基于稀疏表示的电磁兼容测试信号处理技术研究

With the increasement of the electromagnetic compatibility site-test requirements of large industrial automation and weapon system, EMC mechanism analysis, electromagnetic interference source identification and localization are gradually becoming some pressing problems in EMC test signal processing with nonstandard test conditions. This research focuses on the relative theory and technology problems involved in EMC test signal processing with sparse representation theory. The purpose is to put the axe in the helve about weak signal detection and EMC interference source identification and localization in complicated noise environment. The major research contents are included in the following: By the signal representation, EMC test signal can be decomposed into atoms which provide sparse information that represent the signal time-frequency signature. A novel method, which is capable of processing multicomponent chirp signals even in the presence of strong intended interference and colored noise, would be proposed to eliminate the electromagnetic environment in nonstandard test setting. Research on underdetermined blind signal separation based on sparse representation would focus on the electromagnetic interference source identification. Furthermore, a fuzzy observer passive location model using sparse domain information will be studied. These systematic studies aims for improving the location speed and precision while deal with large system. The innovative and pragmatic achievements of this research would enrich the EMC test signal processing means and provide new ideas and methods, especially for EMC site-test signal processing.

随着大型工业自动化系统、武器系统电磁兼容现场测试的需求越来越多，非标准试验环境下的电磁兼容测试、电磁干扰故障诊断、电磁干扰源信号分离和定位逐渐成为电磁兼容测试信号处理迫切需要解决的重点和难点问题。本课题为解决复杂环境背景噪声滤除及电磁干扰盲源分离、定位等难题，针对基于稀疏表示的电磁兼容测试信号处理所涉及的理论和科学技术问题进行系统性和探索性的研究。具体内容包括：研究电磁兼容测试信号稀疏表示方法，解决多分量信号的有色噪声滤波问题，以实现非标准条件下电磁兼容测试中的环境背景噪声消除；研究电磁干扰源欠定盲源分离理论，并利用稀疏域信息进行模糊电磁干扰源定位方法研究，以实现大型系统电磁干扰源的快速、精确定位。课题研究成果将拓展传统的电磁兼容测试信号分析处理方法，从而为复杂系统电磁兼容现场测试提供新的思路和技术途径。

随着大型工业自动化系统、武器系统电磁兼容现场测试的需求越来越多，非标准试验环境下的电磁兼容测试、电磁干扰故障诊断、电磁干扰源信号分离和定位逐渐成为电磁兼容测试信号处理迫切需要解决的重点和难点问题。本项目重点解决非标准环境测试中的环境背景噪声消除及电磁干扰源的分离和定位问题。项目针对电磁兼容现场测试信号在时频域、空域具有的时变、非平稳特征，基于信号稀疏表示理论，解决了电磁兼容测试信号稀疏表示问题，并提出了基于时频原子剪裁的噪声滤波方法，解决了多分量信号的有色噪声滤波关键问题，以实现非标准条件下电磁兼容测试中的环境背景噪声消除；项目进一步提出了稀疏域内电磁干扰源欠定盲源分离理论和方法，实现用尽可能少的测试天线从电磁兼容测试信号中分离出电磁干扰源信号；利用电磁干扰源信号的时频原子特征，借鉴传统无源定位的思想，提出基于时频原子参数、参数变化率等信息的电磁干扰源模糊定位方法，以减少测试天线数目、提高干扰源定位精度。项目提出的方法和结论，具有重要的理论意义。对大型工业自动化系统、武器系统非标准试验环境下的电磁兼容测试具有实际工程参考价值。在下一步的研究中，将在项目研究成果的基础上，加强理论成果和实际测试数据的验证和工程化。