This paper outlines our ongoing work towards developing a system for extracting patterns embedded in heterogeneous data streams that contain people’s recorded movements in both physical and virtual spaces. Examples of such spatial data sources are satellite-based sensors (GPS), ultrasound acoustic trackers, radio frequency (WLAN, Bluetooth, UWB) and infrared-based sensors. The core work on pattern extraction relies on the spatial data fusion component aiming to bring various data types to a common format. The additional benefit of this system will consist in the graphical interface that will enable interactive visualisation of the extracted patterns. The rationale of this work is outlined through the relevance of location aware system in the context of ubiquitous computing, which so far have received limited benefits from fields such as Human Computer Interaction (HCI) and interactive visualisation.