主要依赖于雷达信号的到达时间(TOA)、脉冲宽度(PW)、载频(CF)、脉冲幅度(PA)、到达方位角(AOA)等常规参数,对于宽频率、高捷变、数量巨大且交叠严重的真实战场环境雷达脉冲而言,其分选性能非常有限,已不能满足复杂电磁环境的雷达信号侦察需求。采用高性能GPU,利用其数量巨大的处理单元,在信号分选之前即对雷达信号脉冲开展分析处理,获取其调制样式及调制参数,并将其引入雷达信号分选,可以极大提升雷达信号分选的性能。Traditional methods of radar signal sorting is mainly depended on conventional parameters, like the time of arrival(TOA),pulse width(PW),carrier frequency(CF),pulse amplitude(PA)and the angle of arrival(AOA). its sorting performance is very limited for the realistic battlefield environment where the radar pulses have wide frequency,quick variation,huge numbers and heavily overlapping, and can’t meet the requirements of the radar signal reconnaissance in the complex electromagnetic environments. Using the high performance GPU,which has the huge numbers of processing units,and doing the process and analysis of the radar signal pulses before the signal sorting to get the pulses’ modulation types and modulation parameters,can highly improve the performance of the radar signal sorting where the parameters can be used.

关 键 词：实时脉内分析 信号分选 图形处理器 调制识别 

分 类 号：TN971[电子电信—信号与信息处理]