10 Common Software Filtering Methods In Embedded

The following is a detailed introduction to the 10 common software filtering methods commonly embedded:

10 common software filtering methods of embedded type.png

1.Limited filtering method: This method is mainly filtered according to the changing amplitude of the signal.If the change of the signal exceeds the set threshold, it is considered that the signal is the interference signal and filter it out.The advantage of this method is simple and easy, but it may be filtered out of some valid signals with large changes.

2.Middle value filtering method: Middle value filtering is a non -linear filtering method. The basic principle is to sort the signal value in a window and take the median number as the output.This method is very effective for eliminating the occasional abnormal values and can effectively filter some random noise.

3.Arithmetic average filtering method: The average filtering of the arithmetic is a linear filtering method. The basic principle is to perform an arithmetic value on the signal value in a window to obtain the signal value after the filter.This method can effectively reduce the interference of random noise, but it will produce a certain smooth effect on the mutation of the signal.

4. Require the average filtering method: The average filtering is a real -time filtering method. The basic principle is that each time it samples to a new data, first of all with the previous data, and then take a new average as averages as averages as averages as averages as average value as averages as average value as averages as average value as average value.Output.This method can effectively reduce the interference of random noise, but it will still produce a certain smooth effect on the mutation of the signal.

5. Middle value average filtering method: The median average filtering is a filter method that combines the two methods of the median value and the average method.First of all, sort the signal values in the window, then take the median number as the output, and average the medium number.This method can effectively remove abnormal values and reduce the interference of random noise.

6. Liming average filtering method: Liming average filtering is a filtering method for improvement on the basis of arithmetic average filtering.It compares each sampling value with the sampling value of the previous moment. If the difference exceeds a certain threshold, it is considered that the sampling value is an abnormal value and is not used.

7. First -order lag filtering method: First -order lag filtering is a simple time domain filtering method.It enters the sampling value of the current sampling value with the previous momentCompare and take the average value of the two as the output.This method can effectively reduce the interference of random noise, but it will still produce a certain smooth effect on the mutation of the signal.

8.Weighted recursive average filtering method: The weighted average filtering is a filter method for improving the improvement of the average filter.It has a weighted average of each sampling value based on different weights to obtain more accurate filtering results.

9.Trembling filtering method: Tibetan filtering is a filter method for eliminating mechanical switching.When the mechanical switch is disturbed by the outside world, the trembling filter detects this jitter and eliminates it.So as to get a stable and reliable output signal.

10.Limited Sharing Filtering Method: Limited Shapeing Filter is a filtering method for improving on the basis of trembling filtering.It compares each sampling value with the sampling value of the previous moment. If the difference exceeds a certain threshold, it is considered that the sampling value is an abnormal value (that is, the switch jitter), and it will not be used.Trends trembling.This method can effectively eliminate switching and obtain stable and reliable output signals.