Building a Test Station for ANC Benchmarking and Automatic Filter Tuning

Active Noise Cancellation (ANC) is widely used in headphones and earables to reduce unwanted environmental noise. Its performance strongly depends on the acoustic properties of the device, microphone and loudspeaker placement, and the design of the underlying digital filters. Developing effective ANC systems therefore requires accurate acoustic measurements and systematic filter optimization.

This project focuses on developing an automated test station for OpenEarable, an open-source earable research platform. The test setup will use a GRAS 43AG artificial ear and an external sound source to measure the acoustic response of the earable under controlled conditions. By simultaneously recording signals from the OpenEarable microphones and the reference microphone of the artificial ear, the system will enable the characterization of acoustic transfer functions and the evaluation of ANC performance.

A key challenge is achieving accurate synchronization between the different microphone signals, which is essential for phase-sensitive measurements and ANC filter design. The project will investigate suitable synchronization approaches, potentially using a shared multichannel audio interface for data acquisition.

Based on the measured acoustic transfer functions, algorithms for automatically optimizing feedforward ANC filters will be developed. The resulting system should provide a reproducible framework for benchmarking ANC performance and automatically tuning filters for different acoustic conditions.

• Design and build an automated acoustic measurement station using a GRAS 43AG artificial ear, OpenEarable, and an external sound source.
• Develop a multichannel audio acquisition system to record signals from the OpenEarable microphones and the artificial ear reference microphone.
• Investigate and implement suitable microphone synchronization methods for accurate time- and phase-aligned measurements.
• Measure and characterize acoustic transfer functions in the time and frequency domains.
• Develop and implement algorithms for automatic feedforward ANC filter optimization.
• Evaluate the performance, accuracy, and reproducibility of the measurement and filter optimization system.

• Basic knowledge of digital signal processing, particularly frequency-domain analysis and digital filter design.
• Programming experience in Python and/or C++.
• Interest in audio signal processing, acoustics, and hardware development.
• Motivation to work with physical measurement setups and experimental hardware.
• Experience with soldering and basic electronics is beneficial.
• Familiarity with ANC, adaptive filtering, or audio measurement techniques is beneficial but not required.
• Ability to independently develop, test, and evaluate technical solutions.