Non-linear, adaptive array processing for acoustic interference suppression

Area of Science:

• Acoustics
• Signal Processing
• Array Signal Processing

Background:

• Non-Gaussian, broadband interferers pose challenges for traditional spatial processing techniques.
• Existing methods like generalized sidelobe cancellers often rely on spatial separation, which may not be optimal for all scenarios.

Purpose of the Study:

• To develop and evaluate a new method combining blind source separation (BSS) with spatial processing.
• To remove non-Gaussian, broadband interferers from space-time displays, such as bearing track recorder displays.
• To compare the proposed algorithm's performance against adaptive beamforming techniques.

Main Methods:

• The proposed method integrates blind source separation of acoustical sources with spatial processing.
• Algorithm performance was evaluated using simulations and experiments with two acoustic sources.
• Various signal-to-interference (SIR), signal-to-noise (SNR), and array geometry conditions were simulated.
• A voice activity detection (VAD) algorithm was employed for benchmarking source isolation performance.

Main Results:

• The developed algorithm demonstrates effective removal of non-Gaussian, broadband interferers.
• Performance was benchmarked against adaptive beamforming techniques, including minimum variance distortionless response (MVDR) beamforming.
• Simulations confirmed the algorithm's effectiveness under diverse acoustic conditions and array configurations.

Conclusions:

• The combined BSS and spatial processing approach offers a viable alternative for mitigating complex acoustic interference.
• The method shows promise for improving the clarity of space-time displays in challenging acoustic environments.
• Further research can explore optimizations and applications in real-world scenarios.