Abstract

The term ''smart city'' refers to an instrumented, interconnected, and intelligent city built by leveraging Information and Communication Technologies (ICT). In such a city, a combination of embedded hardware and software involving sensors, actuators, and a host of mobile devices and wearables that are connected to the Internet of Things (IoT) networks will sense data in different contexts and automatically drive desired adaptations through actuators. Through adaptations, city planners, professionals, and researchers aim to optimize resource consumption and cost of providing services while improving the quality of life for the ever increasing urban population. To fully realize this goal, a context-aware and data-centric inference is a necessity. A system is said to be context-aware if it is able to adapt its operations to the current context without explicit user intervention. This thesis proposes a generic context-aware system architecture for development of smart city applications. The proposed architecture puts special emphasis on privacy and security, incorporating mechanisms to protect the system and sensitive information at each layer of the architecture. Furthermore, this architecture integrates with a reasoning component, whose inference engine can be driven by logic or other formalisms. A prototype implementation and a case study done in this thesis indicate the practical merits of the proposed architecture and provide a proof of concept.