Localized Kernel Methods for Signal Processing

TL;DR

This dissertation introduces two innovative signal processing methods using localized kernels that outperform classical algorithms in noisy conditions, enabling accurate parameter recovery and signal separation with fewer samples and no prior component knowledge.

Contribution

It presents novel localized kernel-based algorithms for multidimensional frequency estimation and chirp separation, improving robustness and efficiency over existing methods.

Findings

Outperforms MUSIC and ESPRIT at low SNR in univariate frequency estimation.

Achieves high recovery accuracy with fewer samples in multivariate cases.

Successfully separates intersecting and discontinuous chirps at -30 dB SNR.

Abstract

This dissertation presents two signal processing methods using specially designed localized kernels for parameter recovery under noisy condition. The first method addresses the estimation of frequencies and amplitudes in multidimensional exponential models. It utilizes localized trigonometric polynomial kernels to detect the multivariate frequencies, followed by a more detailed parameter estimation. We compare our method with MUSIC and ESPRIT, which are classical subspace-based algorithms widely used for estimating the parameters of exponential signals. In the univariate case, the method outperforms MUSIC and ESPRIT under low signal-to-noise ratios. For the multivariate case, we develop a coordinate-wise projection and registration approach that achieves high recovery accuracy using significantly fewer samples than other methods. The second method focuses on separating linear chirp components from time-localized signal segments. A variant of the Signal Separation Operator (SSO) is constructed using a localized kernel. Instantaneous frequency estimates are obtained via FFT-based filtering, then clustered and fitted with piecewise linear regression. The method operates without prior knowledge of the number of components and is shown to recover intersecting and discontinuous chirps at SNR levels as low as -30 dB. Both methods share an idea based on localized kernels and efficient FFT-based implementation, and neither requires subspace decomposition or sparsity regularization. Experimental results confirm the robustness and tractability of the proposed approaches across a range of simulated data conditions. Potential extensions include application to nonlinear chirps, adaptive kernel design, and signal classification using extracted features.