Dissertation Defense: Arun Nair
Note: This is a virtual presentation. Check this page at a later date for the Zoom link to where the presentation will be taking place.
Title: Machine Learning for Beamforming in Audio, Ultrasound, and Radar
Abstract: Multi-sensor signal processing plays a crucial role in the working of several everyday technologies, from correctly understanding speech on smart home devices to ensuring aircraft fly safely. A specific type of multi-sensor signal processing called beamforming forms a central part of this thesis. Beamforming works by combining the information from several spatially distributed sensors to directionally filter information, boosting the signal from a certain direction but suppressing others. The idea of beamforming is key to the domains of audio, ultrasound, and radar.
Machine learning is the other central part of this thesis. Machine learning, and especially its sub-field of deep learning, has enabled breakneck progress in tackling several problems that were previously thought intractable. Today, machine learning powers much of the cutting edge systems we see on the internet for image classification, speech recognition, language translation, and more.
In this dissertation, we look at beamforming pipelines in audio, ultrasound, and radar from a machine learning lens and endeavor to improve different parts of the pipelines using ideas from machine learning. We start off in the audio domain and derive a machine learning inspired beamformer to tackle the problem of ensuring the audio captured by a camera matches its visual content, a problem we term audiovisual zooming. Staying in the audio domain, we then demonstrate how deep learning can be used to improve the perceptual qualities of speech by denoising speech clipping, codec distortions, and gaps in speech.
Transitioning to the ultrasound domain, we improve the performance of short-lag spatial coherence ultrasound imaging by exploiting the differences in tissue texture at each short lag value by applying robust principal component analysis. Next, we use deep learning as an alternative to beamforming in ultrasound and improve the information extraction pipeline by simultaneously generating both a segmentation map and B-mode image of high quality directly from raw received ultrasound data.
Finally, we move to the radar domain and study how deep learning can be used to improve signal quality in ultra-wideband synthetic aperture radar by suppressing radio frequency interference, random spectral gaps, and contiguous block spectral gaps. By training and applying the networks on raw single-aperture data prior to beamforming, it can work with myriad sensor geometries and different beamforming equations, a crucial requirement in synthetic aperture radar.
- Trac Tran, Department of Electrical and Computer Engineering
- Muyinatu Bell, Department of Electrical and Computer Engineering
- Vishal Patel, Department of Electrical and Computer Engineering