Research On Adaptive Noise Control And Algorithm Based On Blind Source Separation

As an important method for military investigation, the acoustic direction technology is an effective anti-electronic-interference and anti-break on low altitude means. The signal that passive acoustic detection system detected is inevitably polluted by noise in transmitting and receiving processes. How to effectively reduce the impact of noise to extract useful signal is an important aspect in sound detection. Therefore noise control will be particularly important. The Adaptive Active Noise Control (AANC) is an advanced noise control technology in recent decade.The dissertation firstly give the criteria of adaptive active noise control performance under the condition of the existence of useful signal. Under the guidance of this criteria, the Blind Source Separation (BSS) algorithm, as the pre-processor of AANC system, is introduced to this system. BSS theory is a new research direction in the signal processing field, whose basic model and performance evaluation standards are presented in this dissertation. On the basis of natural gradient blind source separation algorithm from information theory, the power factor and variable step length factor are introduced to improve the robustness of the adaptive algorithm. The adaptive algorithm used in this dissertation, which is on the basis of the standard LMS iteration formula and is improved on the basis of variable step LMS iteration formula, can speed up convergence of the algorithm and can get the effectiveness of the algorithm through computer simulation. At last, the simulation result is given to proved the inclusion that the introduction of the pre-processor can improve the performance of the AANC system and this algorithm is verified using experiment data.