End-To-End Deep Learning-Based Adaptation Control for Frequency-Domain Adaptive System Identification
This addresses adaptation control for system identification in signal processing, but it appears incremental as it applies deep learning to an existing bottleneck.
The paper tackles the problem of controlling step-sizes in frequency-domain adaptive system identification by using a deep neural network to map signal features to step-sizes, achieving fast convergence and robust steady-state performance in noisy, non-stationary environments.
We present a novel end-to-end deep learning-based adaptation control algorithm for frequency-domain adaptive system identification. The proposed method exploits a deep neural network to map observed signal features to corresponding step-sizes which control the filter adaptation. The parameters of the network are optimized in an end-to-end fashion by minimizing the average normalized system distance of the adaptive filter. This avoids the need of explicit signal power spectral density estimation as required for model-based adaptation control and further auxiliary mechanisms to deal with model inaccuracies. The proposed algorithm achieves fast convergence and robust steady-state performance for scenarios characterized by high-level, non-white and non-stationary additive noise signals, abrupt environment changes and additional model inaccuracies.