Open Access

ARTICLE

A Trusted Edge Resource Allocation Framework for Internet of Vehicles

Yuxuan Zhong¹, Siya Xu¹, Boxian Liao¹, Jizhao Lu², Huiping Meng², Zhili Wang¹, Xingyu Chen^1,*, Qinghan Li³

1 State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing, 100876, China
2 State Grid Henan Information & Telecommunication Company, Zhengzhou, 450018, China
3 School of Information Studies, Syracuse University, Syracuse, 13244, USA

* Corresponding Author: Xingyu Chen. Email:

Computers, Materials & Continua 2023, 77(2), 2629-2644. https://doi.org/10.32604/cmc.2023.035526

Received 24 August 2022; Accepted 08 December 2022; Issue published 29 November 2023

Abstract

With the continuous progress of information technique, assisted driving technology has become an effective technique to avoid traffic accidents. Due to the complex road conditions and the threat of vehicle information being attacked and tampered with, it is difficult to ensure information security. This paper uses blockchain to ensure the safety of driving information and introduces mobile edge computing technology to monitor vehicle information and road condition information in real time, calculate the appropriate speed, and plan a reasonable driving route for the driver. To solve these problems, this paper proposes a trusted edge resource allocation framework for assisted driving service, which includes two stages: the blockchain generation stage (the first stage) and assisted driving service stage (the second stage). Furthermore, in the first stage, a delay-and-throughput-oriented block generation model for the mobile terminal is designed. In the second stage, a balanced offloading algorithm for assisted driving service based on edge collaboration is proposed to solve the problems of unbalanced load of cluster mobile edge computing (MEC) servers and low resource utilization of the system. And this paper optimizes the throughput of blockchain and delay of the transportation network through deep reinforcement learning (DRL) algorithm. Finally, compared with joint computation and communication resources’ allocation (JCCR) and resource allocation method based on binary offloading (RAB), our proposed scheme can optimize the delay by 7.4% and 26.7%, and support various application services of the vehicular networks more effectively.

---

Keywords

---