SKM 2021 – scientific programme
Parts | Days | Selection | Search | Updates | Downloads | Help
DY: Fachverband Dynamik und Statistische Physik
DY 4: Poster Session II: Nonlinear Dynamics, Simulations and Machine Learning
DY 4.10: Poster
Tuesday, September 28, 2021, 17:30–19:30, P
Modelling a highly adaptive, nonlinear acoustic sensor — •Philipp Hövel1, Thomas Meurer2, Martin Ziegler3, and Claudia Lenk3 — 1University College Cork, Ireland — 2Kiel University, Germany — 3Technische Universität Ilmenau, Germany
Hearing is a remarkable sense both in terms of physiology and signal processing. In biology, hearing exhibits amazing sensing properties, in particular for low-volume sounds and noisy environments, which is known as the "cocktail party effect". For many state-of-the-art technological implementations, speech recognition remains a challenging task in these hard-to-hear situations and varying surroundings.
In this contribution, we present a mathematical model of a novel, adaptive, bio-inspired acoustic sensor with integrated signal processing functionality, whose sensing and processing properties can be widely tuned using real-time feedback. We show that dynamical switching between linear and nonlinear characteristics improves detection of signals in noisy conditions, increases the dynamic range of the sensor, and enables adaptation to changing acoustic environments. We demonstrate that the dynamical switching can be attributed to a Hopf bifurcation, and its dependence of sensor and feedback parameters is validated in experiments, and highlight the applicability and conceptual advantages of the acoustic sensor.