题目：Approximating Signals from Random Sampling in a Reproducing Kernel Subspace of Homogeneous Type
报告人：冼军 教授 （中山大学）
摘要：The problem of approximating a signal in a reproducing kernel subspace of homogeneous type from its random samples will be shown in this talk. We adopt the iterative frame algorithm to solve this problem. The approximation error is derived. Due to randomness of the samples, the magnitude of the approximation error is a probabilistic estimate and the associated probability tends to one when the sampling size is sufficiently large. We also consider the approximation error when samples are corrupted by noise. Numerical examples are presented to illustrate the effectiveness of the algorithm.
Copyright © 2018 同济大学数学科学学院 版权所有.