Pub Date : 2024-09-06DOI: 10.1109/TNSE.2024.3454459
Xhensilda Allka;Pau Ferrer-Cid;Jose M. Barcelo-Ordinas;Jorge Garcia-Vidal
Sensor networks play an essential role in today's air quality monitoring platforms. Nevertheless, sensors often malfunction, leading to data anomalies. In this paper, an unsupervised pattern-based attention recurrent autoencoder for anomaly detection (PARAAD) is proposed to detect and locate anomalies in a network of air quality sensors. The novelty of the proposal lies in the use of temporal patterns, i.e., blocks of data, instead of point values. By looking at temporal patterns and through an attention mechanism, the architecture captures data dependencies in the feature space and latent space, enhancing the model's ability to focus on the most relevant parts. Its performance is evaluated with two categories of anomalies, bias fault and drift anomalies, and compared with baseline models such as a feed-forward autoencoder and a transformer architecture, as well as with models not based on temporal patterns. The results show that PARAAD achieves anomalous sensor detection and localization rates higher than 80%, outperforming existing baseline models in air quality sensor networks for both bias and drift faults.
{"title":"Pattern-Based Attention Recurrent Autoencoder for Anomaly Detection in Air Quality Sensor Networks","authors":"Xhensilda Allka;Pau Ferrer-Cid;Jose M. Barcelo-Ordinas;Jorge Garcia-Vidal","doi":"10.1109/TNSE.2024.3454459","DOIUrl":"10.1109/TNSE.2024.3454459","url":null,"abstract":"Sensor networks play an essential role in today's air quality monitoring platforms. Nevertheless, sensors often malfunction, leading to data anomalies. In this paper, an unsupervised pattern-based attention recurrent autoencoder for anomaly detection (PARAAD) is proposed to detect and locate anomalies in a network of air quality sensors. The novelty of the proposal lies in the use of temporal patterns, i.e., blocks of data, instead of point values. By looking at temporal patterns and through an attention mechanism, the architecture captures data dependencies in the feature space and latent space, enhancing the model's ability to focus on the most relevant parts. Its performance is evaluated with two categories of anomalies, bias fault and drift anomalies, and compared with baseline models such as a feed-forward autoencoder and a transformer architecture, as well as with models not based on temporal patterns. The results show that PARAAD achieves anomalous sensor detection and localization rates higher than 80%, outperforming existing baseline models in air quality sensor networks for both bias and drift faults.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6372-6381"},"PeriodicalIF":6.7,"publicationDate":"2024-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191987","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 : 2024-09-03DOI: 10.1109/TNSE.2024.3453959
Licheng Chen;Yunquan Dong
In this paper, we estimate the average age of information (AoI) of the status updating over a wireless channel with an unknown fading model. Different from most related works which take the distributions of the inter-arrival time and transmission time of updates as known information, we approximate the average AoI of the system by using their first and second-order moments. Note that these distributions are often not accessible or known with inevitable errors while their moments are much easier to obtain, e.g., by using counting and statistics. We model the communications over the fading channel with a continuous transmission model and a discrete transmission model, which use the variable-rate scheme and the fixed-rate scheme, respectively. We assume that the arrival of the continuous transmission model is a Bernoulli process and make no assumptions about the arrival process of the discrete transmission model. Based on these information, we present two pairs of tight lower and upper bounds for the AoI of the two models. We show that obtained bounds are the tightest when the inter-arrival time (or transmission time) follows the degenerate distribution and are the loosest when it follows the two-point distribution, which randomly takes value from two possible outcomes. We also show that tighter bounds can be obtained by using higher order moments.
{"title":"Estimating Age of Information in Wireless Systems With Unknown Distributions of Inter-Arrival/Service Time","authors":"Licheng Chen;Yunquan Dong","doi":"10.1109/TNSE.2024.3453959","DOIUrl":"10.1109/TNSE.2024.3453959","url":null,"abstract":"In this paper, we estimate the average age of information (AoI) of the status updating over a wireless channel with an unknown fading model. Different from most related works which take the distributions of the inter-arrival time and transmission time of updates as known information, we approximate the average AoI of the system by using their first and second-order moments. Note that these distributions are often not accessible or known with inevitable errors while their moments are much easier to obtain, e.g., by using counting and statistics. We model the communications over the fading channel with a continuous transmission model and a discrete transmission model, which use the variable-rate scheme and the fixed-rate scheme, respectively. We assume that the arrival of the continuous transmission model is a Bernoulli process and make no assumptions about the arrival process of the discrete transmission model. Based on these information, we present two pairs of tight lower and upper bounds for the AoI of the two models. We show that obtained bounds are the tightest when the inter-arrival time (or transmission time) follows the degenerate distribution and are the loosest when it follows the two-point distribution, which randomly takes value from two possible outcomes. We also show that tighter bounds can be obtained by using higher order moments.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6090-6104"},"PeriodicalIF":6.7,"publicationDate":"2024-09-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191989","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 : 2024-09-02DOI: 10.1109/TNSE.2024.3447893
Tarmo Nurmi;Mikko Kivelä
To understand the structure of a network, it can be useful to break it down into its constituent pieces. This is the approach taken in a multitude of successful network analysis methods, such as motif analysis. These methods require one to enumerate or sample small connected subgraphs of a network. Efficient algorithms exists for both enumeration and uniform sampling of subgraphs, and here we generalize the esu algorithm for a very general notion of multilayer networks. We show that multilayer network subnetwork enumeration introduces nontrivial complications to the existing algorithm, and present two different generalized algorithms that preserve the desired features of unbiased sampling and scalable, communication-free parallelization. In addition, we introduce a straightforward aggregation-disaggregation-based enumeration algorithm that leverages existing subgraph enumeration algorithms. We evaluate these algorithms in synthetic networks and with real-world data, and show that none of the algorithms is strictly more efficient but rather the choice depends on the features of the data. Having a general algorithm for finding subnetworks makes advanced multilayer network analysis possible, and enables researchers to apply a variety of methods to previously difficult-to-handle multilayer networks in a variety of domains and across many different types of multilayer networks.
{"title":"Subnetwork Enumeration Algorithms for Multilayer Networks","authors":"Tarmo Nurmi;Mikko Kivelä","doi":"10.1109/TNSE.2024.3447893","DOIUrl":"10.1109/TNSE.2024.3447893","url":null,"abstract":"To understand the structure of a network, it can be useful to break it down into its constituent pieces. This is the approach taken in a multitude of successful network analysis methods, such as motif analysis. These methods require one to enumerate or sample small connected subgraphs of a network. Efficient algorithms exists for both enumeration and uniform sampling of subgraphs, and here we generalize the \u0000<sc>esu</small>\u0000 algorithm for a very general notion of multilayer networks. We show that multilayer network subnetwork enumeration introduces nontrivial complications to the existing algorithm, and present two different generalized algorithms that preserve the desired features of unbiased sampling and scalable, communication-free parallelization. In addition, we introduce a straightforward aggregation-disaggregation-based enumeration algorithm that leverages existing subgraph enumeration algorithms. We evaluate these algorithms in synthetic networks and with real-world data, and show that none of the algorithms is strictly more efficient but rather the choice depends on the features of the data. Having a general algorithm for finding subnetworks makes advanced multilayer network analysis possible, and enables researchers to apply a variety of methods to previously difficult-to-handle multilayer networks in a variety of domains and across many different types of multilayer networks.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"5803-5817"},"PeriodicalIF":6.7,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10663535","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191991","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"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 : 2024-08-28DOI: 10.1109/TNSE.2024.3451118
Junfeng Ma;Shan Ma;Xiaotian Xie;Weihua Gui
In this study, we model the high-speed railway (HSR) network as a directed multi-layer network. Specifically, each node is viewed as a tuple of a train with a station it passes through. A directed edge within a layer means that a train passes through two consecutive stations in its scheduled train route, while an edge between different layers means that two trains pass through the same station sequentially. Then we assess the robustness against cascading failures of these multi-layer networks by introducing metrics such as network efficiency and the ratio of failed nodes under disturbances. Furthermore, we propose a cascading failure model based on train delay propagation to investigate the cascading dynamics within the multi-layer HSR network. To better characterize the delay propagation patterns in the network, train delays at each station are treated as the load of the corresponding node, while the time supplements and buffer time are considered as the capacities of the edges. Finally, we propose two strategies to enhance the robustness of HSR networks against cascading failures. Numerical experiments are conducted to demonstrate the effectiveness of these strategies.
{"title":"Robustness Analysis of High-Speed Railway Networks Against Cascading Failures: From a Multi-Layer Network Perspective","authors":"Junfeng Ma;Shan Ma;Xiaotian Xie;Weihua Gui","doi":"10.1109/TNSE.2024.3451118","DOIUrl":"10.1109/TNSE.2024.3451118","url":null,"abstract":"In this study, we model the high-speed railway (HSR) network as a directed multi-layer network. Specifically, each node is viewed as a tuple of a train with a station it passes through. A directed edge within a layer means that a train passes through two consecutive stations in its scheduled train route, while an edge between different layers means that two trains pass through the same station sequentially. Then we assess the robustness against cascading failures of these multi-layer networks by introducing metrics such as network efficiency and the ratio of failed nodes under disturbances. Furthermore, we propose a cascading failure model based on train delay propagation to investigate the cascading dynamics within the multi-layer HSR network. To better characterize the delay propagation patterns in the network, train delays at each station are treated as the load of the corresponding node, while the time supplements and buffer time are considered as the capacities of the edges. Finally, we propose two strategies to enhance the robustness of HSR networks against cascading failures. Numerical experiments are conducted to demonstrate the effectiveness of these strategies.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6522-6534"},"PeriodicalIF":6.7,"publicationDate":"2024-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191990","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}
As the next generation of the internet, Web 3.0 is expected to revolutionize the Internet and enable users to have greater control over their data and privacy. Federated learning (FL) enables data to be usable yet invisible during its use, thereby facilitating the transfer of data ownership and value. However, the issues of data size and blockchain computing power are of paramount importance for FL in Web 3.0. Due to the openness of Web 3.0, individuals can freely join or leave training and adjust data size, creating population uncertainty and making it difficult to design incentive mechanisms. Therefore, we propose a Poisson game-based FL incentive mechanism that motivates participants to contribute more data and computing power, considering the variability of data size and computing power requirements, and provides a feasible solution to the uncertainty of the number of participants using a Poisson game model. Additionally, our proposed FL architecture in Web 3.0 integrates FL with Decentralized Autonomous Organizations (DAO), utilizing smart contracts for contribution calculation and revenue distribution. This enables an open, free, and autonomous federated learning environment. Experimental evaluation shows that our incentive mechanism is feasible in blockchain with efficiency, robustness, and low overhead.
作为下一代互联网，Web 3.0 预计将彻底改变互联网，使用户能够更好地控制自己的数据和隐私。联合学习（Federated Learning，FL）使数据在使用过程中既可使用又不可见，从而促进了数据所有权和价值的转移。然而，数据规模和区块链计算能力问题对于 Web 3.0 中的联合学习至关重要。由于 Web 3.0 的开放性，个人可以自由加入或退出培训，也可以自由调整数据规模，这就造成了群体的不确定性，使激励机制的设计变得困难。因此，我们提出了一种基于泊松博弈的 FL 激励机制，考虑到数据规模和计算能力需求的可变性，激励参与者贡献更多的数据和计算能力，并利用泊松博弈模型为参与者数量的不确定性提供了可行的解决方案。此外，我们在 Web 3.0 中提出的 FL 架构将 FL 与去中心化自治组织（DAO）相结合，利用智能合约进行贡献计算和收入分配。这就实现了一个开放、自由和自主的联合学习环境。实验评估表明，我们的激励机制在区块链中是可行的，而且高效、稳健、开销低。
{"title":"A Poisson Game-Based Incentive Mechanism for Federated Learning in Web 3.0","authors":"Mingshun Luo;Yunhua He;Tingli Yuan;Bin Wu;Yongdong Wu;Ke Xiao","doi":"10.1109/TNSE.2024.3450932","DOIUrl":"10.1109/TNSE.2024.3450932","url":null,"abstract":"As the next generation of the internet, Web 3.0 is expected to revolutionize the Internet and enable users to have greater control over their data and privacy. Federated learning (FL) enables data to be usable yet invisible during its use, thereby facilitating the transfer of data ownership and value. However, the issues of data size and blockchain computing power are of paramount importance for FL in Web 3.0. Due to the openness of Web 3.0, individuals can freely join or leave training and adjust data size, creating population uncertainty and making it difficult to design incentive mechanisms. Therefore, we propose a Poisson game-based FL incentive mechanism that motivates participants to contribute more data and computing power, considering the variability of data size and computing power requirements, and provides a feasible solution to the uncertainty of the number of participants using a Poisson game model. Additionally, our proposed FL architecture in Web 3.0 integrates FL with Decentralized Autonomous Organizations (DAO), utilizing smart contracts for contribution calculation and revenue distribution. This enables an open, free, and autonomous federated learning environment. Experimental evaluation shows that our incentive mechanism is feasible in blockchain with efficiency, robustness, and low overhead.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"5576-5588"},"PeriodicalIF":6.7,"publicationDate":"2024-08-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142224878","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 : 2024-08-27DOI: 10.1109/TNSE.2024.3450628
Ajit Kumar;Sudhir Kumar
This paper proposes a joint localization and synchronization method in the presence of a 3-D (cuboidal-bounded) channel. Many biologically relevant structures, such as epithelium cell membranes, tissues, and blood vessel networks (particularly capillaries), can be effectively modeled as 3-D systems. Localization and synchronization among nanomachines play an important role in the optimal transmission rate, information exchange, and collaboration among nanomachines. Clock synchronization without localization or localization without clock synchronization affects the accuracy of the system. However, the existing methods consider that nanomachines are already synchronized for localization and vice-versa. Hence, the proposed method considers a combined model for location parameters, clock offset, and clock skew. Unlike the existing method, we consider this combined model in bounded environments, which are relevant for long-range molecular communication where released molecules need to be confined within a certain range to optimize power efficiency. However, deriving an analytical channel characterization for a constrained domain is challenging. We provide an analytical equation for the probability distribution function of the propagation delay of the molecules, taking into account the presence of both single and multiple absorbing walls.
{"title":"Joint Localization and Clock Synchronization in Cuboid Bounded Diffusive Channel With Absorbing and Reflecting Boundaries","authors":"Ajit Kumar;Sudhir Kumar","doi":"10.1109/TNSE.2024.3450628","DOIUrl":"10.1109/TNSE.2024.3450628","url":null,"abstract":"This paper proposes a joint localization and synchronization method in the presence of a 3-D (cuboidal-bounded) channel. Many biologically relevant structures, such as epithelium cell membranes, tissues, and blood vessel networks (particularly capillaries), can be effectively modeled as 3-D systems. Localization and synchronization among nanomachines play an important role in the optimal transmission rate, information exchange, and collaboration among nanomachines. Clock synchronization without localization or localization without clock synchronization affects the accuracy of the system. However, the existing methods consider that nanomachines are already synchronized for localization and vice-versa. Hence, the proposed method considers a combined model for location parameters, clock offset, and clock skew. Unlike the existing method, we consider this combined model in bounded environments, which are relevant for long-range molecular communication where released molecules need to be confined within a certain range to optimize power efficiency. However, deriving an analytical channel characterization for a constrained domain is challenging. We provide an analytical equation for the probability distribution function of the propagation delay of the molecules, taking into account the presence of both single and multiple absorbing walls.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6511-6521"},"PeriodicalIF":6.7,"publicationDate":"2024-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191993","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}
Hierarchical federated learning (HFL) in wireless networks significantly saves communication resources due to edge aggregation conducted in edge mobile computing (MEC) servers. Taking into account the spatially correlated characteristics of data in wireless networks, in this paper, we analyze the performance of HFL with hybrid data distributions, i.e. intra-MEC independent and identically distributed (IID) and inter-MEC non-IID data samples. We derive the upper bound of the difference between the achieved loss and the minimum one, which reveals the impacts of data heterogeneity and global aggregation frequency on the performance of HFL. On this basis, we propose an algorithm named FedHelo which optimizes the aggregation weights and edge/global aggregation frequencies under the constraints of training delay and clients' energy consumption. Our experiments i) verify the obtained theoretical results; ii) demonstrate the performance improvement achieved by FedHelo with the optimal aggregation weights and training/aggregation frequencies, especially in the scenario with high data heterogeneity; and iii) show the preference for edge aggregation in the scenario with a tight delay or client's energy constraint.
{"title":"FedHelo: Hierarchical Federated Learning With Loss-Based-Heterogeneity in Wireless Networks","authors":"Yuchuan Ye;Youjia Chen;Junnan Yang;Ming Ding;Peng Cheng;Haifeng Zheng","doi":"10.1109/TNSE.2024.3447904","DOIUrl":"10.1109/TNSE.2024.3447904","url":null,"abstract":"Hierarchical federated learning (HFL) in wireless networks significantly saves communication resources due to edge aggregation conducted in edge mobile computing (MEC) servers. Taking into account the spatially correlated characteristics of data in wireless networks, in this paper, we analyze the performance of HFL with hybrid data distributions, i.e. intra-MEC independent and identically distributed (IID) and inter-MEC non-IID data samples. We derive the upper bound of the difference between the achieved loss and the minimum one, which reveals the impacts of data heterogeneity and global aggregation frequency on the performance of HFL. On this basis, we propose an algorithm named FedHelo which optimizes the aggregation weights and edge/global aggregation frequencies under the constraints of training delay and clients' energy consumption. Our experiments \u0000<italic>i)</i>\u0000 verify the obtained theoretical results; \u0000<italic>ii)</i>\u0000 demonstrate the performance improvement achieved by FedHelo with the optimal aggregation weights and training/aggregation frequencies, especially in the scenario with high data heterogeneity; and \u0000<italic>iii)</i>\u0000 show the preference for edge aggregation in the scenario with a tight delay or client's energy constraint.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6066-6079"},"PeriodicalIF":6.7,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191992","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 : 2024-08-26DOI: 10.1109/TNSE.2024.3449146
Shiwen Zhang;Zhixue Li;Wei Liang;Kuan-Ching Li;Zakirul Alam Bhuiyan
Mobile Crowd Sensing (MCS) has emerged as a new sensing paradigm due to its cost efficiency, mobility, and expandability. However, user selection for task allocation is a significant challenge in MCS. Most previous studies concentrate on two selection modes, opportunistic and participatory selection. Recent research has proposed a hybrid user selection mode that combines both advantages. However, existing hybrid user selection systems all rely on a centralized architecture, which is vulnerable to malicious attacks, and they do not consider the reliability of users and data availability. Moreover, they cannot ensure the individual rationality of users. To overcome these shortcomings, we propose a blockchain-based hybrid reliable user selection scheme for task allocation in MCS. Specifically, we replace the traditional central server with the blockchain and handle various sensing task operations using smart contracts on the blockchain to ensure system reliability and security. In addition, we design a user reputation calculation algorithm based on semi-Markov and a sensing data anomaly detection algorithm based on Long Short-Term Memory (LSTM) to ensure user reliability and data availability, and also a novel hybrid user selection algorithm, especially in the participatory user selection stage, where we use a user selection algorithm based on reverse auction to ensure the individual rationality of each user. Experimental results demonstrate the effectiveness of the proposed scheme through simulation experiments on GeoLife and sound-sensing public datasets.
{"title":"Blockchain-Based Hybrid Reliable User Selection Scheme for Task Allocation in Mobile Crowd Sensing","authors":"Shiwen Zhang;Zhixue Li;Wei Liang;Kuan-Ching Li;Zakirul Alam Bhuiyan","doi":"10.1109/TNSE.2024.3449146","DOIUrl":"10.1109/TNSE.2024.3449146","url":null,"abstract":"Mobile Crowd Sensing (MCS) has emerged as a new sensing paradigm due to its cost efficiency, mobility, and expandability. However, user selection for task allocation is a significant challenge in MCS. Most previous studies concentrate on two selection modes, opportunistic and participatory selection. Recent research has proposed a hybrid user selection mode that combines both advantages. However, existing hybrid user selection systems all rely on a centralized architecture, which is vulnerable to malicious attacks, and they do not consider the reliability of users and data availability. Moreover, they cannot ensure the individual rationality of users. To overcome these shortcomings, we propose a blockchain-based hybrid reliable user selection scheme for task allocation in MCS. Specifically, we replace the traditional central server with the blockchain and handle various sensing task operations using smart contracts on the blockchain to ensure system reliability and security. In addition, we design a user reputation calculation algorithm based on semi-Markov and a sensing data anomaly detection algorithm based on Long Short-Term Memory (LSTM) to ensure user reliability and data availability, and also a novel hybrid user selection algorithm, especially in the participatory user selection stage, where we use a user selection algorithm based on reverse auction to ensure the individual rationality of each user. Experimental results demonstrate the effectiveness of the proposed scheme through simulation experiments on GeoLife and sound-sensing public datasets.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6494-6510"},"PeriodicalIF":6.7,"publicationDate":"2024-08-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191994","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 : 2024-08-23DOI: 10.1109/TNSE.2024.3448609
Zeyuan Xu;Zhe Yang;Danwei Wang;Zhe Wu
Contextual information has been widely used to improve results of semantic segmentation. However, most approaches investigate contextual dependencies through self-attention and lack guidance on which pixels should have strong (or weak) relationships. In this paper, a category-guided graph convolution network (CGGCN) is proposed to reveal the relationships among pixels. First, we train a coarse segmentation map under the supervision of the ground truth and use it to construct an adjacency matrix among pixels. It turns out that the pixels belonging to the same category have strong connections, and those belonging to different categories have weak connections. Second, a GCN is exploited to enhance the representation of pixels by aggregating contextual information among pixels. The feature of each pixel is represented by node, and the relationship among pixels is denoted by edge. Subsequently, we design four different kinds of network structures by leveraging the CGGCN module and determine the most accurate segmentation result by comparing them. Finally, we reimplement the CGGCN module to refine the final prediction from coarse to fine. The results of extensive evaluations demonstrate that the proposed approach is superior to the existing semantic segmentation approaches and has better convergence.
{"title":"Category-Guided Graph Convolution Network for Semantic Segmentation","authors":"Zeyuan Xu;Zhe Yang;Danwei Wang;Zhe Wu","doi":"10.1109/TNSE.2024.3448609","DOIUrl":"10.1109/TNSE.2024.3448609","url":null,"abstract":"Contextual information has been widely used to improve results of semantic segmentation. However, most approaches investigate contextual dependencies through self-attention and lack guidance on which pixels should have strong (or weak) relationships. In this paper, a category-guided graph convolution network (CGGCN) is proposed to reveal the relationships among pixels. First, we train a coarse segmentation map under the supervision of the ground truth and use it to construct an adjacency matrix among pixels. It turns out that the pixels belonging to the same category have strong connections, and those belonging to different categories have weak connections. Second, a GCN is exploited to enhance the representation of pixels by aggregating contextual information among pixels. The feature of each pixel is represented by node, and the relationship among pixels is denoted by edge. Subsequently, we design four different kinds of network structures by leveraging the CGGCN module and determine the most accurate segmentation result by comparing them. Finally, we reimplement the CGGCN module to refine the final prediction from coarse to fine. The results of extensive evaluations demonstrate that the proposed approach is superior to the existing semantic segmentation approaches and has better convergence.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6080-6089"},"PeriodicalIF":6.7,"publicationDate":"2024-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191995","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}
Data outsourcing is a key service of cloud computing. While data encryption ensures confidentiality, it limits the ability to search encrypted data. Recently, ciphertext-policy attribute-based keyword search (CP-ABKS) schemes, which combine symmetric searchable encryption (SSE) and ciphertext-policy attribute-based encryption (CP-ABE), have gained attention. However, most CP-ABKS schemes depend on an independent key management server (KMS) for key distribution, risking key leakage if the KMS is compromised. Additionally, these schemes lack secure update operations and efficient search result verification. To address these issues, we propose VKSA, a verifiable encrypted keyword search scheme with authorization for cloud-based multi-client environments. VKSA features a novel policy-hidden index for proxy-free authorized searches, a state-based secure update strategy for forward and backward security, and a delegated search result verification mechanism to ensure efficient and privacy-preserving verification. We further optimize VKSA for improved computational and enclave-storage efficiency. Security analysis and experiments confirm the security and efficiency of our schemes.
{"title":"Multi-Client Verifiable Encrypted Keyword Search Scheme With Authorization Over Outsourced Encrypted Data","authors":"Xu Yang;Qiuhao Wang;Saiyu Qi;Ke Li;Jianfeng Wang;Wenjia Zhao;Yong Qi","doi":"10.1109/TNSE.2024.3445343","DOIUrl":"10.1109/TNSE.2024.3445343","url":null,"abstract":"Data outsourcing is a key service of cloud computing. While data encryption ensures confidentiality, it limits the ability to search encrypted data. Recently, ciphertext-policy attribute-based keyword search (CP-ABKS) schemes, which combine symmetric searchable encryption (SSE) and ciphertext-policy attribute-based encryption (CP-ABE), have gained attention. However, most CP-ABKS schemes depend on an independent key management server (KMS) for key distribution, risking key leakage if the KMS is compromised. Additionally, these schemes lack secure update operations and efficient search result verification. To address these issues, we propose VKSA, a verifiable encrypted keyword search scheme with authorization for cloud-based multi-client environments. VKSA features a novel policy-hidden index for proxy-free authorized searches, a state-based secure update strategy for forward and backward security, and a delegated search result verification mechanism to ensure efficient and privacy-preserving verification. We further optimize VKSA for improved computational and enclave-storage efficiency. Security analysis and experiments confirm the security and efficiency of our schemes.","PeriodicalId":54229,"journal":{"name":"IEEE Transactions on Network Science and Engineering","volume":"11 6","pages":"6356-6371"},"PeriodicalIF":6.7,"publicationDate":"2024-08-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142191997","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：