Two-Layer Computing Resource Management for Blockchain-Enabled Internet of Vehicles

Weibo Qin,Yuanfeng Li,Haipeng Yao,Shan Huang,Tianle Mai,Yunjie Liu
DOI: https://doi.org/10.1109/tvt.2024.3390690
2024-01-01
Abstract:With the growing prominence of the Internet of Vehicles (IoV), there is an unprecedented generation of vehicular data, which harbors immense potential for diverse applications. Considering the convergence of blockchain technology and IoV, blockchain can be used as a distributed ledger to establish a decentralised and autonomous trading platform for IoV networks. However, the limitations in the power and computational capacities of IoV devices hinder their direct involvement in the Proof-of-Work (PoW) process. Addressing this challenge, offloading the computational demands of PoW to Mobile Edge Computing (MEC) servers emerges as a feasible solution. This scenario has prompted us to conceptualize the computational resource management in blockchain consensus as a two-layered resource game, aimed at optimizing both the profits of computing power providers and the utility for vehicles. In the Blockchain Market Layer subgame, we employ a game-theoretic framework to analyze the interactions between MEC servers and vehicles within a PoW-based blockchain network. To optimize the system utility, we introduce a centralized evolutionary game-based pool selection algorithm. Simultaneously, in the Edge Cloud Network Layer subgame, a reverse auction algorithm is implemented to align the computing power supply with the demand within the blockchain market. Here, the cloud data center functions as the seller of computing power, while the edge servers aggregate the computing demand from IoV devices. A broker is then tasked with the allocation and pricing of computing resources, in accordance with the bids, ensuring adherence to economic principles of authenticity and individual rationality. Simulation results demonstrate that our mechanism enhances system efficiency while ensuring revenue, thereby enabling an effective distribution of computing resources between vehicles and cloud data centers.
What problem does this paper attempt to address?