Digital Signal Processing for Laser Printer Noise Source Detection and Identification
Yutong Xue, Nicholas Kim, Xihui Wang, Jan P. Allebach, J. Stuart Bolton, George T.‐C. Chiu, Patricia Marybelle Davies, Katy Ferguson, David Kaisle, Mark Q. Shaw · Purdue e-Pubs (Purdue University System) · 2019
Presented here is the description of a software-based noise source detector that was developed to simplify the printing noise source identification process. The typical noise sources studied here were high frequency, stick-slip (squeaking) noises associated with rollers along the paper path. The detector was built by combining several digital signal processing procedures to first create a tone detector, and then to calculate the modulation spectrum of the tone. The latter is particularly important because it has been found that features appearing in the modulation spectrum can be directly associated with the rotational speed of various components in the printer; this allows the faulty part to be identified. By passing different printing noise samples through the detector, it was found that the detector accurately returned the acoustical characteristics of each noise sample, and then by matching those acoustical characteristics to the rotational speed of different parts, the detector was proven capable of providing robust and precise fault identification results.