Pub Date : 2024-08-14DOI: 10.1109/TNSE.2024.3440930
Xinliang Wei;Lei Fan;Yuanxiong Guo;Yanmin Gong;Zhu Han;Yu Wang
Scheduling multiple federated learning (FL) models within a distributed network, especially in large-scale scenarios, poses significant challenges since it involves solving NP-hard mixed-integer nonlinear programming (MINLP) problems. However, it's imperative to optimize participant selection and learning rate determination for these FL models to avoid excessive training costs and prevent resource contention. While some existing methods focus solely on optimizing a single global FL model, others struggle to achieve optimal solutions as the problem grows more complex. In this paper, exploiting the potential of quantum computing, we introduce the Hybrid Quantum-Classical Benders' Decomposition (HQCBD) algorithm to effectively tackle the joint MINLP optimization problem for multi-model FL training. HQCBD combines quantum and classical computing to solve the joint participant selection and learning scheduling problem. It decomposes the optimization problem into a master problem with binary variables and small subproblems with continuous variables, then leverages the strengths of both quantum and classical computing to solve them respectively and iteratively. Furthermore, we propose the Hybrid Quantum-Classical Multiple-cuts Benders' Decomposition (MBD) algorithm, which utilizes the inherent capabilities of quantum algorithms to produce multiple cuts in each round, to speed up the proposed HQCBD algorithm. Extensive simulation on the commercial quantum annealing machine demonstrates the effectiveness and robustness of the proposed methods (both HQCBD and MBD), with improvements of up to 70.3% in iterations and 81% in computation time over the classical Benders' decomposition algorithm on classical CPUs, even at modest scales.
{"title":"Hybrid Quantum–Classical Benders' Decomposition for Federated Learning Scheduling in Distributed Networks","authors":"Xinliang Wei;Lei Fan;Yuanxiong Guo;Yanmin Gong;Zhu Han;Yu Wang","doi":"10.1109/TNSE.2024.3440930","DOIUrl":"10.1109/TNSE.2024.3440930","url":null,"abstract":"Scheduling multiple federated learning (FL) models within a distributed network, especially in large-scale scenarios, poses significant challenges since it involves solving NP-hard mixed-integer nonlinear programming (MINLP) problems. However, it's imperative to optimize participant selection and learning rate determination for these FL models to avoid excessive training costs and prevent resource contention. While some existing methods focus solely on optimizing a single global FL model, others struggle to achieve optimal solutions as the problem grows more complex. In this paper, exploiting the potential of quantum computing, we introduce the Hybrid Quantum-Classical Benders' Decomposition (HQCBD) algorithm to effectively tackle the joint MINLP optimization problem for multi-model FL training. HQCBD combines quantum and classical computing to solve the joint participant selection and learning scheduling problem. It decomposes the optimization problem into a master problem with binary variables and small subproblems with continuous variables, then leverages the strengths of both quantum and classical computing to solve them respectively and iteratively. Furthermore, we propose the Hybrid Quantum-Classical Multiple-cuts Benders' Decomposition (MBD) algorithm, which utilizes the inherent capabilities of quantum algorithms to produce multiple cuts in each round, to speed up the proposed HQCBD algorithm. Extensive simulation on the commercial quantum annealing machine demonstrates the effectiveness and robustness of the proposed methods (both HQCBD and MBD), with improvements of up to 70.3% in iterations and 81% in computation time over the classical Benders' decomposition algorithm on classical CPUs, even at modest scales.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6038-6051"},"PeriodicalIF":6.7,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191800","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-08DOI: 10.1109/TNSE.2024.3436616
Kai Peng;Jintao He;Jialu Guo;Yuan Liu;Jianwen He;Wei Liu;Menglan Hu
Microservices have exerted a profound impact on the development of internet applications. Meanwhile, the growing number of mobile terminal user requests has made the communication between microservices extremely complex, significantly impacting the quality of user service experience in mobile edge computing. Therefore, the joint optimization of microservice deployment and request routing is necessary to alleviate server pressure and enhance overall performance of large-scaled MEC applications. However, most existing work studies the microservice deployment and request routing as two isolated problems and neglects the dependencies between microservices. This paper focuses on the data dependency relationship of request and multi-instance processing problem, and then formulate the joint problem of microservice deployment and request routing as an integer nonlinear program and queuing optimization model under complex constraints. To address this problem, we propose a fine-grained reinforcement learning-based algorithm named Reward Memory Shaping Deep Deterministic Policy Gradient (RMS $_$ DDPG). Furthermore, we introduce the Long Short-Term Memory (LSTM) block into the actor network and critical network to make actions memorable. Finally, our experiments demonstrate that our algorithm is more superior in terms of delay target, load balancing and algorithm robustness compared with four baseline algorithms.
{"title":"Delay-Aware Optimization of Fine-Grained Microservice Deployment and Routing in Edge via Reinforcement Learning","authors":"Kai Peng;Jintao He;Jialu Guo;Yuan Liu;Jianwen He;Wei Liu;Menglan Hu","doi":"10.1109/TNSE.2024.3436616","DOIUrl":"10.1109/TNSE.2024.3436616","url":null,"abstract":"Microservices have exerted a profound impact on the development of internet applications. Meanwhile, the growing number of mobile terminal user requests has made the communication between microservices extremely complex, significantly impacting the quality of user service experience in mobile edge computing. Therefore, the joint optimization of microservice deployment and request routing is necessary to alleviate server pressure and enhance overall performance of large-scaled MEC applications. However, most existing work studies the microservice deployment and request routing as two isolated problems and neglects the dependencies between microservices. This paper focuses on the data dependency relationship of request and multi-instance processing problem, and then formulate the joint problem of microservice deployment and request routing as an integer nonlinear program and queuing optimization model under complex constraints. To address this problem, we propose a fine-grained reinforcement learning-based algorithm named Reward Memory Shaping Deep Deterministic Policy Gradient (RMS \u0000<inline-formula><tex-math>$_$</tex-math></inline-formula>\u0000 DDPG). Furthermore, we introduce the Long Short-Term Memory (LSTM) block into the actor network and critical network to make actions memorable. Finally, our experiments demonstrate that our algorithm is more superior in terms of delay target, load balancing and algorithm robustness compared with four baseline algorithms.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6024-6037"},"PeriodicalIF":6.7,"publicationDate":"2024-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141934722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-07DOI: 10.1109/tnse.2024.3439744
Mohammadreza Doostmohammadian, Zulfiya R. Gabidullina, Hamid R. Rabiee
{"title":"Nonlinear Perturbation-based Non-Convex Optimization over Time-Varying Networks","authors":"Mohammadreza Doostmohammadian, Zulfiya R. Gabidullina, Hamid R. Rabiee","doi":"10.1109/tnse.2024.3439744","DOIUrl":"https://doi.org/10.1109/tnse.2024.3439744","url":null,"abstract":"","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"25 1","pages":""},"PeriodicalIF":6.6,"publicationDate":"2024-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141934723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-06DOI: 10.1109/tnse.2024.3438846
Liezhuo Zhang, Xianwei Lv, Chen Yu, Jiang Xiao, Kai Liu, Hai Jin
{"title":"UHA: An Intelligent Uncertainty Map Based Hierarchical Attention Network System for Building Segmentation","authors":"Liezhuo Zhang, Xianwei Lv, Chen Yu, Jiang Xiao, Kai Liu, Hai Jin","doi":"10.1109/tnse.2024.3438846","DOIUrl":"https://doi.org/10.1109/tnse.2024.3438846","url":null,"abstract":"","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"111 1","pages":""},"PeriodicalIF":6.6,"publicationDate":"2024-08-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141934724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-08-02DOI: 10.1109/TNSE.2024.3434643
Ruoli Zhao;Yong Xie;Debiao He;Kim-Kwang Raymond Choo;Zoe L. Jiang
Using Convolutional Neural Network (CNN) model to analyze monitoring data in Body Area Network (BAN) has become an important way to solve health related issues in the current large sub-health population and aging population. However, the inference and analysis process of BAN data needs to ensure efficiency and security. At present, ensuring a balance of efficiency and security in the inference of CCN models is challenging. Therefore, an efficient and secure CNN inference scheme is proposed based on two Edge-Cloud-Servers (CS $_{0}$ and CS $_{1}$ ). By analyzing the CNN model and combining two secret sharing semantics, we optimize the communication overhead of inference. Specifically, a new non-interactive secure convolutional layer computation protocol is designed to significantly reduce the number of interactions between CS $_{0}$ and CS $_{1}$ . For non-linear layers, we propose a simpler secure comparison computation functionality to reduce the communication overhead. Moreover, we also design some lightweight secure building blocks based on secret sharing to improve computing efficiency. We implement our proposed scheme on two standard datasets. Through the theoretical analysis and experimental comparison, our scheme improves the computational efficiency.
{"title":"Practical Privacy-Preserving Convolutional Neural Network Inference Framework With Edge Computing for Health Monitoring","authors":"Ruoli Zhao;Yong Xie;Debiao He;Kim-Kwang Raymond Choo;Zoe L. Jiang","doi":"10.1109/TNSE.2024.3434643","DOIUrl":"10.1109/TNSE.2024.3434643","url":null,"abstract":"Using Convolutional Neural Network (CNN) model to analyze monitoring data in Body Area Network (BAN) has become an important way to solve health related issues in the current large sub-health population and aging population. However, the inference and analysis process of BAN data needs to ensure efficiency and security. At present, ensuring a balance of efficiency and security in the inference of CCN models is challenging. Therefore, an efficient and secure CNN inference scheme is proposed based on two Edge-Cloud-Servers (CS\u0000<inline-formula><tex-math>$_{0}$</tex-math></inline-formula>\u0000 and CS\u0000<inline-formula><tex-math>$_{1}$</tex-math></inline-formula>\u0000). By analyzing the CNN model and combining two secret sharing semantics, we optimize the communication overhead of inference. Specifically, a new non-interactive secure convolutional layer computation protocol is designed to significantly reduce the number of interactions between CS\u0000<inline-formula><tex-math>$_{0}$</tex-math></inline-formula>\u0000 and CS\u0000<inline-formula><tex-math>$_{1}$</tex-math></inline-formula>\u0000. For non-linear layers, we propose a simpler secure comparison computation functionality to reduce the communication overhead. Moreover, we also design some lightweight secure building blocks based on secret sharing to improve computing efficiency. We implement our proposed scheme on two standard datasets. Through the theoretical analysis and experimental comparison, our scheme improves the computational efficiency.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"5995-6006"},"PeriodicalIF":6.7,"publicationDate":"2024-08-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141886914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-07-31DOI: 10.1109/tnse.2024.3436085
Shuai Sui, Dongyu Shen, Shaocheng Tong, C. L. Philip Chen
{"title":"State-Observer-Based Event-Trigger Bipartite Consensus Secure Control for NMASs with Deception Attacks","authors":"Shuai Sui, Dongyu Shen, Shaocheng Tong, C. L. Philip Chen","doi":"10.1109/tnse.2024.3436085","DOIUrl":"https://doi.org/10.1109/tnse.2024.3436085","url":null,"abstract":"","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"42 1","pages":""},"PeriodicalIF":6.6,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Based on both the kinematic and the dynamic models of Wheeled Mobile Robots (WMRs), an optimal circumnavigation controller around moving targets is proposed by integrating backstepping control with adaptive dynamic programming (ADP) techniques. Initially, the cooperative circumnavigation challenge at the kinematic level is converted into a tracking task for the desired relative velocity by establishing a relative velocity error model between the robot and the target. Then, a dynamic-level error model is formulated to characterize the positional and directional errors between the robot's trajectory and the trajectory derived from the kinematic analysis. The control input is designed through the integration of the backstepping control and ADP. Ultimately, the proposed control strategy is proven to ensure both closed-loop system stability and the minimization of the cost function through Lyapunov’s method. Simulation comparisons with traditional methods confirm both the feasibility and superiority of the proposed controller.
{"title":"Optimal Moving-Target Circumnavigation Control of Multiple Wheeled Mobile Robots Based on Adaptive Dynamic Programming","authors":"Yanhong Luo;Yannan Li;Jinliang Ding;Huaguang Zhang","doi":"10.1109/TNSE.2024.3434633","DOIUrl":"10.1109/TNSE.2024.3434633","url":null,"abstract":"Based on both the kinematic and the dynamic models of Wheeled Mobile Robots (WMRs), an optimal circumnavigation controller around moving targets is proposed by integrating backstepping control with adaptive dynamic programming (ADP) techniques. Initially, the cooperative circumnavigation challenge at the kinematic level is converted into a tracking task for the desired relative velocity by establishing a relative velocity error model between the robot and the target. Then, a dynamic-level error model is formulated to characterize the positional and directional errors between the robot's trajectory and the trajectory derived from the kinematic analysis. The control input is designed through the integration of the backstepping control and ADP. Ultimately, the proposed control strategy is proven to ensure both closed-loop system stability and the minimization of the cost function through Lyapunov’s method. Simulation comparisons with traditional methods confirm both the feasibility and superiority of the proposed controller.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 5","pages":"4679-4688"},"PeriodicalIF":6.7,"publicationDate":"2024-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866351","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
LowEarth orbit (LEO) satellite network can complement terrestrial networks for achieving global wireless coverage and improving delay-sensitive Internet services. This paper proposes an integrated satellite-terrestrial network (ISTN) architecture to provide ground users with seamless and reliable content delivery services. For optimal service provisioning in this architecture, we formulate an optimization model to maximize the network throughput by jointly optimizing content delivery policy, cache placement, and transmission power allocation. The resulting optimization model is a large-scale mixed-integer nonlinear program (MINLP) that is intractable for classical computer solvers. Inspired by quantum computing techniques, we propose a hybrid quantum-classical generalized Benders' decomposition (HQCGBD) algorithm to address this challenge. Specifically, we first exploit the generalized Benders' decomposition (GBD) to decompose the problem into a master problem and a subproblem and then leverage the state-of-the-art quantum annealer to solve the challenging master problem. Furthermore, a multi-cut strategy is designed in HQCGBD to accelerate the solution process by leveraging the quantum advantages in parallel computing. Simulation results demonstrate the superiority of the proposed HQCGBD algorithm and validate the effectiveness of the proposed cache-enabled ISTN architecture.
{"title":"Quantum-Assisted Joint Caching and Power Allocation for Integrated Satellite-Terrestrial Networks","authors":"Yu Zhang;Yanmin Gong;Lei Fan;Yu Wang;Zhu Han;Yuanxiong Guo","doi":"10.1109/TNSE.2024.3435444","DOIUrl":"10.1109/TNSE.2024.3435444","url":null,"abstract":"LowEarth orbit (LEO) satellite network can complement terrestrial networks for achieving global wireless coverage and improving delay-sensitive Internet services. This paper proposes an integrated satellite-terrestrial network (ISTN) architecture to provide ground users with seamless and reliable content delivery services. For optimal service provisioning in this architecture, we formulate an optimization model to maximize the network throughput by jointly optimizing content delivery policy, cache placement, and transmission power allocation. The resulting optimization model is a large-scale mixed-integer nonlinear program (MINLP) that is intractable for classical computer solvers. Inspired by quantum computing techniques, we propose a hybrid quantum-classical generalized Benders' decomposition (HQCGBD) algorithm to address this challenge. Specifically, we first exploit the generalized Benders' decomposition (GBD) to decompose the problem into a master problem and a subproblem and then leverage the state-of-the-art quantum annealer to solve the challenging master problem. Furthermore, a multi-cut strategy is designed in HQCGBD to accelerate the solution process by leveraging the quantum advantages in parallel computing. Simulation results demonstrate the superiority of the proposed HQCGBD algorithm and validate the effectiveness of the proposed cache-enabled ISTN architecture.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"5163-5174"},"PeriodicalIF":6.7,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866353","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
This paper delves into the synchronization of factional uncertain reaction-diffusion complex network. An adaptive scheme composed of time $t$ and space $x$ is utilized to handle unknown couplings. An output-strict passivity lemma is established by means of Green theorem, Kronecker product and the Lyapunov stability theorem. Different from classical synchronous approaches by constructing controllers, a criterion in terms of linear matrix inequality is built on the passivity lemma, Laplace transform and inverse transform to make the resultant closed-loop system be synchronization. Two examples are provided to validate the validity of the proposed methods.
{"title":"Synchronization of Fractional Reaction-Diffusion Complex Networks With Unknown Couplings","authors":"Mouquan Shen;Chen Wang;Qing-Guo Wang;Yonghui Sun;Guangdeng Zong","doi":"10.1109/TNSE.2024.3432997","DOIUrl":"10.1109/TNSE.2024.3432997","url":null,"abstract":"This paper delves into the synchronization of factional uncertain reaction-diffusion complex network. An adaptive scheme composed of time \u0000<inline-formula><tex-math>$t$</tex-math></inline-formula>\u0000 and space \u0000<inline-formula><tex-math>$x$</tex-math></inline-formula>\u0000 is utilized to handle unknown couplings. An output-strict passivity lemma is established by means of Green theorem, Kronecker product and the Lyapunov stability theorem. Different from classical synchronous approaches by constructing controllers, a criterion in terms of linear matrix inequality is built on the passivity lemma, Laplace transform and inverse transform to make the resultant closed-loop system be synchronization. Two examples are provided to validate the validity of the proposed methods.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 5","pages":"4503-4512"},"PeriodicalIF":6.7,"publicationDate":"2024-07-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141866354","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","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：