Online secondary path modeling method with auxiliary noise power scheduling strategy for multi-channel adaptive active noise control system

Based on the analysis of the characteristics of practical noises in a driver's cab, and aiming at the real time signal processing requirements, the high speed digital signal processor is used in the active noise control system. And combined with the nonlinear characteristics of the noise, an adaptive active noise control project based on the radial basis function neural network is proposed, at the same time an filter algorithm is given. The result proves the proposed adaptive active noise control system to be feasible and effective.

The secondary path is essential in the design of adaptive active noise control systems, and it varies with several factors in practical implementation. In this paper, the acoustic responses from secondary loudspeakers installed at the shoulder of the front passenger seat to the error microphones at the passenger's ears in an automobile cabin are measured, when the factors including the passenger movement, the seat posture and the passenger number change. The upper limit frequency where the phase variation is within 90 deg. compared to the reference position is analyzed. The results show that the state of both the passenger head and the seat back determines the lowest upper limit frequency among all the factors under investigation. If multiple factors change at the same time, the upper limit frequency will be further reduced.

Soviet Physics—Acoustics

Online secondary path modeling algorithm without auxiliary noise for narrowband active noise control

Integrated active noise control for open-fit hearing aids with customized filter

In this paper properties of adaptive algorithms utilising ideas of stochastic gradient search optimisation applied to noise reduction are discussed. A focus on their average rate of convergence as well as on the obtained noise reduction is given. The presented discussion is based on a simulation case study devoted to noise reduction using a feedforward adaptive active noise control system.

The frequency-domain implementation of the adaptive LMS algorithm offers several advantages over the time-domain LMS algorithm. The primary advantage is that the orthogonality of the individual bins in the signal spectra permit the implementation of a single complex LMS filter for each FFT bin independently, each with a separate convergence parameter. The result is a uniform rate of convergence for all FFT bins as opposed to the time-domain algorithm which converges only as fast as the dominant eigenvalue of the input data. An additional parameter which affects frequency-domain LMS convergence is the overlap between successive FFT buffers. It is shown that the maximum advisable overlap between FFT buffers is 50%. Greater overlap between FFT buffers can lead to oscillations (and even instability) in the adaptive algorithm. Experimental results are presented which illustrate the convergence characteristics of a frequency-domain adaptive active noise control system for several different types of reference inputs.

This papers aims to develop an Adaptive Feedback Active Noise Control (AFB-ANC) system comprising two adaptive filters. The first adaptive filter is the same as found in the traditional AFB-ANC systems and is adapted using the classical Filtered-x Least Mean Square (FxLMS) algorithm. The stability of the FxLMS algorithm requires filtering the reference signal via a `good' estimate of the cancellation path (CP) present between the cancellation loudspeaker and the error microphone. Since the CP may be time-varying in the practical situations, the task of the second adaptive filter is to provide estimation of the CP during operation of the AFB-ANC system. The Cancellation-Path Estimation (CPE) filter is excited by a probe signal and is adapted using a delay-based Normalized Least Mean Square (NLMS) algorithm. The proposed system is equipped with a simple strategy to monitor the convergence status of the CPE filter. The developed strategy allows to inject a large (small) probe signal during the transient (steady) state of the AFB-ANC system. Computer simulation are carried out which demonstrate the effective performance of the developed system.

An efficient online secondary path estimation for feedback active noise control systems

In active noise control (ANC) systems, the online secondary path modeling (OSPM) methods that use additive random noise are often applied. The additive random noise, however, contributes to the residual noise, and thus deteriorates the noise control performance of ANC systems. This paper proposes a new OSPM method with power scheduling of additive random noise. Here the OSPM filter is adapted using a variable step size (VSS) LMS algorithm already proposed by authors. Furthermore, the additive-random-noise power is scheduled based on the convergence status of an ANC system. Computer simulations demonstrate the effectiveness of the proposed method.

This paper proposes a new modular design for implementation of active noise control (ANC) systems with online secondary path modeling. This design comprises of four similar modules to reduce the unwanted primary noise. One of these modules performs as the main adaptive control filter that uses an FxLMS algorithm to converge the tap-weights. Two other modules work as the adaptive modeling filters that use an LMS algorithm to model the secondary path filter. The last module works as a digital filter to produce the antinoise output signal. This innovative modular implementation of the ANC system results in a fast design and fast convergence with capability to expand the four adaptive filters and data bus when better noise reduction for ANC system is desired.

The original hybrid active noise control (HANC) is designated to suppress two uncorrelated noise sources. Recently, the HANC is modified to deal with cases where the two noise sources are statistically correlated, by separating the broadband and narrowband noise components within the first noise source reference signal and adding subsystems to make the entire system function well. However, the variations with the secondary path (SP) may cause catastrophic damage to the system. In this paper, a new HANC is proposed that works in the presence of not only two correlated noise sources but also time-varying SP. An online SP modeling (OSPM) subsystem is included to take care of the SP variations. Another linear prediction filter (LPF) is added to separate the remaining narrowband target noise component from the SP output due to the injected auxiliary white Gaussian noise (AWGN). The LPF output and error are set up to update the feedback subcontroller and to serve as the OSPM desired signal, respectively. Consequently, the coupling is substantially mitigated among the feedforward subcontrollers, the feedback subcontroller, and the OSPM subsystem, rendering the entire system not only robust against the SP variations but also more effective. Extensive simulations are provided to reveal the effectiveness and capabilities of the proposed HANC system.

Beyond the delay barrier in adaptive feedforward active noise control using Youla–Kučera parametrization

The convergence speed of multichannel adaptive active noise control algorithms can be improved by pre-whitening the reference signals and reducing their cross-correlations. When considering nonstationary processes, like road noise, it proves advantageous to use a short time-based cross-correlation estimation of the reference signal, which can be obtained as part of the bin-normalised frequency domain least-means-squares algorithm. However, the regularisation of such an algorithm is crucial to its success and can further improve the convergence rate. In addition, virtual sensors can be employed by using the additional filter method, which provides targets for the measured error signals to follow, generated from the reference signals. Still, the performance of this virtual sensing is known to be degraded if the properties of the reference signals change, so special consideration should be taken when the reference signals are nonstationary.

In many applications of active noise control (ANC), an online secondary path modelling method using a white noise as a training signal is required to ensure convergence of the system. The modelling accuracy and the convergence rate increase when a white noise with larger variance is used, however larger the variance increases the residual noise, which decreases performance of the system. The proposed algorithm uses the advantages of the white noise with larger variance to model the secondary path, but the injection is stopped at the optimum point to increase performance of the system. In this approach, instead of continuous injection of the white noise, a sudden change in secondary path during the operation makes the algorithm to reactivate injection of the white noise to adjust the secondary path estimation. Comparative simulation results shown in this paper indicate effectiveness of the proposed method.KeywordsActive noise controlfeedforwardFxLMSonline secondary path modelling