With the support of edge computing,the synergy and collaboration among central cloud,edge cloud,and terminal devices form an integrated computing ecosystem known as the cloud-edge-client architecture.This integration ...With the support of edge computing,the synergy and collaboration among central cloud,edge cloud,and terminal devices form an integrated computing ecosystem known as the cloud-edge-client architecture.This integration unlocks the value of data and computational power,presenting significant opportunities for large-scale 3D scene modeling and XR presentation.In this paper,we explore the perspectives and highlight new challenges in 3D scene modeling and XR presentation based on point cloud within the cloud-edge-client integrated architecture.We also propose a novel cloud-edge-client integrated technology framework and a demonstration of municipal governance application to address these challenges.展开更多
Federated learning(FL)is a technology that allows multiple devices to collaboratively train a global model without sharing original data,which is a hot topic in distributed intelligent systems.Combined with satellite ...Federated learning(FL)is a technology that allows multiple devices to collaboratively train a global model without sharing original data,which is a hot topic in distributed intelligent systems.Combined with satellite network,FL can overcome the geographical limitation and achieve broader applications.However,it also faces the issues such as straggler effect,unreliable network environments and non-independent and identically distributed(Non-IID)samples.To address these problems,we propose an intelligent hierarchical FL system based on semi-asynchronous and scheduled synchronous control strategies in cloud-edge-client structure for satellite network.Our intelligent system effectively handles multiple client requests by distributing the communication load of the central cloud to various edge clouds.Additionally,the cloud server selection algorithm and the edge-client semi-asynchronous control strategy minimize clients’waiting time,improving the overall efficiency of the FL process.Furthermore,the center-edge scheduled synchronous control strategy ensures the timeliness of partial models.Based on the experiment results,our proposed intelligent hierarchical FL system demonstrates a distinct advantage in global accuracy over traditional FedAvg,achieving 2%higher global accuracy within the same time frame and reducing 52%training time to achieve the target accuracy.展开更多
基金the National Natural Science Foundation of China(U22B2034)the Fundamental Research Funds for the Central Universities(226-2022-00064).
文摘With the support of edge computing,the synergy and collaboration among central cloud,edge cloud,and terminal devices form an integrated computing ecosystem known as the cloud-edge-client architecture.This integration unlocks the value of data and computational power,presenting significant opportunities for large-scale 3D scene modeling and XR presentation.In this paper,we explore the perspectives and highlight new challenges in 3D scene modeling and XR presentation based on point cloud within the cloud-edge-client integrated architecture.We also propose a novel cloud-edge-client integrated technology framework and a demonstration of municipal governance application to address these challenges.
基金supported by Operations Services Department,China Satellite Network Application Co.,Ltd.
文摘Federated learning(FL)is a technology that allows multiple devices to collaboratively train a global model without sharing original data,which is a hot topic in distributed intelligent systems.Combined with satellite network,FL can overcome the geographical limitation and achieve broader applications.However,it also faces the issues such as straggler effect,unreliable network environments and non-independent and identically distributed(Non-IID)samples.To address these problems,we propose an intelligent hierarchical FL system based on semi-asynchronous and scheduled synchronous control strategies in cloud-edge-client structure for satellite network.Our intelligent system effectively handles multiple client requests by distributing the communication load of the central cloud to various edge clouds.Additionally,the cloud server selection algorithm and the edge-client semi-asynchronous control strategy minimize clients’waiting time,improving the overall efficiency of the FL process.Furthermore,the center-edge scheduled synchronous control strategy ensures the timeliness of partial models.Based on the experiment results,our proposed intelligent hierarchical FL system demonstrates a distinct advantage in global accuracy over traditional FedAvg,achieving 2%higher global accuracy within the same time frame and reducing 52%training time to achieve the target accuracy.
