Abstract:
The human thirst for data has become increasingly potent. This has now permeated into the field of sensors, where the thirst is being satiated by an explosion of more people using devices such as wearable smart watches, ambient environment sensors, and health monitoring. With all this data at our fingertips, there is undoubtedly a need for solutions that help us make sense of the large volume of information. The current research space however, is lacking in a unified interface of information. Instead, the data is spread through proprietary dash-boards limited to a sole device provider or data type. This makes it difficult for users to correlate data from different devices, and does not allow for unique trends to be identified. Heterogeneous sources should be abstracted so a user can combine different data to glean more useful information. For example, there is no way to see if heart rate from a wearable sensor is somehow related to the humidity of an environment. To address the gap, this thesis proposes a consolidated view of multiple sensor data sources allowing for novel comparisons and relationships to be drawn. The use of cloud-based technologies makes it highly scalable and performant, and allows for around-the-world accessibility. A proto-type solution has been implemented with user scenarios in mind, and is described as three layers of functionality: Data Acquisition, Data Trans-formation, and Data Visualisation. The proposed solution was then evaluated against other frameworks using key functional and non-functional criteria to deliver a novel and unified framework for the visualisation of sensor data.