Lightweight IoT platform for rapid application development and deployment

• Main Author: Mohamed Ibrahim, Abdikarim
• Format: Thesis
• Language: English
• Published: 2019
• Subjects: TK Electrical engineering. Electronics Nuclear engineering
• Online Access: http://eprints.utm.my/id/eprint/85732/1/AbdikarimMohamedIbrahimMSKE2019.pdf; http://eprints.utm.my/id/eprint/85732/; http://dms.library.utm.my:8080/vital/access/manager/Repository/vital:132661

Sensors connected to the cloud services equipped with data analytics has created a plethora of new type of applications from personal to industrial levels. In other words, the smart devices, the network, and the data come together to form Internet-of-Things (IoT). In this context, IoT provides an opportunity to increase efficiency in how things are done. IoT-based system normally follows a pattern of data collection, data analytics, automation, and system improvement recommendations. However, most application would have its own unique requirements in terms of smart devices, communication technologies as well as its application provisioning service. Although various services are commercially available that provide services such as Backend-as-a-service (BaaS) and Software-as-a-service (SaaS) hosted on the cloud, this, in turn, raises the issues of security and privacy. Individuals and organizations alike would like to protect their sensitive information for various reasons. Therefore, in this project, a lightweight and secure IoT platform is proposed. The platform consists of Raspberry Pi as an IoT device with a pre-configured image that contains hotspot module, user login, PHP, Apache server, MySQL database, Node.js, and Domain Name Server (DNS). The platform also contains a middleware that provides Application Programming Interfaces (API) for both the sensor layer and the application layer. Moreover, the platform has a Graphical User Interface (GUI) designed using Angular to provide management tools and to enable data display sent by the IoT device for the end-user. The middleware is designed using JavaScript programming language in Node.js development framework to provide a lightweight and scalable features which is proven to save up to 45% of memory. The middleware is connected to NoSQL database that allows the platform to be distributed and thus, enhance security and privacy. The performance analysis of the system shows the developed platform has a Hypertext Transfer Protocol (HTTP) operation which is around 600 Bytes, with the system processor not exceeding 6% of usage. It also demonstrates a reduction by 53% and 41% of byte size and time consumed, respectively, for GET operation over a Local Area Network in UTM campus.