arXiv - EE - Signal Processing最新文献

Self-Supervised Learning via VICReg Enables Training of EMG Pattern Recognition Using Continuous Data with Unclear Labels 通过 VICReg 进行自我监督学习，利用标签不明确的连续数据训练肌电图模式识别能力

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11632

Shriram Tallam Puranam Raghu, Dawn T. MacIsaac, Erik J. Scheme

{"title":"Self-Supervised Learning via VICReg Enables Training of EMG Pattern Recognition Using Continuous Data with Unclear Labels","authors":"Shriram Tallam Puranam Raghu, Dawn T. MacIsaac, Erik J. Scheme","doi":"arxiv-2409.11632","DOIUrl":"https://doi.org/arxiv-2409.11632","url":null,"abstract":"In this study, we investigate the application of self-supervised learning via\u0000pre-trained Long Short-Term Memory (LSTM) networks for training surface\u0000electromyography pattern recognition models (sEMG-PR) using dynamic data with\u0000transitions. While labeling such data poses challenges due to the absence of\u0000ground-truth labels during transitions between classes, self-supervised\u0000pre-training offers a way to circumvent this issue. We compare the performance\u0000of LSTMs trained with either fully-supervised or self-supervised loss to a\u0000conventional non-temporal model (LDA) on two data types: segmented ramp data\u0000(lacking transition information) and continuous dynamic data inclusive of class\u0000transitions. Statistical analysis reveals that the temporal models outperform\u0000non-temporal models when trained with continuous dynamic data. Additionally,\u0000the proposed VICReg pre-trained temporal model with continuous dynamic data\u0000significantly outperformed all other models. Interestingly, when using only\u0000ramp data, the LSTM performed worse than the LDA, suggesting potential\u0000overfitting due to the absence of sufficient dynamics. This highlights the\u0000interplay between data type and model choice. Overall, this work highlights the\u0000importance of representative dynamics in training data and the potential for\u0000leveraging self-supervised approaches to enhance sEMG-PR models.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"54 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251311","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Covert Communications Without Pre-Sharing of Side Information and Channel Estimation Over Quasi-Static Fading Channels 准静态衰减信道上无侧信息预共享和信道估计的隐蔽通信

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11755

Hiroki Fukada, Hiroki Iimori, Chandan Pradhan, Szabolcs Malomsoky, Naoki Ishikawa

引用次数: 0

End-to-End Learning of Transmitter and Receiver Filters in Bandwidth Limited Fiber Optic Communication Systems 带宽受限光纤通信系统中发射器和接收器滤波器的端到端学习

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11980

Søren Føns Nielsen, Francesco Da Ros, Mikkel N. Schmidt, Darko Zibar

引用次数: 0

Atmospheric Turbulence-Immune Free Space Optical Communication System based on Discrete-Time Analog Transmission 基于离散时间模拟传输的大气湍流-免疫自由空间光通信系统

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11928

Hongyu Huang, Zhenming Yu, Yi Lei, Wei Zhang, Yongli Zhao, Shanguo Huang, Kun Xu

{"title":"Atmospheric Turbulence-Immune Free Space Optical Communication System based on Discrete-Time Analog Transmission","authors":"Hongyu Huang, Zhenming Yu, Yi Lei, Wei Zhang, Yongli Zhao, Shanguo Huang, Kun Xu","doi":"arxiv-2409.11928","DOIUrl":"https://doi.org/arxiv-2409.11928","url":null,"abstract":"To effectively mitigate the influence of atmospheric turbulence, a novel\u0000discrete-time analog transmission free-space optical (DTAT-FSO) communication\u0000scheme is proposed. It directly maps information sources to discrete-time\u0000analog symbols via joint source-channel coding and modulation. Differently from\u0000traditional digital free space optical (TD-FSO) schemes, the proposed DTAT-FSO\u0000approach can automatically adapt to the variation of the channel state, with no\u0000need to adjust the specific modulation and coding scheme. The performance of\u0000the DTAT-FSO system was evaluated in both intensity modulation/direct detection\u0000(IM/DD) and coherent FSO systems for high-resolution image transmission. The\u0000results show that the DTAT-FSO reliably transmits images at low received\u0000optical powers (ROPs) and automatically enhances quality at high ROPs, while\u0000the TD-FSO experiences cliff and leveling effects when the channel state\u0000varies. With respect to the TD-FSO scheme, the DTAT-FSO scheme improved\u0000receiver sensitivity by 2.5 dB in the IM/DD FSO system and 0.8 dB in the\u0000coherent FSO system, and it achieved superior image fidelity under the same\u0000ROP. The automatic adaptation feature and improved performance of the DTAT-FSO\u0000suggest its potential for terrestrial, airborne, and satellite optical\u0000networks, addressing challenges posed by atmospheric turbulence.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"20 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251266","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

User Subgrouping in Scalable Cell-Free Massive MIMO Multicasting Systems 可扩展无小区大规模 MIMO 多播系统中的用户分组

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11871

Alejandro de la Fuente, Guillem Femenias, Felip Riera-Palou, Giovanni Interdonato

{"title":"User Subgrouping in Scalable Cell-Free Massive MIMO Multicasting Systems","authors":"Alejandro de la Fuente, Guillem Femenias, Felip Riera-Palou, Giovanni Interdonato","doi":"arxiv-2409.11871","DOIUrl":"https://doi.org/arxiv-2409.11871","url":null,"abstract":"Cell-free massive multiple-input multiple-output (CF-mMIMO) is a breakthrough\u0000technology for beyond-5G systems, designed to significantly boost the energy\u0000and spectral efficiencies of future mobile networks while ensuring a consistent\u0000quality of service for all users. Additionally, multicasting has gained\u0000considerable attention recently because physical-layer multicasting offers an\u0000efficient method for simultaneously serving multiple users with identical\u0000service demands by sharing radio resources. Typically, multicast services are\u0000delivered either via unicast transmissions or a single multicast transmission.\u0000This work, however, introduces a novel subgroup-centric multicast CF-mMIMO\u0000framework that divides users into several multicast subgroups based on the\u0000similarities in their spatial channel characteristics. This approach allows for\u0000efficient sharing of the pilot sequences used for channel estimation and the\u0000precoding filters used for data transmission. The proposed framework employs\u0000two scalable precoding strategies: centralized improved partial MMSE (IP-MMSE)\u0000and distributed conjugate beam-forming (CB). Numerical results show that for\u0000scenarios where users are uniformly distributed across the service area,\u0000unicast transmissions using centralized IP-MMSE precoding are optimal. However,\u0000in cases where users are spatially clustered, multicast subgrouping\u0000significantly improves the sum spectral efficiency (SE) of the multicast\u0000service compared to both unicast and single multicast transmission. Notably, in\u0000clustered scenarios, distributed CB precoding outperforms IP-MMSE in terms of\u0000per-user SE, making it the best solution for delivering multicast content.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"47 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251267","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Accelerated Algorithms for Source Orientation Detection (AORI) and Spatiotemporal LCMV (ALCMV) Beamforming in EEG Source Localization 脑电图源定位中的源方向检测 (AORI) 和时空 LCMV (ALCMV) 波束成形加速算法

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11751

Ava Yektaeian Vaziri, Bahador Makkiabadi

{"title":"Accelerated Algorithms for Source Orientation Detection (AORI) and Spatiotemporal LCMV (ALCMV) Beamforming in EEG Source Localization","authors":"Ava Yektaeian Vaziri, Bahador Makkiabadi","doi":"arxiv-2409.11751","DOIUrl":"https://doi.org/arxiv-2409.11751","url":null,"abstract":"This paper illustrates the development of two efficient source localization\u0000algorithms for electroencephalography (EEG) data, aimed at enhancing real-time\u0000brain signal reconstruction while addressing the computational challenges of\u0000traditional methods. Accurate EEG source localization is crucial for\u0000applications in cognitive neuroscience, neurorehabilitation, and brain-computer\u0000interfaces (BCIs). To make significant progress toward precise source\u0000orientation detection and improved signal reconstruction, we introduce the\u0000Accelerated Linear Constrained Minimum Variance (ALCMV) beamforming toolbox and\u0000the Accelerated Brain Source Orientation Detection (AORI) toolbox. The ALCMV\u0000algorithm speeds up EEG source reconstruction by utilizing recursive covariance\u0000matrix calculations, while AORI simplifies source orientation detection from\u0000three dimensions to one, reducing computational load by 66% compared to\u0000conventional methods. Using both simulated and real EEG data, we demonstrate\u0000that these algorithms maintain high accuracy, with orientation errors below\u00000.2% and signal reconstruction accuracy within 2%. These findings suggest that\u0000the proposed toolboxes represent a substantial advancement in the efficiency\u0000and speed of EEG source localization, making them well-suited for real-time\u0000neurotechnological applications.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"178 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251309","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Real-Time Sound Event Localization and Detection: Deployment Challenges on Edge Devices 实时声音事件定位和检测：边缘设备的部署挑战

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11700

Jun Wei Yeow, Ee-Leng Tan, Jisheng Bai, Santi Peksi, Woon-Seng Gan

引用次数: 0

User Subgrouping and Power Control for Multicast Massive MIMO over Spatially Correlated Channels 空间相关信道上多播大规模 MIMO 的用户分组和功率控制

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11891

Alejandro de la Fuente, Giovanni Interdonato, Giuseppe Araniti

{"title":"User Subgrouping and Power Control for Multicast Massive MIMO over Spatially Correlated Channels","authors":"Alejandro de la Fuente, Giovanni Interdonato, Giuseppe Araniti","doi":"arxiv-2409.11891","DOIUrl":"https://doi.org/arxiv-2409.11891","url":null,"abstract":"Massive multiple-input-multiple-output (MIMO) is unquestionably a key enabler\u0000of the fifth-generation (5G) technology for mobile systems, enabling to meet\u0000the high requirements of upcoming mobile broadband services. Physical-layer\u0000multicasting refers to a technique for simultaneously serving multiple users,\u0000demanding for the same service and sharing the same radio resources, with a\u0000single transmission. Massive MIMO systems with multicast communications have\u0000been so far studied under the ideal assumption of uncorrelated Rayleigh fading\u0000channels. In this work, we consider a practical multicast massive MIMO system\u0000over spatially correlated Rayleigh fading channels, investigating the impact of\u0000the spatial channel correlation on the favorable propagation, hence on the\u0000performance. We propose a subgrouping strategy for the multicast users based on\u0000their channel correlation matrices' similarities. The proposed subgrouping\u0000approach capitalizes on the spatial correlation to enhance the quality of the\u0000channel estimation, and thereby the effectiveness of the precoding. Moreover,\u0000we devise a max-min fairness (MMF) power allocation strategy that makes the\u0000spectral efficiency (SE) among different multicast subgroups uniform. Lastly,\u0000we propose a novel power allocation for uplink (UL) pilot transmission to\u0000maximize the SE among the users within the same multicast subgroup. Simulation\u0000results show a significant SE gain provided by our user subgrouping and power\u0000allocation strategies. Importantly, we show how spatial channel correlation can\u0000be exploited to enhance multicast massive MIMO communications.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"25 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251308","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Electromagnetic Property Sensing and Channel Reconstruction Based on Diffusion Schrödinger Bridge in ISAC 基于 ISAC 扩散薛定谔桥的电磁特性传感和通道重构

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.11651

Yuhua Jiang, Feifei Gao, Shi Jin

{"title":"Electromagnetic Property Sensing and Channel Reconstruction Based on Diffusion Schrödinger Bridge in ISAC","authors":"Yuhua Jiang, Feifei Gao, Shi Jin","doi":"arxiv-2409.11651","DOIUrl":"https://doi.org/arxiv-2409.11651","url":null,"abstract":"Integrated sensing and communications (ISAC) has emerged as a transformative\u0000paradigm for next-generation wireless systems. In this paper, we present a\u0000novel ISAC scheme that leverages the diffusion Schrodinger bridge (DSB) to\u0000realize the sensing of electromagnetic (EM) property of a target as well as the\u0000reconstruction of the wireless channel. The DSB framework connects EM property\u0000sensing and channel reconstruction by establishing a bidirectional process: the\u0000forward process transforms the distribution of EM property into the channel\u0000distribution, while the reverse process reconstructs the EM property from the\u0000channel. To handle the difference in dimensionality between the\u0000high-dimensional sensing channel and the lower-dimensional EM property, we\u0000generate latent representations using an autoencoder network. The autoencoder\u0000compresses the sensing channel into a latent space that retains essential\u0000features, which incorporates positional embeddings to process spatial context.\u0000The simulation results demonstrate the effectiveness of the proposed DSB\u0000framework, which achieves superior reconstruction of the targets shape,\u0000relative permittivity, and conductivity. Moreover, the proposed method can also\u0000realize high-fidelity channel reconstruction given the EM property of the\u0000target. The dual capability of accurately sensing the EM property and\u0000reconstructing the channel across various positions within the sensing area\u0000underscores the versatility and potential of the proposed approach for broad\u0000application in future ISAC systems.","PeriodicalId":501034,"journal":{"name":"arXiv - EE - Signal Processing","volume":"116 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142251315","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Blind Deconvolution on Graphs: Exact and Stable Recovery 图形上的盲解卷：精确稳定的恢复

arXiv - EE - Signal Processing Pub Date : 2024-09-18 DOI: arxiv-2409.12164

Chang Ye, Gonzalo Mateos

引用次数: 0