Chuang Shi’s Personal Homepage

About Me

I am a Lecturer in Active Noise and Vibration Control at the Institute of Sound and Vibration Research (ISVR), University of Southampton, UK.

My research interests are in the development and application of signal processing and machine learning approaches for the manipulation of sound waves. More specifically, I have been researching in the fields of active noise control and parametric acoustic array. These two topics can be brought together under the broader human-centered target to create a favorable sound environment. Both the studies of active noise control and parametric acoustic array involve series of ingenious designs of acoustics, circuits, algorithms, systems and prototypes.

As such, my research vision is to make cutting-edge technology easy to understand and ready to apply.

Research Contributions

Global Active Noise Control through Multi-Channel Systems: There are three basic principles of noise control. They are namely isolating the noise source, blocking the noise propagation path, and protecting the noise receiver. Global active noise control generally cooperates with the first two principles in applications of duct noise control, active windows, virtual sound barriers, etc. My research contributions are mainly in developing the differential error sensing in an open-end duct to mitigate interferences from its downstream, the acoustic design and algorithm optimization of the active window to enlarge the opening size to a usable extent, and the virtual sensing for the virtual sound barrier to move the quiet zone to the far field.

Local Active Noise Control to Form Personal Listening Zones: Local active noise control is found to be more effective in applications that protect people from hazards of traffic noise, engine noise, factory noise, etc. Active headset may be the most representational system of local active noise control. However, there are still key challenges in empowering more adaptivity and personalization. Active headrest is another popular system, but issues regarding varying acoustic paths and noise overflowing pose grand challenges. My research contributions are primarily in creating the digital twin architecture to allocate the adaptation of control filters on a cloud server, investigating directional control sources to resolve noise overflowing, and integrating spatial audio reproduction into binaural active noise control.

Directional Sound Generation using the Parametric Acoustic Array: Parametric acoustic array is a nonlinear acoustic phenomenon that can be utilized to generate directional sound from a remarkably small aperture. Its application in air, a.k.a. the parametric array loudspeaker, has thus become a convenient directional sound source in various audio applications. The directivity control feature is long desired, and the nonlinear distortion remains to be a major challenge for the parametric array loudspeaker. My research contributions are recognized as working out the feasible design of the steerable parametric array loudspeaker, deriving the convolution model that provides an accurate yet concise formula to compute the beampattern of the difference-frequency wave, and adapting the Volterra filter for modeling and reducing the nonlinear distortion in the self-demodulated wave.