Experimental synthesis of random pressure fields based on transfer-matrix analysis on 1D arrays

• Transfer matrix prediction based on segmented acoustic centers is presented. • The segmented acoustic center model can represent non-omnidirectional directivity. • The method offers a practical technique for random pressure field synthesis. • This method saves substantial time in pre-experimental system calibration. Synthesizing random pressure fields with loudspeaker arrays in a laboratory setting requires acquiring a global transfer matrix of all channels between the loudspeaker array and the microphone array. This inevitably involves measuring a large and unwieldy number of transfer functions. Therefore, we propose a prediction method for the transfer matrix under free-field conditions, combining a substantially reduced number of measurements with specific predictions based on segmented acoustic centers. In free-field conditions, if only three sets of transfer functions are measured for each loudspeaker and the remaining entries in the global transfer matrix are predicted using analytical expressions, the results show that the normalized error between the predicted and measured transfer matrices can be less than -13 dB. The experimental results indicate that, based on a one-dimensional loudspeaker array in a standard anechoic chamber, this prediction method shows promise for accurately reproducing random pressure fields, such as a diffuse acoustic field and the pressure field in the spanwise direction of a turbulent boundary layer. Additionally, the prediction method demonstrates the potential for extension to two-dimensional synthesis.