Scientific Article

Sparse estimation and global control of low-frequency wall-scattered pressure for a hybrid semi-anechoic room

A comprehensive simulation and optimization study in the frequency domain

¹ Aix Marseille Univ, CNRS, Centrale Med, Laboratoire de Mécanique et d’Acoustique, Marseille, France
² University of Applied Sciences and Arts Western Switzerland, Geneva, Switzerland
³ PHEASO, Roquevaire, France

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Received: 30 January 2026
Accepted: 19 May 2026

Abstract

This paper investigates the simulation and optimization of an active control strategy designed for the hybrid semi-anechoic room currently under development at the Laboratoire de Mécanique et d’Acoustique. Control is intended to minimize the acoustic pressure scattered by the room walls and ceiling at low frequencies. The objective is to obtain a semi-free field, that complies with the inverse square law in a specific measurement volume, despite the presence of room modes. A key objective of the study is to determine the minimal yet sufficient number and spatial distribution of microphones required for the control system, with a view to reducing signal processing load and overall system complexity while preserving control performance. To this end, we employ a three-dimensional analytical modal model to evaluate sparse estimation methods for reconstructing the scattered acoustic field, with particular attention to the performance of group-lasso and least-squares optimization strategies. Complementary numerical simulations are carried out using a Finite Element model calibrated from measurements collected in the physical room. Numerical simulations indicate that, with 36 loudspeakers and 59 microphones positioned in close proximity to the walls, the active device should be capable of providing close to semi-anechoic conditions up to 200 Hz, while requiring only 5 microphone signals to estimate the scattered pressure at each of the 59 locations. The combined results provide guidance for designing an efficient, reduced-order sensing strategy for the practical implementation of active semi-anechoic environments.