在实际的工程测量过程中,噪声与系统固有信号往往是混叠而难以分离。为了对后续测量进行改进,需要对遥测信号进行精处理。如何剔除或抑制噪声,提取出测量信号中的固有信号,是对遥测数据进行精处理的基础。文中提出应用奇异值分解理论对遥测信号中的电量信号进行滤波处理。针对奇异值分解中有效秩对矩阵重构的重要性,应用统计聚类分析方法确定奇异值分解中的有效秩。通过与传统中值滤波结果进行比较,说明文中提出的方法能够应用于遥测信号的滤波处理中。

In practical project measurement, noise and system signal are always mixed together. In order to improve follow-up measurement, it is necessary to make precision processing for telemetry signal. How to eliminate or restrain noise and separate the signal from the noise is the foundation for precision processing. In this paper, singular value decomposition method was applied to process the telemetry signal, statistical clustering analysis method was applied to determine effective order according to importance of effective order on matrix reconstruction, and the filtering result was compared with traditional median filtering algorithm. Results show that the singular value decomposition method is better than the traditional median filtering algorithm. The proposed approach can be applied to remote sensing signal filter processing.