Toward Intelligent Vehicular Networks: A Machine Learning Framework

@article{Liang2019TowardIV,
  title={Toward Intelligent Vehicular Networks: A Machine Learning Framework},
  author={Le Liang and Hao Ye and Geoffrey Y. Li},
  journal={IEEE Internet of Things Journal},
  year={2019},
  volume={6},
  pages={124-135}
}
As wireless networks evolve toward high mobility and providing better support for connected vehicles, a number of new challenges arise due to the resulting high dynamics in vehicular environments and thus motive rethinking of traditional wireless design methodologies. Future intelligent vehicles, which are at the heart of high mobility networks, are increasingly equipped with multiple advanced onboard sensors and keep generating large volumes of data. Machine learning, as an effective approach… 

Figures and Tables from this paper

Federated Learning for Vehicular Networks
TLDR
This article provides a comprehensive analysis of the usage of FL over ML in vehicular network applications to develop intelligent transportation systems and illustrates the superior performance of FLover ML in terms of data transmission complexity for vehicular object detection application.
A Survey on Multi-Agent Reinforcement Learning Methods for Vehicular Networks
TLDR
A survey on research issues related to vehicular networks such as resource allocation, data offloading, cache placement, ultra-reliable low latency communication (URLLC), and high mobility and the potential applications of MARL that enables decentralized and scalable decision making in vehicle-to-everything (V2X) scenarios are provided.
Comprehensive Survey of Machine Learning Approaches in Cognitive Radio-Based Vehicular Ad Hoc Networks
TLDR
An overview of ML, CR, VANET, and CR-VANET is presented, including their architectures, functions, challenges, and open issues, and insights into the use of ML for autonomous or driver-less vehicles are presented.
Distributed Estimation Framework for Beyond 5G Intelligent Vehicular Networks
TLDR
This paper proposes a distributed estimation framework to improve the scalability and robustness, in which vehicles can communicate wirelessly with other vehicles within the communication range, and introduces different consensus operations as a realization of this proposed framework.
How to Train your ITS? Integrating Machine Learning with Vehicular Network Simulation
TLDR
Veins-Gym combines the popular Veins vehicular networking simulator with OpenAI Gym and shows that RL techniques can be easily applied to ITSs with this framework, and enables an agent that outperformed hand-written algorithms.
Federated Machine Learning in Vehicular Networks: A summary of Recent Applications
TLDR
A detailed summary of recent FL applications in VNs is provided and gives insights on current research challenges, including resource management, performance optimization and applications based on VNs.
Deep-Learning-Based Wireless Resource Allocation With Application to Vehicular Networks
TLDR
The key motivations and roadblocks of using deep learning for wireless resource allocation with application to vehicular networks are discussed and the deep reinforcement learning approach to address resource allocation problems that are difficult to handle in the traditional optimization framework is highlighted.
Spectrum Sharing in Vehicular Networks Based on Multi-Agent Reinforcement Learning
TLDR
This paper investigates the spectrum sharing problem in vehicular networks based on multi-agent reinforcement learning and demonstrates that with a proper reward design and training mechanism, the multiple V2V agents successfully learn to cooperate in a distributed way to simultaneously improve the sum capacity of V2I links and payload delivery rate of V1V links.
Role of Machine Learning in Resource Allocation Strategy over Vehicular Networks: A Survey
TLDR
A thorough understanding of how machine learning algorithms are utilized to offer a dynamic resource allocation in vehicular networks is provided in this study.
Deep Reinforcement Learning-aided Transmission Design for Multi-user V2V Networks
TLDR
This paper proposes transforming the power/rate control problem into a Markov decision process and then solving it using the deep deterministic policy gradient (DDPG) algorithm, and results show that in a two-user network the DRL-aided solution can achieve better performance than that with Lyapunov optimization.
...
...

References

SHOWING 1-10 OF 92 REFERENCES
Machine Learning for Vehicular Networks
TLDR
Recent advances in applying machine learning in vehicular networks are reviewed and several open issues that warrant further research are highlighted.
Machine Learning for Vehicular Networks: Recent Advances and Application Examples
TLDR
Recent advances in applying machine learning in vehicular networks are reviewed and an attempt is made to bring more attention to this upcoming area.
User Association for Load Balancing in Vehicular Networks: An Online Reinforcement Learning Approach
TLDR
An online reinforcement learning approach for network load balancing in vehicular networks, called ORLA, which can obtain a good association solution through learning from the dynamic vehicular environment continually and has higher quality load balancing compared with other popular association methods.
Deep reinforcement learning-based scheduling for roadside communication networks
TLDR
This paper presents a deep reinforcement learning model, namely deep Q-network, that learns an energy-efficient scheduling policy from high-dimensional inputs corresponding to the characteristics and requirements of vehicles residing within a RoadSide Unit's (RSU) communication range.
Machine Learning Paradigms for Next-Generation Wireless Networks
TLDR
The goal is to assist the readers in refining the motivation, problem formulation, and methodology of powerful machine learning algorithms in the context of future networks in order to tap into hitherto unexplored applications and services.
Integrated Networking, Caching, and Computing for Connected Vehicles: A Deep Reinforcement Learning Approach
TLDR
This paper proposes an integrated framework that can enable dynamic orchestration of networking, caching, and computing resources to improve the performance of next generation vehicular networks and formulate the resource allocation strategy in this framework as a joint optimization problem.
D2D for Intelligent Transportation Systems: A Feasibility Study
TLDR
For the first time, a feasibility study of D2D for ITS is carried out based on both the features of D1D and the nature of vehicular networks to demonstrate the promising potential of this technology and propose novel remedies necessary to make D 2D technology practical as well as beneficial for ITS.
A Reinforcement Learning Technique for Optimizing Downlink Scheduling in an Energy-Limited Vehicular Network
TLDR
A reinforcement learning technique, i.e., protocol for energy-efficient adaptive scheduling using reinforcement learning (PEARL), is proposed for the purpose of optimizing the RSU's downlink traffic scheduling during a discharge period to guarantee the operation of the vehicular network during the discharge cycle while fulfilling the largest number of service requests.
Fuzzy Q-learning based vertical handoff control for vehicular heterogeneous wireless network
TLDR
A fuzzy Q-learning based vertical handoff (FQVH) control strategy for supporting the mobility management that has online learning ability and can give optimal handoff decisions adaptively with no need for prior knowledge on handoff behavior is proposed.
Reinforcement learning for resource provisioning in the vehicular cloud
TLDR
The benefits of reinforcement-learning-based techniques for resource provisioning in the vehicular cloud are shown and the learning techniques can perceive long-term benefits and are ideal for minimizing the overhead of resource Provisioning for vehicular clouds.
...
...