arXiv - MATH - Information Theory最新文献_第9页

Nuclear Atomic Norm for parametric estimation of sparse channels 用于稀疏信道参数估计的核原子规范

arXiv - MATH - Information Theory Pub Date : 2024-09-02 DOI: arxiv-2409.01166

Álvaro Callejas-Ramos, Matilde Sánchez-Fernández, Antonia Tulino, Jaime Llorca

引用次数: 0

Decoding Quantum LDPC Codes Using Graph Neural Networks 利用图神经网络解码量子 LDPC 代码

arXiv - MATH - Information Theory Pub Date : 2024-08-09 DOI: arxiv-2408.05170

Vukan Ninkovic, Ognjen Kundacina, Dejan Vukobratovic, Christian Häger, Alexandre Graell i Amat

引用次数: 0

Towards Effective and Interpretable Semantic Communications 实现有效且可解释的语义通信

arXiv - MATH - Information Theory Pub Date : 2024-08-09 DOI: arxiv-2408.04825

Youlong Wu, Yuanmin Shi, Shuai Ma, Chunxiao Jiang, Wei Zhang, Khaled B. Letaief

{"title":"Towards Effective and Interpretable Semantic Communications","authors":"Youlong Wu, Yuanmin Shi, Shuai Ma, Chunxiao Jiang, Wei Zhang, Khaled B. Letaief","doi":"arxiv-2408.04825","DOIUrl":"https://doi.org/arxiv-2408.04825","url":null,"abstract":"With the exponential surge in traffic data and the pressing need for\u0000ultra-low latency in emerging intelligence applications, it is envisioned that\u00006G networks will demand disruptive communication technologies to foster\u0000ubiquitous intelligence and succinctness within the human society. Semantic\u0000communication, a novel paradigm, holds the promise of significantly curtailing\u0000communication overhead and latency by transmitting only task-relevant\u0000information. Despite numerous efforts in both theoretical frameworks and\u0000practical implementations of semantic communications, a substantial\u0000theory-practice gap complicates the theoretical analysis and interpretation,\u0000particularly when employing black-box machine learning techniques. This article\u0000initially delves into information-theoretic metrics such as semantic entropy,\u0000semantic distortions, and semantic communication rate to characterize the\u0000information flow in semantic communications. Subsequently, it provides a\u0000guideline for implementing semantic communications to ensure both theoretical\u0000interpretability and communication effectiveness.","PeriodicalId":501082,"journal":{"name":"arXiv - MATH - Information Theory","volume":"5 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141943447","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

High-Efficiency Urban 3D Radio Map Estimation Based on Sparse Measurements 基于稀疏测量的高效城市 3D 无线电地图估算

arXiv - MATH - Information Theory Pub Date : 2024-08-08 DOI: arxiv-2408.04205

Xinwei Chen, Xiaofeng Zhong, Zijian Zhang, Linglong Dai, Shidong Zhou

{"title":"High-Efficiency Urban 3D Radio Map Estimation Based on Sparse Measurements","authors":"Xinwei Chen, Xiaofeng Zhong, Zijian Zhang, Linglong Dai, Shidong Zhou","doi":"arxiv-2408.04205","DOIUrl":"https://doi.org/arxiv-2408.04205","url":null,"abstract":"Recent widespread applications for unmanned aerial vehicles (UAVs) -- from\u0000infrastructure inspection to urban logistics -- have prompted an urgent need\u0000for high-accuracy three-dimensional (3D) radio maps. However, existing methods\u0000designed for two-dimensional radio maps face challenges of high measurement\u0000costs and limited data availability when extended to 3D scenarios. To tackle\u0000these challenges, we first build a real-world large-scale 3D radio map dataset,\u0000covering over 4.2 million m^3 and over 4 thousand data points in complex urban\u0000environments. We propose a Gaussian Process Regression-based scheme for 3D\u0000radio map estimation, allowing us to realize more accurate map recovery with a\u0000lower RMSE than state-of-the-art schemes by over 2.5 dB. To further enhance\u0000data efficiency, we propose two methods for training point selection, including\u0000an offline clustering-based method and an online maximum a posterior\u0000(MAP)-based method. Extensive experiments demonstrate that the proposed scheme\u0000not only achieves full-map recovery with only 2% of UAV measurements, but also\u0000sheds light on future studies on 3D radio maps.","PeriodicalId":501082,"journal":{"name":"arXiv - MATH - Information Theory","volume":"39 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141943450","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

On many-to-one mappings over finite fields 关于有限域上的多对一映射

arXiv - MATH - Information Theory Pub Date : 2024-08-08 DOI: arxiv-2408.04218

Yanbin Zheng, Yanjin Ding, Meiying Zhang, Pingzhi Yuan, Qiang Wang

引用次数: 0

Hybrid Quantum-Classical Neural Networks for Downlink Beamforming Optimization 用于下行波束成形优化的混合量子-经典神经网络

arXiv - MATH - Information Theory Pub Date : 2024-08-08 DOI: arxiv-2408.04747

Juping Zhang, Gan Zheng, Toshiaki Koike-Akino, Kai-Kit Wong, Fraser Burton

{"title":"Hybrid Quantum-Classical Neural Networks for Downlink Beamforming Optimization","authors":"Juping Zhang, Gan Zheng, Toshiaki Koike-Akino, Kai-Kit Wong, Fraser Burton","doi":"arxiv-2408.04747","DOIUrl":"https://doi.org/arxiv-2408.04747","url":null,"abstract":"This paper investigates quantum machine learning to optimize the beamforming\u0000in a multiuser multiple-input single-output downlink system. We aim to combine\u0000the power of quantum neural networks and the success of classical deep neural\u0000networks to enhance the learning performance. Specifically, we propose two\u0000hybrid quantum-classical neural networks to maximize the sum rate of a downlink\u0000system. The first one proposes a quantum neural network employing parameterized\u0000quantum circuits that follows a classical convolutional neural network. The\u0000classical neural network can be jointly trained with the quantum neural network\u0000or pre-trained leading to a fine-tuning transfer learning method. The second\u0000one designs a quantum convolutional neural network to better extract features\u0000followed by a classical deep neural network. Our results demonstrate the\u0000feasibility of the proposed hybrid neural networks, and reveal that the first\u0000method can achieve similar sum rate performance compared to a benchmark\u0000classical neural network with significantly less training parameters; while the\u0000second method can achieve higher sum rate especially in presence of many users\u0000still with less training parameters. The robustness of the proposed methods is\u0000verified using both software simulators and hardware emulators considering\u0000noisy intermediate-scale quantum devices.","PeriodicalId":501082,"journal":{"name":"arXiv - MATH - Information Theory","volume":"21 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141943446","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

pyBregMan: A Python library for Bregman Manifolds pyBregMan: 用于布雷格曼曼形图的 Python 库

arXiv - MATH - Information Theory Pub Date : 2024-08-08 DOI: arxiv-2408.04175

Frank Nielsen, Alexander Soen

引用次数: 0

Trustworthy Image Semantic Communication with GenAI: Explainablity, Controllability, and Efficiency 利用 GenAI 进行可信的图像语义交流：可解释性、可控性和效率

arXiv - MATH - Information Theory Pub Date : 2024-08-07 DOI: arxiv-2408.03806

Xijun Wang, Dongshan Ye, Chenyuan Feng, Howard H. Yang, Xiang Chen, Tony Q. S. Quek

引用次数: 0

Limitations of the decoding-to-LPN reduction via code smoothing 通过平滑代码减少解码到 LPN 的局限性

arXiv - MATH - Information Theory Pub Date : 2024-08-07 DOI: arxiv-2408.03742

Madhura Pathegama, Alexander Barg

引用次数: 0

On Fast SC-based Polar Decoders: Metric Polarization and a Pruning Technique 基于 SC 的快速极化解码器：公制极化和剪枝技术

arXiv - MATH - Information Theory Pub Date : 2024-08-07 DOI: arxiv-2408.03840

Mohsen Moradi, Hessam Mahdavifar

{"title":"On Fast SC-based Polar Decoders: Metric Polarization and a Pruning Technique","authors":"Mohsen Moradi, Hessam Mahdavifar","doi":"arxiv-2408.03840","DOIUrl":"https://doi.org/arxiv-2408.03840","url":null,"abstract":"Short- to medium-block-length polar-like and polarization-adjusted\u0000convolutional (PAC) codes have demonstrated exceptional error-correction\u0000performance through sequential decoding. Successive cancellation list (SCL)\u0000decoding of polar-like and PAC codes can potentially match the performance of\u0000sequential decoding though a relatively large list size is often required. By\u0000benefiting from an optimal metric function, sequential decoding can find the\u0000correct path corresponding to the transmitted data by following almost one path\u0000on average at high Eb/N0 regimes. When considering a large number of paths in\u0000SCL decoding, a main bottleneck emerges that is the need for a rather expensive\u0000sorting operation at each level of decoding of data bits. In this paper, we\u0000propose a method to obtain the optimal metric function for each depth of the\u0000polarization tree through a process that we call polarization of the metric\u0000function. One of the major advantages of the proposed metric function is that\u0000it can be utilized in fast SC-based (FSC) and SCL-based (FSCL) decoders, i.e.,\u0000decoders that opt to skip the so-called rate-1 and rate-0 nodes in the binary\u0000tree representation for significantly more efficient implementation.\u0000Furthermore, based on the average value of the polarized metric function of\u0000FSC-based decoders, we introduce a pruning technique that keeps only the paths\u0000whose metric values are close to the average value. As a result, our proposed\u0000technique significantly reduces the number of required sorting operations for\u0000FSCL-based decoding algorithms. For instance, for a high-rate PAC(128,99) code,\u0000SCL decoding with a list size of 32 achieves error-correction performance\u0000comparable to the Fano algorithm. Our method reduces the number of sorting\u0000operations of FSCL decoding to 33%, further decreasing latency.","PeriodicalId":501082,"journal":{"name":"arXiv - MATH - Information Theory","volume":"43 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141943452","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