A Low-Cost Robust Distributed Linearly Constrained Beamformer for Wireless Acoustic Sensor Networks with Arbitrary Topology

Koutrouvelis, A.; Sherson, T.W.; Heusdens, R.; Hendriks, R.C.

doi:10.1109/TASLP.2018.2829405

A Low-Cost Robust Distributed Linearly Constrained Beamformer for Wireless Acoustic Sensor Networks with Arbitrary Topology

Title

A Low-Cost Robust Distributed Linearly Constrained Beamformer for Wireless Acoustic Sensor Networks with Arbitrary Topology

Author

Koutrouvelis, A. (TU Delft Signal Processing Systems)
Sherson, T.W. (TU Delft Signal Processing Systems)
Heusdens, R. (TU Delft Signal Processing Systems)
Hendriks, R.C. (TU Delft Signal Processing Systems)

Date

2018

Abstract

We propose a new robust distributed linearly constrained beamformer which utilizes a set of linear equality constraints to reduce the cross power spectral density matrix to a block-diagonal form. The proposed beamformer has a convenient objective function for use in arbitrary distributed network topologies while having identical performance to a centralized implementation. Moreover, the new optimization problem is robust to relative acoustic transfer function (RATF) estimation errors and to target activity detection (TAD) errors. Two variants of the proposed beamformer are presented and evaluated in the context of multi-microphone speech enhancement in a wireless acoustic sensor network, and are compared with other state-of-the-art distributed beamformers in terms of communication costs and robustness to RATF estimation errors and TAD errors.

Subject

Distributed beamforming
Estimation error
LCMV
Microphones
MVDR
Network topology
Noise measurement
Reverberation
robust beamforming
Robustness
speech enhancement
WASN