Abstract:
The Internet of Things has transformed modern life by changing how humans, systems, and devices interact with the physical world. The number of IoT devices is expected to increase from 10 billion in 2020 to approximately 30 billion in 2030. With the increasing reliance on the Internet of Things (IoT) as a primary technology for developing smart cities, several technical challenges have emerged due to the heterogeneous and distributed nature of IoT networks. One of the most critical challenges lies in choosing an efficient communication protocol. It is a complex mission and needs more effort from the developers. Therefore, one of the primary objectives of this thesis is to examine the performance of application layer protocols for the Internet of Things in the context of smart city development. The research work in this thesis is organized into two parts: a theoretical part and a practical part. A detailed literature review was applied as a methodology in the theoretical phase of this study. The strengths and weaknesses of five existing IoT application layer protocols—HTTP, MQTT, CoAP, XMPP, and AMQP—in smart city IoT environments were evaluated, and a comparative performance analysis was conducted. The philosophy of each protocol has been addressed separately regarding its applicability in the smart city scenario. Based on the results obtained from the theoretical phase, MQTT emerges as an effective and ideal compromise protocol that balances the strengths and limitations of these protocols in terms of performance and efficiency; it combines lightweight and ease of use with an acceptable level of reliability and quality of service. Since the MQTT protocol has been suggested as the main communication protocol in IoT, the performance of this protocol was evaluated under various scenarios in the context of smart city applications, focusing on two primary scenarios: the transmission of small payloads and large payloads, each evaluated under varying QoS levels (0, 1, and 2). Using a Raspberry Pi as the publisher and Hive MQ as the broker, key performance metrics, including latency and message loss, were analyzed to assess protocol efficiency. Ultimately, the study emphasizes the importance of selecting an appropriate protocol tailored to the specific requirements of smart city applications, underscoring that no single solution is suitable for all IoT deployments. Additionally, the conclusions drawn from this extensive review revealed that various factors can impact a protocol evaluation.
