Sound field reconstruction based on constrained L2/3 norm sparse regularization

The near-field acoustic holography based on compressive sensing is an effective method for identifying sources and reconstructing sound fields. However, when reconstructing sound fields with spatially sparse sources, conventional equivalent source methods based on L₂ and L₁ norm regularization suffer from inadequate source amplitude and poor algorithm stability. Thus, based on compressive sensing near-field acoustic holography, this work proposes an L_2/3 norm sparse regularization technique for sound field reconstruction. The method establishes an enhanced Lagrangian model by applying constraints to the objective function using the augmented Lagrange multiplier approach. Then the equivalent source strengths can be obtained by using the alternate direction approach. The effectiveness and superiority of the L_2/3 norm regularization algorithm are demonstrated through simulation tests, which compare the reconstruction performance of the proposed method with that of three other methods.