Research On Active Noise Control Methods Based On Adaptive Filtering Theory

Normally the passive noise control (PNC) approaches are more efficient for higher frequencyrange rather than lower frequency, while the active noise control (ANC) approaches have a goodability to control the lower frequency noise oppositely. The system of ANC mainly involves threeparts: controller link, secondary source link and error sensing link, in which the controller is the keypart. The system of ANC is studied in this paper based on the theory of adaptive filter, and the threeaspects including secondary path model, neutralization of secondary acoustic feedback and adaptivecontrol algorithms are analyzed.For the secondary path model, three different seconary path model algorithms are adopted tosuppress the internal systemic interference by using online additive random noise technique.For theneutralization of the secondary acoustic feedback, a method is applied by using online additiverandom noise technique, based on which the acoustic fedback path model is setup and the secondaryacoustic feedback signals are neutralized by the feedback neutralized signals. Three different kinds offeedback neutralization methods are then compared with each other. For the adaptive controller, it isthe most important part for the ANC system, which consists of two distinct parts: a digital filter toperform the desired signal processing, and an adaptie algorithm using a reference signal and residualerror to adjust the coefficients of that filter. Four diferent approaches are compared firstly, includingincluding the Filtered-x Least Mean square (FxLMS) algorithm, Filtered-x Recursive Least Square(FxRLS) algorithm, Filtered-x Gradient Adaptive Lattice (FxGAL) algorithm and Filtered-x GradientAdaptive Laguerre Lattice (FxGALL) algorithm. Based on four different types of filters, which areLaguerre transversal filter, Moment Gradient Adaptive Lattice filter, QR Decomposition based LeastSquare Lattice filter and Recursive Least Square Lattice filter using a Posteriori Estimation Errors anda priori Estimation Errors with Error Feedback, new structures and algorithms for ANC controller arethen proposed in this part. The new algorithms satisfy the requirements of ANC controller byreasonable deformation of the adaptive filters, and simplify the structure of ANC system, whichimproved the noise controlling ability finally.The noise characteristic in the cabin of airplane isintroduced at last, simulation cases are studied correspondingly, through which all the algorithmsproposed are vitified.