Time Domain Beamforming (TDBF) and Frequency Domain Beamforming (FDBF) are two widely adopted techniques in the realm of acoustic signal processing. Their primary goals include refining source localization, enhancing the quality of recorded audio signals and facilitating the separation of multiple sound sources within a particular environment.
In the table below, the commonalities of and differences between TDBF and FDBF are presented, while the specific advantages of each technique and their corresponding applications areas are highlighted.
| Time Domain Beamforming (TDBF) | Frequency Domain Beamforming (FDBF) | |
|---|---|---|
| Commonalities | Equal imaging methods, achieve the same results | |
| Frequency-selective analysis, TDBF requires band-pass filtering of the time signal | ||
Both use time correlation between the microphone signals | ||
| Both realise spatial filters to enhance or attenuate sounds at mapped locations | ||
| CLEAN deconvolution algorithms exist: TD: CLEAN-T FD: CLEAN-SC | ||
| Advantages | Fast for short recordings | Efficient processes of long-time signals by pre-calculating the Cross-Spectral Matrix (CSM) |
| Small data volumes when creating/saving acoustic movies | Fast and uncomplicated frequency-dependent analysis | |
| Low computational complexity | Multitude of extended algorithms: Power BF, Orthogonal BF…, all based on the CSM | |
| Suitable for processing strongly time-varying/transient signals | High localization accuracy even at low sampling rates e. g. 24 kHz | |
| Disadvantages | Frequency analysis requires additional band-pass filtering in advance (multiple-band analysis is more demanding) | Suitable for efficient processing of stationary signals |
| Almost no advanced algorithms applicable | Spectral leakage due to Short-Time Fourier Transform (STFT) process may affect FDBF results | |
| Application Areas | Sound source localization | |
| Engine analysis | ||
| Virtual microphone | Wind tunnel | |
| Pass-by | Leakage detection | |
| Wind turbines | Room acoustics | |
| Squeak & rattle test | ||
Jaeckel, O.: "Strengths and weaknesses of calculating beamforming in the time domain", BeBeC 2006, Berlin.
Visit the website Berlin Beamforming Conference held by GFaI e. V. https://www.bebec.eu.