To improve the agility, dynamics, composability, reusability, and development efficiency restricted by monolithic federation object model （FOM）, a modular FOM is proposed by high level architecture （HLA） evolved p...To improve the agility, dynamics, composability, reusability, and development efficiency restricted by monolithic federation object model （FOM）, a modular FOM is proposed by high level architecture （HLA） evolved product development group. This paper reviews the state-of-the-art of HLA evolved modular FOM. In particular, related concepts, the overall impact on HLA standards, extension principles, and merging processes are discussed. Also permitted and restricted combinations, and merging rules are provided, and the influence on HLA interface specification is given. The comparison between modular FOM and base object model （BOM） is performed to illustrate the importance of their combination. The applications of modular FOM are summarized. Finally, the significance to facilitate compoable simulation both in academia and practice is presented and future directions are pointed out.展开更多
Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure ...Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.展开更多
Heterogeneous federated learning(HtFL)has gained significant attention due to its ability to accommodate diverse models and data from distributed combat units.The prototype-based HtFL methods were proposed to reduce t...Heterogeneous federated learning(HtFL)has gained significant attention due to its ability to accommodate diverse models and data from distributed combat units.The prototype-based HtFL methods were proposed to reduce the high communication cost of transmitting model parameters.These methods allow for the sharing of only class representatives between heterogeneous clients while maintaining privacy.However,existing prototype learning approaches fail to take the data distribution of clients into consideration,which results in suboptimal global prototype learning and insufficient client model personalization capabilities.To address these issues,we propose a fair trainable prototype federated learning(FedFTP)algorithm,which employs a fair sampling training prototype(FSTP)mechanism and a hyperbolic space constraints(HSC)mechanism to enhance the fairness and effectiveness of prototype learning on the server in heterogeneous environments.Furthermore,a local prototype stable update(LPSU)mechanism is proposed as a means of maintaining personalization while promoting global consistency,based on contrastive learning.Comprehensive experimental results demonstrate that FedFTP achieves state-of-the-art performance in HtFL scenarios.展开更多
The process of ground vehicle dynamic gravimetry is inevitably affected by the carrier’s maneuvering acceleration,which makes the result contain a large amount of dynamic error.In this paper,we propose a dynamic erro...The process of ground vehicle dynamic gravimetry is inevitably affected by the carrier’s maneuvering acceleration,which makes the result contain a large amount of dynamic error.In this paper,we propose a dynamic error suppression method of gravimetry based on the high-precision acquisition of external velocity for compensating the horizontal error of the inertial plat-form.On the basis of platform gravity measurement,firstly,the dynamic performance of the system is enhanced by optimizing the horizontal damping network of the inertial platform and selecting its parameter.Secondly,an improved federal Kalman filtering algorithm and a fault diagnosis method are designed using strapdown inertial navigation system(SINS),odometer(OD),and laser Doppler velocimeter(LDV).Simulation validates that these methods can improve the accuracy and robustness of the external velocity acquisition.Three survey lines are selected in Tianjin,China,for the gravimetry experiments with different maneuvering levels,and the results demonstrate that after dynamic error suppression,the internal coincidence accuracies of smooth and uniform operation,obvious acceleration and deceleration operation,and high-dynamic operation are improved by 70.2%,73.6%,and 77.9%to reach 0.81 mGal,1.30 mGal,and 1.94 mGal,respectively,and the external coinci-dence accuracies during smooth and uniform operation are improved by 48.6%up to 1.66 mGal.It is shown that the pro-posed method can effectively suppress the dynamic error,and that the accuracy improvement increases with carrier maneuver-ability.However,the amount of residual error that can not be entirely eliminated increases as well,so the ground vehicle dynamic gravimetry should be maintained in the carrier for smooth and uniform operation.展开更多
To solve the problem of increased computation and communication costs caused by using homomorphic encryption(HE) to protect all gradients in traditional cryptographic aggregation(cryptoaggregation) schemes,a fast cryp...To solve the problem of increased computation and communication costs caused by using homomorphic encryption(HE) to protect all gradients in traditional cryptographic aggregation(cryptoaggregation) schemes,a fast crypto-aggregation scheme called RandomCrypt was proposed.RandomCrypt performed clipping and quantization to fix the range of gradient values and then added two types of noise on the gradient for encryption and differential privacy(DP) protection.It conducted HE on noise keys to revise the precision loss caused by DP protection.RandomCrypt was implemented based on a FATE framework,and a hacking simulation experiment was conducted.The results show that the proposed scheme can effectively hinder inference attacks while ensuring training accuracy.It only requires 45%~51% communication cost and 5%~23% computation cost compared with traditional schemes.展开更多
随着21世纪初欧盟禁止在动物饲料中使用抗生素,以及近期美国联邦食品和药物管理局（Federal Drug Administration,FDA）从美国市场注销舍砷药物Histostat Nitarsone,对疾病的新型控制策略的需求显而易见。自Histostat于2015年底退出市...随着21世纪初欧盟禁止在动物饲料中使用抗生素,以及近期美国联邦食品和药物管理局（Federal Drug Administration,FDA）从美国市场注销舍砷药物Histostat Nitarsone,对疾病的新型控制策略的需求显而易见。自Histostat于2015年底退出市场以来,美国家禽生产商发现由于组织滴虫病（又称黑头病）造成的家禽死亡率正在攀升。展开更多
A new nonlinear algorithm is proposed for strapdown inertial navigation system （SINS）/celestial navigation system （CNS）/global positioning system （GPS） integrated navigation systems. The algorithm employs a nonl...A new nonlinear algorithm is proposed for strapdown inertial navigation system （SINS）/celestial navigation system （CNS）/global positioning system （GPS） integrated navigation systems. The algorithm employs a nonlinear system error model which can be modified by unscented Kalman filter （UKF） to give predictions of local filters. And these predictions can be fused by the federated Kalman filter. In the system error model, the rotation vector is introduced to denote vehicle＇s attitude and has less variables than the quaternion. Also, the UKF method is simplified to estimate the system error model, which can both lead to less calculation and reduce algorithm implement time. In the information fusion section, a modified federated Kalman filter is proposed to solve the singular covariance problem. Specifically, the new algorithm is applied to maneuvering vehicles, and simulation results show that this algorithm is more accurate than the linear integrated navigation algorithm.展开更多
In order to take full advantage of federated filter in fault-tolerant design of integrated navigation system, the limitation of fault detection algorithm for gradual changing fault detection and the poor fault toleran...In order to take full advantage of federated filter in fault-tolerant design of integrated navigation system, the limitation of fault detection algorithm for gradual changing fault detection and the poor fault tolerance of global optimal fusion algorithm are the key problems to deal with. Based on theoretical analysis of the influencing factors of federated filtering fault tolerance, global fault-tolerant fusion algorithm and information sharing algorithm are proposed based on fuzzy assessment. It achieves intelligent fault-tolerant structure with two-stage and feedback, including real-time fault detection in sub-filters, and fault-tolerant fusion and information sharing in main filter. The simulation results demonstrate that the algorithm can effectively improve fault-tolerant ability and ensure relatively high positioning precision of integrated navigation system when a subsystem having gradual changing fault.展开更多
In order to improve the autonomous navigation capability of satellite,a pulsar/CNS（celestial navigation system） integrated navigation method based on federated unscented Kalman filter（UKF） is proposed.The celestia...In order to improve the autonomous navigation capability of satellite,a pulsar/CNS（celestial navigation system） integrated navigation method based on federated unscented Kalman filter（UKF） is proposed.The celestial navigation is a mature and stable navigation method.However,its position determination performance is not satisfied due to the low accuracy of horizon sensor.Single pulsar navigation is a new navigation method,which can provide highly accurate range measurements.The major drawback of single pulsar navigation is that the system is completely unobservable.As two methods are complementary to each other,the federated UKF is used here for fusing the navigation data from single pulsar navigation and CNS.Compared to the traditional celestial navigation method and single pulsar navigation,the integrated navigation method can provide better navigation performance.The simulation results demonstrate the feasibility and effectiveness of the navigation method.展开更多
基金supported by the National Natural Science Foundation of China(6067406960574056).
文摘To improve the agility, dynamics, composability, reusability, and development efficiency restricted by monolithic federation object model （FOM）, a modular FOM is proposed by high level architecture （HLA） evolved product development group. This paper reviews the state-of-the-art of HLA evolved modular FOM. In particular, related concepts, the overall impact on HLA standards, extension principles, and merging processes are discussed. Also permitted and restricted combinations, and merging rules are provided, and the influence on HLA interface specification is given. The comparison between modular FOM and base object model （BOM） is performed to illustrate the importance of their combination. The applications of modular FOM are summarized. Finally, the significance to facilitate compoable simulation both in academia and practice is presented and future directions are pointed out.
基金supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region(Grant No.2022D01B 187)。
文摘Federated learning(FL)is a distributed machine learning paradigm for edge cloud computing.FL can facilitate data-driven decision-making in tactical scenarios,effectively addressing both data volume and infrastructure challenges in edge environments.However,the diversity of clients in edge cloud computing presents significant challenges for FL.Personalized federated learning(pFL)received considerable attention in recent years.One example of pFL involves exploiting the global and local information in the local model.Current pFL algorithms experience limitations such as slow convergence speed,catastrophic forgetting,and poor performance in complex tasks,which still have significant shortcomings compared to the centralized learning.To achieve high pFL performance,we propose FedCLCC:Federated Contrastive Learning and Conditional Computing.The core of FedCLCC is the use of contrastive learning and conditional computing.Contrastive learning determines the feature representation similarity to adjust the local model.Conditional computing separates the global and local information and feeds it to their corresponding heads for global and local handling.Our comprehensive experiments demonstrate that FedCLCC outperforms other state-of-the-art FL algorithms.
基金supported by the Natural Science Foundation of Xinjiang Uygur Autonomous Region(No.2022D01B187).
文摘Heterogeneous federated learning(HtFL)has gained significant attention due to its ability to accommodate diverse models and data from distributed combat units.The prototype-based HtFL methods were proposed to reduce the high communication cost of transmitting model parameters.These methods allow for the sharing of only class representatives between heterogeneous clients while maintaining privacy.However,existing prototype learning approaches fail to take the data distribution of clients into consideration,which results in suboptimal global prototype learning and insufficient client model personalization capabilities.To address these issues,we propose a fair trainable prototype federated learning(FedFTP)algorithm,which employs a fair sampling training prototype(FSTP)mechanism and a hyperbolic space constraints(HSC)mechanism to enhance the fairness and effectiveness of prototype learning on the server in heterogeneous environments.Furthermore,a local prototype stable update(LPSU)mechanism is proposed as a means of maintaining personalization while promoting global consistency,based on contrastive learning.Comprehensive experimental results demonstrate that FedFTP achieves state-of-the-art performance in HtFL scenarios.
基金supported by the Shanxi Provincial Natural Science Basic Research Program Young Talent Project(S2019-JC-QN-2408).
文摘The process of ground vehicle dynamic gravimetry is inevitably affected by the carrier’s maneuvering acceleration,which makes the result contain a large amount of dynamic error.In this paper,we propose a dynamic error suppression method of gravimetry based on the high-precision acquisition of external velocity for compensating the horizontal error of the inertial plat-form.On the basis of platform gravity measurement,firstly,the dynamic performance of the system is enhanced by optimizing the horizontal damping network of the inertial platform and selecting its parameter.Secondly,an improved federal Kalman filtering algorithm and a fault diagnosis method are designed using strapdown inertial navigation system(SINS),odometer(OD),and laser Doppler velocimeter(LDV).Simulation validates that these methods can improve the accuracy and robustness of the external velocity acquisition.Three survey lines are selected in Tianjin,China,for the gravimetry experiments with different maneuvering levels,and the results demonstrate that after dynamic error suppression,the internal coincidence accuracies of smooth and uniform operation,obvious acceleration and deceleration operation,and high-dynamic operation are improved by 70.2%,73.6%,and 77.9%to reach 0.81 mGal,1.30 mGal,and 1.94 mGal,respectively,and the external coinci-dence accuracies during smooth and uniform operation are improved by 48.6%up to 1.66 mGal.It is shown that the pro-posed method can effectively suppress the dynamic error,and that the accuracy improvement increases with carrier maneuver-ability.However,the amount of residual error that can not be entirely eliminated increases as well,so the ground vehicle dynamic gravimetry should be maintained in the carrier for smooth and uniform operation.
基金Beijing Natural Science Foundation (L233005)National Key Research and Development Program of China (2023YFB3308200)。
文摘To solve the problem of increased computation and communication costs caused by using homomorphic encryption(HE) to protect all gradients in traditional cryptographic aggregation(cryptoaggregation) schemes,a fast crypto-aggregation scheme called RandomCrypt was proposed.RandomCrypt performed clipping and quantization to fix the range of gradient values and then added two types of noise on the gradient for encryption and differential privacy(DP) protection.It conducted HE on noise keys to revise the precision loss caused by DP protection.RandomCrypt was implemented based on a FATE framework,and a hacking simulation experiment was conducted.The results show that the proposed scheme can effectively hinder inference attacks while ensuring training accuracy.It only requires 45%~51% communication cost and 5%~23% computation cost compared with traditional schemes.
文摘随着21世纪初欧盟禁止在动物饲料中使用抗生素,以及近期美国联邦食品和药物管理局（Federal Drug Administration,FDA）从美国市场注销舍砷药物Histostat Nitarsone,对疾病的新型控制策略的需求显而易见。自Histostat于2015年底退出市场以来,美国家禽生产商发现由于组织滴虫病（又称黑头病）造成的家禽死亡率正在攀升。
基金supported by the National Natural Science Foundation of China (60535010)
文摘A new nonlinear algorithm is proposed for strapdown inertial navigation system （SINS）/celestial navigation system （CNS）/global positioning system （GPS） integrated navigation systems. The algorithm employs a nonlinear system error model which can be modified by unscented Kalman filter （UKF） to give predictions of local filters. And these predictions can be fused by the federated Kalman filter. In the system error model, the rotation vector is introduced to denote vehicle＇s attitude and has less variables than the quaternion. Also, the UKF method is simplified to estimate the system error model, which can both lead to less calculation and reduce algorithm implement time. In the information fusion section, a modified federated Kalman filter is proposed to solve the singular covariance problem. Specifically, the new algorithm is applied to maneuvering vehicles, and simulation results show that this algorithm is more accurate than the linear integrated navigation algorithm.
基金supported by the National Natural Science Foundationof China (60902055)
文摘In order to take full advantage of federated filter in fault-tolerant design of integrated navigation system, the limitation of fault detection algorithm for gradual changing fault detection and the poor fault tolerance of global optimal fusion algorithm are the key problems to deal with. Based on theoretical analysis of the influencing factors of federated filtering fault tolerance, global fault-tolerant fusion algorithm and information sharing algorithm are proposed based on fuzzy assessment. It achieves intelligent fault-tolerant structure with two-stage and feedback, including real-time fault detection in sub-filters, and fault-tolerant fusion and information sharing in main filter. The simulation results demonstrate that the algorithm can effectively improve fault-tolerant ability and ensure relatively high positioning precision of integrated navigation system when a subsystem having gradual changing fault.
基金supported by the National High Technology Research and Development Program of China(2006AAJ109)Aviation Science Fund(20070818001)
文摘In order to improve the autonomous navigation capability of satellite,a pulsar/CNS（celestial navigation system） integrated navigation method based on federated unscented Kalman filter（UKF） is proposed.The celestial navigation is a mature and stable navigation method.However,its position determination performance is not satisfied due to the low accuracy of horizon sensor.Single pulsar navigation is a new navigation method,which can provide highly accurate range measurements.The major drawback of single pulsar navigation is that the system is completely unobservable.As two methods are complementary to each other,the federated UKF is used here for fusing the navigation data from single pulsar navigation and CNS.Compared to the traditional celestial navigation method and single pulsar navigation,the integrated navigation method can provide better navigation performance.The simulation results demonstrate the feasibility and effectiveness of the navigation method.
