【作者】 王益明；

【导师】 茅玉龙；

【作者基本信息】 南京信息工程大学 ， 信号与信息处理， 2011， 硕士

【摘要】 雷达信号分选是电子战领域的一项关键技术,信号分选能力决定了雷达侦察系统能否适应现代电子战环境。因此,提高侦察系统的信号分选能力,尤其是在信号密集且复杂,脉冲丢失率较大的情况下进行实时分选和识别,显得非常重要。本文主要研究现代电子战环境中的雷达信号分选,首先介绍了雷达信号分选的研究背景与意义以及国内外研究发展与现状。然后再介绍了雷达信号分选参数及其测量精度,随后阐述了雷达信号分选的主要方法,重点研究了未知雷达信号的分选流程,包括雷达信号预分选及主分选两个部分,预分选算法采用DOA, PW, RF联合分选,采用的是动态聚类法,具体实现算法是：改进的K-均值聚类算法。最大的优点是：无须事先划分存储空间,无须事先设定类别数目,这正好符合了现代电子战中脉冲密度未知和雷达辐射源数目未知的特点。主分选是基于PRI的分选,主要研究了相关函数法,CDIF, SDIF, PRI变换法,修正的PRI变换法,平面变换法。对于相关函数法子谐波问题,CDIF和SDIF采用避免的策略,PRI变换法采用抑制的策略,对PRI变换法作了两点修正：利用可变的时间起点,利用交迭的PRI箱。使其适应对抖动PRI的估计,仿真实验证明在抖动10%,脉丢失率20%下,估计结果正确,随后进行了序列检索和参差鉴别。在以上工作的基础上,提出了综合分选方案,方案顺序如下：已知辐射源的匹配与扣除,对剩余脉冲串聚类,对每一类进行SDIF分选,无法分选的,采用修正PRI变换,脉冲的类型鉴别,常规,抖动,还是参差。仿真实验表明,综合分选方案对参数的测量和分选非常准确。更多还原

【Abstract】 Radar signal sorting is a key area of electronic warfare, signal sorting capacity to determine whether the radar reconnaissance system to adapt to modern electronic warfare environment. Therefore, improving the detection signal sorting system capacity, especially in the dense and complex signal, pulse the case of the larger loss rate for real-time sorting and identification, it is very important.This paper studies the environment of modern electronic warfare radar signal sorting, first introduced the study of radar signal sorting and significance of the background and current situation of research and development at home and abroad. And then introduces the radar signal sorting parameters and measurement accuracy, and then describes the main radar signal sorting method and sorting system composition. Focus on the unknown radar signal sorting process, including pre-sorting and radar signal separation of two main parts, pre-sorting algorithm of DOA, PW, RF combined sorting, using a dynamic clustering method, the specific algorithm is: Improved K-means clustering algorithm. Biggest advantages are:no prior division of storage space, not the number of pre-set categories, which fits well with the modern electronic warfare and radar emitter pulse density unknown number of unknown characteristics. Sorting is based on the PRI primary sorting, correlation function are studied, CDIF, SDIF, PRI transform, modified PRI transform, plane transformation. Another way for the correlation function of harmonic problems, CDIF and SDIF strategy used to avoid, PRI transform using suppression strategy, made two o’clock on the PRI transform Fixed:use of variable time point, the use of overlap in the PRI case. To adapt the estimated PRI jitter, simulation results show that 10% of the jitter, pulse rate of 20% loss, the estimated results are basically correct. Subsequent identification of sequence retrieval and varied.Based on the work in the above proposed program of integrated sorting, program in the following order:the matching and less known radiation sources, the remaining burst cluster, the SDIF for each type of separation, not separation, and the modified PRI transform, the type of pulse identification, conventional, jitter, or mixed. Simulation results show that the comprehensive program of separation and sorting of the parameters in the basic accuracy.更多还原