Limitations of the equivalence between spatial and ensemble estimators in the case of a single-tone excitation

Florian Monsef (1) , Andrea Cozza (2)

1 L2S - Laboratoire des signaux et systèmes
2 DRE - Département de Recherche en Electromagnétisme ; EMG - Département Electromagnétisme - L2S

The ensemble-average value of the mean-square pressure is often assessed by using the spatial average technique, underlying an equivalence principle between spatial and ensemble estimators. Using the ideal-diffuse-field model, the accuracy of the spatial-average method has been studied theoretically forty years ago in the case of a single-tone excitation. This study is revisited in the present work on the basis of a more realistic description of the sound field accounting for a finite number of plane waves. The analysis of the spatial-average estimator is based on the study of its convergence rate. Using experimental data from practical examples, it is shown that the classical expression underestimates the estimator uncertainty even for frequencies greater than Schroeder's frequency, and that the number of plane waves may act as lower bound on the spatial-average estimator accuracy. The comparison of the convergence rate with an ensemble-estimator shows that the two statistics cannot be regarded as equivalent in a general case.

Domains: Engineering Sciences [physics]; Electromagnetism

Florian Monsef, Andrea Cozza. Limitations of the equivalence between spatial and ensemble estimators in the case of a single-tone excitation. Journal of the Acoustical Society of America, 2011, 130 (4), pp.1943-1953. ⟨hal-00614712⟩