Engine Noise of Identification using an Improved Blind Sources Separation Algorithm
Lin Yongman, Lin Tu-sheng · 2007
Engine noise of identification is a key technique on controlling engine noise. Traditionally, each of noises is separated to test. That is complicated, and experiment is needed many tested instruments. The paper proposes a novel approach to identify engine noise. The approach utilizes blind sources separation (BSS) to separated engine noise according to mixing noise of engine. Two experiment results indicate the effectiveness of this new algorithm that has low requirements for experiment environments and the amount of measure instruments. We approach an improved AC algorithm that is lesser time for separated mixing noise than the AC algorithm.