OWNERSHIP AND TRUST MANAGEMENT IN SOCIAL INTERNET OF THINGS BASED ON BLOCKCHAIN TECHNOLOGYMIHELJ, JERNEJ (Avtor) Sedlar, Urban (Mentor) Zhang, Yuan (Komentor) trustownershipreputationtrust and ownership managementsocial internet of thingsvehicular trafficIn vehicular traffic, we see a fast deployment of automated vehicles; manned vehicles and human actors involved in traffic are nevertheless going to continue to play important roles for quite some time to come. Therefore, the vehicular traffic domain is a great example of solutions that have to be designed with human behaviour in mind. Additionally, including people in a device-centered domain introduces additional uncertainties, nondeterminism, as well as a complex mix of automation, artificial intelligence and human-in-the-loop systems. In such environments, managing trust and reputation between all the entities involved is crucial. Often, people not only interact with devices, but also own and control them. When managed centrally, people lose direct ownership and control over their own devices as well as their device data privacy. Therefore, in certain scenarios, distributed solutions are preferred. Since these systems are of a distributed and decentralized nature, the proposed solution is designed to operate on a decentralized blockchain platform, thus eliminating a single central authority and increasing resilience to intentional and unintentional data manipulation. We compare and contrast such a decentralized design with a traditional centralized approach. To further evaluate the proposed solution, we designed a dedicated traffic event reporting system based on real-life traffic reports, mobility simulations and driving simulator study results. The results show that the proposed solution can accurately detect both events as well as manipulative behaviour even in resource-constrained decentralized blockchain implementation. Both centralized and decentralized solutions implemented the same deterministic algorithm. Since in the decentralized blockchain solution data is spread across mining nodes, the data requirements are higher in comparison to a centralized solution; however, in such cases, tamper-proof storage is provided. The proposed trust and reputation assessment solution was also assessed from the personal vehicle sharing perspective using interviews and an online survey. This part of the research coincided with the first wave of the COVID-19 pandemic in Slovenia. Because of this, it yielded interesting results, especially regarding social distancing recommendations and lockdown policies. The results give insights about the perception and attitude towards car-sharing. An important takeaway was that the majority of people would heavily rely on their existing social networks rather than on the driving capabilities and behaviour of potential vehicle user. This informed our further design of the user-centered social IoT (SIoT) solution where we proposed an ownership management system that provides support for users to gather existing human-layer social networks data and trust emerging from it, so they can use it to inform their decisions regarding borrowing and lending vehicles.20222022-09-22 08:15:00Doktorsko delo/naloga141008VisID: 43166COBISS_ID: 122731779sl