Digital Signal Processing最新文献_第8页

S2VSNet: Single stage V-shaped network for image deraining & dehazing S2VSNet：用于图像去毛刺和去细化的单级 V 型网络

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-30 DOI: 10.1016/j.dsp.2024.104786

Thatikonda Ragini , Kodali Prakash , Ramalinga Swamy Cheruku

Producing high-quality, noise-free images from noisy or hazy inputs relies on essential tasks such as single image deraining and dehazing. In many advanced multi-stage networks, there is often an imbalance in contextual information, leading to increased complexity. To address these challenges, we propose a simplified method inspired by a U-Net structure, resulting in the “Single-Stage V-Shaped Network” (

S^{2} VSNet

), capable of handling both deraining and dehazing tasks. A key innovation in our approach is the introduction of a Feature Fusion Module (FFM), which facilitates the sharing of information across multiple scales and hierarchical layers within the encoder-decoder structure. As the network progresses towards deeper layers, the FFM gradually integrates insights from higher levels, ensuring that spatial details are preserved while contextual feature maps are balanced. This integration enhances the image processing capability, producing noise-free, high-quality outputs. To maintain efficiency and reduce system complexity, we replaced or removed several non-essential non-linear activation functions, opting instead for simple multiplication operations. Additionally, we introduced a “Multi-Head Attention Integrated Module” (MHAIM) as an intermediary layer between encoder-decoder levels. This module addresses the limited receptive fields of traditional Convolutional Neural Networks (CNNs), allowing for the capture of more comprehensive feature-map information. Our focus on deraining and dehazing led to extensive experiments on a wide range of synthetic and real-world datasets. To further validate the robustness of our network, we implemented

S^{2} VSNet

on a low-end edge device, achieving deraining in 2.46 seconds.

要从嘈杂或朦胧的输入图像中生成高质量、无噪音的图像，需要完成一些基本任务，如单幅图像去毛刺和去阴影。在许多先进的多级网络中，上下文信息往往不平衡，导致复杂性增加。为了应对这些挑战，我们提出了一种受 U 型网络结构启发的简化方法，即 "单级 V 型网络"（S2VSNet），它能够同时处理去毛刺和去雾化任务。我们的方法的一个关键创新是引入了特征融合模块（FFM），该模块有助于在编码器-解码器结构中的多个尺度和层次层之间共享信息。随着网络向更深层次发展，FFM 会逐渐整合来自更高层次的洞察力，确保保留空间细节，同时平衡上下文特征图。这种整合增强了图像处理能力，产生无噪声的高质量输出。为了保持效率并降低系统复杂性，我们替换或删除了几个非必要的非线性激活函数，转而使用简单的乘法运算。此外，我们还引入了 "多头注意力集成模块"（MHAIM），作为编码器-解码器层之间的中间层。该模块解决了传统卷积神经网络（CNN）感受野有限的问题，从而可以捕捉到更全面的特征图信息。我们将重点放在了去毛刺和去马赛克上，并在大量的合成数据集和真实数据集上进行了广泛的实验。为了进一步验证我们网络的鲁棒性，我们在低端边缘设备上实施了 S2VSNet，在 2.46 秒内实现了去链。

{"title":"S2VSNet: Single stage V-shaped network for image deraining & dehazing","authors":"Thatikonda Ragini , Kodali Prakash , Ramalinga Swamy Cheruku","doi":"10.1016/j.dsp.2024.104786","DOIUrl":"10.1016/j.dsp.2024.104786","url":null,"abstract":"<div><div>Producing high-quality, noise-free images from noisy or hazy inputs relies on essential tasks such as single image deraining and dehazing. In many advanced multi-stage networks, there is often an imbalance in contextual information, leading to increased complexity. To address these challenges, we propose a simplified method inspired by a U-Net structure, resulting in the “Single-Stage V-Shaped Network” (<span><math><msup><mrow><mtext>S</mtext></mrow><mrow><mn>2</mn></mrow></msup><mtext>VSNet</mtext></math></span>), capable of handling both deraining and dehazing tasks. A key innovation in our approach is the introduction of a Feature Fusion Module (FFM), which facilitates the sharing of information across multiple scales and hierarchical layers within the encoder-decoder structure. As the network progresses towards deeper layers, the FFM gradually integrates insights from higher levels, ensuring that spatial details are preserved while contextual feature maps are balanced. This integration enhances the image processing capability, producing noise-free, high-quality outputs. To maintain efficiency and reduce system complexity, we replaced or removed several non-essential non-linear activation functions, opting instead for simple multiplication operations. Additionally, we introduced a “Multi-Head Attention Integrated Module” (MHAIM) as an intermediary layer between encoder-decoder levels. This module addresses the limited receptive fields of traditional Convolutional Neural Networks (CNNs), allowing for the capture of more comprehensive feature-map information. Our focus on deraining and dehazing led to extensive experiments on a wide range of synthetic and real-world datasets. To further validate the robustness of our network, we implemented <span><math><msup><mrow><mtext>S</mtext></mrow><mrow><mn>2</mn></mrow></msup><mtext>VSNet</mtext></math></span> on a low-end edge device, achieving deraining in 2.46 seconds.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104786"},"PeriodicalIF":2.9,"publicationDate":"2024-09-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428114","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Quantized kernel recursive q-Rényi-like algorithm 量化核递归 q-Rényi-like 算法

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-27 DOI: 10.1016/j.dsp.2024.104790

Wenwen Zhou , Yanmin Zhang , Chunlong Huang , Sergey V. Volvenko , Wei Xue

This paper introduces the kernel recursive q-Rényi-like (KRqRL) algorithm, based on the q-Rényi kernel function and the kernel recursive least squares (KRLS) algorithm. To reduce the computational complexity and memory requirements of the KRqRL algorithm, an online vector quantization (VQ) method is employed to quantize the network size to a codebook size, resulting in the quantized KRqRL (QKRqRL) algorithm. This paper provides a detailed analysis of the convergence and computational complexity of the QKRqRL algorithm. In the simulation experiments, the network size of each algorithm is reduced to 25% of its original size. The performance of the QKRqRL algorithm is evaluated in terms of convergence speed, prediction error, and computation time under non-Gaussian noise conditions. Finally, the QKRqRL algorithm is further validated using sunspot data, demonstrating its superior stability and online prediction performance.

本文介绍了基于 q-Rényi 核函数和核递归最小二乘法（KRLS）的核递归 q-Rényi-like 算法（KRqRL）。为了降低 KRqRL 算法的计算复杂度和内存需求，本文采用了在线矢量量化（VQ）方法，将网络大小量化为编码本大小，从而形成了量化 KRqRL（QKRqRL）算法。本文详细分析了 QKRqRL 算法的收敛性和计算复杂度。在仿真实验中，每种算法的网络规模都缩小到原来的 25%。在非高斯噪声条件下，从收敛速度、预测误差和计算时间等方面评估了 QKRqRL 算法的性能。最后，利用太阳黑子数据对 QKRqRL 算法进行了进一步验证，证明了其卓越的稳定性和在线预测性能。

引用次数: 0

DEDBNet: DoG-enhanced dual-branch object detection network for remote sensing object detection DEDBNet：用于遥感物体探测的 DoG 增强型双分支物体探测网络

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-27 DOI: 10.1016/j.dsp.2024.104789

Dongbo Pan, Jingfeng Zhao, Tianchi Zhu, Jianjun Yuan

With the improvement of spatial resolution of remote sensing images, object detection of remote sensing images has gradually become a difficult task. Extracted object features are usually hidden in a large amount of interference information in the background due to the complexity and large area of backgrounds, as well as the multi-scale nature of objects in remote sensing images. Still, many existing background weakening methods face difficulties in practical applications and are prone to high rates of false positives and false negatives. Therefore, remote sensing object detection has become increasingly challenging. To address these challenges, a novel background weakening method called Difference of Gaussian (DoG) to weaken background (DWB) module is proposed. Then, we develop a dual-branch network, named DoG-Enhanced Dual-Branch Object Detection Network (DEDBNet) for Remote Sensing Object Detection. The base branch network is responsible for detecting objects, while the DWB's branch network corrects the detected objects using feature-level attention. To combine the features of these branches, we propose two new methods Self-Mutual-Correcter with Detect heads (SMCD) for corrective learning and Map Channel Attention (MCA) for channel attention. Self-Corrector (SC) enables modification and integration of features, while the Mutual-Corrector (MC) enhances the features and further fuses them. We evaluate our proposed network, DEDBNet, through extensive experiments on four public datasets (DOTA with an mAP of 0.836, DIOR with an mAP of 0.871, NWPU VHR-10 with an mAP of 0.973, and RSOD with an mAP of 0.975). The results demonstrate that our method outperforms other state-of-the-art object detection methods significantly for remote sensing images.

随着遥感图像空间分辨率的提高，遥感图像的目标检测逐渐成为一项艰巨的任务。由于遥感图像中背景的复杂性和大面积性以及物体的多尺度性，提取的物体特征通常隐藏在背景的大量干扰信息中。然而，现有的许多背景弱化方法在实际应用中都面临着困难，容易产生较高的假阳性和假阴性。因此，遥感物体检测变得越来越具有挑战性。为了应对这些挑战，我们提出了一种新颖的背景弱化方法--高斯差（DoG）弱化背景（DWB）模块。然后，我们开发了一种双分支网络，名为 DoG 增强双分支目标检测网络（DEDBNet），用于遥感目标检测。基础分支网络负责检测物体，而 DWB 的分支网络则利用特征级关注修正检测到的物体。为了结合这些分支的特点，我们提出了两种新方法：用于矫正学习的带检测头的自互矫正器（SMCD）和用于通道关注的地图通道关注（MCA）。自校正器（SC）可对特征进行修改和整合，而互校正器（MC）可增强特征并进一步将其融合。我们在四个公开数据集（DOTA 的 mAP 为 0.836，DIOR 的 mAP 为 0.871，NWPU VHR-10 的 mAP 为 0.973，RSOD 的 mAP 为 0.975）上进行了大量实验，评估了我们提出的网络 DEDBNet。结果表明，在遥感图像方面，我们的方法明显优于其他最先进的物体检测方法。

{"title":"DEDBNet: DoG-enhanced dual-branch object detection network for remote sensing object detection","authors":"Dongbo Pan, Jingfeng Zhao, Tianchi Zhu, Jianjun Yuan","doi":"10.1016/j.dsp.2024.104789","DOIUrl":"10.1016/j.dsp.2024.104789","url":null,"abstract":"<div><div>With the improvement of spatial resolution of remote sensing images, object detection of remote sensing images has gradually become a difficult task. Extracted object features are usually hidden in a large amount of interference information in the background due to the complexity and large area of backgrounds, as well as the multi-scale nature of objects in remote sensing images. Still, many existing background weakening methods face difficulties in practical applications and are prone to high rates of false positives and false negatives. Therefore, remote sensing object detection has become increasingly challenging. To address these challenges, a novel background weakening method called Difference of Gaussian (DoG) to weaken background (DWB) module is proposed. Then, we develop a dual-branch network, named DoG-Enhanced Dual-Branch Object Detection Network (DEDBNet) for Remote Sensing Object Detection. The base branch network is responsible for detecting objects, while the DWB's branch network corrects the detected objects using feature-level attention. To combine the features of these branches, we propose two new methods Self-Mutual-Correcter with Detect heads (SMCD) for corrective learning and Map Channel Attention (MCA) for channel attention. Self-Corrector (SC) enables modification and integration of features, while the Mutual-Corrector (MC) enhances the features and further fuses them. We evaluate our proposed network, DEDBNet, through extensive experiments on four public datasets (DOTA with an mAP of 0.836, DIOR with an mAP of 0.871, NWPU VHR-10 with an mAP of 0.973, and RSOD with an mAP of 0.975). The results demonstrate that our method outperforms other state-of-the-art object detection methods significantly for remote sensing images.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104789"},"PeriodicalIF":2.9,"publicationDate":"2024-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142428112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-modal signal adaptive time-reassigned multisynchrosqueezing transform of mechanism 多模式信号自适应时间分配多同步阙值变换的机制

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-24 DOI: 10.1016/j.dsp.2024.104788

Qiqiang Wu , Xianmin Zhang , Bo Zhao

High-end mechanical equipment often operates under non-stationary conditions, such as varying loads, changing speeds, and transient impacts, which can lead to failures. Time-frequency analysis (TFA) integrates time and frequency parameters, allowing for detailed signal analysis and is widely used in this context. To improve the accuracy of assessing the operational status of mechanical equipment, this paper proposed a multi-modal signal adaptive time reassignment multiple synchrosqueezing transform (MSST) TFA method. This method enhances the MSST method by using a local maximum technique to address energy ambiguity in TFA. Additionally, the optimal window width for each function is determined through iterative processes to better concentrate energy in the TFA. Multi-modal signals are jointly analyzed using an impulse feature extraction method for signal reconstruction, enabling multi-dimensional fault analysis. The proposed method is validated with both simulation and experimental data from a planar parallel mechanism (PPM) and is compared against classical and advanced techniques. The results show that the method effectively captures shock features in multi-modal signals, offering a more consolidated time-frequency representation (TFR) than existing TFA algorithms.

高端机械设备通常在非稳态条件下运行，如负载变化、速度变化和瞬态冲击，这些都可能导致故障。时频分析（TFA）集成了时间和频率参数，可以进行详细的信号分析，在这方面得到了广泛应用。为了提高评估机械设备运行状态的准确性，本文提出了一种多模态信号自适应时间重分配多重同步阙值变换（MSST）TFA 方法。该方法通过使用局部最大值技术来解决 TFA 中的能量模糊问题，从而增强了 MSST 方法。此外，通过迭代过程确定每个函数的最佳窗宽，以更好地集中 TFA 中的能量。使用脉冲特征提取方法对多模态信号进行联合分析，以重建信号，从而实现多维故障分析。利用平面并联机构 (PPM) 的模拟和实验数据对所提出的方法进行了验证，并与经典和先进技术进行了比较。结果表明，与现有的 TFA 算法相比，该方法能有效捕捉多模态信号中的冲击特征，提供更全面的时频表示（TFR）。

{"title":"Multi-modal signal adaptive time-reassigned multisynchrosqueezing transform of mechanism","authors":"Qiqiang Wu , Xianmin Zhang , Bo Zhao","doi":"10.1016/j.dsp.2024.104788","DOIUrl":"10.1016/j.dsp.2024.104788","url":null,"abstract":"<div><div>High-end mechanical equipment often operates under non-stationary conditions, such as varying loads, changing speeds, and transient impacts, which can lead to failures. Time-frequency analysis (TFA) integrates time and frequency parameters, allowing for detailed signal analysis and is widely used in this context. To improve the accuracy of assessing the operational status of mechanical equipment, this paper proposed a multi-modal signal adaptive time reassignment multiple synchrosqueezing transform (MSST) TFA method. This method enhances the MSST method by using a local maximum technique to address energy ambiguity in TFA. Additionally, the optimal window width for each function is determined through iterative processes to better concentrate energy in the TFA. Multi-modal signals are jointly analyzed using an impulse feature extraction method for signal reconstruction, enabling multi-dimensional fault analysis. The proposed method is validated with both simulation and experimental data from a planar parallel mechanism (PPM) and is compared against classical and advanced techniques. The results show that the method effectively captures shock features in multi-modal signals, offering a more consolidated time-frequency representation (TFR) than existing TFA algorithms.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104788"},"PeriodicalIF":2.9,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142359606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

DeConformer-SENet: An efficient deformable conformer speech enhancement network DeConformer-SENet：高效的可变形保形语音增强网络

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-24 DOI: 10.1016/j.dsp.2024.104787

Man Li, Ya Liu, Li Zhou

The Conformer model has demonstrated superior performance in speech enhancement by combining the long-range relationship modeling capability of self-attention with the local information processing ability of convolutional neural networks (CNNs). However, existing Conformer-based speech enhancement models struggle to balance performance and model complexity. In this work, we propose, DeConformer-SENet, an end-to-end time-domain deformable Conformer speech enhancement model, with modifications to both the self-attention and CNN components. Firstly, we introduce the time-frequency-channel self-attention (TFC-SA) module, which compresses information from each dimension of the input features into a one-dimensional vector. By calculating the energy distribution, this module models long-range relationships across three dimensions, reducing computational complexity while maintaining performance. Additionally, we replace standard convolutions with deformable convolutions, aiming to expand the receptive field of the CNN and accurately model local features. We validate our proposed DeConformer-SENet on the WSJ0-SI84 + DNS Challenge dataset. Experimental results demonstrate that DeConformer-SENet outperforms existing Conformer and Transformer models in terms of ESTOI and PESQ metrics, while also being more computationally efficient. Furthermore, ablation studies confirm that DeConformer-SENet improvements enhance the performance of conventional Conformer and reduce model complexity without compromising the overall effectiveness.

Conformer 模型将自我注意的长程关系建模能力与卷积神经网络（CNN）的局部信息处理能力相结合，在语音增强方面表现出卓越的性能。然而，现有的基于 Conformer 的语音增强模型很难在性能和模型复杂度之间取得平衡。在这项工作中，我们提出了端到端时域可变形 Conformer 语音增强模型 DeConformer-SENet，并对自注意和 CNN 部分进行了修改。首先，我们引入了时频信道自注意（TFC-SA）模块，它将输入特征的每个维度的信息压缩为一维向量。通过计算能量分布，该模块对三个维度的长程关系进行建模，从而在保持性能的同时降低了计算复杂度。此外，我们还用可变形卷积取代了标准卷积，旨在扩大 CNN 的感受野，并对局部特征进行精确建模。我们在 WSJ0-SI84 + DNS Challenge 数据集上验证了我们提出的 DeConformer-SENet。实验结果表明，DeConformer-SENet 在 ESTOI 和 PESQ 指标方面优于现有的 Conformer 和 Transformer 模型，同时计算效率也更高。此外，消融研究证实，DeConformer-SENet 的改进提高了传统 Conformer 的性能，降低了模型的复杂性，同时不影响整体效果。

{"title":"DeConformer-SENet: An efficient deformable conformer speech enhancement network","authors":"Man Li, Ya Liu, Li Zhou","doi":"10.1016/j.dsp.2024.104787","DOIUrl":"10.1016/j.dsp.2024.104787","url":null,"abstract":"<div><div>The Conformer model has demonstrated superior performance in speech enhancement by combining the long-range relationship modeling capability of self-attention with the local information processing ability of convolutional neural networks (CNNs). However, existing Conformer-based speech enhancement models struggle to balance performance and model complexity. In this work, we propose, DeConformer-SENet, an end-to-end time-domain deformable Conformer speech enhancement model, with modifications to both the self-attention and CNN components. Firstly, we introduce the time-frequency-channel self-attention (TFC-SA) module, which compresses information from each dimension of the input features into a one-dimensional vector. By calculating the energy distribution, this module models long-range relationships across three dimensions, reducing computational complexity while maintaining performance. Additionally, we replace standard convolutions with deformable convolutions, aiming to expand the receptive field of the CNN and accurately model local features. We validate our proposed DeConformer-SENet on the WSJ0-SI84 + DNS Challenge dataset. Experimental results demonstrate that DeConformer-SENet outperforms existing Conformer and Transformer models in terms of ESTOI and PESQ metrics, while also being more computationally efficient. Furthermore, ablation studies confirm that DeConformer-SENet improvements enhance the performance of conventional Conformer and reduce model complexity without compromising the overall effectiveness.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104787"},"PeriodicalIF":2.9,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142322923","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Robustness enhancement in neural networks with alpha-stable training noise 利用阿尔法稳定训练噪声增强神经网络的鲁棒性

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-20 DOI: 10.1016/j.dsp.2024.104778

Xueqiong Yuan , Jipeng Li , Ercan Engin Kuruoglu

With the increasing use of deep learning on data collected by non-perfect sensors and in non-perfect environments, the robustness of deep learning systems has become an important issue. A common approach for obtaining robustness to noise has been to train deep learning systems with data augmented with Gaussian noise. In this work, the common choice of Gaussian noise is challenged and the possibility of stronger robustness for non-Gaussian impulsive noise is explored, specifically alpha-stable noise. Justified by the Generalized Central Limit Theorem and evidenced by observations in various application areas, alpha-stable noise is widely present in nature. By comparing the testing accuracy of models trained with Gaussian noise and alpha-stable noise on data corrupted by different noise, it is found that training with alpha-stable noise is more effective than Gaussian noise, especially when the dataset is corrupted by impulsive noise, thus improving the robustness of the model. Moreover, in the testing on the common corruption benchmark dataset, training with alpha-stable noise also achieves promising results, improving the robustness of the model to other corruption types and demonstrating comparable performance with other state-of-the-art data augmentation methods. Consequently, a novel data augmentation method is proposed that replaces Gaussian noise, which is typically added to the training data, with alpha-stable noise.

随着深度学习越来越多地用于非完美传感器和非完美环境中收集的数据，深度学习系统的鲁棒性已成为一个重要问题。获得噪声鲁棒性的一种常见方法是用添加了高斯噪声的数据训练深度学习系统。在这项工作中，人们对高斯噪声的常见选择提出了质疑，并探索了非高斯脉冲噪声（特别是阿尔法稳定噪声）具有更强鲁棒性的可能性。根据广义中心极限定理，并通过在不同应用领域的观察证明，α-稳定噪声广泛存在于自然界中。通过比较用高斯噪声和阿尔法稳定噪声训练的模型在不同噪声破坏的数据上的测试精度，发现用阿尔法稳定噪声训练比高斯噪声更有效，特别是当数据集被脉冲噪声破坏时，从而提高了模型的鲁棒性。此外，在对常见腐败基准数据集进行测试时，使用阿尔法稳定噪声进行训练也取得了可喜的结果，提高了模型对其他腐败类型的鲁棒性，其性能与其他最先进的数据增强方法不相上下。因此，我们提出了一种新的数据增强方法，用阿尔法稳定噪声取代通常添加到训练数据中的高斯噪声。

{"title":"Robustness enhancement in neural networks with alpha-stable training noise","authors":"Xueqiong Yuan , Jipeng Li , Ercan Engin Kuruoglu","doi":"10.1016/j.dsp.2024.104778","DOIUrl":"10.1016/j.dsp.2024.104778","url":null,"abstract":"<div><div>With the increasing use of deep learning on data collected by non-perfect sensors and in non-perfect environments, the robustness of deep learning systems has become an important issue. A common approach for obtaining robustness to noise has been to train deep learning systems with data augmented with Gaussian noise. In this work, the common choice of Gaussian noise is challenged and the possibility of stronger robustness for non-Gaussian impulsive noise is explored, specifically alpha-stable noise. Justified by the Generalized Central Limit Theorem and evidenced by observations in various application areas, alpha-stable noise is widely present in nature. By comparing the testing accuracy of models trained with Gaussian noise and alpha-stable noise on data corrupted by different noise, it is found that training with alpha-stable noise is more effective than Gaussian noise, especially when the dataset is corrupted by impulsive noise, thus improving the robustness of the model. Moreover, in the testing on the common corruption benchmark dataset, training with alpha-stable noise also achieves promising results, improving the robustness of the model to other corruption types and demonstrating comparable performance with other state-of-the-art data augmentation methods. Consequently, a novel data augmentation method is proposed that replaces Gaussian noise, which is typically added to the training data, with alpha-stable noise.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104778"},"PeriodicalIF":2.9,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142322920","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Explicit entropy error bound for compressive DOA estimation in sensor array 传感器阵列中压缩 DOA 估计的显式熵误差约束

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-19 DOI: 10.1016/j.dsp.2024.104785

Nan Wang , Han Zhang , Xiaolong Kong , Dazhuan Xu

Compressive sensing (CS) simplifies software and hardware by sampling below the Nyquist rate, making it widely used in array signal processing. For the assessment of the compressive direction-of-arrival (DOA) estimation, a lower bound on the mean square error is essential. However, the most widely utilized Cramér-Rao bound (CRB) is only asymptotically tight. This paper proposes a globally tight bound with a closed-form expression for compressive DOA estimation in the uniform linear arrays employing Shannon information theory. Based on the a posteriori probability density function, we propose the indicator of the entropy error (EE) with compression to assess the DOA estimation. The theoretical EE bounds the compressive DOA estimation performance. Moreover, the explicit EE is derived by approximating the normalized differential entropy, which is comprehensive and captures the effect of the compression ratio, the SNR, the number of elements, and the mean square bandwidth. In Particular, the compression ratio has almost no influence on the EE in low SNR. Additionally, the asymptotic lower bound of the theoretical EE is identical to the CRB. Simulation results illustrate the superiority of EE over CRB in evaluating and predicting compressive DOA estimation performance in the uniform linear arrays.

压缩传感（CS）通过低于奈奎斯特速率的采样简化了软件和硬件，因此在阵列信号处理中得到了广泛应用。对于压缩到达方向（DOA）估计的评估，均方误差的下限至关重要。然而，最广泛使用的克拉梅尔-拉奥约束（CRB）只是渐近紧密的。本文利用香农信息论，为均匀线性阵列中的压缩 DOA 估计提出了一个具有闭式表达的全局紧约束。基于后验概率密度函数，我们提出了压缩熵误差（EE）指标来评估 DOA 估计。理论 EE 对压缩 DOA 估计性能进行了约束。此外，显式 EE 是通过近似归一化差分熵得出的，它很全面，能捕捉到压缩比、信噪比、元素数和均方带宽的影响。特别是在低信噪比情况下，压缩比对熵几乎没有影响。此外，理论 EE 的渐进下限与 CRB 相同。仿真结果表明，在评估和预测均匀线性阵列中的压缩 DOA 估计性能时，EE 优于 CRB。

{"title":"Explicit entropy error bound for compressive DOA estimation in sensor array","authors":"Nan Wang , Han Zhang , Xiaolong Kong , Dazhuan Xu","doi":"10.1016/j.dsp.2024.104785","DOIUrl":"10.1016/j.dsp.2024.104785","url":null,"abstract":"<div><div>Compressive sensing (CS) simplifies software and hardware by sampling below the Nyquist rate, making it widely used in array signal processing. For the assessment of the compressive direction-of-arrival (DOA) estimation, a lower bound on the mean square error is essential. However, the most widely utilized Cramér-Rao bound (CRB) is only asymptotically tight. This paper proposes a globally tight bound with a closed-form expression for compressive DOA estimation in the uniform linear arrays employing Shannon information theory. Based on the <em>a posteriori</em> probability density function, we propose the indicator of the entropy error (EE) with compression to assess the DOA estimation. The theoretical EE bounds the compressive DOA estimation performance. Moreover, the explicit EE is derived by approximating the normalized differential entropy, which is comprehensive and captures the effect of the compression ratio, the SNR, the number of elements, and the mean square bandwidth. In Particular, the compression ratio has almost no influence on the EE in low SNR. Additionally, the asymptotic lower bound of the theoretical EE is identical to the CRB. Simulation results illustrate the superiority of EE over CRB in evaluating and predicting compressive DOA estimation performance in the uniform linear arrays.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104785"},"PeriodicalIF":2.9,"publicationDate":"2024-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142312703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Recursive state estimation for delayed complex networks with random link failures and stochastic inner coupling under cyber attacks 网络攻击下具有随机链路故障和随机内部耦合的延迟复杂网络的递归状态估计

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-19 DOI: 10.1016/j.dsp.2024.104784

Hui Qi , Huaiyu Wu , Xiujuan Zheng

This paper focuses on the filtering issue regarding a class of uncertain time-delayed complex networks (CNs) subject to random link failures (RLF) and stochastic inner coupling (SIC) under cyber attacks, where the attacks occur in a random way with uncertain occurrence probabilities. The phenomena of RLF and SIC are described via a Bernoulli distributed random variable and a Gaussian noise, respectively. The former reflects the presence or absence of connections between different nodes, while the latter presents the uncertainty of the inner coupling strength. In addition, denial of service attacks (DoSAs) and false data injection attacks (FDIAs) are both taken into account in the discussed cyber space, where the switching behavior of attack modes is described by two independent Bernoulli distributed random variables. A novel recursive estimator is constructed with consideration of RLF, uncertain time-delay, SIC and cyber attacks, wherein a state estimation error covariance upper bound (SEECUB) is obtained. And such upper bound is minimized via reasonable design of the estimator gain. Through theoretical deduction and analysis, it is proved that the presented SEECUB is uniformly bounded under certain conditions. Finally, the availability of the proposed filtering strategy and the gained results are verified through simulation examples.

本文重点研究了一类不确定的延时复杂网络（CN）在网络攻击下的过滤问题，这类网络受到随机链路故障（RLF）和随机内部耦合（SIC）的影响，攻击以随机方式发生，发生概率不确定。RLF 和 SIC 现象分别通过伯努利分布式随机变量和高斯噪声来描述。前者反映了不同节点之间是否存在连接，而后者则显示了内部耦合强度的不确定性。此外，在所讨论的网络空间中，拒绝服务攻击（DoSA）和虚假数据注入攻击（FDIA）都被考虑在内，攻击模式的切换行为由两个独立的伯努利分布式随机变量来描述。考虑到 RLF、不确定时延、SIC 和网络攻击，构建了一种新的递归估计器，从而获得了状态估计误差协方差上界（SEECUB）。并通过合理设计估计器增益使该上界最小化。通过理论推导和分析，证明了所提出的 SEECUB 在一定条件下是均匀有界的。最后，通过仿真实例验证了所提滤波策略的可用性和所得结果。

{"title":"Recursive state estimation for delayed complex networks with random link failures and stochastic inner coupling under cyber attacks","authors":"Hui Qi , Huaiyu Wu , Xiujuan Zheng","doi":"10.1016/j.dsp.2024.104784","DOIUrl":"10.1016/j.dsp.2024.104784","url":null,"abstract":"<div><div>This paper focuses on the filtering issue regarding a class of uncertain time-delayed complex networks (CNs) subject to random link failures (RLF) and stochastic inner coupling (SIC) under cyber attacks, where the attacks occur in a random way with uncertain occurrence probabilities. The phenomena of RLF and SIC are described via a Bernoulli distributed random variable and a Gaussian noise, respectively. The former reflects the presence or absence of connections between different nodes, while the latter presents the uncertainty of the inner coupling strength. In addition, denial of service attacks (DoSAs) and false data injection attacks (FDIAs) are both taken into account in the discussed cyber space, where the switching behavior of attack modes is described by two independent Bernoulli distributed random variables. A novel recursive estimator is constructed with consideration of RLF, uncertain time-delay, SIC and cyber attacks, wherein a state estimation error covariance upper bound (SEECUB) is obtained. And such upper bound is minimized via reasonable design of the estimator gain. Through theoretical deduction and analysis, it is proved that the presented SEECUB is uniformly bounded under certain conditions. Finally, the availability of the proposed filtering strategy and the gained results are verified through simulation examples.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104784"},"PeriodicalIF":2.9,"publicationDate":"2024-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142359605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Joint beamforming design for STAR-RIS-assisted integrated sensing, communication, and power transfer systems STAR-RIS 辅助集成传感、通信和电力传输系统的联合波束成形设计

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-18 DOI: 10.1016/j.dsp.2024.104783

Xingxing Huang, Guoping Zhang, Hongbo Xu, Dong Wang, Kunyu Li, Ze Wang

In the sixth-generation network, many low-power devices are predicted to be integrated into tasks incorporating communication and sensing. The integrated sensing and communication (ISAC) technology reuses a wireless signal for data transfer and radar sensing. Furthermore, wireless signals have the capacity to transmit energy, allowing for the simultaneous wireless information and power transfer (SWIPT). To enhance spectrum utilization, the deeper integration of SWIPT and ISAC has opened up novel investigate directions for integrated sensing, communication, and power transfer (ISCPT). This paper investigates a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) supported ISCPT system that solves the problem of traditional RIS not achieving full space wireless signal coverage. In particular, since the serious path loss issue of sensing, we propose a novel architecture for installing dedicated sensors on STAR-RIS. Under this setting, we jointly optimized the passive beamforming at STAR-RIS and the active beamforming at base station to minimize the Cramér-Rao bound (CRB) used to estimate the sensing target's two-dimensional direction of arrivals, while being constrained by the lowest signal-to-interference-plus-noise-ratio (SINR) of communication users (CUs), the lowest energy harvesting (EH) of energy users (EUs), and the maximum transmission power at base station. For complex non-convex problems, a proposed two-layer cyclic algorithm utilizes penalty dual decomposition (PDD) and block coordinate descent (BCD). Finally, the numerical outcomes verify the efficacy of our suggested design, which reveals the performance trade-off between communication, power transmission and sensing. Furthermore, compared to traditional RIS, the estimated CRB of this design is lower.

在第六代网络中，许多低功耗设备预计将被集成到通信和传感任务中。集成传感与通信（ISAC）技术可重复利用无线信号进行数据传输和雷达传感。此外，无线信号还具有传输能量的能力，可同时进行无线信息和功率传输（SWIPT）。为了提高频谱利用率，SWIPT 和 ISAC 的深度集成为集成传感、通信和功率传输（ISCPT）开辟了新的研究方向。本文研究了一种同时发射和反射的可重构智能表面（STAR-RIS）支持的 ISCPT 系统，该系统解决了传统 RIS 无法实现全空间无线信号覆盖的问题。特别是，由于传感存在严重的路径损耗问题，我们提出了一种在 STAR-RIS 上安装专用传感器的新型架构。在此背景下，我们联合优化了 STAR-RIS 上的无源波束成形和基站上的有源波束成形，以最小化用于估计传感目标二维到达方向的克拉梅尔-拉奥约束（CRB），同时受到通信用户（CUs）的最低信噪比（SINR）、能源用户（EUs）的最低能量收集（EH）和基站最大传输功率的限制。对于复杂的非凸问题，提出的双层循环算法利用了惩罚二元分解（PDD）和块坐标下降（BCD）。最后，数值结果验证了我们建议的设计的有效性，揭示了通信、功率传输和传感之间的性能权衡。此外，与传统的 RIS 相比，该设计的估计 CRB 更低。

{"title":"Joint beamforming design for STAR-RIS-assisted integrated sensing, communication, and power transfer systems","authors":"Xingxing Huang, Guoping Zhang, Hongbo Xu, Dong Wang, Kunyu Li, Ze Wang","doi":"10.1016/j.dsp.2024.104783","DOIUrl":"10.1016/j.dsp.2024.104783","url":null,"abstract":"<div><div>In the sixth-generation network, many low-power devices are predicted to be integrated into tasks incorporating communication and sensing. The integrated sensing and communication (ISAC) technology reuses a wireless signal for data transfer and radar sensing. Furthermore, wireless signals have the capacity to transmit energy, allowing for the simultaneous wireless information and power transfer (SWIPT). To enhance spectrum utilization, the deeper integration of SWIPT and ISAC has opened up novel investigate directions for integrated sensing, communication, and power transfer (ISCPT). This paper investigates a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS) supported ISCPT system that solves the problem of traditional RIS not achieving full space wireless signal coverage. In particular, since the serious path loss issue of sensing, we propose a novel architecture for installing dedicated sensors on STAR-RIS. Under this setting, we jointly optimized the passive beamforming at STAR-RIS and the active beamforming at base station to minimize the Cramér-Rao bound (CRB) used to estimate the sensing target's two-dimensional direction of arrivals, while being constrained by the lowest signal-to-interference-plus-noise-ratio (SINR) of communication users (CUs), the lowest energy harvesting (EH) of energy users (EUs), and the maximum transmission power at base station. For complex non-convex problems, a proposed two-layer cyclic algorithm utilizes penalty dual decomposition (PDD) and block coordinate descent (BCD). Finally, the numerical outcomes verify the efficacy of our suggested design, which reveals the performance trade-off between communication, power transmission and sensing. Furthermore, compared to traditional RIS, the estimated CRB of this design is lower.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104783"},"PeriodicalIF":2.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142322921","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel FOTD-FRSET for optimization TFF analysis under impulsive noise 用于脉冲噪声下 TFF 优化分析的新型 FOTD-FRSET

IF 2.9 3区工程技术 Q2 ENGINEERING, ELECTRICAL & ELECTRONIC

Digital Signal Processing

Pub Date : 2024-09-18 DOI: 10.1016/j.dsp.2024.104743

Yong Guo , Houyou Wang , Lidong Yang

Impulsive noise is characterized by large amplitude and short duration, causing significant interference to the non-stationary signal representation and characteristic extraction. In response to the inadequacy of existing time-frequency analysis (TFA) methods in accurately representing the signal under impulsive noise, a novel time-fractional-frequency (TFF) analysis method based on FOTD-FRSET is proposed in this paper. This method effectively suppresses impulsive noise through fractional order tracking differentiator (FOTD), and then establishes the non-stationary signal TFF distribution by fractional synchroextraction transform (FRSET). Experimental results demonstrate that FOTD-FRSET can construct high-resolution TFF spectrum under impulsive noise, with superior energy concentration and ridge extraction over some existing methods. Furthermore, a noise correction algorithm is utilized to address the signal representation and characteristic extraction in the presence of non-standard symmetric α-stable distribution impulsive noise, enhancing the practicality of the proposed method for measured noise. Ultimately, the developed FOTD-FRSET method is effectively employed for linear frequency modulation (LFM) signal parameter estimation, and shows superior performance in the estimation accuracy, noise robustness, and practicality compared with existing methods.

脉冲噪声具有振幅大、持续时间短的特点，对非稳态信号的表示和特征提取造成很大干扰。针对现有时频分析（TFA）方法在准确表示脉冲噪声下信号方面的不足，本文提出了一种基于 FOTD-FRSET 的新型时分频（TFF）分析方法。该方法通过分数阶跟踪微分器（FOTD）有效抑制脉冲噪声，然后通过分数同步提取变换（FRSET）建立非稳态信号 TFF 分布。实验结果表明，FOTD-FRSET 可以构建脉冲噪声下的高分辨率 TFF 频谱，其能量集中和脊提取效果优于现有的一些方法。此外，在非标准对称 α 稳定分布脉冲噪声下，利用噪声校正算法解决信号表示和特征提取问题，增强了所提方法在测量噪声时的实用性。最终，所开发的 FOTD-FRSET 方法被有效地用于线性频率调制（LFM）信号参数估计，与现有方法相比，在估计精度、噪声鲁棒性和实用性方面都表现出了卓越的性能。

{"title":"A novel FOTD-FRSET for optimization TFF analysis under impulsive noise","authors":"Yong Guo , Houyou Wang , Lidong Yang","doi":"10.1016/j.dsp.2024.104743","DOIUrl":"10.1016/j.dsp.2024.104743","url":null,"abstract":"<div><div>Impulsive noise is characterized by large amplitude and short duration, causing significant interference to the non-stationary signal representation and characteristic extraction. In response to the inadequacy of existing time-frequency analysis (TFA) methods in accurately representing the signal under impulsive noise, a novel time-fractional-frequency (TFF) analysis method based on FOTD-FRSET is proposed in this paper. This method effectively suppresses impulsive noise through fractional order tracking differentiator (FOTD), and then establishes the non-stationary signal TFF distribution by fractional synchroextraction transform (FRSET). Experimental results demonstrate that FOTD-FRSET can construct high-resolution TFF spectrum under impulsive noise, with superior energy concentration and ridge extraction over some existing methods. Furthermore, a noise correction algorithm is utilized to address the signal representation and characteristic extraction in the presence of non-standard symmetric <em>α</em>-stable distribution impulsive noise, enhancing the practicality of the proposed method for measured noise. Ultimately, the developed FOTD-FRSET method is effectively employed for linear frequency modulation (LFM) signal parameter estimation, and shows superior performance in the estimation accuracy, noise robustness, and practicality compared with existing methods.</div></div>","PeriodicalId":51011,"journal":{"name":"Digital Signal Processing","volume":"156 ","pages":"Article 104743"},"PeriodicalIF":2.9,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142322922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0