Introducing RIFT: A Hierarchical Entropic Filtering Scheme for Ideal Time-Frequency Reconstruction
In this paper, we introduce the Reconstructive Ideal Fractional Transform (RIFT), an entropy-based probabilistic filtering algorithm formulated to reconstruct the Ideal Time-Frequency Representation (ITFR). RIFT surpasses the limitations imposed by the Gabor uncertainty principle for linear transforms, achieving the bilinear transform accuracy present in the Wigner-Ville Distribution (WVD) while effectively suppressing cross-terms. The algorithm utilises a hierarchical fractional wavelet-based scheme to account for localised time-frequency trajectory curvature. This scheme is optimised through an entropic-based filtering method that probabilistically extracts auto-terms while retaining the resolution of the WVD. This is achieved by employing a spatially varying, positively constrained deconvolution algorithm (Lucy-Richardson) with regularisation for adequate noise suppression. Additionally, the optimisation yields an Instantaneous Phase Direction field, which allows the localised curvature in speech or music extracts to be visualised and utilised within a Kalman tracking scheme, enabling the extraction of signal component trajectories. Evaluation demonstrates the algorithm's ability to eradicate cross-terms effectively and achieve superior time-frequency precision. This advance holds significant potential for a wide range of applications requiring high-resolution cross-term-free time-frequency analysis.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Chaining Meets Chain Rule: Multilevel Entropic Regularization and Training of Neural Nets
We derive generalization and excess risk bounds for neural nets using a family of complexity measures based on a multilevel relative entropy. The bounds are obtained by introducing the notion of generated hierarchical co…
Plug-in estimation of Schrödinger bridges
We propose a procedure for estimating the Schr\"odinger bridge between two probability distributions. Unlike existing approaches, our method does not require iteratively simulating forward and backward diffusions or trai…
Concept Drift Detection and Adaptation with Hierarchical Hypothesis Testing
A fundamental issue for statistical classification models in a streaming environment is that the joint distribution between predictor and response variables changes over time (a phenomenon also known as concept drifts), …
Drift DetectionGeneral ClassificationTwo-sample testingEmpirical Study of Diachronic Word Embeddings for Scarce Data
Word meaning change can be inferred from drifts of time-varying word embeddings. However, temporal data may be too sparse to build robust word embeddings and to discriminate significant drifts from noise. In this paper, …
Diachronic Word EmbeddingsWord EmbeddingsFunction approximation by neural nets in the mean-field regime: Entropic regularization and controlled McKean-Vlasov dynamics
We consider the problem of function approximation by two-layer neural nets with random weights that are "nearly Gaussian" in the sense of Kullback-Leibler divergence. Our setting is the mean-field limit, where the finite…