IoT-based smart home security system

OData support
Dr. Dudás Ákos
Department of Automation and Applied Informatics

IoT devices take an increasingly important part both in our daily lives, and our homes. Utilizing the capabilities of these devices can help us in the process of creating a “smart home”. Besides enabling the remote monitoring of our living space, these smart home systems make it possible to receive instant notifications about various events, such as accidents or break-ins. However, in order to provide a pleasant experience and gain the users’ trust, the system must provide these services with an adequate level of comfort, safety and reliability.

The appearance of high performance, low cost IoT devices – such as the Raspberry Pi – make smart home systems available to a broader audience. In this thesis new features are introduced to the PCA home security framework – designed to run on such devices –, to increase the performance of the system with regards to the points mentioned above.

First, a new module capable of detecting human falls in a video stream is introduced to the system. Accidents resulting from falls at home are a major health risk to the elderly. The module aims to shorten the time they need to wait for medical assistance by enabling the notification of their relatives or nurses in case of an accident.

The reliability of the system can be increased by decreasing the number of its false alarms. To achieve this, the system is extended with a Passive Infrared motion sensor. The sensor can increase the accuracy of motion detection and eliminate the false alarms caused for example by pets or a moving curtain, among many others.

Finally, a solution for the convenient and secure remote monitoring of the PCA system is presented. The solution consists of a cloud component based on Google Cloud Platform, which is capable of communicating with the PCA system via a secure MQTT connection, and a cross-platform, React Native mobile application enabling the user to remotely monitor the system and receive alerts in the form of push notifications.


Please sign in to download the files of this thesis.