The Theory And Method Of Novel Adaptive Filters And Secondary Path Modeling For Nonlinear Active Noise Control

Noise control technology has been widely used in aircraft,military,industrial and transportation,et al.Passive noise control technology based on the absorption and/or reflection characteristics of materials was first used in various fields.However,due to the disadvantages of large,costly and inefficient at low-frequency,it is difficult to meet the increasing noise control requirements.Active noise control(ANC)technology based on the superposition principle has been widely studied and applied due to its low cost,effectiveness at low-frequency,and simple implementation.It is likely to become a standard technology for noise reduction in closed space in the future.This dissertation focuses on the theory and method of novel adaptive filters and secondary path modeling for nonlinear active noise control.In the research part of the novel adaptive filter theory and method,we proposed three types of new filters,those are,the even mirror Fourier nonlinear(EMFN)with linear finite impulse response(FIR)section(EMFNL)filter,the recursive even mirror Fourier nonlinear with linear infinite impulse response(IIR)section(REMFNL)filter and the function expansion bilinear(FEB)filter.Associated adaptive filtered-x least mean square(FXLMS)algorithms were derived for the proposed filters.In the research part of the secondary path modeling,we proposed two secondary path models,one is nonlinear saturation active noise control(NSANC)model for saturation nonlinearity,the other is sparse secondary path(SSP)model for reducing computaitional complexity of second-order Volterra secondary path.Finally,we built a feedforward duct active noise control system to verified some of our proposed theories.The main research contents and innovations of this paper are as follows:(1)To futher improve the performance of EMFN filter,a EMFNL filter is proposed.The associated adaptive filtered error least mean square(FELMS)(termed FFELMS)algorithm is derived for the proposed EMFNL filter.The computational complexity is analyzed and the computer simulations are performed to demonstrate the satisfied performance.(2)We improve the recursive EMFN(REMFN)filter by adding an additional linear IIR section termed REMFNL filter together with its channel reduced form(CRREMFNL)to explore better control performance.As for the linear feedback section introduced,we studied the bound input bound output(BIBO)stability condition for the REMFNL filter and design a stable control scheme to guarantee the filter satisfying the stability criteria.The performance of the proposed filters equipped with the associated FXLMS algorithm is validated through analyses of computational complexity and simulations of various nonlinearities for nonlinear ANC(NANC)systems.Simulation results demonstrate that the proposed REMFNL and CRREMFNL filters can achieve better performance over the standard REMFN filters based on the FXLMS algorithm.(3)Motivated by recursive filters,a general function expansion bilinear(FEB)filter with a 3-D diagonal-structure is proposed to deal with the problem of saturation nonlinearity in active noise control(ANC)systems.However,the huge computational burden and BIBO unstable problem hinder its practical applications.Therefore,we propose a general channel reduced efficient implementation algorithm and stable control method for the FEB filter to further reduce the computational loads and overcome the unstable problem.The performance of the proposed FEB filters equipped with the associated FXLMS algorithm is validated through analysis of computational complexity and simulations of various nonlinearities.(4)To overcome the saturation nonlinearity happened when when the acoustic noise and/or the controller output exceeds the dynamic limits of electronic devices,we use a new nonlinear saturation active noise control(NSANC)model and derive a function expansion diagonal-structure bilinear FXLMS(FEDBFXLMS)algorithm.The performance of the proposed FEB filters using the NSANC model is validated through analysis of computational complexity and simulations of various nonlinearities for NSANC systems.Computer simulation results demonstrate that the proposed FEB filter can achieve significant performance improvement in reducing saturation effects in comparison with the diagonal-channel bilinear(DB)filter and the recursive secondorder Volterra(RSOV)filter based on the FXLMS algorithm,which often outperform the conventional nonlinear Volterra and functional link artificial neural network(FLANN)filters.(5)To reduce the computationl complexity of nonlinear secondary path(NSP)model,we proposed a sparse secondary path(SSP)model with saprse kernels for second-order Volterra secondary path.The sparse FXLMS and FELMS algoriths were derived.We also provide an analysis of the NANC system to show that the noise at the cancelling point could be approximated by the function expansion filters when the secondary path is modeled as the second-order Volterra series.The computational complexity analysis,the simulations and actual system experiments validate that the new control algorithms with a SSP model for the secondary path can significantly reduce computational load without sacrificing the control performance.