Broadband active noise control using modal properties of 3-dimensional acoustic field (reduction of required memories for a filter map)

On active noise control in 3-dimensional acoustic field, a control effect becomes worse when an evaluation point moves. Against this problem, we proposed a control algorithm using filter map. Validity of this algorithm was shown by simulations and experiments. To use the method, however, the adaptive filters must be accumulated in a controller so that the required memory becomes large. Therefore, reduction of required memory is necessary. In this paper, a method for the reduction is proposed. In the method, adaptive FIR filters is regarded as superpositions of several modes. At first, modal parameters of an optimal filter are identified. Prony's method and least-square method are used in this identification process. An ANC simulation using the identified parameters is carried out. According to the result, the required memory is reduced about eighty percent. Moreover, an adaptive algorithm is proposed. The algorithm calculates amplitudes of modal parameters. Validity of the adaptive method is shown by simulations and an experiment.