Combinatorial Resources Auction in Decentralized Edge-Thing Systems Using Blockchain and Differential Privacy

TL;DR

This paper proposes a blockchain-based combinatorial double auction mechanism with differential privacy for edge-thing systems, enhancing resource allocation fairness, security, and revenue in IoT edge computing.

Contribution

It introduces a novel differentially private auction mechanism using blockchain and smart contracts to secure and optimize resource allocation in decentralized edge-thing systems.

Findings

Ensures approximate truthfulness and high revenue.

Guarantees privacy security against inference attacks.

Validated effectiveness through theoretical analysis and simulations.

Abstract

With the continuous expansion of Internet of Things (IoT) devices, edge computing mode has emerged in recent years to overcome the shortcomings of traditional cloud computing mode, such as high delay, network congestion, and large resource consumption. Thus, edge-thing systems will replace the classic cloud-thing/cloud-edge-thing systems and become mainstream gradually, where IoT devices can offload their tasks to neighboring edge nodes. A common problem is how to utilize edge computing resources. For the sake of fairness, double auction can be used in the edge-thing system to achieve an effective resource allocation and pricing mechanism. Due to the lack of third-party management agencies and mutual distrust between nodes, in our edge-thing systems, we introduce blockchains to prevent malicious nodes from tampering with transaction records and smart contracts to act as an auctioneer to realize resources auction. Since the auction results stored in this blockchain-based system are transparent, they are threatened with inference attacks. Thus in this paper, we design a differentially private combinatorial double auction mechanism by exploring the exponential mechanism such that maximizing the revenue of edge computing platform, in which each IoT device requests a resource bundle and edge nodes compete with each other to provide resources. It can not only guarantee approximate truthfulness and high revenue, but also ensure privacy security. Through necessary theoretical analysis and numerical simulations, the effectiveness of our proposed mechanisms can be validated.