Robust risk estimators and their minimization for signal processing

鲁棒风险估计器及其信号处理最小化

• 批准号: 24800022
• 负责人: YAMAGISHI Masao
• 金额: $ 1.91万
• 依托单位: Tokyo Institute of Technology
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Research Activity Start-up
• 财政年份: 2012
• 资助国家: 日本
• 起止时间: 2012-08-31 至 2014-03-31
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/en/grant/KAKENHI-PROJECT-24800022/
• 关键词: 信号処理; リスク推定量; 画像処理; 適応フィルタ; 最適化工学; パラメータ推定問題; 凸最適化; 適応システム同定; 信号処理工学; 線形パラメータ推定問題; SURE; 悪条件; 低階数最小分散擬似不偏推定法; 画像復元

Many signal processing problems can be reduced to estimation problems in linear models: estimating an unknown vector with an observation that is the multiplication of a known model matrix and the unknown vector in the presence of additive noise. In this context, many algorithms attempt to minimize the Mean Squared Error (MSE). However, since evaluation of the MSE requires the exact knowledge of the unknown vector, estimation techniques for the MSE are essential in many practical situations. In this study, we propose robust estimators for the MSE and utilize their efficient minimization techniques in the context of image restoration and parameter tuning of adaptive filtering algorithms.

许多信号处理问题可以简化为线性模型中的估计问题：在存在加性噪声的情况下，用已知模型矩阵与未知向量相乘的观测值估计未知向量。在这种情况下，许多算法试图最小化均方误差（MSE）。然而，由于评估的MSE需要的未知向量的确切知识，估计技术的MSE是必不可少的，在许多实际情况下。在这项研究中，我们提出了强大的估计的MSE，并利用其有效的最小化技术的背景下，图像恢复和自适应滤波算法的参数调整。

项目成果

Variants of the Alternating Direction Method of Multipliers with Rate of Convergence O(1/k)

收敛速度为 O(1/k) 的乘法器交替方向法的变体

• 发表时间: 2012
• 作者: Masao Yamagishi;Shunsuke Ono;and Isao Yamada
• 通讯作者: and Isao Yamada

Sparsity-Aware Feed-Forward Active Noise Control with the Adaptive Douglas-Rachford Splitting

采用自适应 Douglas-Rachford 分裂的稀疏感知前馈主动噪声控制

• 发表时间: 2013
• 期刊: IEICE Technical Workshop on Signal Processing, Technical Report
• 作者: Masao Yamagishi;Isao Yamada
• 通讯作者: Isao Yamada

Adaptive projected subgradient method for expanded real systems

扩展真实系统的自适应投影次梯度法

• 发表时间: 2012
• 作者: Takehiko Mizoguchi;Masao Yamagishi;and Isao Yamada
• 通讯作者: and Isao Yamada

An image restoration by the MV-PURE with SURE based rank selection

使用基于 SURE 的等级选择的 MV-PURE 进行图像恢复

• 发表时间: 2012
• 作者: Masao Yamagishi;and Isao Yamada
• 通讯作者: and Isao Yamada

A RANK SELECTION OF MV-PURE WITH AN UNBIASED PREDICTED-MSE CRITERION AND ITS EFFICIENT IMPLEMENTATION IN IMAGE RESTORATION

具有无偏预测MSE准则的MV-PURE排序选择及其在图像恢复中的高效实现

• 发表时间: 2013
• 作者: Masao Yamagishi;Isao Yamada
• 通讯作者: Isao Yamada