Abstract: The aim of signal de-noising is recovering the original signal from the noisy environment. A new de-noising method for telemetry signal based on Principal Component Analysis (PCA) was proposed. First a one-dimension time series can be embedded to equivalent multi-dimensions through the Reconstructed Phase Space (RPS) method. Then it can extract the telemetry signal by PCA because the telemetry signal was the principal component. The experimental results show that the de-noising performance of the proposed method is good, and also is very similar to the wavelet de-noising method. Simulations verify that the proposed method possesses good performance and better self-adaptive ability compared to other methods.

Key words: Reconstructed Phase Space (RPS), Principal Component Analysis (PCA), telemetry signal, noise cancellation

摘要： 信号消噪的目的在于有效地滤除噪声干扰以达到对原始信号的真实复现。提出了一种基于主分量分析的遥测信号噪声消除方法。首先利用相空间重构(RPS)将一维时间序列信号嵌入为等效的多维时间序列信号，然后利用主分量分析(PCA)实现信号的提取，其实质是利用了真实信号在混合信号中占主导地位的特性。实验结果表明，该方法去噪效果明显，且与小波去噪效果比较接近。与一般滤波方法相比，具有效果好，自适应较强等特点。

关键词: 相空间重构, 主分量分析, 遥测信号, 噪声消除