We install our wireless sensors at the edges and/or inside of a large-scale environment and analyze the received signal strength (RSS) value differences between measurements taken in an unoccupied environment (calibration measurements) and live measurements during which crowds can be present. The average attenuation experienced by all communication links within the entire network when compared to the calibration measurements has proven to be a unique indicator to estimate the actual crowd size. A patent is currently pending.
The hardware part exists of wireless nodes, that will be deployed in a specific environment at an average waist height, and a gateway. Each node is battery powered and has a custom-made electronics board that can transmit and receive sub 1-GHz wireless signals according to specific firmware. This embedded software enables the sensor network to apply all measurements and to communicate with each other in a synchronized manner. The gateway on the other hand receives all messages and passes the result towards the software layer that can estimate the crowd densities, flows, and flux of the different areas. As a result of the real-time processing, the estimations are visualized in a graphical dashboard or can be integrated with other systems from customers or partners.
Our technology provides a privacy non-instrusive way of measuring crowds without making any connection to a particular device that users would carry or taking any pictures of individuals.
Our wireless sensor network provides a new measurement every 150ms.
Compared to a bracelet scanning system, we can predict crowd sizes with 95%.
Based on 6 years of research at the University of Antwerp and Imec, our technology was validated at different large-scale events such as Tomorrowland.
An intuïtive crowd density map of diiferent sub-regions is processed real-time.