Service selection middleware using blockchain for IoT applications

Abstract

In today’s world, the Internet of Things (IoT) has ushered in an unprecedented revolution in information technology and is used in many industries, including retail, manufacturing and healthcare. Each individual IoT application is different when it comes to functional and nonfunctional requirements, defined by the application owner or industry. Service provider selection in a decentralized environments have become a critical issue for different IoT applications, in particular when services providers having similar functionality but different Quality of Service (QoS). As a result, selecting a high quality service that best suits IoT application requirements from a large list of functionally equivalent services is a challenging task. Furthermore, it is challenging to secure the private and business-critical information of the IoT application, as the sharing of service parameters with third-party solutions or even with the service provider itself can create serious security and privacy concerns, since the service requirements could contain sensitive information regarding the application, that could be misused or sold by the third-party or service provider. Therefore, a decentralized service selection mechanisms are required to select the best service provider for IoT applications, while ensuring end-to-end security and privacy of the associated sensitive data. In this thesis, we first consider an IoT data storage selection problem, which focuses on a middleware design for blockchain-based intelligent data storage selection for large-scale IoT applications. The proposed framework extends the current IoT cloud architecture and considers peer-to-peer (P2P) and blockchain-based storage solutions along with cloud and multi-cloud technologies. We model IoT data storage selection problem as a decision optimization problem and propose two polynomial-time algorithms as a solution. We also propose a verifiable blockchain-based storage selection protocol, which enables the IoT applications as well as the storage technologies to verify the correctness of the data placement decision made by the middleware design without the need of any trusted third party. Finally, we propose an intelligent maintenance strategy, which takes into account and learns the dynamically evolving features of the IoT applications service requirements to optimizes the computational complexity along with the blockchain storage and transactions overhead in the middleware design. The second aspect of the work in the thesis is the development of a blockchain-based efficient and secure charging station (CS) selection protocol for electric vehicles (EV) charging networks. We propose a decentralized blockchain-based EV charging architecture and theoretically model a decentralized decision optimization problem, which eliminates the need of any trusted third party and enables the EVs and the CSs to communicate in a decentralized manner, through the smart contracts, running on a blockchain network. It also enables EVs to select a CS and make a remote reservation with a CS without sharing any private information with CSs or any central management entity. The third part of this thesis extends the second aspect of this thesis, and concerns the security and privacy problems arisen by the linkability of the public blockchain addresses with the EV owner’s physical identity. To solve this problem, we propose a blockchain-based end-to-end privacy-preserving CS reservation protocol, which enables EVs to reserve a charging slot at the selected CS privately, without sharing their private information and by dissociating its real identity from the blockchain address, thereby preserving the EV’s privacy. As the information provided by CS cannot be trusted, we propose an SMC-based CS information verification protocol that allows EVs to collaboratively verify the availability of charging slots from the CS by securely sharing reservations in an untrusted environment. Finally, we propose a smart contract design based on time-lock deposit protocols, which aggregates the balance in such a way that no CS is able to link charging service payments with EV users’ blockchain addresses, while still ensuring that CSs receive full payment for the services they provide. The major contribution of the thesis is providing a secure and efficient service selection mechanisms tailored for different IoT applications with different service requirements and security needs. With the proposed approach consumers will be able to securely select a suitable service provider based on their requirements. Experimental results showed that proposed approaches achieved better performance and efficiency compared to state-of-the-art.

Date	18 Aug 2022
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Kaiwen Zhang (Supervisor) & Hans Arno Jacobsen (Co-supervisor)