The Study On Algorithm And Experiment Of Active Noise Control System

From testing cabin noise, researcher found not only the medium and high frequency component, but also large proportion of the low frequency component in living and working cabin which are mainly distributed in 30Hz~500Hz, however, it's hard for traditional noise control technique to attenuate low frequency noise, so the purpose of this paper is to study the Active Noise Control (ANC) technique.Firstly, simply analyzed the technique of Single Channel Active Noise Control and Multi-channel Active Noise Control in the mechanism of physics, then simulate the influence of the position of sub-source. Simply introduced the algorithm of Filter Least Mean Square variance (FLMS) based on the simple system model, and analyze the quantity of operation, simply deduced the algorithm of filtering recursive least mean square variance (FRLS),filtering normalized least mean square variance (FNLMS) and so on, introduced the theory of error channel modeling. Simulated these algorithms, respectively compared the capability of attenuation from different angle such as the frequency of noise source, the length of filter, the coefficient of convergence, the rate of single and noise.Then, propose the concept of Active noise control with virtual sensors, this algorithm overcome the difficulty of depositing , and simulate the capability of this algorithm. Respectively compared the capability of attenuation of two sensors system and three sensors system.Finally, proved the algorithm in experiment by VC5509A development board, and analyze the data collected by the system, found that the system has the capability of attenuation and there are some influences which severely affect the system.