检索结果-内蒙古大学图书馆

12th IEEE International Conference on Information and Communication Networks, ICICN 2024

作者： Di, Shijie Yuan, Renzhi Peng, Mugen Wang, Siming Chu, Xinyi Chen, Zhibei Sun, Kailin Li, Yuchai Li, Junwei Guo, Jia Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350355802

Displacement based quantum receivers have gained increasing attention because they can beat the standard quantum limit. The displacement value in quantum receivers can be calibrated by using number states as probes. Current estimating methods assumed that ideal number states can be prepared perfectly, which is impossible in practical implementations. In this work, we focus on the impacts of thermal noise when preparing the number states on the estimation performance of displacement operations. We first derive the probability density of photon numbers in a thermal noise contaminated number state after a displacement operation, which we call the noisy displaced number state (NDNS). Based on the derived probability density, we then analyze the performance of estimating displacement operations with NDNS probes by using the maximum likelihood estimation (MLE) method. Numerical results found that the thermal noise in preparing the number states has great influence on the MLE accuracy of the displacement operations. © 2024 IEEE.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

LongMuEL: An entity disambiguation method targeting multiple sources of data and long text 7

LongMuEL: An entity disambiguation method targeting multiple...

引用

7th IEEE International Conference on Big Data and Artificial Intelligence, BDAI 2024

作者： Zhang, Erxuan Shuang, Kai Wang, Xianhao Beijing University of Posts and Telecommunications State Key Laboratory of Networking & Switching Technology Beijing China

ISBN: (纸本)9798350352009

Entity disambiguation is an important method for completing knowledge graphs. The data used for knowledge graph completion often integrates various data types and formats. When designing models for entity disambiguation, a key task is to design a knowledge fusion module to efficiently obtain embeddings of candidate entities from the descriptive text and structured data of the candidates. Since the length of candidate entity descriptive text usually far exceeds the limit of input text for pre-trained models, how to incorporate truncated text knowledge into the embeddings of candidate entities is also a problem to be considered when designing entity disambiguation models. In this paper, we propose a novel entity disambiguation model, LongMuEL, which can represent the long descriptive text of candidate entities and integrate structured text from candidate entity knowledge bases. In the long text embedding module, we use multiple attention modules and weight pooling for all truncated text. When obtaining the embedding of structured text, we adopt two self-supervised tasks to pre-train the pre-trained model and generate negative examples through negative sampling. We demonstrate that pre-training the model with two self-supervised tasks results in better embedding effects for structured data. Finally, we fuse the embedding vectors generated from the two data types to calculate the ranking of candidate entities. LongMuEL is a knowledge graph completion model adapted to existing data features, and through ablation experiments, we verify the effectiveness of each part of LongMuEL. © 2024 IEEE.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

AUTOMATIC DESIGN OF ADAPTER ARCHITECTURES FOR ENHANCED PARAMETER-EFFICIENT FINE-TUNING 49

AUTOMATIC DESIGN OF ADAPTER ARCHITECTURES FOR ENHANCED PARAM...

引用

49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Xu, Siya Wen, Xinyan Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350344868;9798350344851

Adapters, one of the most important parameter-efficient fine-tuning (PEFT) methods, achieve state-of-the-art (SOTA) performance through manual architecture design. To unlock the full potential of adapter tuning, we introduce the AutoAdapter framework, designated to design novel adapter architectures automatically. First, we discuss the adapter design choices and define a search space. Second, we propose C-DARTS, an contribution-based differentiable neural architecture search (NAS) method, to enhance the search results. We conduct comprehensive experiments and analysis on the GLUE and SuperCLUE benchmark tasks, demonstrating that AutoAdapter effectively designs novel adapters that outperform recent baseline PEFT methods.

关键词： Adapters Parameter-efficient fine-tuning Neural architecture search pretrained language models

来源：评论

学校读者我要写书评

暂无评论

Efficient Collaborative Perception with Adaptive Communication in Bandwidth-Constrained Scenarios 6

Efficient Collaborative Perception with Adaptive Communicati...

引用

6th International Conference on Robotics, Intelligent Control and Artificial Intelligence, RICAI 2024

作者： Guo, Junbo Luo, Guiyang Yuan, Quan Li, Jinglin State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9798331541699

In autonomous driving, collaborative perception significantly improves safety by addressing occlusions and expanding the perceptual range of individual vehicles. Nevertheless, effectively exchanging data among vehicles under bandwidth constraints presents a major challenge, especially regarding adaptive communication strategies. To tackle this, we introduce Adacomm (Adaptive Communication for Collaborative Multi-agent Perception), a network-adaptive framework that enhances multi-agent collaborative perception by optimizing data exchange to improve perception accuracy and reduce communication costs. The Adacomm framework consists of modular operations such as blind spot identification, multi-scale feature generation, data scheduling optimization, shared message fusion, and detection decoding. Experimental results demonstrate that Adacomm enhances multi-vehicle perception by 16.8% in AP@IoU=0.5, compared to traditional single-vehicle perception, and lowers average communication costs by 29.5% relative to the existing Where2comm method, with a minimal performance decline of 5.6%. Our analysis across different feature layers indicates a trade-off between communication volume and perception accuracy, with medium-resolution features providing a balanced solution. Overall, Adacomm effectively addresses critical challenges in collaborative perception, showing substantial benefits and advancements in scenarios with limited communication resources. © 2024 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

SoundSpring: Loss-Resilient Audio Transceiver With Dual-Functional Masked Language Modeling

引用

IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS 2025年第4期43卷 1308-1322页

作者： Yao, Shengshi Dai, Jincheng Qin, Xiaoqi Wang, Sixian Wang, Siye Niu, Kai Zhang, Ping Beijing Univ Posts & Telecommun Key Lab Universal Wireless Commun Minist Educ Beijing 100876 Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

In this paper, we propose "SoundSpring", a cutting-edge error-resilient audio transceiver that marries the robustness benefits of joint source-channel coding (JSCC) while also being compatible with current digital communication systems. Unlike recent deep JSCC transceivers, which learn to directly map audio signals to analog channel-input symbols via neural networks, our SoundSpring adopts the layered architecture that delineates audio compression from digital coded transmission, but it sufficiently exploits the impressive in-context predictive capabilities of large language (foundation) models. Integrated with the casual-order mask learning strategy, our single model operates on the latent feature domain and serve dual-functionalities: as efficient audio compressors at the transmitter and as effective mechanisms for packet loss concealment at the receiver. By jointly optimizing towards both audio compression efficiency and transmission error resiliency, we show that mask-learned language models are indeed powerful contextual predictors, and our dual-functional compression and concealment framework offers fresh perspectives on the application of foundation language models in audio communication. Through extensive experimental evaluations, we establish that SoundSpring apparently outperforms contemporary audio transmission systems in terms of signal fidelity metrics and perceptual quality scores. These new findings not only advocate for the practical deployment of SoundSpring in learning-based audio communication systems but also inspire the development of future audio semantic transceivers.

关键词： Transceivers Codecs Resilience Context modeling Packet loss Decoding Audio coding Robustness Receivers Propagation losses Semantic-aware transceiver audio communication error-resilient coding masked language models

来源：评论

学校读者我要写书评

暂无评论

Proactive Platooning Based on C-V2X to Relieve Congestion at a Signalized Intersection

引用

China Communications 2023年第2期20卷 155-167页

作者： Wei Gao Yan Shi Shanzhi Chen The State Key Laboratory of Networking and Switching Technology Beijing University of Posts and TelecommunicationsBeijing 100876China The State Key Laboratory of Wireless Mobile Communications China Academy of Telecommunications TechnologyBeijing 100083China

Nowadays urban traffic congestion becomes a major issue due to the increase in vehicles that lead to more commuting time,accidents,traffic violations,and high fuel *** is a good way to solve traffic congestion because it drives with a smaller inter-vehicle distance than *** paper proposes proactive platooning based on C-V2X(cellular vehicle-to-everything)to relieve *** platooning reduces the inter-vehicle distance and increases the throughput of signalized intersections through integrating the networked control platooning and computation-centralized *** and simulation experiments of proactive platooning were conducted to verify the impact of inter-vehicle distance on platooning latency,platooning safety,and traffic *** results show that the optimal inter-vehicle distance under proactive platooning is less than half of human-driving at a signalized intersection.

关键词： platooning communication platooning safety C-V2X signalized intersection

来源：评论

学校读者我要写书评

暂无评论

INCS: Intent-driven network-wide configuration synthesis based on deep reinforcement learning

引用

COMPUTER NETWORKS 2024年 251卷

作者： Dai, Yuqi Zhang, Hua Wang, Jingyu Liao, Jianxin Zhang, Penghui Southeast Univ Natl Mobile Commun Res Lab Nanjing Peoples R China Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

Configuring modern networks is a complex task due to the intricate low-level configuration languages and diverse routing protocols used across numerous interconnected devices. Network configuration synthesis has been proposed to solve these challenges. However, existing synthesizers encounter several limitations: (i) They cannot scale to large topologies with more than a few tens of routers. (ii) They only support specific routing protocols. (iii) They require prolonged synthesis times. (iv) They lack configuration optimization. To address these limitations, this paper introduces INCS, an end-to-end network configuration synthesizer designed to generate configurations that comply with any hard and soft specifications. It comprises a fact base graph constructor using graph embedding techniques, a Graph Neural Network (GNN)-based predictor employing a Graph Attention Network (GAT) model, and a Deep Reinforcement Learning (DRL)-based optimizer trained with the Deep Deterministic Policy Gradient (DDPG) algorithm. The graph constructor translates network topology, hard specifications, and configuration sketches into a fact base graph. The predictor then generates initial configurations, which are further optimized by the optimizer to meet predefined soft specifications. We compare INCS with two benchmark synthesizers: NetComplete and a neural algorithmic reasoning (NAR)-based synthesizer. The evaluation shows that INCS has superior synthesis speed and specification consistency. It is 27.9x faster than the benchmarks. Even in the presence of unsatisfiable hard specifications, INCS maintains 100% specification consistency, whereas Netcomplete fails to synthesize, and the NAR-based synthesizer achieves only 90% consistency.

关键词： Network configuration synthesis Deep reinforcement learning Network management Intent-driven

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Multi-aerial Base Station Deployment Via DDPG-Mix Algorithm 13th

Energy Efficient Multi-aerial Base Station Deployment Via D...

引用

13th International Conference on Computer Engineering and Networks, CENet 2023

作者： He, Haoran Zhao, Yikun Zhang, Jinli Zhou, Fanqin Li, Wenjing Feng, Lei State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing China

ISBN: (纸本)9789819992386

Compared to ground base stations, aerial base stations (AeBSs) offer advantages in terms of overcoming limitations posed by complex terrain and transportation constraints, providing broader coverage and faster response speeds. In this paper, we investigate the problem of energy efficiency maximization in a multiple AeBSs network and adopt multi-agent deep reinforcement learning (MADRL) for the deployment of AeBSs. Specifically, considering the partial observation range of AeBSs, the multi-AeBS deployment problem is modeled as a decentralized partially observable markov decision process (Dec-POMDP) and a deep deterministic policy gradient mix (DDPG-MIX) algorithm is designed to solve the problem. The proposed algorithm uses the value decomposition framework to solve the lazy agent problem. Simulation results show that our proposed DDPG-MIX algorithm performs better than other baseline algorithms in the multi-AeBS deployment scenario. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Real-Time Power Optimal Schedule Method for Energy Internet Based on LSTM Encoding 19

Real-Time Power Optimal Schedule Method for Energy Internet ...

引用

19th IEEE International Symposium on Broadband Multimedia Systems and Broadcasting (BMSB)

作者： Wang, Jiaxin Xu, Siya Qiu, Xuesong Yu, Peng Beijing Univ Posts & Telecommun State Key Lab Networking & Switching Technol Beijing Peoples R China

ISBN: (纸本)9798350364279;9798350364262

With the rapid growth of green energy and customer-side intermittent loads, dealing with the uncertainty and variability of the energy internet has become an urgent problem. In addition, privacy issues are often overlooked in the power optimal scheduling process. Virtual power plant (VPP), which is based on power IoT technology to aggregate distributed resources (DERs), is an important technical solution to improve the flexible operation performance of new power systems. Thus, this paper introduces a real-time power optimal schedule method for energy internet based on LSTM encoding. First, the cloud-edge collaborative VPP optimal schedule framework is proposed according to the characteristics of the distribution network. Then, a multi-participant power schedule model containing the generation, the consumption, the energy storage, and the market is developed. Considering the privacy of power terminal data and the insufficient feature extraction capability of traditional deep reinforcement learning(DRL) algorithms, a DDPG algorithm based on LSTM encoding (LDDPG) is proposed for power schedule policy making. Finally, comparative experiments demonstrate the superiority of the proposed method.

关键词： cloud edge collaboration deep reinforcement learning optimal schedule privacy virtual power plant(VPP)

来源：评论

学校读者我要写书评

暂无评论

Locating the Root Cause of Large-scale BGP Anomaly with Routing Dependence

Locating the Root Cause of Large-scale BGP Anomaly with Rout...

引用

2024 IEEE International Performance, Computing, and Communications Conference, IPCCC 2024

作者： Zhao, Shiqi Huang, Xiaohong Zhang, Pei Beijing University of Posts and Telecommunications State Key Laboratory of Networking and Switching Technology Beijing China

ISBN: (纸本)9798350367942

BGP anomaly has a major impact on the stability, availability, and efficiency of the Internet, thus locating the network element which initiates the anomaly is crucial for network recovery. However, BGP is an information-hiding protocol and BGP anomalies are heterogeneous, which limit the ability to perform root cause localization. This paper proposes an approach to automatically detect large-scale BGP anomaly and locate the root cause in an efficient and generalized fashion. Firstly, time-series analysis and clustering strategy are performed on BGP traffic to ascertain "when"and "where"anomaly happened to start the investigation. Next, based on interdomain routing's propagation property, a subtopology is constructed which not only reveals the dynamic evolution during anomaly, but also helps narrow down the scale of the Internet for further root cause analysis. Last, by taking deep analysis in the routing characteristics of the root cause from different anomalous scenarios, a routing dependence-based algorithm is developed to realize the localization goal. The proposed method is validated on three different real-world, large-scale BGP anomalies. Results demonstrate its effectiveness and robustness, and in most cases the root causes identified with maximum likelihood are consistent with ground-truths. Further investigations reveal some insights which are helpful for network administrations. © 2024 IEEE.

关键词： anomaly detection BGP anomaly root cause localization routing dependence

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：