IEEE Signal Processing Letters最新文献_第5页

Enhancing Noise Robustness in Focus Measure Using Tight Framelet Features 利用紧框架特征增强焦点测量中的噪声鲁棒性

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-20 DOI: 10.1109/LSP.2025.3553425

Yan-Ran Li;Junwei Liu;Zhangtao Ye;Lixin Shen;Xiaosheng Zhuang

引用次数: 0

Collaborative Adapter Experts for Class-Incremental Learning 类增量学习的协作适配器专家

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-20 DOI: 10.1109/LSP.2025.3553431

Sunyuan Qiang;Xinxing Yu;Yanyan Liang;Jun Wan;Du Zhang

{"title":"Collaborative Adapter Experts for Class-Incremental Learning","authors":"Sunyuan Qiang;Xinxing Yu;Yanyan Liang;Jun Wan;Du Zhang","doi":"10.1109/LSP.2025.3553431","DOIUrl":"https://doi.org/10.1109/LSP.2025.3553431","url":null,"abstract":"Pre-trained models (PTMs) with parameter-efficient fine-tuning (PEFT) techniques have been extensively utilized in class-incremental learning (CIL) scenarios. However, they still remain susceptible to performance degradation as the individual PEFT module operates as an independent learning entity during the incremental process. To this end, this work proposes a novel class-incremental collaborative adapter experts (CICAE) model, which incorporates multiple adapters operating collaboratively to facilitate CIL. Specifically, our model primarily consists of two phases. Initially, multiple adapters are employed to establish a multi-expert system aimed at acquiring diverse incremental knowledge. Through the collaborative knowledge sharing (CKS) mechanism, the expertise of each adapter expert is transferable, promoting collaborative development and mutual advancement. Subsequently, with the category prototype distributions, collaborative classifier alignment (CCA) is proposed to further align the classifiers with the representation space in a cooperative manner. Extensive experiments on CIL benchmarks validate the superior performance of our model.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1530-1534"},"PeriodicalIF":3.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845330","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Variation Autoencoder of Spatial-Spectral Joint Mask for Hyperspectral Anomaly Detection 用于高光谱异常检测的空间-光谱联合掩模变化自编码器

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-20 DOI: 10.1109/LSP.2025.3553433

Dandan Ma;Zhuozhao Liu;Zhiyu Jiang

{"title":"Variation Autoencoder of Spatial-Spectral Joint Mask for Hyperspectral Anomaly Detection","authors":"Dandan Ma;Zhuozhao Liu;Zhiyu Jiang","doi":"10.1109/LSP.2025.3553433","DOIUrl":"https://doi.org/10.1109/LSP.2025.3553433","url":null,"abstract":"In recent years, autoencoders and their variants have emerged as effective tools for hyperspectral anomaly detection. Nevertheless, owing to the complex distribution of anomalous regions and the similarity in spatial-spectral features, these models often reconstruct anomalies and backgrounds simultaneously, hindering their ability to distinguish between them and reducing detection accuracy. To address this issue, we propose a novel hyperspectral anomaly detection method based on a spatial-spectral joint mask variational autoencoder (VAE). By combining the probabilistic modeling capabilities of VAEs with a masking-based attention mechanism, our method enables more precise extraction of essential background information in localized regions. Specifically, the spatial-spectral joint masking technique is proposed to guide the network to concentrate on background features across multiple dimensions, tackling issues of spatial structure approximation and spectral redundancy. To further enhance robustness in noisy and complex environments, we iteratively refine the reconstructed residual image through recursive filtering. Extensive comparative experiments and ablation studies on multiple public datasets demonstrate that our approach consistently outperforms existing methods in detection accuracy.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1535-1539"},"PeriodicalIF":3.2,"publicationDate":"2025-03-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845481","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

Flow-PLC: Towards Efficient Packet Loss Concealment With Flow Matching 流plc：基于流匹配的高效丢包隐藏

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-20 DOI: 10.1109/LSP.2025.3553421

Da-Hee Yang;Joon-Hyuk Chang

引用次数: 0

Blind Deconvolution of Graph Signals: Robustness to Graph Perturbations 图信号的盲反卷积：对图扰动的鲁棒性

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-19 DOI: 10.1109/LSP.2025.3553064

Chang Ye;Gonzalo Mateos

引用次数: 0

Factorization-Based Information Reconstruction for Enhancing Missing Modality Robustness 基于因子分解的信息重构增强缺失模态鲁棒性

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-18 DOI: 10.1109/LSP.2025.3552435

Chao Wang;Miaolei Zhou;Xiaofei Yu;Junchao Weng;Yong Zhang

{"title":"Factorization-Based Information Reconstruction for Enhancing Missing Modality Robustness","authors":"Chao Wang;Miaolei Zhou;Xiaofei Yu;Junchao Weng;Yong Zhang","doi":"10.1109/LSP.2025.3552435","DOIUrl":"https://doi.org/10.1109/LSP.2025.3552435","url":null,"abstract":"In recent years, Multimodal Sentiment Analysis (MSA) has emerged as a prominent research area, utilizing multiple signals to better understand human sentiment. Previous studies in MSA have primarily concentrated on performing interaction and fusion with complete signals. However, they have overlooked the issue of missing signals, which commonly arise in real-world scenarios due to factors such as occlusion, privacy concerns, and device malfunctions, leading to reduced generalizability. To this end, we propose a Factorization-based Information Reconstruction Framework (FIRF) to mitigate the modality missing problem in the MSA task. Specifically, we propose a fine-grained complementary factorization module that factorizes modality into synergistic, modality-heterogeneous, and noisy representations and design elaborate constraint paradigms for representation learning. Furthermore, we design a distribution calibration self-distillation module that fully recovers the missing semantics by utilizing bidirectional knowledge transfer. Comprehensive experiments on two datasets indicate that FIRF has a significant performance advantage over previous methods with uncertain missing modalities.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1376-1380"},"PeriodicalIF":3.2,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143748819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 Target Localization and Channel Estimation for ODDM-ISAC Systems ODDM-ISAC系统联合目标定位与信道估计

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-18 DOI: 10.1109/LSP.2025.3552518

Luning Lin;Jun Tong;Hai Lin;Hang Zheng;Zhiguo Shi

{"title":"Joint Target Localization and Channel Estimation for ODDM-ISAC Systems","authors":"Luning Lin;Jun Tong;Hai Lin;Hang Zheng;Zhiguo Shi","doi":"10.1109/LSP.2025.3552518","DOIUrl":"https://doi.org/10.1109/LSP.2025.3552518","url":null,"abstract":"In pursuit of reliable performance for high-mobility integrated sensing and communication (ISAC) scenarios, the orthogonal delay-Doppler division multiplexing (ODDM) modulation can be leveraged to exploit the inherent channel sparsity in the delay-Doppler (DD) domain. In this letter, we propose a joint target localization and channel estimation method for ODDM-ISAC systems that employs a multi-pilot training frame to enhance parameter estimation with accumulated signal energy. By exploiting the block-circulant-like structure of the equivalent sampled DD domain (ESDD) channel matrix, multiple pilots are strategically arranged within a training frame. To facilitate effective energy accumulation from different pilots, a phase compensation strategy is devised, which improves the accuracy of parameter estimation for target localization and channel reconstruction. Moreover, the feasibility of the proposed method is theoretically analyzed when the actual delay and Doppler shift are on or off the quantized DD grid, respectively. Simulation results validate the effectiveness of the proposed method for both target localization and channel estimation.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1525-1529"},"PeriodicalIF":3.2,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143845331","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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

DNGG: Medical Image Lossless Encryption via Deep Network Guided Generative 基于深度网络引导生成的医学图像无损加密

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-18 DOI: 10.1109/LSP.2025.3552528

Lin Fan;Meng Li;Zhenting Hu;Yuan Hong;Dexing Kong

引用次数: 0

Optimizing the Order of Modes in Tensor Train Decomposition 张量列分解中模态阶数的优化

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-17 DOI: 10.1109/LSP.2025.3552005

Petr Tichavský;Ondřej Straka

引用次数: 0

Sensing Within Ultra-Short Duration: Extended Subspace Algorithms With Insufficient Snapshots 超短持续时间内的感知：快照不足的扩展子空间算法

IF 3.2 2区工程技术

IEEE Signal Processing Letters Pub Date : 2025-03-17 DOI: 10.1109/LSP.2025.3551943

Teng Ma;Yuxuan Feng;Yue Xiao;Xia Lei;Vladimir Poulkov

{"title":"Sensing Within Ultra-Short Duration: Extended Subspace Algorithms With Insufficient Snapshots","authors":"Teng Ma;Yuxuan Feng;Yue Xiao;Xia Lei;Vladimir Poulkov","doi":"10.1109/LSP.2025.3551943","DOIUrl":"https://doi.org/10.1109/LSP.2025.3551943","url":null,"abstract":"In pursuit of real-time sensing within ultra-short duration, conventional sensing algorithms are gradually failing to fulfill the stringent latency demands. Specifically, traditional subspace-based methods such as multiple signal classification (MUSIC) are hindered by their need for an extensive number of snapshots to accumulate the rank of the spatial covariance matrix (SCM), resulting in poor real-time performance. Moreover, advanced techniques like compressed sensing and machine learning are constrained by requirements for high signal sparsity or suffer from limited generality. To handle these challenges, this paper proposes an innovative extension of subspace theory tailored to insufficient-snapshot scenarios, leveraging the concept of spatio-temporal exchangeability. Based on the defined spatio-temporal correlation predicated on the space translation invariance characteristic of uniform linear arrays, we engineer a pseudo SCM that inherently possesses sufficient rank. This methodology not only resolves the rank-deficiency issue but also fully exploits the array aperture and significantly reduces the noise level. Simulation results are presented, substantiating the feasibility and enhanced performance of the proposed algorithms, marking a significant advancement over existing methodologies.","PeriodicalId":13154,"journal":{"name":"IEEE Signal Processing Letters","volume":"32 ","pages":"1490-1494"},"PeriodicalIF":3.2,"publicationDate":"2025-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143830431","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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