Proactive Content Caching for Internet-of-Vehicles based on Peer-to-Peer Federated Learning

Yu, Zhengxin; Hu, Jia; Min, Geyong; Han, Xu; Mills, Jed

doi:10.1109/icpads51040.2020.00083

Cited by 21 publications

(7 citation statements)

References 19 publications

(20 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…H. Wang, Muñoz‐González, et al (2021) proposed a novel peer‐to‐peer algorithm named P2PK‐SMOTE, able to train supervised anomaly detection ML models in non‐IID scenarios. A new proactive caching scheme based on peer‐to‐peer federated deep learning was proposed by Z. Yu et al (2020). The global prediction model is trained from data scattered at vehicles to mitigate the privacy risks.…”

Section: Network Topologymentioning

confidence: 99%

Towards federated learning: An overview of methods and applications

Silva

Vinagre

Gama

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

Federated learning (FL) is a collaborative, decentralized privacy-preserving method to attach the challenges of storing data and data privacy. Artificial intelligence, machine learning, smart devices, and deep learning have strongly marked the last years. Two challenges arose in data science as a result. First, the regulation protected the data by creating the General Data Protection Regulation, in which organizations are not allowed to keep or transfer data without the owner's authorization. Another challenge is the large volume of data generated in the era of big data, and keeping that data in one only server becomes increasingly tricky. Therefore, the data is allocated into different locations or generated by devices, creating the need to build models or perform calculations without transferring data to a single location. The new term FL emerged as a sub-area of machine learning that aims to solve the challenge of making distributed models with privacy considerations. This survey starts by describing relevant concepts, definitions, and methods, followed by an indepth investigation of federated model evaluation. Finally, we discuss three promising applications for further research: anomaly detection, distributed data streams, and graph representation.

show abstract

Section: Network Topologymentioning

confidence: 99%

Towards federated learning: An overview of methods and applications

Silva

Vinagre

Gama

2023

WIREs Data Min & Knowl

View full text Add to dashboard Cite

show abstract

“…Furthermore, the combination of FL and blockchain was used to improve data security and shorten the time it takes to upload data. Furthermore, References [137,138] also proposed a proactive caching scheme by leveraging FL in the edge computing paradigm. Edge devices typically have limited computational power and are resource constrained.…”

Section: Computation Offloading and Content Cachingmentioning

confidence: 99%

Federated Learning in Edge Computing: A Systematic Survey

Abreha

Hayajneh

Serhani

2022

Sensors

101

View full text Add to dashboard Cite

Edge Computing (EC) is a new architecture that extends Cloud Computing (CC) services closer to data sources. EC combined with Deep Learning (DL) is a promising technology and is widely used in several applications. However, in conventional DL architectures with EC enabled, data producers must frequently send and share data with third parties, edge or cloud servers, to train their models. This architecture is often impractical due to the high bandwidth requirements, legalization, and privacy vulnerabilities. The Federated Learning (FL) concept has recently emerged as a promising solution for mitigating the problems of unwanted bandwidth loss, data privacy, and legalization. FL can co-train models across distributed clients, such as mobile phones, automobiles, hospitals, and more, through a centralized server, while maintaining data localization. FL can therefore be viewed as a stimulating factor in the EC paradigm as it enables collaborative learning and model optimization. Although the existing surveys have taken into account applications of FL in EC environments, there has not been any systematic survey discussing FL implementation and challenges in the EC paradigm. This paper aims to provide a systematic survey of the literature on the implementation of FL in EC environments with a taxonomy to identify advanced solutions and other open problems. In this survey, we review the fundamentals of EC and FL, then we review the existing related works in FL in EC. Furthermore, we describe the protocols, architecture, framework, and hardware requirements for FL implementation in the EC environment. Moreover, we discuss the applications, challenges, and related existing solutions in the edge FL. Finally, we detail two relevant case studies of applying FL in EC, and we identify open issues and potential directions for future research. We believe this survey will help researchers better understand the connection between FL and EC enabling technologies and concepts.

show abstract

“…It causes reducing the cache hit ratio and the content delivery ratio. To efficiently download the whole content and to enhance the cache hit ratio, some studies [ 25 , 26 , 36 , 37 , 38 , 39 ] on mobility-based precaching that exploit the mobility of the requester vehicle have been researched in CCVNs. They proactively cache the same copy of the whole content (i.e., full precaching) for the requester vehicle to the next RSU where it arrives next by its mobility.…”

Section: Related Workmentioning

confidence: 99%

“…This strategy is suitable for improving cache utilization, hop ratios, and resolved interest ratios. Hu et al [ 39 ] proposed a Peer-to-Peer Federated learning-based proactive Caching scheme (PPFC) that is well suited to the highly dynamic IoV environments. Due to the heterogeneous abilities of vehicles, a dual-weighted model aggregation scheme is designed to reduce the effect of straggler vehicles in order to further improve the accuracy of the trained global model in the designed peer-to-peer FL.…”

Section: Related Workmentioning

confidence: 99%

“…However, since the full precaching schemes [ 25 , 26 , 36 , 37 , 38 , 39 ] precache the whole of an intended content for a requester vehicle to its next RSU, they have the problem to precache unnecessary parts of the content that the vehicle already downloaded in the current RSU or cannot download in the next RSU, because the vehicle can download only the limited amount in the next RSU due to its limited travel time in the RSU. Thus, many studies [ 27 , 28 , 29 , 30 , 31 , 40 , 41 ] have been addressed to solve this problem and proposed partial precaching schemes to precache only the amount that a requester vehicle can download in the communication coverage of the next RSU through the communication with it based on the speed information.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Adaptive Content Precaching Scheme Based on the Predictive Speed of Vehicles in Content-Centric Vehicular Networks

Nam

Choi

Shin

et al. 2021

Sensors

View full text Add to dashboard Cite

Content-Centric Vehicular Networks (CCVNs) are considered as an attractive technology to efficiently distribute and share contents among vehicles in vehicular environments. Due to the large size of contents such as multimedia data, it might be difficult for a vehicle to download the whole of a content within the coverage of its current RoadSide Unit (RSU). To address this issue, many studies exploit mobility-based content precaching in the next RSU on the trajectory of the vehicle. To calculate the amount of the content precaching, they use a constant speed such as the current speed of the vehicle requesting the content or the average speed of vehicles in the next RSU. However, since they do not appropriately reflect the practical speed of the vehicle in the next RSU, they could incorrectly calculate the amount of the content precaching. Therefore, we propose an adaptive content precaching scheme (ACPS) that correctly estimates the predictive speed of a requester vehicle to reflect its practical speed and calculates the amount of the content precaching using its predictive speed. ACPS adjusts the predictive speed to the average speed starting from the current speed with the optimized adaptive value. To compensate for a subtle error between the predictive and the practical speeds, ACPS appropriately adds a guardband area to the precaching amount. Simulation results verify that ACPS achieves better performance than previous schemes with the current or the average speeds in terms of the content download delay and the backhaul traffic overhead.

show abstract

Proactive Content Caching for Internet-of-Vehicles based on Peer-to-Peer Federated Learning

Cited by 21 publications

References 19 publications

Towards federated learning: An overview of methods and applications

Towards federated learning: An overview of methods and applications

Federated Learning in Edge Computing: A Systematic Survey

Adaptive Content Precaching Scheme Based on the Predictive Speed of Vehicles in Content-Centric Vehicular Networks

Contact Info

Product

Resources

About