OPTIMAL PLACEMENT OF SOURCES AND SENSORS WITH THE MINIMAX CRITERION FOR ACTIVE CONTROL OF A ONE-DIMENSIONAL SOUND FIELD

The optimization of the placement of secondary sources and error senso rs can enhance the efficiency of active noise control. Since the class ic quadratic cost functions are not adapted to the search of the optim al placement, it is shown that the ''minimax'' criterion-minimization of the largest squared pressure at a number of distributed points-bett er suits the strategy of selection in the placement of sources and sen sors. Sufficient numbers of sources and sensors, their locations and t he volume velocity of each secondary source are found simultaneously, in a short computation time, by solving a unique linear programming pr oblem. Bounds on the volume velocity are introduced easily. If the num ber of sources is prescribed, a mixed programming problem is solved. T his paper is devoted to the application of this method in a one-dimens ional enclosure with rigid walls. (C) 1997 Academic Press Limited.