A Survey of Collaborative Machine Learning Using 5G Vehicular Communications

Balkus, Salvador V; Wang, Honggang; Cornet, Brian D; Mahabal, Chinmay; Ngo, Hau; Fang, Hua

doi:10.1109/comst.2022.3149714

Cited by 64 publications

(24 citation statements)

References 169 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These devices can collect tons of data that can be utilized to analyze and solve various networking challenges. Various ML algorithms can be useful for these operations [22]. In the traditional centralized ML approaches, the data collected by the devices are required to be transmitted towards a centralized node equipped with more powerful processing capabilities.…”

Section: Distributed Machine Learning Algorithmsmentioning

confidence: 99%

Overview of Distributed Machine Learning Techniques for 6G Networks

2022

View full text Add to dashboard Cite

The main goal of this paper is to survey the influential research of distributed learning technologies playing a key role in the 6G world. Upcoming 6G technology is expected to create an intelligent, highly scalable, dynamic, and programable wireless communication network able to serve many heterogeneous wireless devices. Various machine learning (ML) techniques are expected to be deployed over the intelligent 6G wireless network that provide solutions to highly complex networking problems. In order to do this, various 6G nodes and devices are expected to generate tons of data through external sensors, and data analysis will be needed. With such massive and distributed data, and various innovations in computing hardware, distributed ML techniques are expected to play an important role in 6G. Though they have several advantages over the centralized ML techniques, implementing the distributed ML algorithms over resource-constrained wireless environments can be challenging. Therefore, it is important to select a proper ML algorithm based upon the characteristics of the wireless environment and the resource requirements of the learning process. In this work, we survey the recently introduced distributed ML techniques with their characteristics and possible benefits by focusing our attention on the most influential papers in the area. We finally give our perspective on the main challenges and advantages for telecommunication networks, along with the main scenarios that could eventuate.

show abstract

Section: Distributed Machine Learning Algorithmsmentioning

confidence: 99%

Overview of Distributed Machine Learning Techniques for 6G Networks

2022

View full text Add to dashboard Cite

show abstract

“…Therefore, associations like 5GAA and leading research partners have acknowledged that 5G-V2X will be the leading technology for the advanced automated driving communication for UAVs, and that the present LTE-V2X and ITS-G5 technologies need to be considered for basic safety use cases and thus not for autonomous vehicles [2,11,20,24]. One of the reasons 5G-V2X will be a game changer for vehicular communication is due to its, low latency, extreme throughput, enhanced reliability which are requirements for many aspects of autonomous driving and will allow UAVs to share real-time data [20].…”

Section: Next Generation Communication Technologiesmentioning

confidence: 99%

A proposal on a Connected Automated Mobility (CAM) communication system for (U)AVs

Charpentier

Slamnik-Kriještorac

Márquez-Barja

et al. 2022

Proceedings of the 2022 ACM Conference on Information Technology for Social Good

View full text Add to dashboard Cite

The next generation of vehicular communications has the potential to interconnect Unmanned Automated Vehicles (UAVs), which are the level 5 autonomous vehicles, with its nearby surroundings including Vulnerable Road Users (VRUs), other UAVs, and roadside infrastructure. Given the increased interest in the demand of autonomous driving and Cooperative Connected and Automated Mobility (CCAM), the future UAVs will be safer through enabling communication between UAVs themselves and with their surroundings e.g., VRUs. Communication between UAVs and VRUs will be needed since VRUs will not be able to make a physical eye-contact with the UAVs since their is no physical driver anymore behind the steering wheel. In this way the VRUs can make themselves digitally aware in the UAV traffic and vice versa. Such that UAVs will not be isolated black boxes, i.e., operating and relying fully on their embedded sensors, to its surrounding UAVs and VRUs. This paper shows the current and future needs for autonomous vehicular communication between UAVs and VRUs. Therefore, we present the current trends in the vehicular communication domain and its research challenges. We propose a solution to enable communication between semi-autonomous vehicles, UAVs and their surrounding e.g., VRUs. In our paper, we provide also an early initial proposal to fuse the Vehicle-to-everything (V2X) communication with the embedded autonomous software in the autonomous vehicle.

show abstract

“…For example, the perception accuracy of a single vehicle is often limited by the on-board sensor performance, such as perceive view and scope. There has been suggestions for connected autonomous vehicles (CAVs) [1], [2] to overcome this problem through vehicle-mounted wireless communication technology, such as vehicle-to-vehicle (V2V) communication [3], [4]. Specifically, CAVs use high-definition cameras and other sensors to capture and share images within their own scopes, alleviating object collisions and occlusion problems due to poor views [5], [6].…”

Section: Introductionmentioning

confidence: 99%

Achieving Lightweight and Privacy-Preserving Object Detection for Connected Autonomous Vehicles

Xiong

Tian

et al. 2023

IEEE Internet Things J.

View full text Add to dashboard Cite

Connected autonomous vehicles (CAVs) are capable of capturing high-definition images from onboard sensors, which can be used to facilitate the detection of objects in the vicinity. Such images may, however, contain sensitive information (e.g., human faces and license plates) as well as the indirect location of CAVs. To protect the object privacy of images shared by CAVs, this paper proposes a privacy-preserving object detection (P2OD) framework. Specifically, we propose multiple secure computing protocols designed to construct a privacy-preserving Faster R-CNN model to securely extract features and boundingboxes of objects in an image. By leveraging edge computing (with higher performance computation and lower latency, in comparison to cloud-based solutions), CAVs randomly split the captured images and upload them to two non-collusive edge servers. Both servers will then perform the P2OD framework cooperatively to directly detect objects over random image shares without exposing sensitive information. The theoretical analysis demonstrates the security, correctness and efficiency of the P2OD framework, and the experimental findings show that the P2OD framework can effectively protect the classification and location privacy of image objects for CAVs. Compared with the original Faster R-CNN model, the classification and regression errors of the P2OD framework can be controlled within 10 −12 and 10 −14 , respectively.

show abstract

A Survey of Collaborative Machine Learning Using 5G Vehicular Communications

Cited by 64 publications

References 169 publications

Overview of Distributed Machine Learning Techniques for 6G Networks

Overview of Distributed Machine Learning Techniques for 6G Networks

A proposal on a Connected Automated Mobility (CAM) communication system for (U)AVs

Achieving Lightweight and Privacy-Preserving Object Detection for Connected Autonomous Vehicles

Contact Info

Product

Resources

About