Author: Sasa Coh
University Program Computing and Informatics, Computer Software
Mentor: Asoc. Prof. Dr. Damjan Zazula
Co mentor: Asst. Prof. Dr. Danilo Korze
Date: Jun. 1997

Active Noise Cancellation

Keywords: digital signal processing, adaptive filtering, active noise control, least mean squares algorithm (LMS)

UDK: 681.32:616

Abstract:  

Using active noise control technique we want to remove unwanted components of noise. Conventional methods of suppressing acoustic noise using sound absorbers generally do not work well at low frequencies. Today's problem is the low-frequency noise. For these reasons, a number of practically important acoustic noise problems are dominated by low frequency contributions. The purpose of this work is to consider methods for canceling  the low frequency noise and implementation of practical active noise control system (ANC). Next, we outline the adaptive filtering theory and, in an experiment, we use the well known least mean square algorithm (LMS). In any particular implementation of active noise control it is important that both the physical principles of control as well as the electrical control considerations are thoroughly understood.

In our experiment we use a model with feedforward control. We are trying to minimize mean square error signal on an error microphone output. Program was running on personal computer INTEL P133 platform with built-in SB16 ASP sound card. The results of experiment show that the acoustic field level in surroundings of an error microphone was reduced by about 10 dB. The experiment can be used as presentation of active noise control technique, or as a basis for further study of similar systems.