Pub Date : 2024-03-19DOI: 10.1109/TSIPN.2024.3375611
Sangli Shi;Zhengxin Wang;Min Xiao;Guo-Ping Jiang;Jinde Cao
Distributed secure control is investigated for cooperative-competitive multiagent systems suffered from false data injection attacks (FDIAs) via event-triggered observers. Attack signals are injected into controller-to-actuator channels. A static event-triggered control is first presented, then an auxiliary-variable-based dynamic event-triggered control is further put forward. The dynamic event-triggered control ensures fewer triggering instants and the dynamic variable plays a significant part in the exclusion of Zeno-behavior. Then based on estimated states and attacks calculated by observers, distributed controllers are proposed to resist attacks. Bipartite consensus is ensured in multiagent systems and corresponding sufficient conditions are obtained. Meanwhile, the Zeno-behaviors are proven to be nonexistent. Finally, theoretical analyses are explained by simulations.
{"title":"Consensus Analysis for Cooperative-Competitive Multiagent Systems Under False Data Injection Attacks via Dynamic Event-Triggered Observers","authors":"Sangli Shi;Zhengxin Wang;Min Xiao;Guo-Ping Jiang;Jinde Cao","doi":"10.1109/TSIPN.2024.3375611","DOIUrl":"10.1109/TSIPN.2024.3375611","url":null,"abstract":"Distributed secure control is investigated for cooperative-competitive multiagent systems suffered from false data injection attacks (FDIAs) via event-triggered observers. Attack signals are injected into controller-to-actuator channels. A static event-triggered control is first presented, then an auxiliary-variable-based dynamic event-triggered control is further put forward. The dynamic event-triggered control ensures fewer triggering instants and the dynamic variable plays a significant part in the exclusion of Zeno-behavior. Then based on estimated states and attacks calculated by observers, distributed controllers are proposed to resist attacks. Bipartite consensus is ensured in multiagent systems and corresponding sufficient conditions are obtained. Meanwhile, the Zeno-behaviors are proven to be nonexistent. Finally, theoretical analyses are explained by simulations.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"195-204"},"PeriodicalIF":3.2,"publicationDate":"2024-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140168740","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 : 2024-03-19DOI: 10.1109/TSIPN.2024.3378122
Guangrui Yang;Qing Zhang;Lihua Yang
Signal representations on graphs are at the heart of most graph signal processing techniques, allowing for targeted signal models for tasks such as denoising, compression, sampling, reconstruction and detection. This paper studies the piecewise-constant representation of bandlimited graph signals, thereby establishing the relationship between the bandlimited graph signal and the piecewise-constant graph signal. For this purpose, we first introduce the concept of $epsilon$ -level piecewise-constant representation for a general signal space. Then, using a distance matrix, a single-layer piecewise-constant representation algorithm is proposed to find an $epsilon$ -level piecewise-constant representation for bandlimited graph signals. On this basis, we further propose a multi-layer piecewise-constant representation algorithm, which can find a node partition with as few pieces as possible to represent bandlimited graph signals piecewise within a preset error bound. Finally, as an application, we apply the node partition obtained by the multi-layer algorithm to establish a sampling theory for bandlimited signals, which does not need to compute the eigendecomposition of a variation operator in both sampling and signal reconstruction. Numerical experiments show that the proposed algorithms have good performance.
{"title":"Piecewise-Constant Representation and Sampling of Bandlimited Signals on Graphs","authors":"Guangrui Yang;Qing Zhang;Lihua Yang","doi":"10.1109/TSIPN.2024.3378122","DOIUrl":"10.1109/TSIPN.2024.3378122","url":null,"abstract":"Signal representations on graphs are at the heart of most graph signal processing techniques, allowing for targeted signal models for tasks such as denoising, compression, sampling, reconstruction and detection. This paper studies the piecewise-constant representation of bandlimited graph signals, thereby establishing the relationship between the bandlimited graph signal and the piecewise-constant graph signal. For this purpose, we first introduce the concept of \u0000<inline-formula><tex-math>$epsilon$</tex-math></inline-formula>\u0000-level piecewise-constant representation for a general signal space. Then, using a distance matrix, a single-layer piecewise-constant representation algorithm is proposed to find an \u0000<inline-formula><tex-math>$epsilon$</tex-math></inline-formula>\u0000-level piecewise-constant representation for bandlimited graph signals. On this basis, we further propose a multi-layer piecewise-constant representation algorithm, which can find a node partition with as few pieces as possible to represent bandlimited graph signals piecewise within a preset error bound. Finally, as an application, we apply the node partition obtained by the multi-layer algorithm to establish a sampling theory for bandlimited signals, which does not need to compute the eigendecomposition of a variation operator in both sampling and signal reconstruction. Numerical experiments show that the proposed algorithms have good performance.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"332-346"},"PeriodicalIF":3.2,"publicationDate":"2024-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140168763","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 : 2024-03-17DOI: 10.1109/TSIPN.2024.3402354
Shuaiyu Zhou;Yiheng Wei;Shu Liang;Jinde Cao
This paper investigates the distributed consensus optimization over a class of nabla fractional multi-agent systems (nFMASs). The proposed approach, built upon conventional gradient tracking techniques, addresses the specificity of the studied system by introducing a fractional gradient tracking protocol based on globally differential information of optimization variables. This protocol is applicable to nabla fractional systems of any order less than 1 and can be extended to integer discrete-time systems. The distributed optimization algorithms derived from this protocol ensure globally precise convergence under fixed step-sizes, thereby guaranteeing the feasibility of consensus optimization over nFMASs. Simulation results are presented to validate and substantiate the effectiveness of the proposed algorithms.
{"title":"A Gradient Tracking Protocol for Optimization Over Nabla Fractional Multi-Agent Systems","authors":"Shuaiyu Zhou;Yiheng Wei;Shu Liang;Jinde Cao","doi":"10.1109/TSIPN.2024.3402354","DOIUrl":"10.1109/TSIPN.2024.3402354","url":null,"abstract":"This paper investigates the distributed consensus optimization over a class of nabla fractional multi-agent systems (nFMASs). The proposed approach, built upon conventional gradient tracking techniques, addresses the specificity of the studied system by introducing a fractional gradient tracking protocol based on globally differential information of optimization variables. This protocol is applicable to nabla fractional systems of any order less than 1 and can be extended to integer discrete-time systems. The distributed optimization algorithms derived from this protocol ensure globally precise convergence under fixed step-sizes, thereby guaranteeing the feasibility of consensus optimization over nFMASs. Simulation results are presented to validate and substantiate the effectiveness of the proposed algorithms.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"500-512"},"PeriodicalIF":3.2,"publicationDate":"2024-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141061133","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}
In this paper, we consider remote state estimation for an arbitrary tree topology in cyber-physical systems (CPSs) subject to Denial-of-Service (DoS) attacks. A sensor transmits its local estimation to the root node of the tree, and the root node transmits the optimal estimation to its child nodes until the leaf nodes are reached. In the meanwhile, a malicious attacker can jam all communication channels strategically connected to the attacked node. With the energy constraints in mind, both the defender and attacker adopt strategies that involve allocating energy to determine which nodes to protect or attack at each time step. A Bayesian Stackelberg game (BSG) framework with incomplete information is implemented, where the defender has no access to the available energy of the attacker exactly except for its probability distribution. In addition, a Markov decision process (MDP) and a Stackelberg Q-learning algorithm are presented to obtain the Stackelberg equilibrium (SE) policy over a finite time horizon. Finally, a numerical example is provided to demonstrate our main results.
{"title":"Remote State Estimation Under DoS Attacks in CPSs With Arbitrary Tree Topology: A Bayesian Stackelberg Game Approach","authors":"Yuhan Wang;Wei Xing;Junfeng Zhang;Le Liu;Xudong Zhao","doi":"10.1109/TSIPN.2024.3394776","DOIUrl":"10.1109/TSIPN.2024.3394776","url":null,"abstract":"In this paper, we consider remote state estimation for an arbitrary tree topology in cyber-physical systems (CPSs) subject to Denial-of-Service (DoS) attacks. A sensor transmits its local estimation to the root node of the tree, and the root node transmits the optimal estimation to its child nodes until the leaf nodes are reached. In the meanwhile, a malicious attacker can jam all communication channels strategically connected to the attacked node. With the energy constraints in mind, both the defender and attacker adopt strategies that involve allocating energy to determine which nodes to protect or attack at each time step. A Bayesian Stackelberg game (BSG) framework with incomplete information is implemented, where the defender has no access to the available energy of the attacker exactly except for its probability distribution. In addition, a Markov decision process (MDP) and a Stackelberg Q-learning algorithm are presented to obtain the Stackelberg equilibrium (SE) policy over a finite time horizon. Finally, a numerical example is provided to demonstrate our main results.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"527-538"},"PeriodicalIF":3.2,"publicationDate":"2024-03-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141061136","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}
The monitoring of natural gas pipelines is highly dependent on the information provided by different types of sensors. However, sensors are prone to faults, which results in performance degradation and serious hazards such as leaks or explosions. To prevent catastrophic failures and ensure the safe and efficient operation of the pipelines, it is crucial to timely diagnose sensor faults in natural gas pipelines. This paper investigates model-based sensor fault diagnosis techniques in a natural-gas pipeline under transient flow. A fusing architecture based on distributed data fusion is used for implementing the sensor fault detection, isolation, and accommodation (SFDIA) mechanism. The fusing architecture consists of a set of local filters and an information mixer. The local filters estimate the state variables in parallel, which are subsequently transferred to the information mixer to evaluate the sensor faults and compute fault-free state estimates. In this paper, three different types of fusing filters, namely based on the ensemble Kalman filter (EnKF), fusing unscented Kalman filter (UKF), and fusing extended Kalman filter (EKF) are investigated for fault diagnosis. Results demonstrate that all three filters can successfully detect, isolate, and accommodate sensor faults.
{"title":"Sensor-Fault Detection, Isolation and Accommodation for Natural-Gas Pipelines Under Transient Flow","authors":"Khadija Shaheen;Apoorva Chawla;Ferdinand Evert Uilhoorn;Pierluigi Salvo Rossi","doi":"10.1109/TSIPN.2024.3377134","DOIUrl":"10.1109/TSIPN.2024.3377134","url":null,"abstract":"The monitoring of natural gas pipelines is highly dependent on the information provided by different types of sensors. However, sensors are prone to faults, which results in performance degradation and serious hazards such as leaks or explosions. To prevent catastrophic failures and ensure the safe and efficient operation of the pipelines, it is crucial to timely diagnose sensor faults in natural gas pipelines. This paper investigates model-based sensor fault diagnosis techniques in a natural-gas pipeline under transient flow. A fusing architecture based on distributed data fusion is used for implementing the sensor fault detection, isolation, and accommodation (SFDIA) mechanism. The fusing architecture consists of a set of local filters and an information mixer. The local filters estimate the state variables in parallel, which are subsequently transferred to the information mixer to evaluate the sensor faults and compute fault-free state estimates. In this paper, three different types of fusing filters, namely based on the ensemble Kalman filter (EnKF), fusing unscented Kalman filter (UKF), and fusing extended Kalman filter (EKF) are investigated for fault diagnosis. Results demonstrate that all three filters can successfully detect, isolate, and accommodate sensor faults.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"264-276"},"PeriodicalIF":3.2,"publicationDate":"2024-03-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140125622","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 : 2024-03-11DOI: 10.1109/TSIPN.2024.3375593
Alessio Fascista;Angelo Coluccia;Chiara Ravazzi
Reconstructing bandlimited graph signals from a subset of noisy measurements is a fundamental challenge within the realm of signal processing. Historically, this problem has been approached assuming uniform noise variance across the network. Nevertheless, practical scenarios often present heterogeneous noise landscapes, greatly complicating the signal reconstruction process. This study tackles reconstruction of graph signals across networks where measurements may be affected by heterogeneous noise. A Bayesian model tailored for graph signals is employed, considering the potential existence of node-specific variations in measurement variance, namely different (and unknown) levels of uncertainty. Moreover, a novel uncertainty-aware local graph coherence metric is introduced, capitalizing on estimated parameters to refine the sampling process. By accommodating uncertainty, signal reconstruction accuracy is enhanced, even in demanding noise conditions. The proposed approach revolves around a framework combining maximum likelihood and maximum a-posteriori principles. Specifically, each observation is weighted based on a soft classification of nodes, so incorporating measurements reliability into the reconstruction process. The latter is performed through a novel algorithm coupling re-weighted iterative least squares with expectation-maximization. Such an algorithm can effectively manage heterogeneous noise and features a non-local regularization term, which promotes sparsity in the reconstructed signal while preserving signal discontinuities, crucial for capturing the characteristics of the underlying graph signal. Extensive simulations demonstrate the effectiveness of the proposed approach for various graph topologies and anomalous conditions, revealing substantial enhancements in signal reconstruction compared to existing methods. An illustrative example on PM10 data from the European Copernicus Atmosphere Monitoring Service (CAMS) is also reported.
{"title":"Graph Signal Reconstruction Under Heterogeneous Noise via Adaptive Uncertainty-Aware Sampling and Soft Classification","authors":"Alessio Fascista;Angelo Coluccia;Chiara Ravazzi","doi":"10.1109/TSIPN.2024.3375593","DOIUrl":"10.1109/TSIPN.2024.3375593","url":null,"abstract":"Reconstructing bandlimited graph signals from a subset of noisy measurements is a fundamental challenge within the realm of signal processing. Historically, this problem has been approached assuming uniform noise variance across the network. Nevertheless, practical scenarios often present heterogeneous noise landscapes, greatly complicating the signal reconstruction process. This study tackles reconstruction of graph signals across networks where measurements may be affected by heterogeneous noise. A Bayesian model tailored for graph signals is employed, considering the potential existence of node-specific variations in measurement variance, namely different (and unknown) levels of uncertainty. Moreover, a novel uncertainty-aware local graph coherence metric is introduced, capitalizing on estimated parameters to refine the sampling process. By accommodating uncertainty, signal reconstruction accuracy is enhanced, even in demanding noise conditions. The proposed approach revolves around a framework combining maximum likelihood and maximum a-posteriori principles. Specifically, each observation is weighted based on a soft classification of nodes, so incorporating measurements reliability into the reconstruction process. The latter is performed through a novel algorithm coupling re-weighted iterative least squares with expectation-maximization. Such an algorithm can effectively manage heterogeneous noise and features a non-local regularization term, which promotes sparsity in the reconstructed signal while preserving signal discontinuities, crucial for capturing the characteristics of the underlying graph signal. Extensive simulations demonstrate the effectiveness of the proposed approach for various graph topologies and anomalous conditions, revealing substantial enhancements in signal reconstruction compared to existing methods. An illustrative example on PM10 data from the European Copernicus Atmosphere Monitoring Service (CAMS) is also reported.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"277-293"},"PeriodicalIF":3.2,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140105541","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 : 2024-03-11DOI: 10.1109/TSIPN.2024.3375597
Richard Heusdens;Guoqiang Zhang
In this article, we consider the problem of distributed optimisation of a separable convex cost function over a graph, where every edge and node in the graph could carry both linear equality and/or inequality constraints. We show how to modify the primal-dual method of multipliers (PDMM), originally designed for linear equality constraints, such that it can handle inequality constraints as well. The proposed algorithm does not need any slack variables, which is similar to the recent work (He et al., 2023) which extends the alternating direction method of multipliers (ADMM) for addressing decomposable optimisation with linear equality and inequality constraints. Using convex analysis, monotone operator theory and fixed-point theory, we show how to derive the update equations of the modified PDMM algorithm by applying Peaceman-Rachford splitting to the monotonic inclusion related to the lifted dual problem. To incorporate the inequality constraints, we impose a non-negativity constraint on the associated dual variables. This additional constraint results in the introduction of a reflection operator to model the data exchange in the network, instead of a permutation operator as derived for equality constraint PDMM. Convergence for both synchronous and stochastic update schemes of PDMM are provided. The latter includes asynchronous update schemes and update schemes with transmission losses. Experiments show that PDMM converges notably faster than extended ADMM of (He et al., 2023).
{"title":"Distributed Optimisation With Linear Equality and Inequality Constraints Using PDMM","authors":"Richard Heusdens;Guoqiang Zhang","doi":"10.1109/TSIPN.2024.3375597","DOIUrl":"10.1109/TSIPN.2024.3375597","url":null,"abstract":"In this article, we consider the problem of distributed optimisation of a separable convex cost function over a graph, where every edge and node in the graph could carry both linear equality and/or inequality constraints. We show how to modify the primal-dual method of multipliers (PDMM), originally designed for linear equality constraints, such that it can handle inequality constraints as well. The proposed algorithm does not need any slack variables, which is similar to the recent work (He et al., 2023) which extends the alternating direction method of multipliers (ADMM) for addressing decomposable optimisation with linear equality and inequality constraints. Using convex analysis, monotone operator theory and fixed-point theory, we show how to derive the update equations of the modified PDMM algorithm by applying Peaceman-Rachford splitting to the monotonic inclusion related to the lifted dual problem. To incorporate the inequality constraints, we impose a non-negativity constraint on the associated dual variables. This additional constraint results in the introduction of a reflection operator to model the data exchange in the network, instead of a permutation operator as derived for equality constraint PDMM. Convergence for both synchronous and stochastic update schemes of PDMM are provided. The latter includes asynchronous update schemes and update schemes with transmission losses. Experiments show that PDMM converges notably faster than extended ADMM of (He et al., 2023).","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"294-306"},"PeriodicalIF":3.2,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140105668","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 : 2024-03-11DOI: 10.1109/TSIPN.2024.3375600
Arnab Sarker;Ali Jadbabaie;Devavrat Shah
The COVID-19 pandemic has emphasized the need for a robust understanding of epidemic models. Current models of epidemics are classified as either mechanistic or non-mechanistic: mechanistic models make explicit assumptions on the dynamics of disease, whereas non-mechanistic models make assumptions on the form of observed time series. Here, we introduce a simple mixture-based model which bridges the two approaches while retaining benefits of both. The model represents time series of cases as a mixture of Gaussian curves, providing a flexible function class to learn from data, and we show that it arises as the natural outcome of a stochastic process based on a networked SIR framework. This allows learned parameters to take on a more meaningful interpretation compared to similar non-mechanistic models, and we validate the interpretations using auxiliary mobility data collected during the COVID-19 pandemic. We provide a simple learning algorithm to identify model parameters and establish theoretical results which show the model can be efficiently learned from data. Empirically, we find the model to have low prediction error. Moreover, this allows us to systematically understand the impacts of interventions on COVID-19, which is critical in developing data-driven solutions for controlling epidemics.
COVID-19 大流行强调了对流行病模型进行深入理解的必要性。目前的流行病模型分为机理型和非机理型两种：机理型模型对疾病的动态做出明确假设，而非机理型模型则对观察到的时间序列形式做出假设。在此，我们介绍一种简单的基于混合物的模型，该模型在保留这两种方法优点的同时，也在这两种方法之间架起了一座桥梁。该模型将病例的时间序列表示为高斯曲线的混合物，为从数据中学习提供了一个灵活的函数类别，我们还展示了该模型是基于网络化 SIR 框架的随机过程的自然结果。与类似的非机理模型相比，这使得学习到的参数具有更有意义的解释，我们使用 COVID-19 大流行期间收集的辅助流动数据验证了这些解释。我们提供了一种简单的学习算法来识别模型参数，并建立了理论结果，表明该模型可从数据中有效学习。根据经验，我们发现该模型的预测误差很小。此外，这使我们能够系统地了解干预措施对 COVID-19 的影响，这对开发数据驱动的流行病控制解决方案至关重要。
{"title":"Unifying Epidemic Models With Mixtures","authors":"Arnab Sarker;Ali Jadbabaie;Devavrat Shah","doi":"10.1109/TSIPN.2024.3375600","DOIUrl":"10.1109/TSIPN.2024.3375600","url":null,"abstract":"The COVID-19 pandemic has emphasized the need for a robust understanding of epidemic models. Current models of epidemics are classified as either mechanistic or non-mechanistic: mechanistic models make explicit assumptions on the dynamics of disease, whereas non-mechanistic models make assumptions on the form of observed time series. Here, we introduce a simple mixture-based model which bridges the two approaches while retaining benefits of both. The model represents time series of cases as a mixture of Gaussian curves, providing a flexible function class to learn from data, and we show that it arises as the natural outcome of a stochastic process based on a networked SIR framework. This allows learned parameters to take on a more meaningful interpretation compared to similar non-mechanistic models, and we validate the interpretations using auxiliary mobility data collected during the COVID-19 pandemic. We provide a simple learning algorithm to identify model parameters and establish theoretical results which show the model can be efficiently learned from data. Empirically, we find the model to have low prediction error. Moreover, this allows us to systematically understand the impacts of interventions on COVID-19, which is critical in developing data-driven solutions for controlling epidemics.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"239-252"},"PeriodicalIF":3.2,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140105671","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 : 2024-03-11DOI: 10.1109/TSIPN.2024.3375602
Jiaqi Liang;Xuhui Bu;Zhongsheng Hou
A general finite-time bipartite consensus problem is studied for multi-agent systems with completely unknown nonlinearities. An asymmetric bipartite consensus task is defined by introducing a proportional-related coefficient and a relationship-related index, which arranges that the agents reach an agreement with proportional modulus and opposite signs. With the cooperative-antagonistic interactions, a model-free adaptive bipartite iterative learning consensus protocol is proposed for promoting the accuracy of the performance within a finite-time interval. By employing the matrix transformation and property of the nonnegative matrix, the iteratively asymptotic convergence of the error of the MAS is guaranteed under the structurally balanced digraph has an oriented spanning tree. This differs from MFAILC results that have been proven based on matrix norm and do not require strong connectivity of digraphs. Moreover, the bounds for elements in the estimation-related matrices are presented, followed by providing a graph correlated sufficient condition to guide selection of control parameters. The results further extend to the control of asymmetric bipartite consensus tracking. The simulation examples verify the effectiveness of the distributed learning control protocols.
{"title":"Finite-Time Asymmetric Bipartite Consensus for Multi-Agent Systems Using Data-Driven Iterative Learning Control","authors":"Jiaqi Liang;Xuhui Bu;Zhongsheng Hou","doi":"10.1109/TSIPN.2024.3375602","DOIUrl":"10.1109/TSIPN.2024.3375602","url":null,"abstract":"A general finite-time bipartite consensus problem is studied for multi-agent systems with completely unknown nonlinearities. An asymmetric bipartite consensus task is defined by introducing a proportional-related coefficient and a relationship-related index, which arranges that the agents reach an agreement with proportional modulus and opposite signs. With the cooperative-antagonistic interactions, a model-free adaptive bipartite iterative learning consensus protocol is proposed for promoting the accuracy of the performance within a finite-time interval. By employing the matrix transformation and property of the nonnegative matrix, the iteratively asymptotic convergence of the error of the MAS is guaranteed under the structurally balanced digraph has an oriented spanning tree. This differs from MFAILC results that have been proven based on matrix norm and do not require strong connectivity of digraphs. Moreover, the bounds for elements in the estimation-related matrices are presented, followed by providing a graph correlated sufficient condition to guide selection of control parameters. The results further extend to the control of asymmetric bipartite consensus tracking. The simulation examples verify the effectiveness of the distributed learning control protocols.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"227-238"},"PeriodicalIF":3.2,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140105673","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 : 2024-03-11DOI: 10.1109/TSIPN.2024.3375596
Meiyu Li;Jinling Liang
This paper studies the probability-guaranteed distributed estimation problem for a kind of two-dimensional shift-varying sensor networks under the stochastic access protocol (SAP). The considered system is affected by unknown-but-bounded perturbations and sector bounded nonlinearity. The communication architecture of a multi-node network is expressed by a digraph. Due to the limited communication channel, each moment allows only one adjacent node to send its measurement data and schedules the signal transmission of the addressing system using the SAP, characterized by a series of independent random variables. For each smart sensor, we designed a distributed estimator based on the network topology as well as the SAP and derived sufficient conditions to ascertain the probability of the estimation error located in the desired ellipsoid being not less than the predetermined value. Collection of these probability ellipsoids acquired at each position is then minimized by solving a set of convex optimization problems in the meaning of matrix trace. Finally, efficiency of the estimator design strategy proposed is demonstrated using a numerical example.
本文研究了随机接入协议（SAP）下一种二维位移变化传感器网络的概率保证分布式估计问题。所考虑的系统受到未知但有界的扰动和扇区有界非线性的影响。多节点网络的通信架构用数字图表示。由于通信信道有限，每个时刻只允许一个相邻节点发送测量数据，并使用 SAP 调度寻址系统的信号传输，SAP 由一系列独立随机变量组成。我们为每个智能传感器设计了基于网络拓扑结构和 SAP 的分布式估算器，并推导出充分条件，以确定位于所需椭球体中的估算误差概率不小于预定值。然后，通过求解一系列矩阵跟踪意义上的凸优化问题，最小化在每个位置获取的这些概率椭圆的集合。最后，通过一个数值示例展示了所提出的估算器设计策略的效率。
{"title":"Probability-Guaranteed Distributed Estimation for Two-Dimensional Systems Under Stochastic Access Protocol","authors":"Meiyu Li;Jinling Liang","doi":"10.1109/TSIPN.2024.3375596","DOIUrl":"10.1109/TSIPN.2024.3375596","url":null,"abstract":"This paper studies the probability-guaranteed distributed estimation problem for a kind of two-dimensional shift-varying sensor networks under the stochastic access protocol (SAP). The considered system is affected by unknown-but-bounded perturbations and sector bounded nonlinearity. The communication architecture of a multi-node network is expressed by a digraph. Due to the limited communication channel, each moment allows only one adjacent node to send its measurement data and schedules the signal transmission of the addressing system using the SAP, characterized by a series of independent random variables. For each smart sensor, we designed a distributed estimator based on the network topology as well as the SAP and derived sufficient conditions to ascertain the probability of the estimation error located in the desired ellipsoid being not less than the predetermined value. Collection of these probability ellipsoids acquired at each position is then minimized by solving a set of convex optimization problems in the meaning of matrix trace. Finally, efficiency of the estimator design strategy proposed is demonstrated using a numerical example.","PeriodicalId":56268,"journal":{"name":"IEEE Transactions on Signal and Information Processing over Networks","volume":"10 ","pages":"216-226"},"PeriodicalIF":3.2,"publicationDate":"2024-03-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140105672","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：