Pub Date : 2024-09-11DOI: 10.26599/TST.2023.9010145
Beiteng Yang;Haibin Zhu;Dongning Liu
The Vehicle Allocation Problem (VAP) in the vehicle sales chain has three bottlenecks in practice. The first is to collect relevant cooperation or conflict information, the second is to accurately quantify and analyze other factors affecting the distribution of cars, and the third is to establish a stable and rapid response to the vehicle allocation management method. In order to improve the real-time performance and reliability of vehicle allocation in the vehicle sales chain, it is crucial to find a method that can respond quickly and stabilize the vehicle allocation strategy. Therefore, this paper addresses these issues by extending Group Multi-Role Assignment with Cooperation and Conflict Factors (GMRACCF) from a new perspective. Through the logical reasoning of closure computation, the KD45 logic algorithm is used to find the implicit cognitive Cooperation and Conflict Factors (CCF). Therefore, a collaborative filtering comprehensive evaluation method is proposed to help administrators determine the influence weight of CCFs and Cooperation Scales (CSs) on the all-round performance according to their needs. Based on collaborative filtering, semantic modification is applied to resolve conflicts among qualifications. Large-scale simulation results show that the proposed method is feasible and robust, and provides a reliable decision-making reference in the vehicle sales chain.
{"title":"Improve GMRACCF Qualifications via Collaborative Filtering in Vehicle Sales Chain","authors":"Beiteng Yang;Haibin Zhu;Dongning Liu","doi":"10.26599/TST.2023.9010145","DOIUrl":"https://doi.org/10.26599/TST.2023.9010145","url":null,"abstract":"The Vehicle Allocation Problem (VAP) in the vehicle sales chain has three bottlenecks in practice. The first is to collect relevant cooperation or conflict information, the second is to accurately quantify and analyze other factors affecting the distribution of cars, and the third is to establish a stable and rapid response to the vehicle allocation management method. In order to improve the real-time performance and reliability of vehicle allocation in the vehicle sales chain, it is crucial to find a method that can respond quickly and stabilize the vehicle allocation strategy. Therefore, this paper addresses these issues by extending Group Multi-Role Assignment with Cooperation and Conflict Factors (GMRACCF) from a new perspective. Through the logical reasoning of closure computation, the KD45 logic algorithm is used to find the implicit cognitive Cooperation and Conflict Factors (CCF). Therefore, a collaborative filtering comprehensive evaluation method is proposed to help administrators determine the influence weight of CCFs and Cooperation Scales (CSs) on the all-round performance according to their needs. Based on collaborative filtering, semantic modification is applied to resolve conflicts among qualifications. Large-scale simulation results show that the proposed method is feasible and robust, and provides a reliable decision-making reference in the vehicle sales chain.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"247-261"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676354","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Heterogeneous graphs contain multiple types of entities and relations, which are capable of modeling complex interactions. Embedding on heterogeneous graphs has become an essential tool for analyzing and understanding such graphs. Although these meticulously designed methods make progress, they are limited by model design and computational resources, making it difficult to scale to large-scale heterogeneous graph data and hindering the application and promotion of these methods. In this paper, we propose Restage, a relation structure-aware hierarchical heterogeneous graph embedding framework. Under this framework, embedding only a smaller-scale graph with existing graph representation learning methods is sufficient to obtain node representations on the original heterogeneous graph. We consider two types of relation structures in heterogeneous graphs: interaction relations and affiliation relations. Firstly, we design a relation structure-aware coarsening method to successively coarsen the original graph to the top-level layer, resulting in a smaller-scale graph. Secondly, we allow any unsupervised representation learning methods to obtain node embeddings on the top-level graph. Finally, we design a relation structure-aware refinement method to successively refine the node embeddings from the top-level graph back to the original graph, obtaining node embeddings on the original graph. Experimental results on three public heterogeneous graph datasets demonstrate the enhanced scalability of representation learning methods by the proposed Restage. On another large-scale graph, the speed of existing representation learning methods is increased by up to eighteen times at most.
{"title":"Restage: Relation Structure-Aware Hierarchical Heterogeneous Graph Embedding","authors":"Huanjing Zhao;Pinde Rui;Jie Chen;Shu Zhao;Yanping Zhang","doi":"10.26599/TST.2023.9010147","DOIUrl":"https://doi.org/10.26599/TST.2023.9010147","url":null,"abstract":"Heterogeneous graphs contain multiple types of entities and relations, which are capable of modeling complex interactions. Embedding on heterogeneous graphs has become an essential tool for analyzing and understanding such graphs. Although these meticulously designed methods make progress, they are limited by model design and computational resources, making it difficult to scale to large-scale heterogeneous graph data and hindering the application and promotion of these methods. In this paper, we propose Restage, a relation structure-aware hierarchical heterogeneous graph embedding framework. Under this framework, embedding only a smaller-scale graph with existing graph representation learning methods is sufficient to obtain node representations on the original heterogeneous graph. We consider two types of relation structures in heterogeneous graphs: interaction relations and affiliation relations. Firstly, we design a relation structure-aware coarsening method to successively coarsen the original graph to the top-level layer, resulting in a smaller-scale graph. Secondly, we allow any unsupervised representation learning methods to obtain node embeddings on the top-level graph. Finally, we design a relation structure-aware refinement method to successively refine the node embeddings from the top-level graph back to the original graph, obtaining node embeddings on the original graph. Experimental results on three public heterogeneous graph datasets demonstrate the enhanced scalability of representation learning methods by the proposed Restage. On another large-scale graph, the speed of existing representation learning methods is increased by up to eighteen times at most.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"198-214"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676347","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169642","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Diffusion models are a type of generative deep learning model that can process medical images more efficiently than traditional generative models. They have been applied to several medical image computing tasks. This paper aims to help researchers understand the advancements of diffusion models in medical image computing. It begins by describing the fundamental principles, sampling methods, and architecture of diffusion models. Subsequently, it discusses the application of diffusion models in five medical image computing tasks: image generation, modality conversion, image segmentation, image denoising, and anomaly detection. Additionally, this paper conducts fine-tuning of a large model for image generation tasks and comparative experiments between diffusion models and traditional generative models across these five tasks. The evaluation of the fine-tuned large model shows its potential for clinical applications. Comparative experiments demonstrate that diffusion models have a distinct advantage in tasks related to image generation, modality conversion, and image denoising. However, they require further optimization in image segmentation and anomaly detection tasks to match the efficacy of traditional models. Our codes are publicly available at: https://github.com/hiahub/CodeForDiffusion.
{"title":"Diffusion Models for Medical Image Computing: A Survey","authors":"Yaqing Shi;Abudukelimu Abulizi;Hao Wang;Ke Feng;Nihemaiti Abudukelimu;Youli Su;Halidanmu Abudukelimu","doi":"10.26599/TST.2024.9010047","DOIUrl":"https://doi.org/10.26599/TST.2024.9010047","url":null,"abstract":"Diffusion models are a type of generative deep learning model that can process medical images more efficiently than traditional generative models. They have been applied to several medical image computing tasks. This paper aims to help researchers understand the advancements of diffusion models in medical image computing. It begins by describing the fundamental principles, sampling methods, and architecture of diffusion models. Subsequently, it discusses the application of diffusion models in five medical image computing tasks: image generation, modality conversion, image segmentation, image denoising, and anomaly detection. Additionally, this paper conducts fine-tuning of a large model for image generation tasks and comparative experiments between diffusion models and traditional generative models across these five tasks. The evaluation of the fine-tuned large model shows its potential for clinical applications. Comparative experiments demonstrate that diffusion models have a distinct advantage in tasks related to image generation, modality conversion, and image denoising. However, they require further optimization in image segmentation and anomaly detection tasks to match the efficacy of traditional models. Our codes are publicly available at: https://github.com/hiahub/CodeForDiffusion.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"357-383"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676408","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169643","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2024.9010024
Binbin Feng;Zhijun Ding
Workload prediction is critical in enabling proactive resource management of cloud applications. Accurate workload prediction is valuable for cloud users and providers as it can effectively guide many practices, such as performance assurance, cost reduction, and energy consumption optimization. However, cloud workload prediction is highly challenging due to the complexity and dynamics of workloads, and various solutions have been proposed to enhance the prediction behavior. This paper aims to provide an in-depth understanding and categorization of existing solutions through extensive literature reviews. Unlike existing surveys, for the first time, we comprehensively sort out and analyze the development landscape of workload prediction from a new perspective, i.e., application-oriented rather than prediction methodologies per se. Specifically, we first introduce the basic features of workload prediction, and then analyze and categorize existing efforts based on two significant characteristics of cloud applications: variability and heterogeneity. Furthermore, we also investigate how workload prediction is applied to resource management. Finally, open research opportunities in workload prediction are highlighted to foster further advancements.
{"title":"Application-Oriented Cloud Workload Prediction: A Survey and New Perspectives","authors":"Binbin Feng;Zhijun Ding","doi":"10.26599/TST.2024.9010024","DOIUrl":"https://doi.org/10.26599/TST.2024.9010024","url":null,"abstract":"Workload prediction is critical in enabling proactive resource management of cloud applications. Accurate workload prediction is valuable for cloud users and providers as it can effectively guide many practices, such as performance assurance, cost reduction, and energy consumption optimization. However, cloud workload prediction is highly challenging due to the complexity and dynamics of workloads, and various solutions have been proposed to enhance the prediction behavior. This paper aims to provide an in-depth understanding and categorization of existing solutions through extensive literature reviews. Unlike existing surveys, for the first time, we comprehensively sort out and analyze the development landscape of workload prediction from a new perspective, i.e., application-oriented rather than prediction methodologies per se. Specifically, we first introduce the basic features of workload prediction, and then analyze and categorize existing efforts based on two significant characteristics of cloud applications: variability and heterogeneity. Furthermore, we also investigate how workload prediction is applied to resource management. Finally, open research opportunities in workload prediction are highlighted to foster further advancements.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"34-54"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676402","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169563","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2024.9010063
Yibo Dong;Jin Liu;Jiaqi Ren;Zhe Li;Weili Li
Due to our increasing dependence on infrastructure networks, the attack and defense game in these networks has draw great concerns from security agencies. Moreover, when it comes to evaluating the payoffs in practical attack and defense games in infrastructure networks, the lack of consideration for the fuzziness and uncertainty of subjective human judgment brings forth significant challenges to the analysis of strategic interactions among decision makers. This paper employs intuitionistic fuzzy sets (IFSs) to depict such uncertain payoffs, and introduce a theoretical framework for analyzing the attack and defense game in infrastructure networks based on intuitionistic fuzzy theory. We take the changes in three complex network metrics as the universe of discourse, and intuitionistic fuzzy sets are employed based on this universe of discourse to reflect the satisfaction of decision makers. We employ an algorithm based on intuitionistic fuzzy theory to find the Nash equilibrium, and conduct experiments on both local and global networks. Results show that: (1) the utilization of intuitionistic fuzzy sets to depict the payoffs of attack and defense games in infrastructure networks can reflect the unique characteristics of decision makers' subjective preferences. (2) the use of differently weighted proportions of the three complex network metrics has little impact on decision makers' choices of different strategies.
{"title":"Attack and Defense Game with Intuitionistic Fuzzy Payoffs in Infrastructure Networks","authors":"Yibo Dong;Jin Liu;Jiaqi Ren;Zhe Li;Weili Li","doi":"10.26599/TST.2024.9010063","DOIUrl":"https://doi.org/10.26599/TST.2024.9010063","url":null,"abstract":"Due to our increasing dependence on infrastructure networks, the attack and defense game in these networks has draw great concerns from security agencies. Moreover, when it comes to evaluating the payoffs in practical attack and defense games in infrastructure networks, the lack of consideration for the fuzziness and uncertainty of subjective human judgment brings forth significant challenges to the analysis of strategic interactions among decision makers. This paper employs intuitionistic fuzzy sets (IFSs) to depict such uncertain payoffs, and introduce a theoretical framework for analyzing the attack and defense game in infrastructure networks based on intuitionistic fuzzy theory. We take the changes in three complex network metrics as the universe of discourse, and intuitionistic fuzzy sets are employed based on this universe of discourse to reflect the satisfaction of decision makers. We employ an algorithm based on intuitionistic fuzzy theory to find the Nash equilibrium, and conduct experiments on both local and global networks. Results show that: (1) the utilization of intuitionistic fuzzy sets to depict the payoffs of attack and defense games in infrastructure networks can reflect the unique characteristics of decision makers' subjective preferences. (2) the use of differently weighted proportions of the three complex network metrics has little impact on decision makers' choices of different strategies.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"384-401"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676403","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169638","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2023.9010157
Hao Pan;Lili Qiu
Low Earth Orbit (LEO) satellite communication is vital for wireless systems. The main challenges in designing LEO satellite ground stations include increasing the input signal strength to counteract severe path loss, and adaptively steering the direction of the output signal to accommodate the continuous movement of LEO satellites. To overcome these challenges, we present a novel transceiver system, referred to as MetaLEO. This system integrates a passive metasurface with a small phased array, enabling powerful focusing and adaptive signal steering. By harnessing the metasurface's robust wavefront manipulation capabilities and the programmability of phased arrays, MetaLEO offers an efficient and cost-effective solution that supports both uplink and downlink bands. Specifically, we devise a joint optimization model specifically to obtain the optimal uplink codebook for phased array antennas and metasurface phase profile, which enables electronic steering. In a similar manner, we establish the downlink metasurface phase profile to enhance focusing and signal reception. MetaLEO's efficacy is evaluated via theoretical analysis, simulations, and experiments. Our prototype includes a single metasurface with 21×21 uplink and 22×22 downlink elements, and a 1×4 antenna array for receiving and transmitting. Experimental results show signal strength improvements of 8.32 dB (uplink) and 16.57 dB (downlink).
{"title":"Passive Metasurface-Based Low Earth Orbit Ground Station Design","authors":"Hao Pan;Lili Qiu","doi":"10.26599/TST.2023.9010157","DOIUrl":"https://doi.org/10.26599/TST.2023.9010157","url":null,"abstract":"Low Earth Orbit (LEO) satellite communication is vital for wireless systems. The main challenges in designing LEO satellite ground stations include increasing the input signal strength to counteract severe path loss, and adaptively steering the direction of the output signal to accommodate the continuous movement of LEO satellites. To overcome these challenges, we present a novel transceiver system, referred to as MetaLEO. This system integrates a passive metasurface with a small phased array, enabling powerful focusing and adaptive signal steering. By harnessing the metasurface's robust wavefront manipulation capabilities and the programmability of phased arrays, MetaLEO offers an efficient and cost-effective solution that supports both uplink and downlink bands. Specifically, we devise a joint optimization model specifically to obtain the optimal uplink codebook for phased array antennas and metasurface phase profile, which enables electronic steering. In a similar manner, we establish the downlink metasurface phase profile to enhance focusing and signal reception. MetaLEO's efficacy is evaluated via theoretical analysis, simulations, and experiments. Our prototype includes a single metasurface with 21×21 uplink and 22×22 downlink elements, and a 1×4 antenna array for receiving and transmitting. Experimental results show signal strength improvements of 8.32 dB (uplink) and 16.57 dB (downlink).","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"148-160"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676348","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The exponential advancement witnessed in 5G communication and quantum computing has presented unparalleled prospects for safeguarding sensitive data within healthcare infrastructures. This study proposes a novel framework for healthcare applications that integrates 5G communication, quantum computing, and sensitive data measurement to address the challenges of measuring and securely transmitting sensitive medical data. The framework includes a quantum-inspired method for quantifying data sensitivity based on quantum superposition and entanglement principles and a delegated quantum computing protocol for secure data transmission in 5G-enabled healthcare systems, ensuring user anonymity and data confidentiality. The framework is applied to innovative healthcare scenarios, such as secure 5G voice communication, data transmission, and short message services. Experimental results demonstrate the framework's high accuracy in sensitive data measurement and enhanced security for data transmission in 5G healthcare systems, surpassing existing approaches.
{"title":"Quantum-Inspired Sensitive Data Measurement and Secure Transmission in 5G-Enabled Healthcare Systems","authors":"Xiaohong Lv;Shalli Rani;Shanmuganathan Manimurugan;Adam Slowik;Yanhong Feng","doi":"10.26599/TST.2024.9010122","DOIUrl":"https://doi.org/10.26599/TST.2024.9010122","url":null,"abstract":"The exponential advancement witnessed in 5G communication and quantum computing has presented unparalleled prospects for safeguarding sensitive data within healthcare infrastructures. This study proposes a novel framework for healthcare applications that integrates 5G communication, quantum computing, and sensitive data measurement to address the challenges of measuring and securely transmitting sensitive medical data. The framework includes a quantum-inspired method for quantifying data sensitivity based on quantum superposition and entanglement principles and a delegated quantum computing protocol for secure data transmission in 5G-enabled healthcare systems, ensuring user anonymity and data confidentiality. The framework is applied to innovative healthcare scenarios, such as secure 5G voice communication, data transmission, and short message services. Experimental results demonstrate the framework's high accuracy in sensitive data measurement and enhanced security for data transmission in 5G healthcare systems, surpassing existing approaches.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"456-478"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676360","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169561","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2024.9010020
Wenji He;Haipeng Yao;Huan Chang;Yunjie Liu
With various service types including massive machine-type communication (mMTC) and ultra-reliable low-latency communication (URLLC), fifth generation (5G) networks require advanced resources management strategies. As a method to segment network resources logically, network slicing (NS) addresses the challenges of heterogeneity and scalability prevalent in these networks. Traditional software-defined networking (SDN) technologies, lack the flexibility needed for precise control over network resources and fine-grained packet management. This has led to significant developments in programmable switches, with programming protocol-independent packet processors (P4) emerging as a transformative programming language. P4 endows network devices with flexibility and programmability, overcoming traditional SDN limitations and enabling more dynamic, precise network slicing implementations. In our work, we leverage the capabilities of P4 to forge a groundbreaking closed-loop architecture that synergizes the programmable data plane with an intelligent control plane. We set up a token bucket-based bandwidth management and traffic isolation mechanism in the data plane, and use the generative diffusion model to generate the key configuration of the strategy in the control plane. Through comprehensive experimentation, we validate the effectiveness of our architecture, underscoring its potential as a significant advancement in 5G network traffic management.
第五代（5G）网络拥有多种服务类型，包括大规模机器型通信（mMTC）和超可靠低延迟通信（URLLC），因此需要先进的资源管理策略。作为一种对网络资源进行逻辑分割的方法，网络切片（NS）解决了这些网络中普遍存在的异构性和可扩展性挑战。传统的软件定义网络（SDN）技术缺乏精确控制网络资源和细粒度数据包管理所需的灵活性。这促使可编程交换机取得了重大发展，与编程协议无关的数据包处理器（P4）成为一种变革性的编程语言。P4 赋予网络设备灵活性和可编程性，克服了传统的 SDN 限制，实现了更动态、更精确的网络切片。在我们的工作中，我们利用 P4 的功能打造了一个开创性的闭环架构，将可编程数据平面与智能控制平面协同起来。我们在数据平面建立了基于令牌桶的带宽管理和流量隔离机制，并在控制平面使用生成式扩散模型生成策略的关键配置。通过全面的实验，我们验证了我们的架构的有效性，强调了其作为 5G 网络流量管理的重要进步的潜力。
{"title":"A P4-Based Approach to Traffic Isolation and Bandwidth Management for 5G Network Slicing","authors":"Wenji He;Haipeng Yao;Huan Chang;Yunjie Liu","doi":"10.26599/TST.2024.9010020","DOIUrl":"https://doi.org/10.26599/TST.2024.9010020","url":null,"abstract":"With various service types including massive machine-type communication (mMTC) and ultra-reliable low-latency communication (URLLC), fifth generation (5G) networks require advanced resources management strategies. As a method to segment network resources logically, network slicing (NS) addresses the challenges of heterogeneity and scalability prevalent in these networks. Traditional software-defined networking (SDN) technologies, lack the flexibility needed for precise control over network resources and fine-grained packet management. This has led to significant developments in programmable switches, with programming protocol-independent packet processors (P4) emerging as a transformative programming language. P4 endows network devices with flexibility and programmability, overcoming traditional SDN limitations and enabling more dynamic, precise network slicing implementations. In our work, we leverage the capabilities of P4 to forge a groundbreaking closed-loop architecture that synergizes the programmable data plane with an intelligent control plane. We set up a token bucket-based bandwidth management and traffic isolation mechanism in the data plane, and use the generative diffusion model to generate the key configuration of the strategy in the control plane. Through comprehensive experimentation, we validate the effectiveness of our architecture, underscoring its potential as a significant advancement in 5G network traffic management.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"171-185"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676349","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169560","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2024.9010040
Runjie Miao;Chenchen Wu;Jinjiang Yuan
Capacitated facility location problem (CFLP) is a classical combinatorial optimization problem that has various applications in operations research, theoretical computer science, and management science. In the CFLP, we have a potential facilities set and a clients set. Each facility has a certain capacity and an open cost, and each client has a spliitable demand that need to be met. The goal is to open some facilities and assign all clients to these open facilities so that the total cost is as low as possible. The CFLP is NP-hard (non-deterministic polynomial-hard), and a large amount of work has been devoted to designing approximation algorithms for CFLP and its variants. Following this vein, we introduce a new variant of CFLP called capacitated uniform facility location problem with soft penalties (CUFLPSP), in which the demand of each client can be partially rejected by paying penalty costs. As a result, we present a linear programming-rounding (LP-rounding) based 5.5122-approximation algorithm for the CUFLPSP.
{"title":"LP-Rounding Based Algorithm for Capacitated Uniform Facility Location Problem with Soft Penalties","authors":"Runjie Miao;Chenchen Wu;Jinjiang Yuan","doi":"10.26599/TST.2024.9010040","DOIUrl":"https://doi.org/10.26599/TST.2024.9010040","url":null,"abstract":"Capacitated facility location problem (CFLP) is a classical combinatorial optimization problem that has various applications in operations research, theoretical computer science, and management science. In the CFLP, we have a potential facilities set and a clients set. Each facility has a certain capacity and an open cost, and each client has a spliitable demand that need to be met. The goal is to open some facilities and assign all clients to these open facilities so that the total cost is as low as possible. The CFLP is NP-hard (non-deterministic polynomial-hard), and a large amount of work has been devoted to designing approximation algorithms for CFLP and its variants. Following this vein, we introduce a new variant of CFLP called capacitated uniform facility location problem with soft penalties (CUFLPSP), in which the demand of each client can be partially rejected by paying penalty costs. As a result, we present a linear programming-rounding (LP-rounding) based 5.5122-approximation algorithm for the CUFLPSP.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"279-289"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676352","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-11DOI: 10.26599/TST.2024.9010037
Liangkai Liu;Yanzhi Wang;Weisong Shi
Predictability is an essential challenge for autonomous vehicles (AVs)‘ safety. Deep neural networks have been widely deployed in the AV's perception pipeline. However, it is still an open question on how to guarantee the perception predictability for AV because there are millions of deep neural networks (DNNs) model combinations and system configurations when deploying DNNs in AVs. This paper proposes configurable predictability testbed (CPT), a configurable testbed for quantifying the predictability in AV's perception pipeline. CPT provides flexible configurations of the perception pipeline on data, DNN models, fusion policy, scheduling policies, and predictability metrics. On top of CPT, the researchers can profile and optimize the predictability issue caused by different application and system configurations. CPT has been open-sourced at: https://github.com/Torreskai0722/CPT.
可预测性是自动驾驶汽车（AV）安全性面临的一项重要挑战。深度神经网络已被广泛应用于自动驾驶汽车的感知管道。然而，由于在 AV 中部署深度神经网络时存在数百万种深度神经网络（DNN）模型组合和系统配置，因此如何保证 AV 的感知可预测性仍是一个未决问题。本文提出了可配置可预测性测试平台（CPT），这是一种用于量化 AV 感知管道可预测性的可配置测试平台。CPT 对感知管道的数据、DNN 模型、融合策略、调度策略和可预测性指标进行了灵活配置。在 CPT 的基础上，研究人员可以剖析和优化由不同应用和系统配置引起的可预测性问题。CPT 已开源：https://github.com/Torreskai0722/CPT。
{"title":"CPT: A Configurable Predictability Testbed for DNN Inference in AVs","authors":"Liangkai Liu;Yanzhi Wang;Weisong Shi","doi":"10.26599/TST.2024.9010037","DOIUrl":"https://doi.org/10.26599/TST.2024.9010037","url":null,"abstract":"Predictability is an essential challenge for autonomous vehicles (AVs)‘ safety. Deep neural networks have been widely deployed in the AV's perception pipeline. However, it is still an open question on how to guarantee the perception predictability for AV because there are millions of deep neural networks (DNNs) model combinations and system configurations when deploying DNNs in AVs. This paper proposes configurable predictability testbed (CPT), a configurable testbed for quantifying the predictability in AV's perception pipeline. CPT provides flexible configurations of the perception pipeline on data, DNN models, fusion policy, scheduling policies, and predictability metrics. On top of CPT, the researchers can profile and optimize the predictability issue caused by different application and system configurations. CPT has been open-sourced at: https://github.com/Torreskai0722/CPT.","PeriodicalId":48690,"journal":{"name":"Tsinghua Science and Technology","volume":"30 1","pages":"87-99"},"PeriodicalIF":6.6,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10676407","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169645","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：