Task Offloading for Vehicular Edge Computing Based on Improved Hotstuff under Parking Assistance

TL;DR

This paper introduces a blockchain-based vehicular edge computing framework that improves security, reliability, and efficiency of task offloading in parking-assisted vehicular environments through novel algorithms and game-theoretic models.

Contribution

It proposes a new BPVEC framework with a CDS-based Hotstuff consensus enhancement and a Stackelberg game model for optimized offloading and pricing strategies.

Findings

Enhanced blockchain reliability in vehicular edge computing.

Improved security and trustworthiness of task offloading.

Maximized system revenue through optimized algorithms.

Abstract

Parked-assisted vehicular edge computing (PVEC) fully leverages communication and computing resources of parking vehicles, thereby significantly alleviating the pressure on edge servers. However, resource sharing and trading for vehicular task offloading in the PVEC environment usually occur between untrustworthy entities, which compromises the security of data sharing and transactions by vehicles and edge devices. To address these concerns, blockchain is introduced to provide a secure and trustworthy environment for offloading and transactions in PVEC. Nevertheless, due to the mobility of the vehicles, the processes of computing offloading and blockchain transactions are interrupted, which greatly reduces the reliability of the blockchain in edge computing process. In this paper, we propose a blockchain-based PVEC (BPVEC) offloading framework to enhance the security and reliability of the task offloading and transaction. Specifically, a consensus node selection algorithm based on the connected dominating set (CDS) is designed to improve the Hotstuff consensus according to parking time, computing capability and communication quality, which enhances blockchain reliability in computing offloading and transactions. Meanwhile, a Stackelberg game model, establishing the roadside units (RSUs) and parking vehicles (PVs) as leaders and the requesting vehicles (RVs) as follower, is utilized to optimize the offloading strategy and pricing. Subsequently, a BPVEC offloading strategy algorithm with gradient descent method is designed to maximize system revenue. Simulation results show that the proposed BPVEC offloading scheme is secure and reliable while ensuring maximum benefits.