Lifu Wu

An Active Impulsive Noise Control Algorithm With Logarithmic Transformation

To overcome the limitations of the existing algorithms for active impulsive noise control, an algorithm based on minimizing the squared logarithmic transformation of the error signal is proposed in this correspondence. The proposed algorithm is more robust for impulsive noise control and does not need the parameter selection and thresholds estimation according to the noise characteristics. These are verified by theoretical analysis and numerical simulations.

Time Difference of Arrival Estimation Exploiting Multichannel Spatio-Temporal Prediction

To localize sound sources in room acoustic environments, time differences of arrival (TDOA) between two or more microphone signals must be determined. This problem is often referred to as time delay estimation (TDE). The multichannel cross-correlation-coefficient (MCCC) algorithm, which is an extension of the traditional cross-correlation method from two- to multiple-channel cases, exploits spatial information among multiple microphones to improve the robustness of TDE. In this paper, we propose a multichannel spatio-temporal prediction (MCSTP) algorithm, which can be viewed as a generalization of the MCCC principle from using only spatial information to using both spatial and temporal information. A recursive version of this new algorithm is then developed, which can achieve similar performance as MCSTP, but is computationally more efficient. Experimental results in reverberant and noisy environments demonstrate the advantages of this new method for TDE.

Active impulsive noise control algorithm with post adaptive filter coefficient filtering

Most algorithms for active impulsive noise control employ non-linear transformations to limit the reference and/or error signals and to maintain system stability. From a more direct manner, a new cost function is proposed in this study which is defined as the summation of the squared Euclidean norm of difference between the currently updated filter coefficients vector and all past filter coefficients vector subject to the constraint imposed on the adaptive filter output. A new adaptive algorithm is derived from the cost function and called the filtered weight filtered-x normalised least mean square algorithm because it can be interpreted as a post filter structure which passes the filter coefficients through a first-order infinite impulse response filter. The proposed algorithm is suitable for active control of impulsive noise since it directly limits the dynamic range of the adaptive filter coefficients and prevents heavy fluctuation of the filter coefficients. Simulations compare the performance of the proposed algorithm with the existing algorithms and demonstrate the effectiveness of the proposed algorithm.

Reverberation time estimation from speech signals based on blind room impulse response identification.

A reverberation time (RT) estimation method is presented which consists of three steps, the anechoic speech is first recovered by maximizing the skewness of the linear prediction residual of the reverberant speech, then room impulse response (RIR) is identified using the recovered anechoic and reverberant speech, finally RIR is truncated to compensate the estimation errors and RT is estimated using the Schroeders method. Simulations show that the proposed method successfully estimates RT less than 1.4 s and is insensitive to the speech content such as the number of long pauses and sharp offsets.

An extension to the filtered-x LMS algorithm with logarithmic transformation

Active control of impulsive noise has been of increasing interest due to high impact of such noise on humans. The algorithm with logarithmic transformation, developed by Wu, et al. has been found particularly interesting. In this paper this idea is continued, and an extension to this algorithm is proposed to improve its convergence properties and allow for successful control if the noise has also another type of noise together with the impulses. A number of simulations are performed to validate the algorithm and compare it with algorithms leading in the literature. Additionally to simulated benchmark impulsive noises, real recordings are considered, which bring another insight into efficiency of the algorithms.