Window-Based Constant Beamwidth Beamformer.

Area of Science:

• Signal Processing
• Acoustics
• Array Signal Processing

Background:

• Beamformers are crucial for enhancing desired signals and suppressing interference in various applications.
• Existing constant beamwidth beamformer designs often suffer from high computational complexity, sensitivity to microphone mismatches, or inability to control sidelobe levels.

Purpose of the Study:

• To propose a novel window-based technique for designing constant beamwidth beamformers.
• To achieve constant beamwidth across a wide frequency range while simultaneously controlling sidelobe levels.
• To offer an alternative to computationally intensive optimization algorithms and frequency-dependent sensor adjustments.

Main Methods:

• A window-based technique is proposed, applying different standard window function shapes to microphone weighting coefficients for distinct frequency bins.
• This approach directly manipulates the real weighting coefficients, enabling precise control over beamformer characteristics.

Main Results:

• The proposed method successfully maintains a constant beamwidth over a wide frequency range.
• It effectively controls the sidelobe level, a limitation in some existing methods.
• Simulation results demonstrate superior performance compared to existing techniques, showing a lower sidelobe level, higher directivity factor, and increased white noise gain.

Conclusions:

• The proposed window-based technique offers an effective and computationally efficient method for designing constant beamwidth beamformers.
• This approach provides superior performance in terms of sidelobe level control, directivity, and noise gain.
• It presents a significant advancement over traditional optimization-based and sensor-adjustment methods.