Development of the filtered-U algorithm for active noise control

Abstract

The filtered-X algorithm developed by Widrow and Burgess is an alternate form of the least-mean-square (LMS) algorithm for use when there are transfer functions in the auxiliary path following the adaptive filter. To ensure convergence of the algorithm, the input to the error correlators is filtered by a copy of these auxiliary path transfer functions. More recently, the author has presented a new approach to active noise control in the presence of acoustic feedback that uses an infinite impulse response (IIR) filter structure with an alternate form of the recursive least-mean-square (RLMS) algorithm. This algorithm may be described as a filtered-U algorithm since it uses a copy of the auxiliary path transfer functions to filter the generalized input vector U to the error correlators of both the direct and recursive elements of the filter to ensure convergence. The relationship of the filtered-U to the filtered-X algorithm and other earlier concepts is discussed.

Effect of sound-related activities on human behaviours and acoustic comfort in urban open spaces

“…Considering that the activities themselves may dynamically change, 5-7 survey positions were selected randomly around the activity areas, and each survey position was 5 m away from the activity areas and main sound sources, if any, to avoid any instantaneous error (Eriksson 1991, Vincent et al 2006; the corresponding LAeq of activities was derived as an average from the 5 positions.…”

Application of wave field synthesis to active control of highly non-stationary noise

“…Feedback path modeling and neutralization are well-established issues in the literature [2] and represent crucial aspects in the practical realization of feedforward ANC systems. The feedback path induces a closed loop gain on the RM input and the overall feedforward ANC system must be modeled as an infinite impulse response (IIR) filter; solutions based on offline path estimation [2] and on-line adaptive filtering [24][25] [26] have been proposed.…”

On Active Noise Control Systems With Online Acoustic Feedback Path Modeling

“…(This is obtained by using FxLMS algorithm under no-acoustic-feedback condition. ); • the estimation error for (17) • the error in the reference signal (18) where is the noise source signal and is the reference signal used in adaptation of . For acoustic paths, the experimental data provided by [1] is used.…”

“…The IIR control system, being a pole-zero system, allows the system to dynamically track changes in the feedback and secondary paths during cancelation operation. There are a number of adaptive IIR algorithms: filtered-u recursive LMS (FuRLMS) [17], full-feedback IIR LMS [18], and filtered-v LMS (FvLMS) [19]. In these IIR-based structures, 1) the stability cannot be guaranteed, and 2) the adaptation may converge to a local minimum.…”