低截获概率雷达信号的时频分析

发布时间：2018-01-06 19:26

本文关键词：低截获概率雷达信号的时频分析　出处：《南京理工大学》2014年硕士论文　论文类型：学位论文

【摘要】：低截获概率(Low Probability of Intercept, LPI)雷达具有低峰值功率、大时宽带宽积等出众优点,这导致传统截获接收机难以检测LPI雷达信号；同时LPI雷达多采用复杂信号调制方式,这也使得大部分传统信号参数估计算法的性能严重下降甚至失效。本文从LPI技术原理着手,对多个经典LPI雷达信号进行了性能分析,重点结合LPI信号的非平稳特点,详细研究了时频分布算法对LPI雷达信号进行检测及参数估计的性能,并对算法进行性能优化。主要研究内容包括： 1)研究了LPI原理及技术,并结合截获因子概念详细分析了LPI雷达的工作原理；从时域、频域以及模糊函数等角度对多种典型LPI雷达信号进行了详细的仿真分析。 2)研究了时频分布中的魏格纳-威尔分布(Wigner-Ville Distribution, WVD)及伪魏格纳-威尔分布(Pseudo Wigner-Ville Distribution, PWVD)推导了交叉项表示形式并分析了时频聚集性。不同噪声环境下,运用PWVD对各LPI信号进行仿真,仿真结果表明该算法具有很好的时频聚集性,但存在严重交叉项影响,降低了算法的整体性能。 3)针对WVD中交叉项过大的问题,采用乔-威廉姆斯分布(Choi-Williams Distribution, CWD)对各LPI信号进行分析及调制参数提取,仿真结果表明虽然交叉项得到了很好地抑制,但其时频聚集性发生明显恶化,严重降低了参数提取的精度。 4)针对交叉项过大或时频聚集性恶化的问题,提出采用信源熵对CWD进行参数优化的方法。首先讨论了该优化算法的原理,给出了具体实施步骤框图,并通过仿真得到所需优化参数；其次,采用优化后的CWD对各LPI信号进行仿真,仿真结果表明优化后CWD具有更优秀的性能,不但很好地抑制住了交叉项幅度,在时频聚集性方面也具有远好于一般CWD的表现,显著提高了参数提取的精度。
[Abstract]:Low Probability of Intercept (LPI) radar has the advantages of low peak power, large time broadband wide product and so on. This makes it difficult for traditional interceptor to detect LPI radar signal. At the same time, most of LPI radar adopt complex signal modulation, which makes the performance of most traditional signal parameter estimation algorithm seriously decline or even fail. This paper starts with the principle of LPI technology. The performance of several classical LPI radar signals is analyzed, and the performance of time-frequency distribution algorithm for LPI radar signal detection and parameter estimation is studied in detail in combination with the non-stationary characteristics of LPI signal. And the performance optimization of the algorithm is carried out. The main research contents include: 1) the principle and technology of LPI are studied, and the working principle of LPI radar is analyzed in detail with the concept of interception factor. In this paper, several typical LPI radar signals are simulated and analyzed in time domain, frequency domain and ambiguity function. 2) Wigner-Ville Distribution in time-frequency distribution is studied. WVD) and pseudo Wigner-Ville Distribution. PWVD is used to simulate the LPI signals in different noise environments. The simulation results show that the algorithm has good time-frequency aggregation, but there are serious cross-term effects, which reduces the overall performance of the algorithm. 3) aiming at the problem of excessive cross term in WVD, Choi-Williams Distribution is adopted. LPI signals were analyzed and modulation parameters were extracted. The simulation results show that although the crossover term is well suppressed, the time-frequency aggregation is obviously deteriorated. The precision of parameter extraction is seriously reduced. 4) aiming at the problem of excessive crossover or aggravation of time-frequency aggregation, a method for parameter optimization of CWD using source entropy is proposed. Firstly, the principle of the optimization algorithm is discussed, and the block diagram of concrete implementation steps is given. The optimal parameters are obtained by simulation. Secondly, the optimized CWD is used to simulate the LPI signals. The simulation results show that the optimized CWD has better performance, and not only suppress the amplitude of crossover terms well. In the aspect of time-frequency aggregation, the performance of CWD is much better than that of general CWD, and the precision of parameter extraction is improved significantly.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN957.51

【参考文献】