Skip to main content
main-content

Über dieses Buch

This book introduces the Internet access for vehicles as well as novel communication and computing paradigms based on the Internet of vehicles.

To enable efficient and reliable Internet connection for mobile vehicle users, this book first introduces analytical modelling methods for the practical vehicle-to-roadside (V2R) Internet access procedure, and employ the interworking of V2R and vehicle-to-vehicle (V2V) to improve the network performance for a variety of automotive applications.

In addition, the wireless link performance between a vehicle and an Internet access station is investigated, and a machine learning based algorithm is proposed to improve the link throughout by selecting an efficient modulation and coding scheme.

This book also investigates the distributed machine learning algorithms over the Internet access of vehicles. A novel broadcasting scheme is designed to intelligently adjust the training users that are involved in the iteration rounds for an asynchronous federated learning scheme, which is shown to greatly improve the training efficiency. This book conducts the fully asynchronous machine learning evaluations among vehicle users that can utilize the opportunistic V2R communication to train machine learning models.

Researchers and advanced-level students who focus on vehicular networks, industrial entities for internet of vehicles providers, government agencies target on transportation system and road management will find this book useful as reference. Network device manufacturers and network operators will also want to purchase this book.

Inhaltsverzeichnis

Frontmatter

Chapter 1. Introduction of Internet Access of Vehicular Networks

Abstract
Connected vehicles are changing the modern transportation. Based on the wireless communication between vehicles with sophisticated radio interfaces, vehicles in the mobility world can exchange information with neighbors as well as remote transportation center, which can enable the vehicle to understand both the in-vehicle status and the road situation. Based on such capability, a lot of smart road applications can be realized, including the safety-related application, intelligent transportation system (ITS) and in-vehicle infotainment, etc. The Internet access for vehicles can further extend the spatial scope and temporal range of the vehicular communication, which can help all road users to conduct both the long-term evaluation and short-time response to all situations. In this chapter, we first introduce the overview of Internet of vehicles (IoV), then we present the Internet access procedure for a vehicle to connect to a wireless access station that deployed along the roadside. We then explain the aim of the book, covering the topic of Internet access performance evaluation, data traffic offloading, Internet link management and intelligent machine learning (ML) paradigm over IoV.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen

Chapter 2. Internet Access Modeling for Vehicular Connection

Abstract
In this chapter, we focus on the analytical modeling of the Internet access procedure for vehicles. Specifically, the Markov chain model is applied to describe the management frame exchange between the vehicle and the roadside access point. Due to the non-negligible overhead, the access delay is analyzed, which can determine the overall data throughput that can be achieved by the drive-by vehicle. Such access delay and throughput performance is crucial for future IoV network protocol design. We have demonstrated the accuracy of our analysis via both simulation and experimental verification methods.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen

Chapter 3. V2X Interworking via Vehicular Internet Access

Abstract
In this chapter, we considered to utilize the V2V and V2R communication simultaneously to improve the Internet access performance. To strike a balance between the throughput and the time delay for a data tasks, we propose queueing based theory to analyze the tradeoff between the data volume that can be transferred and the time needed. We have analyzed the unique characteristics of the V2V and V2R communication, and apply the M/G/1/K queue to analyze their property in fulfilling the data tasks, which is used to conduct the interworking of V2V and V2R communication for the vehicular offloading, which is shown to significantly improve the Internet access performance.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen

Chapter 4. Intelligent Link Management for Vehicular Internet Access

Abstract
In this chapter, we focus on the link layer between a vehicle and an access point. Due to the high mobility of vehicles, the channel condition is highly dynamic due to the varying path loss, shadowing, multi-path fading, etc., which requires an accurate tracking for the link capacity for the proper rate selection for the data packets sent to the air. Traditional model-based link management schemes cannot well follow the various channel variation patterns, and can lead to significantly performance loss especially in high mobility conditions. We propose to utilize multiple model-free ML enabled intelligent schemes that can learn from the previous network trace, which can provide efficient link prediction and near-optimal link rate choice. We analyze different spectrum access for vehicles, including unlicensed bands, TVWS, etc. And more categories of vehicles, e.g., maritime vehicles, flying vehicles, etc., are considered to evaluate the performance gain via the intelligent link management based on machine learning algorithms.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen

Chapter 5. Intelligent Networking enabled Vehicular Distributed Learning

Abstract
Mobile computing has emerged as an important paradigm to envision the ‘last mile’ of computing services to mobile users. There are many novel distributed computing methods which can be applied for vehicle users to let them cooperatively train ML models for future AI applications, However, traditional centralized training methods are not suitable for vehicle users since they are not connected by a reliable and bandwidth-rich Internet access, which is highly dynamic and often suffers a lot from interruption, interference, etc. In this chapter, we exploit two learning paradigms and analyze their performance based on the IoV connectivity and exploit the vehicles’ mobility. Due to the specific mobility pattern and communication characteristics in IoV, excessive training latency can be caused by the communication bandwidth constraints in vehicular environments, non-negligible volumes of iteration parameters and heterogeneity in computing capacities of distributed workers, etc. We propose novel computing methods to seek the possibility to provision artificial intelligence (AI) to the mobility world with the help of IoV, which has great potential to bring the power of AI to all road users, to support a variety of intelligent applications, e.g., autonomous driving, road safety, ITS, etc.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen

Chapter 6. Conclusions and Future Workers

Abstract
In this chapter, we conclude the monograph with the overall summary, and provide some promising directions regarding the Internet access of vehicles and novel architectures for further development of new generation IoV Internet access mechanisms.
Wenchao Xu, Haibo Zhou, Xuemin (Sherman) Shen
Weitere Informationen

Premium Partner