AD转换器数据处理（AD converter data processing）.doc

AD转换器数据处理（AD converter data processing） Theoretically SCM from A/D chip acquisition signals is to quantify, but due to interference of the circuit, power supply noise interference and electromagnetic interference, on the A/D chip analog input signal superposition of periodic or aperiodic jamming signal, and will be attached to the quantization value, to signal degradation. Considering the real-time and security of data collection, it is sometimes necessary to soft handle the data collected and minimize the influence of interference signals. This process is called data acquisition filtering. Here are ten methods and programming examples of data collection filtering. These 10 methods have different performance for different noise and sampling signals, and provide a wide selection space for applications of different occasions. When choosing these methods, it is necessary to understand the main types of noise in the circuit, mainly including the aspects: * noise is a random noise or periodic noise * the frequency of noise * the type of the sampling signal is a block change signal or a slow change signal * also consider the resources available for the system Through the analysis of noise and sampling performance, the best method and reasonable parameters are selected to achieve good results. Currently used in the main method of filtering the data collection has the following 10, the 10 ways are processed in time domain, from the view point of frequency domain relative to the design of IIR and FIR filter, its computational complexity is small, and easy to implement and performance can meet the application requirement of most of the occasion. 1. Limited filter method (also known as program judgment filtering method) Implementation methods: (1) determine the maximum deviation value (set as A) of the two sampling permits according to the experience judgment; (2) when the new value is detected, the value of this value is valid if it is less than or equal to the value of the p