Pub Date : 2024-08-27DOI: 10.1109/JSYST.2024.3440472
Xihe Qiu;Haoyu Wang;Xiaoyu Tan
Accurately understanding intentions is crucial in various real-world multiagent scenarios, which helps comprehend motives and predict actions within these contexts. Existing methods tend to either concentrate too much on single agents' isolated characteristics or model complex interactions among multiple agents, failing to adequately address both aspects simultaneously. To address this challenge, we propose a novel framework called integrative multiagent behavior prediction framework to systematically incorporate individual features and interagent relational dynamics. Our approach not only models multiagent interactions by learning from visual data, but also integrates mining of imagery and videos to leverage intrinsic invariant and variant qualities of each agent's extrinsic morphology. Meanwhile, inspired by time -series forecasting, we represent interagent history and connections as seasonal and trend features in time-series patterns, capturing past behavioral influences that are often ignored. We also design an encoder that efficiently learns time-dependencies and concatenates individual invariant–variant feature learning modules with multiagent interaction representations to accurately infer intentions and trajectory predictions. Our approach not only models multiagent interactions by learning from visual data, but also integrates mining of imagery and videos to leverage intrinsic invariant qualities of each agent's extrinsic morphology (e.g., body shape, color) and variant qualities (e.g., pose, expression, attire). Extensive experiments demonstrate that, compared to current state-of-the-art intention analysis models, our framework improves behavioral prediction performance in multiagent environments.
{"title":"Inferring Intents From Equivariant–Invariant Representations and Relational Learning in Multiagent Systems","authors":"Xihe Qiu;Haoyu Wang;Xiaoyu Tan","doi":"10.1109/JSYST.2024.3440472","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3440472","url":null,"abstract":"Accurately understanding intentions is crucial in various real-world multiagent scenarios, which helps comprehend motives and predict actions within these contexts. Existing methods tend to either concentrate too much on single agents' isolated characteristics or model complex interactions among multiple agents, failing to adequately address both aspects simultaneously. To address this challenge, we propose a novel framework called integrative multiagent behavior prediction framework to systematically incorporate individual features and interagent relational dynamics. Our approach not only models multiagent interactions by learning from visual data, but also integrates mining of imagery and videos to leverage intrinsic invariant and variant qualities of each agent's extrinsic morphology. Meanwhile, inspired by time -series forecasting, we represent interagent history and connections as seasonal and trend features in time-series patterns, capturing past behavioral influences that are often ignored. We also design an encoder that efficiently learns time-dependencies and concatenates individual invariant–variant feature learning modules with multiagent interaction representations to accurately infer intentions and trajectory predictions. Our approach not only models multiagent interactions by learning from visual data, but also integrates mining of imagery and videos to leverage intrinsic invariant qualities of each agent's extrinsic morphology (e.g., body shape, color) and variant qualities (e.g., pose, expression, attire). Extensive experiments demonstrate that, compared to current state-of-the-art intention analysis models, our framework improves behavioral prediction performance in multiagent environments.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1765-1775"},"PeriodicalIF":4.0,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169666","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-26DOI: 10.1109/JSYST.2024.3442017
Deemah H. Tashman;Soumaya Cherkaoui;Walaa Hamouda
This article presents a reinforcement learning-based approach to improve the physical layer security of an underlay cognitive radio network over cascaded channels. These channels are utilized in highly mobile networks such as cognitive vehicular networks (CVN). In addition, an eavesdropper aims to intercept the communications between secondary users (SUs). The SU receiver has full-duplex and energy harvesting capabilities to generate jamming signals to confound the eavesdropper and enhance security. Moreover, the SU transmitter extracts energy from ambient radio frequency signals in order to power subsequent transmissions to its intended receiver. To optimize the privacy and reliability of the SUs in a CVN, a deep Q-network (DQN) strategy is utilized where multiple DQN agents are required such that an agent is assigned at each SU transmitter. The objective for the SUs is to determine the optimal transmission power and decide whether to collect energy or transmit messages during each time period in order to maximize their secrecy rate. Thereafter, we propose a DQN approach to maximize the throughput of the SUs while respecting the interference threshold acceptable at the receiver of the primary user. According to our findings, our strategy outperforms two other baseline strategies in terms of security and reliability.
{"title":"Optimizing Cognitive Networks: Reinforcement Learning Meets Energy Harvesting Over Cascaded Channels","authors":"Deemah H. Tashman;Soumaya Cherkaoui;Walaa Hamouda","doi":"10.1109/JSYST.2024.3442017","DOIUrl":"10.1109/JSYST.2024.3442017","url":null,"abstract":"This article presents a reinforcement learning-based approach to improve the physical layer security of an underlay cognitive radio network over cascaded channels. These channels are utilized in highly mobile networks such as cognitive vehicular networks (CVN). In addition, an eavesdropper aims to intercept the communications between secondary users (SUs). The SU receiver has full-duplex and energy harvesting capabilities to generate jamming signals to confound the eavesdropper and enhance security. Moreover, the SU transmitter extracts energy from ambient radio frequency signals in order to power subsequent transmissions to its intended receiver. To optimize the privacy and reliability of the SUs in a CVN, a deep Q-network (DQN) strategy is utilized where multiple DQN agents are required such that an agent is assigned at each SU transmitter. The objective for the SUs is to determine the optimal transmission power and decide whether to collect energy or transmit messages during each time period in order to maximize their secrecy rate. Thereafter, we propose a DQN approach to maximize the throughput of the SUs while respecting the interference threshold acceptable at the receiver of the primary user. According to our findings, our strategy outperforms two other baseline strategies in terms of security and reliability.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 4","pages":"1839-1848"},"PeriodicalIF":4.0,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-26DOI: 10.1109/JSYST.2024.3443332
Yongsheng Chen;Shi Li;Yujing Yan;Guobao Liu
In this letter, a distributed periodic event-triggered (PET) consensus protocol is proposed for nonlinear multiagent systems under directed communication topology switchings. Only at the sampling instants, the information exchanges and the modes of topologies are used. Then, a distributed PET controller with a discrete event-triggering mechanism is proposed to reduce the burden of communication. By deriving a novel average dwell time constraint and constructing multiple Lyapunov functions, the leader–following consensus objective can be achieved. Finally, the proposed method's effectiveness is demonstrated by simulation results.
本文针对定向通信拓扑切换下的非线性多代理系统提出了一种分布式周期事件触发（PET）共识协议。该协议仅在采样时刻使用信息交换和拓扑模式。然后，提出了一种具有离散事件触发机制的分布式 PET 控制器，以减轻通信负担。通过推导新的平均停留时间约束条件和构建多个 Lyapunov 函数，可以实现领导-跟随共识目标。最后，模拟结果证明了所提方法的有效性。
{"title":"Distributed Periodic Event-Triggered Consensus of Second-Order Nonlinear Multiagent Systems With Switching Topologies","authors":"Yongsheng Chen;Shi Li;Yujing Yan;Guobao Liu","doi":"10.1109/JSYST.2024.3443332","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3443332","url":null,"abstract":"In this letter, a distributed periodic event-triggered (PET) consensus protocol is proposed for nonlinear multiagent systems under directed communication topology switchings. Only at the sampling instants, the information exchanges and the modes of topologies are used. Then, a distributed PET controller with a discrete event-triggering mechanism is proposed to reduce the burden of communication. By deriving a novel average dwell time constraint and constructing multiple Lyapunov functions, the leader–following consensus objective can be achieved. Finally, the proposed method's effectiveness is demonstrated by simulation results.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1821-1824"},"PeriodicalIF":4.0,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169714","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-23DOI: 10.1109/JSYST.2024.3437232
Han Deng;Dengfeng Xia;Han Cai;Qifa Yan;Peng Xu;Bin Dai
In this article, the relay network (RN) with receiver-transmitter feedback is investigated. First, we propose an efficient feedback coding scheme for the additive white Gaussian noise (AWGN) RN with noiseless receiver-transmitter feedback, which generalizes the Schalkwijk–Kailath (SK) type scheme for the AWGN channel with a single relay and noiseless receiver-transmitter feedback. The corresponding achievable rate of our proposed scheme is obtained under fixed coding block length and the receiver's decoding error probability, and it is shown that channel feedback significantly enhances the achievable rate of the same model without feedback. Then, we extend the above scheme to the same model with an AWGN feedback channel, where a modulo lattice function is applied to mitigate the impact of the feedback channel noise on the performance of the SK-type scheme. Finally, we further extend our proposed scheme to the quasi-static Rayleigh fading RN by using a precoding strategy. The results of this article are further explained via numerical examples, and this article provides a new method to construct low complexity coding schemes for relay networks.
{"title":"An Efficient Coding Scheme for the AWGN Relay Network With Receiver–Transmitter Feedback","authors":"Han Deng;Dengfeng Xia;Han Cai;Qifa Yan;Peng Xu;Bin Dai","doi":"10.1109/JSYST.2024.3437232","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3437232","url":null,"abstract":"In this article, the relay network (RN) with receiver-transmitter feedback is investigated. First, we propose an efficient feedback coding scheme for the additive white Gaussian noise (AWGN) RN with noiseless receiver-transmitter feedback, which generalizes the Schalkwijk–Kailath (SK) type scheme for the AWGN channel with a single relay and noiseless receiver-transmitter feedback. The corresponding achievable rate of our proposed scheme is obtained under fixed coding block length and the receiver's decoding error probability, and it is shown that channel feedback significantly enhances the achievable rate of the same model without feedback. Then, we extend the above scheme to the same model with an AWGN feedback channel, where a modulo lattice function is applied to mitigate the impact of the feedback channel noise on the performance of the SK-type scheme. Finally, we further extend our proposed scheme to the quasi-static Rayleigh fading RN by using a precoding strategy. The results of this article are further explained via numerical examples, and this article provides a new method to construct low complexity coding schemes for relay networks.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1717-1728"},"PeriodicalIF":4.0,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Edge caching is expected to alleviate the traffic consumption in next-generation communications. In this article, we consider the transmission delay in wideband communications deteriorated by rapid user movements, where the frequency-selective wideband fading channels become fast time-varying and hence doubly-selective due to the user movements. To preferably allocate the caching resource in such circumstance, we introduce a coordinated caching network and accordingly formulate an allocation problem. However, the formulated problem is shown to be NP-hard. By considering the extremely high computational complexity to solve the NP-hard problem by traditional optimization algorithm, and considering only a few samples can be obtained for each training instance due to shortened coherence-time in the dynamical doubly selective fading channels, we propose a model-agnostic meta-reinforcement learning method to address the formulated problem. Particularly, the proposed method can efficiently recognize the unstable mobile channels and accordingly cache to reduce the overall transmission delay while only requires a few training samples. Numerical simulations are performed to verify the effectiveness of the proposed method and results show that the proposed one outperforms the commonly adopted existing method of deep-deterministic-policy-gradient learning in terms of average delay and cache hit rate.
{"title":"Caching for Doubly Selective Fading Channels via Model-Agnostic Meta-Reinforcement Learning","authors":"Weibao He;Fasheng Zhou;Dong Tang;Fang Fang;Wei Chen","doi":"10.1109/JSYST.2024.3442958","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3442958","url":null,"abstract":"Edge caching is expected to alleviate the traffic consumption in next-generation communications. In this article, we consider the transmission delay in wideband communications deteriorated by rapid user movements, where the frequency-selective wideband fading channels become fast time-varying and hence \u0000<italic>doubly-selective</i>\u0000 due to the user movements. To preferably allocate the caching resource in such circumstance, we introduce a coordinated caching network and accordingly formulate an allocation problem. However, the formulated problem is shown to be NP-hard. By considering the extremely high computational complexity to solve the NP-hard problem by traditional optimization algorithm, and considering only a few samples can be obtained for each training instance due to shortened coherence-time in the dynamical doubly selective fading channels, we propose a model-agnostic meta-reinforcement learning method to address the formulated problem. Particularly, the proposed method can efficiently recognize the unstable mobile channels and accordingly cache to reduce the overall transmission delay while only requires a few training samples. Numerical simulations are performed to verify the effectiveness of the proposed method and results show that the proposed one outperforms the commonly adopted existing method of deep-deterministic-policy-gradient learning in terms of average delay and cache hit rate.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1776-1785"},"PeriodicalIF":4.0,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169716","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-21DOI: 10.1109/JSYST.2024.3437769
Siyi Li;Xiaojie Fang;Heng Dong;Wei Wu;Fugang Liu;Zhuoming Li
The integration of communication technologies is regarded as a potential direction of future network. Combining cell-free massive multiple-input multiple-output (CF-mMIMO) with nonorthogonal multiple access (NOMA) can achieve substantial access and uniform coverage, but it also poses challenges on channel state information (CSI) acquisition. Although systems operating in time division duplex can approximate uplink CSI for downlink precoding design, the performance degradation resulting from nonreciprocity is still nonnegligible. This article proposes a calibration method to improve system sum rate of CF-mMIMO-NOMA with nonreciprocal channels. The proposed method exploits the impact of all access points (APs) on achievable rate to provide a prioritization and calibrates the selected APs orderly, finding an optimal compromise between pilot overhead and performance gain. The order and number of calibrated APs can be adjusted according to the parameters including coherence interval, number of users, etc. Based on the calibration method, a power allocation algorithm focusing on maximizing the minimum rate is presented. The nonconvex initial model is transformed into a standard second-order cone programming problem through slack variables for bisection solution. Simulation results validate the importance of calibration and indicate that the proposed method provides enhanced achievable rate with flexibility and adaptability.
{"title":"Cell-Free Massive MIMO-NOMA Systems With Nonreciprocal Channels: Performance Analysis and Calibration Method Design","authors":"Siyi Li;Xiaojie Fang;Heng Dong;Wei Wu;Fugang Liu;Zhuoming Li","doi":"10.1109/JSYST.2024.3437769","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3437769","url":null,"abstract":"The integration of communication technologies is regarded as a potential direction of future network. Combining cell-free massive multiple-input multiple-output (CF-mMIMO) with nonorthogonal multiple access (NOMA) can achieve substantial access and uniform coverage, but it also poses challenges on channel state information (CSI) acquisition. Although systems operating in time division duplex can approximate uplink CSI for downlink precoding design, the performance degradation resulting from nonreciprocity is still nonnegligible. This article proposes a calibration method to improve system sum rate of CF-mMIMO-NOMA with nonreciprocal channels. The proposed method exploits the impact of all access points (APs) on achievable rate to provide a prioritization and calibrates the selected APs orderly, finding an optimal compromise between pilot overhead and performance gain. The order and number of calibrated APs can be adjusted according to the parameters including coherence interval, number of users, etc. Based on the calibration method, a power allocation algorithm focusing on maximizing the minimum rate is presented. The nonconvex initial model is transformed into a standard second-order cone programming problem through slack variables for bisection solution. Simulation results validate the importance of calibration and indicate that the proposed method provides enhanced achievable rate with flexibility and adaptability.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1729-1740"},"PeriodicalIF":4.0,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169673","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-19DOI: 10.1109/JSYST.2024.3426541
Joana L. F. P. Cardoso;Paul T. Grogan;Michael J. Pennock
Reuse of cyber-physical modules is critical to accelerate the development times and reduce design costs of cyber-physical systems (CPS). Successful module reuse depends on the ability to mitigate interoperability problems. Over the years, different communities have worked to understand and mitigate these problems leading to different interoperability models, oftentimes tied to particular domains. One important question, tackled in this article, is whether these models can help with the design of CPS. Specifically, this article evaluates a real-world example of a CPS against the levels of conceptual interoperability model (LCIM). The approach followed leverages natural language processing techniques to narrow an initial set of system changes down to a subset of problems, some of which are interoperability problems. This effort enables a comparison between the empirical evidence found and the interoperability model considered. The results obtained show that the observed interoperability problems fit well under the multilevel categories proposed by the LCIM. Furthermore, the LCIM seems to provide a complete description for the interoperability problems found, checking for fit and completeness. A significant impact of the results obtained is the enabling of more focused discussions of the key attributes of said problems and subsequent identification and assessment of the effectiveness of existing mitigation approaches.
{"title":"Classifying Interoperability Problems in Cyber-Physical Systems: Empirical Cases From OpenWrt","authors":"Joana L. F. P. Cardoso;Paul T. Grogan;Michael J. Pennock","doi":"10.1109/JSYST.2024.3426541","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3426541","url":null,"abstract":"Reuse of cyber-physical modules is critical to accelerate the development times and reduce design costs of cyber-physical systems (CPS). Successful module reuse depends on the ability to mitigate interoperability problems. Over the years, different communities have worked to understand and mitigate these problems leading to different interoperability models, oftentimes tied to particular domains. One important question, tackled in this article, is whether these models can help with the design of CPS. Specifically, this article evaluates a real-world example of a CPS against the levels of conceptual interoperability model (LCIM). The approach followed leverages natural language processing techniques to narrow an initial set of system changes down to a subset of problems, some of which are interoperability problems. This effort enables a comparison between the empirical evidence found and the interoperability model considered. The results obtained show that the observed interoperability problems fit well under the multilevel categories proposed by the LCIM. Furthermore, the LCIM seems to provide a complete description for the interoperability problems found, checking for fit and completeness. A significant impact of the results obtained is the enabling of more focused discussions of the key attributes of said problems and subsequent identification and assessment of the effectiveness of existing mitigation approaches.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1658-1668"},"PeriodicalIF":4.0,"publicationDate":"2024-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169654","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-15DOI: 10.1109/JSYST.2024.3435687
Tianming Feng;Chenyu Wu;Xiaodong Zheng;Peilin Chen;Yilong Liu;Shuai Han
This article provides a fine-grained performance analysis of a multiantenna cache-enabled network with interference nulling (IN) employed at base stations. Two IN schemes, namely, the fixed IN scheme and the flexible IN scheme are considered to improve the received signal-to-interference ratio (SIR) at users. To thoroughly explore the effects of the caching parameter and the IN parameters on the network performance, we focus on the analysis of not only the successful transmission probability (STP) but the link reliability distribution. First, the expression for the STP of each IN scheme is derived. Then, for the link reliability analysis of each IN scheme, an approximation for the SIR metadistribution is obtained by deriving the first and second moments of an upper bound of the link reliability and utilizing the beta distribution, and an approximated expression for the variance of link reliability is also derived. With this analytical framework, we compare the performance of these two IN schemes and gain some useful system design guidelines from the perspectives of STP and the link reliability distribution.
{"title":"Fine-Grained Analysis of Cache-Enabled Multiantenna Networks With Interference Nulling","authors":"Tianming Feng;Chenyu Wu;Xiaodong Zheng;Peilin Chen;Yilong Liu;Shuai Han","doi":"10.1109/JSYST.2024.3435687","DOIUrl":"10.1109/JSYST.2024.3435687","url":null,"abstract":"This article provides a fine-grained performance analysis of a multiantenna cache-enabled network with interference nulling (IN) employed at base stations. Two IN schemes, namely, the fixed IN scheme and the flexible IN scheme are considered to improve the received signal-to-interference ratio (SIR) at users. To thoroughly explore the effects of the caching parameter and the IN parameters on the network performance, we focus on the analysis of not only the successful transmission probability (STP) but the link reliability distribution. First, the expression for the STP of each IN scheme is derived. Then, for the link reliability analysis of each IN scheme, an approximation for the SIR metadistribution is obtained by deriving the first and second moments of an upper bound of the link reliability and utilizing the beta distribution, and an approximated expression for the variance of link reliability is also derived. With this analytical framework, we compare the performance of these two IN schemes and gain some useful system design guidelines from the perspectives of STP and the link reliability distribution.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 4","pages":"2097-2108"},"PeriodicalIF":4.0,"publicationDate":"2024-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142183283","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-15DOI: 10.1109/JSYST.2024.3439746
Peng Qin;Yang Fu;Rui Ding;Haoting He
Vehicular edge computing is poised to meet the requirements of emerging applications in Internet of Vehicles (IoV) by offloading computation tasks from resource-limited vehicles to edge. However, the space-time-dynamic offloading demands of vehicle users (VUs) can hardly be satisfied only by road side units (RSUs) due to their fixed resource deployment and incomplete coverage. To this end, in this article, we design a multitier IoV system, where RSU, parked cars, and unmanned aerial vehicles (UAVs) serve as edge platforms to offer computing services. To fully utilize the multitier resources, the tasks generated by VUs can be split into multiple parts and executed in parallel on local processors as well as edge servers. Under this arrangement, we formulate a joint UAV deployment and partial task offloading problem to minimize the system cost, which includes processing delay, energy consumption, and rental price. We then develop a heuristic UAV deployment method to optimize the coverage of multitier network. Moreover, a distributed task offloading approach based on multiagent deep reinforcement learning is proposed to achieve cooperative decision makings and load balancing, thereby overcoming the adversarial competition among VUs. Experimental evaluations reveal that compared to state-of-the-art schemes that rely on a centralized controller, the proposed approach achieves superior performance with higher implementation efficiency while avoiding extra information exchange overhead.
通过将计算任务从资源有限的车辆卸载到边缘，车载边缘计算有望满足车联网（IoV）新兴应用的要求。然而，由于路侧单元（RSU）的固定资源部署和不完全覆盖，仅靠路侧单元很难满足车辆用户（VU）的时空动态卸载需求。为此，本文设计了一种多层物联网系统，将 RSU、停放的汽车和无人机（UAV）作为提供计算服务的边缘平台。为了充分利用多层资源，可以将 VU 生成的任务分成多个部分，在本地处理器和边缘服务器上并行执行。在这种安排下，我们提出了一个联合无人机部署和部分任务卸载问题，以最小化系统成本，其中包括处理延迟、能耗和租赁价格。然后，我们开发了一种启发式无人机部署方法，以优化多层网络的覆盖范围。此外，我们还提出了一种基于多代理深度强化学习的分布式任务卸载方法，以实现合作决策和负载平衡，从而克服无人机之间的对抗性竞争。实验评估表明，与依赖于集中控制器的最先进方案相比，所提出的方法在避免额外信息交换开销的同时，以更高的实施效率实现了更优越的性能。
{"title":"Competition-Awareness Partial Task Offloading and UAV Deployment for Multitier Parallel Computational Internet of Vehicles","authors":"Peng Qin;Yang Fu;Rui Ding;Haoting He","doi":"10.1109/JSYST.2024.3439746","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3439746","url":null,"abstract":"Vehicular edge computing is poised to meet the requirements of emerging applications in Internet of Vehicles (IoV) by offloading computation tasks from resource-limited vehicles to edge. However, the space-time-dynamic offloading demands of vehicle users (VUs) can hardly be satisfied only by road side units (RSUs) due to their fixed resource deployment and incomplete coverage. To this end, in this article, we design a multitier IoV system, where RSU, parked cars, and unmanned aerial vehicles (UAVs) serve as edge platforms to offer computing services. To fully utilize the multitier resources, the tasks generated by VUs can be split into multiple parts and executed in parallel on local processors as well as edge servers. Under this arrangement, we formulate a joint UAV deployment and partial task offloading problem to minimize the system cost, which includes processing delay, energy consumption, and rental price. We then develop a heuristic UAV deployment method to optimize the coverage of multitier network. Moreover, a distributed task offloading approach based on multiagent deep reinforcement learning is proposed to achieve cooperative decision makings and load balancing, thereby overcoming the adversarial competition among VUs. Experimental evaluations reveal that compared to state-of-the-art schemes that rely on a centralized controller, the proposed approach achieves superior performance with higher implementation efficiency while avoiding extra information exchange overhead.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1753-1764"},"PeriodicalIF":4.0,"publicationDate":"2024-08-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169717","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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 : 2024-08-14DOI: 10.1109/JSYST.2024.3437185
Ying Zhang;Wenyuan Zhang;Xiaoyu Jiang;Yuzhong Sun;Baiming Feng;Naixue Xiong;Tianyu Wo
As a pivotal component of Industry 4.0, the Industrial Internet of Things has significantly propelled the intelligent evolution of industrial systems. However, this advancement has led to increased system complexity and scale, consequently increasing the likelihood of operational failures and potential security threats. Performing an effective analysis of log information and accurately identifying system fault categories has become a substantial challenge for system administrators. To extract valuable insights from edge device logs more efficiently and ensure system security, we propose an intelligent method for system fault detection and localization. Our approach begins with an analysis of the system's source code to extract message and fault classification templates. Subsequently, real-time preprocessing of the log stream occurs, employing techniques, such as pattern matching and statistical grouping, to construct a feature vector–matrix. The detection and identification module then discerns abnormal feature vectors, using a fast classification algorithm to categorize these anomalies and determine fault types. The proposed methodology undergoes testing on our edge cloud platform. The experimental results demonstrate that the method achieves a fault detection and localization accuracy that exceeds 98%.
{"title":"An Intelligent Secure Fault Classification and Identification Scheme for Mining Valuable Information in IIoT","authors":"Ying Zhang;Wenyuan Zhang;Xiaoyu Jiang;Yuzhong Sun;Baiming Feng;Naixue Xiong;Tianyu Wo","doi":"10.1109/JSYST.2024.3437185","DOIUrl":"https://doi.org/10.1109/JSYST.2024.3437185","url":null,"abstract":"As a pivotal component of Industry 4.0, the Industrial Internet of Things has significantly propelled the intelligent evolution of industrial systems. However, this advancement has led to increased system complexity and scale, consequently increasing the likelihood of operational failures and potential security threats. Performing an effective analysis of log information and accurately identifying system fault categories has become a substantial challenge for system administrators. To extract valuable insights from edge device logs more efficiently and ensure system security, we propose an intelligent method for system fault detection and localization. Our approach begins with an analysis of the system's source code to extract message and fault classification templates. Subsequently, real-time preprocessing of the log stream occurs, employing techniques, such as pattern matching and statistical grouping, to construct a feature vector–matrix. The detection and identification module then discerns abnormal feature vectors, using a fast classification algorithm to categorize these anomalies and determine fault types. The proposed methodology undergoes testing on our edge cloud platform. The experimental results demonstrate that the method achieves a fault detection and localization accuracy that exceeds 98%.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1705-1716"},"PeriodicalIF":4.0,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142169660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"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：