An integrated approach to the design of digital signal processing algorithms for hearing instruments

A key component to the successful design of hearing instruments is the design of digital signal processing (DSP) algorithms. These algorithms realize specific functionalities such as noise reduction, dereverberation, dynamic range compression, adaptive feedback cancellation, and active noise control. While there has been a significant advance in the technology of such DSP functionalities, different DSP functionalities are mostly developed and analyzed in isolation. Whenever such DSP functionalities are cascaded in a complete system, their interaction -even counteraction- is rarely taken into consideration. This in particular compounds the general problem of robustness in typical dynamic acoustic environments, as observed by a hearing instrument. Hence, the objective of this project is to address this problem by following an integrated approach: integrating individual DSP functionalities in a common acoustic control problem statement and solution strategy. From this, suitable integrated algorithms will be designed and compared against the state-of-the-art based on off-line simulations, as well as based on real-time testing and perceptual evaluations.