# How to design a filter that can filter out noise accurately, after setting the parameters of the filter using standard signal?

I meet a problem with designing a filter. I have two different instruments that could measure the same AC signal (usually ~200hz, always <1kHz), A and B. A can carry out signal measurement during the normal operation of the instrument. B can only be used during instrument calibration, and A can also be used at this time. B can accurately measure the signal. The measurement accuracy of A is lower than that of B.

I want to design a real-time filter for A to filter out the interference noise. The parameters of the filter are adjustable. When the instrument is calibrated, both A and B are used to measure the signal. The measurement result of B is taken as the standard result to calculate the parameters of the filter adaptively. After setting the parameters of the filter, A is used to measure the signal without B.

I have investigated a variety of filters and found that the adaptive filter is quite suitable for my needs. However, the adaptive filter needs to input two signals, while I can only provide the measurement result of A during the operation of the instrument. So the adaptive filter is not completely suitable for my needs.

I don't know how to solve this problem. Is there any method that suits my needs? Thank you very much!

• Is the signal you measure is a pure sine with unknown frequency in the range (0, 1000) [Hz]? – Royi Oct 19 at 6:08