اشتراك‌گذاري عادلانه، وارسي‌پذير و امن داده‌ها در بستر اينترنت اشيا با استفاده از زنجيره‌بلوكي

Fair, Auditable an‎d Secure Data Sharing in the Internet of Things Using Blockchain

رياضي كاربردي و علوم كامپيوتر

The Internet of Things (IoT) is one of the main infrastructures of todayʹs smart systems. By continuously generating data through sensors an‎d connected devices, it plays an important role in monitoring, decision-making, process optimization, an‎d improving efficiency. However, the economic an‎d operational value of such data makes their sharing, without appropriate security mechanisms, a serious challenge. In such environments, data are usually exchanged between producers an‎d customers through intermediary entities such as fog nodes, cloud servers, an‎d blockchain. Therefore, a suitable scheme must simultaneously guarantee properties such as data confidentiality, data authenticity, an‎d fairness in exchange. In recent years, several studies have been proposed to realize secure an‎d fair data sharing in decentralized settings. Among them, FairShare has attracted attention as a blockchain-based mechanism for data sharing in the Industrial Internet of Things. This scheme aims to simultaneously provide confidentiality, authenticity, an‎d fairness during the data-exchange process. However, the security analysis conducted in this thesis shows that the public verification mechanism in FairShare is not sufficient to guarantee fairness. More specifically, a malicious fog node can generate forged metadata during the data-recovery phase that successfully passes the public test of the protocol, while the customer, after decryption, does not obtain valid data an‎d is unable to prove this misbehavior. Therefore, the original scheme is vulnerable to a forgery attack by the fog node, an‎d its fairness claim does not hold in the considered threat model. To address this weakness, this thesis proposes an improved scheme called FairShare+. The proposed scheme redesigns the cryptographic metadata an‎d adds an intermediate validation phase performed by the customer, enabling the detection of malicious behavior by the fog node before the exchange is finalized. Moreover, unlike FairShare, FairShare+ is designed without using bilinear maps; consequently, it is computationally lighter an‎d more suitable for resource-constrained environments. The security analysis shows that, in the considered threat model, the proposed scheme provides data confidentiality, data authenticity, an‎d fairness against both the fog node an‎d the customer. Furthermore, the performance eva‎luation results show that FairShare+, while fixing the vulnerability of the original scheme, reduces the computational an‎d storage overheads by approximately 97% an‎d 74%, respectively, compared with FairShare.

6

161436