Sharing Security Information Between Resource Constrained Network Nodes

Challenges in resource constrained IoT networks are driven by the miniaturization of purpose-specific hardware and the imperative for cost-efficiency. These characteristics make traditional security solutions, which focus on securing individual devices, unrealistic. Instead, decentralized solutions that leverage the collective resources of a network must be considered. For these solutions, key enablers include self-configuration, scalability, resilience to extreme dynamics, and the overall sustainability and reliability of the network, suggesting the ad hoc paradigm, an attractive foundation for development. Therefore, the objective of this master’s thesis is to examine the feasibility of ad hoc networking in providing a reliable communication framework for sharing security information between resource constrained devices. The research consists of the examination of major challenges in resource constrained IoT systems and exploration of promising solutions from academic literature. It details various categories of ad hoc networking, key standards, and routing protocols. Additionally, a simulation artifact was developed to assess the feasibility of using ad hoc networking to distribute security information among resource constrained devices across different network sizes. The findings indicate that ad hoc networking is an efficient and scalable framework for sharing security information in terms of throughput, latency, and the overall reliability of the introduced security functionality. The key factors affecting these values could be identified as the physical topology of the network and the availability of existing routes between nodes. Furthermore, the alignment of ad hoc networking with academically proposed solutions was affirmed. The results from this thesis can provide a reference base for further steps in validating ad hoc networking as a suitable communication framework upon which more complex, decentralized security functionalities can be implemented.