The Low-rank Gaussian Mixture Model with Interference Reference in the Acoustic Array Measurement for Background Interference Suppression

Background interference suppression for acoustic array measurements has essential applications in the aircraft industry, particularly during wind tunnel tests where interference from flow and various other measurement devices may affect measurement data. The low-rank Gaussian mixture model (LRGMM) has emerged as a potential method to suppress the strong and complex inference in the measurement. However, the performance and computational efficiency of the algorithm can be significantly affected by the number of Gaussian components in the model. This paper proposes a method for adaptively determining the number of Gaussian components in the Gaussian mixture model (GMM) using Bayesian information criteria (BIC) when interference reference has been measured. The model with fewer parameters is chosen by BIC, which improves computational efficiency while ensuring performance. The performance of the proposed method is validated by numerical simulation.