The Internet of Things (IoT) is a technology aimed at developing a global network
of machines and devices that can interact and communicate with each other.
Supporting IoT, therefore, requires revisiting the Internet's best e ort service model
and reviewing its complex communication patterns.
In this dissertation, we explore the unique characteristics of IoT tra c and examine
IoT systems. Our work is motivated by the new capabilities o ered by modern
Software De ned Networks (SDN) and blockchain technology. We evaluate IoT Quality
of Service (QoS) in traditional networking. We obtain mathematical expressions
to calculate end-to-end delay, and dropping. Our results provide insight into the
advantages of an intelligent edge serving as a detection mechanism. Subsequently,
we propose SADIQ, SDN-based Application-aware Dynamic Internet of things QoS.
SADIQ provides context-driven QoS for IoT applications by allowing applications to
express their requirements using a high-level SQL-like policy language. Our results
show that SADIQ improves the percentage of regions with an error in their reported
temperature for the Weather Signal application up to 45 times; and it improves the
percentage of incorrect parking statuses for regions with high occupancy for the Smart
Parking application up to 30 times under the same network conditions and drop rates.
Despite centralization and the control of data, IoT systems are not safe from
cyber-crime, privacy issues, and security breaches. Therefore, we explore blockchain
technology. In the context of IoT, Byzantine fault tolerance-based consensus protocols are used. However, the blockchain consensus layer contributes to the most remarkable
performance overhead especially for IoT applications subject to maximum delay
constraints. In order to capture the unique requirements of the IoT, consensus mechanisms
and block formation need to be redesigned. To this end, we propose Synopsis, a
novel hierarchical blockchain system. Synopsis introduces a wireless-optimized Byzantine
chain replication protocol and a new probabilistic data structure. The results
show that Synopsis successfully reduces the memory footprint from Megabytes to a
few Kilobytes with an improvement of 1000 times. Synopsis also enables reductions
in message complexity and commitment delay of 85% and 99.4%, respectively.
|Date of Award||Apr 13 2021|
|Original language||English (US)|
- Computer, Electrical and Mathematical Science and Engineering
|Supervisor||Basem Shihada (Supervisor)|
- Internet of Things
- Quality of Service
- Fault Tolerance
- Software Defined Networks
- Distributed Ledger