Title: A cancellation method of additive external noise and state dependent noise in sound environment systems

Authors: Akira Ikuta; Nazmul H. Siddique

Addresses: Department of Management Information Systems, Prefectural University of Hiroshima, Hiroshima, 734-8558, Japan ' School of Computing and Intelligent Systems, University of Ulster, Londonderry, BT48 7JL, UK

Abstract: In the actual situation of sound environment systems, the observed data often contain several fluctuating noises. Therefore, it is necessary to estimate the specific signal based on the observed data by introducing a noise cancellation method. In this study, a sound environment system considering both noises of the external noise based on the additive property of energy variable and the internal noise dependent on the state variable is first paid our attention. Next, a noise cancellation method to estimate recursively the specific signal is derived on the basis of Bayes' theorem in an expansion series form suitable to a recursive signal processing. Finally, the validity of the proposed method is experimentally confirmed by applying it to the actual data observed in real sound environment.

Keywords: additive external noise; state dependent noise; sound environment systems; noise cancellation method; recursive signal processing.

DOI: 10.1504/IJAPR.2014.065766

International Journal of Applied Pattern Recognition, 2014 Vol.1 No.3, pp.224 - 241

Received: 28 Mar 2013
Accepted: 08 Aug 2013
Published online: 29 Nov 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article