The current trend towards the use of smart tools within universities open up new opportunities. Insight in the location and daily rhythms of users within the building form valuable input for many decision-making processes within campus management. Unlike the outdoor environment, where the position is easily obtained through omnipresent satellite-based positioning systems such as GPS, the use of these systems in indoor environments is limited. Various implementations of indoor positioning systems try to fill this gap and provide indoor positioning capabilities. However, the complex indoor environment presents its own challenges regarding positioning. The effectiveness of current implementations varies depending on techniques and methods used, while often being limited to function only within small test bed environments. Privacy, cost, scalability, ease of integration in the environment and many other decisive factors steer the choice for certain indoor positioning systems. Within this research the focus is on the development of a non-intrusive network-based indoor positioning system using Wi-Fi. Wi-Fi has clear advantages over other measuring techniques in that these systems are ubiquitous, cost-effective and their use is multi-faceted. Now over 85% of the users carry a smartphone. These off-the-shelf, unmodified smartphones and other Wi-Fi-enabled mobile devices can be used as mobile stations to indicate user locations through an iterative positioning process. Besides the mobile stations, existing Wi-Fi infrastructures and networking equipment can be re-used with minor adaptations. The following research question was addressed in this research: 'To what extent can indoor Wi-Fi positioning be used for indoor localisation in order to determine occupancy rhythms and movement patterns within and between rooms to support campus management?'. The research question was approached starting with exploring the different techniques and methods suited for indoor positioning or localisation. The case study was used to review and develop a system suitable for indoor campus environments. As the case study requires a non-intrusive and low-cost solution that functions without active user participation a signal propagation-based technique was chosen. Distances were estimated employing the signal strength metadata extracted from various 802.11 management frames originating from user mobile devices. Various influences on the Wi-Fi signals were identified and a solution to counteract those influences was designed in the form of a differential Wi-Fi system using a grid of interconnected reference stations. With the use differential Wi-Fi correction parameters the positioning improved by mitigating for temporal environmental influences that are continuously present in indoor environments. Next, based on a comparison of different methods, a multilateration method was developed as the core of the positioning system. The multilateration algorithms combined with various filtering algorithms provide for a highly maintenance-free system with minimal manual set-up required. As the environment changes new training data is automatically acquired for ready-use. All algorithms were designed to work both in (near) real-time as well as non real-time, with configurable processing delays of at least 10 seconds. The longer delays allow for more observations, while the shorter delays are better for moving users and fast updates. Accuracies of 2 meters at the lower bounds are reported, ranging up to the 10 meter mark due to Wi-Fi characteristics. Various filtering and post-processing mechanisms were designed to generate accurate mappings from mobile devices to user counts. Furthermore, in the preparation phase different geometries were tested with reference stations 10 to 20 meters apart from each other. Preferred spatial geometries were matches as closely as possible for the deployment of the system in a real-world test case scenario. The designed reference stations were deployed in the Architecture faculty building to monitor occupancy and movement patterns before, during and after a fire drill. Ground truth was acquired using BLE wrist bands and manual counting. The fire drill had different elements: counting the number of users per m2 and per room, the occupancy rhythms over time and the movement patterns. Overall during 48% of the uncontrolled testing a 100% match between BLE wrist band equipped user counts and Wi-Fi counts was achieved. In 77% of the cases there was a 1 person difference in counting, where in more than 95% of the cases the maximum difference was 4 persons. In conclusion, with this localisation system the occupancy rhythms and movement patterns of users can be measured to a large extent to aid in the various tasks of campus management, e.g. planning, and safety evacuations.