论文部分内容阅读
在设计通常的(固定参数的)最佳滤波器时,必须对信号和噪声的统计特性具备先验知识。但是,在实际上,往往无法预先知道这些统计特性,或者,它们是随时间而变化的,这时,便无法实现最佳滤波。自适应滤波器的特点是。它可以“自动”地调节其自身的参数;而在设计这种自适应滤波器时,需要很少的、或根本不需要任何信号和噪声的先验(统计)知识。自适应噪声消除器(ANC)就是利用自适应的方法来实现一维纳(Wiener)滤波器。它可以在对信号和噪声不具备先验知识的条件下,利用一自适应滤波器(AF)来“自动”地实现维纳滤波。 本文简述和分析了AF、ANC和自适应运算的工作原理;根据实际需要(诊断一种疾病),采用多路ANC来提取强噪声背景下的有用信号。在计算机上进行了单路、多路ANC的模拟试验。试验结果表明:利用多路ANC,可以在对信号和噪声具备极少先验知识的条件下,有效地提取在强噪声背景下的有用信号。
When designing the usual (fixed-parameter) best filter, prior knowledge of the statistical properties of the signal and noise must be available. However, in practice, these statistical properties are often not known in advance, or they change over time, at which point optimal filtering can not be achieved. The characteristic of adaptive filter is. It can adjust its own parameters “automatically”, whereas in the design of such adaptive filters little or no prior knowledge of signal and noise is required. Adaptive noise canceller (ANC) is the use of adaptive methods to achieve a Wiener filter. It can “automatically” implement Wiener filtering using an adaptive filter (AF) without prior knowledge of the signal and noise. This paper briefly describes and analyzes the working principle of AF, ANC and adaptive computing. Based on the actual needs (diagnosis of a disease), using multiple ANC to extract the useful signal in the context of strong noise. In the computer on a single, multiple ANC simulation test. The experimental results show that the multi-channel ANC can effectively extract the useful signal under strong noise under the condition of little prior knowledge of signal and noise.