The Study Of Active Noise Control System Based On DSP

Noise pollution has attracted much attention as the Third Environmental Pollution ofthe world. It is harmful to the health of the person working or studying under it. Noise alsohas a relationship with the character of weapon in the military area.. Active Noise Control(ANC) is a good method to reduce the noise using the interference of the sound wave, howto reduce noise is widely studied recentlyIn the ANC, we need that output has the same frequency and amplitude with the noisesource and the phase difference is180. However, the source of noise may be unknowned,we make the filter adjust parameters adaptively according to the previous value time toproduce what we want. In this paper, we design a single-channel ANC system based onLMS adaptive algorithm which is easy to realize and a single-channel ANC systemcontrolled by host computer. Local space noise is generally seemed as the synthesis ofcertain number of complex frequency source signals in this space range from the wholedirections and different spread channels. All these noise sources are strong correlationsignals with some bandwidth, in other words, they have the same frequency composition.We design a Multiple-channel control system based on the blind signal separation(BSS)due to the ineffective reducing noise of Dual-channel control system based on LMSalgorithm.In our paper, we choose the TMS320VC5509A DSP digital signal processor, which iseffective in the digital signal process algorithm of voice identification and voice synthesis.Experimental results show that our Multiple-channel control system can obtain noisereduction with8~10dB(A), better than the1dB(A) of single-channel ANC system based onLMS adaptive algorithm and13~16dB(A) of single-channel ANC system controlled byhost computer. At last part, we give an outlook of these different systems and proposesome improved methods.