Wi-Fi Visualization

Wi-Fi is one of the most common technologies in our daily life. It is essential to find a suitable way of visualizing Wi-Fi signals to understand how a Wifi network covers humans’ working and living areas. This research project aims at exploring an easily understandable way of displaying Wifi signals in a 3D space.

By analyzing Wi-Fi signal strength (represented by RSSI) in several places in HKUST campus using specific software, the research team has successfully produced a set of visualized graph of Wifi signal strengths.

The findings enable researchers to further study how humans interact with Wifi, for example, effect on health by exposure from Wifi, optimization of Wifi Access Point to maximize coverage, etc.

Data Collection

Before making a data model of Wi-Fi coverage, we need to collect a set of data representing the Wi-Fi signal strength first. We have samples more than 180 points along the corridors on each floor. Each measuring point is 2 meters away from the adjacent points. Three points (left, middle and right) are taken for each crosssection area of the corridor. At each point, we sampled 100 connections to the Wi-Fi Access Points and record the RSSI value and APs information for identification purpose.

Geographical Mapping

After collecting the data, five Wi-Fi APs are chosen to analyze. For each AP on a floor, we can create 2.5D model representing the RSSI (a 3D object with single horizontal layer of surface). Moreover, after combining the data from the 2nd floor and the 3rd floor, a truly 3D model can be created. The data will be processed by Excel for organization purpose. RSSI collected from every Access Points at a particular position will be taken average. After that, the data is output to Origin, a program used to generate 3D data model.

Visualization

The 3D model can be adjusted to a suitable viewpoint that matches the real photos. We produce the visualized graph by adjusting the points of view of the 3D model and render it into a corresponding photo of the building.

Check out the research paper.