This paper presents a multichannel adaptive filtering algorithm for real-time full-band adaptive transaural reproduction on general-purpose hardware. It is based on a multichannel frequency-domain FxLMS algorithm using an overlap-save framework for both filtering and adaptation, and is extended with (i) online plant identification for fully adaptive operation, (ii) frequency-dependent normalization for faster and more uniform convergence across frequency, and (iii) frequency-dependent regularization to limit the control filter magnitude at frequencies where the plant is ill-conditioned. The proposed algorithm is implemented in C on a standard desktop PC and evaluated on a 42 transaural configuration running in real time at 48 kHz with 2048-tap control filters. Four experiments are conducted. The first two assess crosstalk cancellation performance with uncorrelated white-noise reference signals, and yield an average crosstalk cancellation of 32 dB over 100 Hz20 kHz. The last two consider binaural signal reproduction as a more realistic use case of the algorithm. In both cases, performance is assessed under both a static and a moving listener scenario, demonstrating the algorithms ability to re-adapt.
(See document for exact affiliation information.)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
