Related Experiment Videos
Reduction of computational complexity in the butterfly search technique
1Department of Radiology, University of Texas Medical School, Houston 77030, USA.
IEEE Transactions on Bio-Medical Engineering
|July 1, 1996
Summary
Butterfly search, a velocity estimation technique, is enhanced for reduced complexity. This method offers improved noise immunity and simpler implementation for signal processing applications.
Area of Science:
- Signal Processing
- Biomedical Engineering
- Ultrasound Imaging
Background:
- The butterfly search technique estimates target velocity using signal echoes.
- It utilizes predetermined trajectories called "butterfly lines" for resampling.
- Existing methods can be computationally intensive.
Purpose of the Study:
- To present modifications for reducing butterfly search computational complexity.
- To simplify the implementation of butterfly search on quadrature components.
- To maintain or improve performance with reduced complexity.
Main Methods:
- Resampling transducer echoes along constant velocity "butterfly lines".
- Estimating target velocity from signal trajectory slopes.
- Focusing on butterfly search using quadrature components.
- Implementing modifications to reduce computational load without complex multipliers.
Main Results:
- Demonstrated significant reduction in computational complexity.
- Maintained or improved noise immunity and performance.
- Showcased simpler implementation using elementary digital signal processing hardware.
- Outperformed common time domain and Doppler techniques in noisy conditions.
Conclusions:
- The modified butterfly search offers a simpler and computationally efficient approach.
- This technique provides excellent noise immunity, particularly on quadrature components.
- The enhancements make butterfly search more practical for real-time applications.