复杂环境下未知辐射源信号分选算法研究

发布时间：2018-05-31 22:02

本文选题：信号分选 + 改进PRI算法　；参考：《上海应用技术大学》2017年硕士论文

【摘要】：随着雷达信号调制技术的不断发展,辐射源的工作体制不断变化,信号分选坏境变得日益复杂,分选的难度也就不断增加。本文对复杂环境下未知辐射源信号的分选展开研究,注意到单一算法分选效果不理想的缺点,联合不同类型的算法进行信号分选。首先,分别从信号的频域、时域、空域三组参数出发分析信号的脉间信号特征,其中重点介绍时域参数及其主要的几种调制方式。详细介绍几种不同调制类型的辐射源信号,使用Matlab对信号源进行仿真,构建信号分选模型。简述传统PRI算法的工作流程,重点介绍PRI变换。对于大抖动下PRI变换算法存在的问题进行改进,介绍改进后的算法并进行仿真实验。通过观察各算法实验结果,判断各自优缺点。其次,简述盲信号分离及ICA (独立分量分析)算法的基本工作原理,重点介绍基于负熵最大化的FastICA算法,将此算法应用到信号分选领域。实验证明,此算法对不同调制类型的信号分选效果显著,但是对含噪信号较敏感。因此,运用小波去噪算法对信号进行去噪。综合两种算法进行实验仿真,可得出结论:基于FastICA和小波去噪的综合算法,在较低信噪比的情况下,信号分选效果较好。最后,对于复杂环境下信号的分选,本章给出了一种新的信号分选算法,此算法综合了正弦波抽取特性和改进后的PRI变换这两个算法。首先基于正弦波抽取特性的算法对信号进行调制类型的分类,在辐射源脉冲数减少的情况下,再使用改进后的PRI算法对相同类型下不同参数的信号进一步的分选。仿真实验表明此综合算法精准有效、性能较高。
[Abstract]:With the continuous development of radar signal modulation technology, the working system of emitter is constantly changing, the bad condition of signal sorting becomes more and more complex, and the difficulty of sorting becomes more and more difficult. In this paper, the signal sorting of unknown emitter in complex environment is studied, and the shortcoming of single algorithm is noticed, which combines different algorithms for signal sorting. Firstly, the characteristics of interpulse signal are analyzed from three groups of parameters in frequency domain, time domain and spatial domain, respectively, in which time domain parameters and their main modulation modes are introduced. Several emitter signals of different modulation types are introduced in detail. The signal source is simulated by Matlab and the signal sorting model is constructed. The workflow of the traditional PRI algorithm is briefly described, and the PRI transformation is emphasized. The problem of PRI transform algorithm under large jitter is improved. The improved algorithm is introduced and simulated. By observing the experimental results of each algorithm, the merits and demerits of each algorithm are judged. Secondly, the basic principle of blind signal separation and ICA (Independent component Analysis) algorithm is briefly introduced. The FastICA algorithm based on negative entropy maximization is introduced, and the algorithm is applied to the field of signal sorting. Experimental results show that the algorithm is effective for signal sorting with different modulation types, but sensitive to noisy signals. Therefore, wavelet denoising algorithm is used to denoise the signal. By synthesizing the two algorithms for experimental simulation, it can be concluded that the synthetic algorithm based on FastICA and wavelet denoising has better signal sorting effect under the condition of low signal-to-noise ratio (SNR). Finally, for the signal sorting in complex environment, a new signal sorting algorithm is presented in this chapter, which combines the sinusoidal wave extraction characteristics and the improved PRI transform. Firstly, the modulation type of signals is classified based on the algorithm of sinusoidal extraction, and then the improved PRI algorithm is used to further separate the signals with different parameters under the same type when the number of emitter pulses is reduced. The simulation results show that the algorithm is accurate and effective.
【学位授予单位】：上海应用技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN957.51

【相似文献】