Active microphonic noise cancellation in radiation detectors
View Patent ↗A new adaptive filtering technique to reduce microphonic noise in radiation detectors is presented. The technique is based on system identification that actively cancels the microphonic noise. A sensor is used to measures mechanical disturbances that cause vibration on the detector assembly, and the digital adaptive filtering estimates the impact of these disturbances on the microphonic noise. The noise then can be subtracted from the actual detector measurement. In this paper the technique is presented and simulations are used to support this approach.
1. A method of reducing microphonic noise comprising:
providing a radiation detector that receives radiation and generates a radiation detector output signal;
providing a sensor coupled to the radiation detector;
measuring a mechanical disturbance utilizing the sensor coupled to the radiation detector, wherein the sensor produces a mechanical disturbance signal;
identifying a system transfer function H(k) of an electro/mechanical system that injects microphonic noise into the radiation detector output signal by utilizing an identification algorithm that further utilizes a digitized mechanical disturbance signal that has been digitized by an analog to digital converter (ADC) and a digitized radiation detector output signal that has been digitized by a second ADC as inputs, wherein the identification algorithm modifies the output of the system transfer function H(k);
applying the digitized mechanical disturbance signal to the system transfer function H(k) to generate a digitized output signal; and
subtracting the digitized output signal from the digitized radiation detector output signal to reduce the microphonic noise.
2. The method of claim 1 , wherein the identification algorithm comprises a least means square (LMS) algorithm.