SearchPublisher and Institutes
You are here: Home :: Area NEM :: Engineering :: Machine construction
Mirko Knaak, Dieter Filbert
Multi-channel Signal Estimation for Machine Diagnosis
The goal of acoustic machine diagnosis is to assess the state of a machine by the sound emitted. Since this sound is frequently interfered by jammer sound, this paper proposes to suppress the jammer by multi channel signal processing. The proposed algorithm is based on ICA (Independent Component Analysis). In contrast to blind application of that method the algorithm employs geometrical information. It solves the permutation problem of the blind source separation of acoustic mixtures, and is significantly less sensitive to the precision of the geometrical constraint than an adaptive beamformer. A high degree of robustness is very important since the steering vector can only be roughly estimated in a reverberant environment, even when the direction of reception is precise. It is theoretically and experimentally analyzed with respect to the roughness of the steering vector estimation by using impulse responses of a real room. The effectiveness of the algorithms for real-world mixtures is also shown for three sources and three microphones.
tm – Technisches Messen, Oldenbourg Wissenschaftsverlag
Print ISSN: 0171-8096
Volume: 71, 04/2004
Pages: 269 - 277