Input-output linearization of loudspeaker dynamics via automatic differentiation

Heuchel FM, Agerkvist FT. Input-output linearization of loudspeaker dynamics via automatic differentiation. In: Convention Paper. Audio Engineering Society; 2026. Paper 10307. Available from: https://aes.org/publications/elibrary-page/?id=23200

@inproceedings{Heuchel2026_23200,
  author = {Heuchel, Franz M. and Agerkvist, Finn T.},
  title = {{Input-output linearization of loudspeaker dynamics via automatic differentiation}},
  note = {Paper 10307},
  year = {2026},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=23200}
}

TY  - CPAPER
TI  - Input-output linearization of loudspeaker dynamics via automatic differentiation
AU  - Heuchel, Franz M.
AU  - Agerkvist, Finn T.
M1  - Paper 10307
PY  - 2026
DA  - 2026/05/28
UR  - https://aes.org/publications/elibrary-page/?id=23200
PB  - Audio Engineering Society
LA  - en
AB  - Input-output linearization is a technique for compensating nonlinear distortion in loudspeakers. To apply it to complex loudspeaker models, we describe an end-to-end framework for estimating model parameters from data and deriving the linearizing control laws using automatic differentiation. The parameter estimation approach combines frequency-domain linear parameter estimation with a time-domain prediction-error method for the nonlinear parameters. The linearization approach supports nonlinear reference systems and stabilization of the control law using trajectory tracking. We implement the framework in dynax, an open-source Python package based on JAX, and validate it experimentally as a feed-forward controller on a closed-box loudspeaker. Results demonstrate validation errors of 15%NRMSE and total harmonic distortion reductions of 612 dB The framework enables researchers and engineers to rapidly prototype and validate complex loudspeaker models for distortion compensation without manual symbolic derivations.
ER  -