Initial trust establishment for personal space IoT systems

Publication Type:
Thesis
Issue Date:
2019
Full metadata record
Internet of Things (IoT) is becoming a reality with innovative applications, and IoT platforms have been developed to transfer technologies from research to business solutions. With IoT applications, we have greater control over personal devices and achieve more insights into the resource consumption habits; business processes can be streamlined; people are also better connected to each other. Despite the benefits derived from the IoT systems, users are concerned about the trustworthiness of their collected data and offered services. Security controls can prevent user’s data from being compromised during transmission, storage or unauthorized access, but do not provide a guarantee against the misbehaved devices that report incorrect information and poor services or avoid conducting a common task. Establishing trust relationship among devices and continuously monitoring their trust is the key to guarantee a reliable IoT system and hence mitigate user’s concerns. In this dissertation, we propose and investigate a novel initial trust establishment architecture for personal space IoT systems. In the initial trust establishment architecture, we propose a trust evidence generation module based on a challenge-response mechanism to generate the trust evidence relying on the device’s responses to the challenges, a trust knowledge assessment module to obtain the knowledge about the device from the generated trust evidence, and a trust evaluation scheme to quantify the initial trust level of the devices. We design and investigate a challenge-response information design to determine feasible designs of the challenge-response mechanism that ensure meaningful and related trust knowledge about the device’s trustworthiness captured from the challenge-response operations. A new trust-aware communication protocol is designed and implemented by incorporating the proposed initial trust establishment architecture into existing Bluetooth Low Energy (BLE) protocol to demonstrate the feasibility and efficiency of the proposed initial trust establishment architecture in practice. In this work, we first study building blocks and possible architectures of the IoT and analyze key requirements of an IoT system. Based on the analysis, we identify the critical role of the initial trust establishment model and the challenges of establishing initial trust in IoT systems due to the lack of knowledge for the trust assessment to work. To address the challenges, we propose a novel initial trust establishment architecture that can generate trust evidence for assessing the initial trust level of new devices by conducting challenge-response operations within a limited time window before they are admitted to the system. We propose three new initial trust establishment models based on the proposed architecture. An implicit relationship between the responses and the challenges is assumed for the system to judge the initial trustworthiness of the devices. The first model assesses the initial trust value based on a probability associated with the device’s behavior captured from the challenge-response process. The second model investigates the initial trust value based on a binary outcome set, and the third model quantifies the initial trust level based on a multiple-component outcome set from the challenge-response process. Subsequently, we propose the challenge-response information design where the challenge-response process is investigated and designed to determine the information space of the challenger’s view on its environment so that the challenge can invite relevant responses from the target environment. Based on the design of the challenge-response mechanism, the system can capture meaningful trust knowledge about the devices from challenge-response operations at their admission phase. We finally design and implement the initial trust-aware BLE protocol which incorporates the proposed initial trust establishment architecture into the existing BLE protocol. The simulation results show the efficiency, feasibility, and dependability of using initial trust-aware BLE protocol for building a trustworthy personal space IoT systems. The novelty of this research lies in assessing the devices’ initial trust level within a limited time window, before their admission to the personal space IoT system, without requiring prior experience or recommendations. The major contribution of this thesis is that it helps the IoT business solution providers to build secure and trustworthy IoT systems by admitting dependable devices, monitoring the trust of admitted devices, detecting maligned devices, and building long-term trust among. As a result, it mitigates the user’s concerns about the trustworthiness of IoT systems and encourages broader adoption of IoT applications.
Please use this identifier to cite or link to this item: