Play the video by hitting Play icon. You can render the video in full-screen by hitting the X icon (below the Play icon)

Alternatively, download or stream the video from the following links, for rendering in your favorite media player (e.g. vlc)

Demo Description

The current demo exists of four parts:

  1. A Backoffice, running on a Server
  2. A Proxy hardware module, connecting infrastructure to the Backoffice
  3. Beacons, this is the fixed part of the infrastructure
  4. Tags are localized within the infrastructure.

Backoffice

The backoffice part, running the Analytics software and the Presentation software, This software is running on the Liferay server which renders the pages you are now looking at.

The Analytics software takes the distances measured between the beacons and the tags, and determines the position of the tags, knowing the absolute positioning of the tags in the infrastructure. This positioning determination is done by a process called triangulation. The analytics software optimizes the results by exploiting redundancy: the more beacons there are, the more choice to optimize. Choice is done based on quality of the signal and geometric quality of the triangles.

The presentation layer shows the end result. This can be just visualization, or triggering events such as personal alarm or service transaction. In the demo movie, you can see how the analytics results are currently presented.

Proxy

This proxy is currently built using a generic Raspberry Pi board.

The proxy collects the data from  the beacons using a serial link between Raspberry and one of the Beacons, The beacons which are not connected to a proxy, route their data to the proxied beacon using mesh networking.

The proxy then sends the data of all the beacons in the infrastructure to the Backoffice analytics using a standard WiFi or LAN connection.

Beacons and Tags

Beacons and tags are actually different software on the same hardware board. In the picture to the left, the actual hardware is the small "seal" to which the antenna is connected, the remaining of the board is for development and debug purposes.

Tags send radio signals to the beacons, which are then "echoed" back to the tag. The travel time of the signals are a measure of the distance between both, since light speed is constant. Beacons collect the results and forward it to the proxy, using the "mesh network" between tags and beacons.

The challenge of the technology is that:

  1. radio waves may be reflected by metal objects such as metalized windows or metal structures in a building. This leads to a longer travel path than the actual distance. Many travel paths may co-exist.
  2. radio waves may travel through other materials, such as water (bodies!), wood (furniture), walls or glass. These materials slow down the travel time of the radio signal
  3. times to be measured are extremely short: light travels 1 meter in 3 picoseconds (3x10-12 seconds, or 0.000000000003 second, count 12 zeros behind the comma). Compare that to an ultra-fast 3GHz PC: during one single  clock cycle, light travels already 100m!

Results

In the demo movie to the side, we show an infrastructure in a normal living room, with 5 beacons, positioning the tag with an accuracy of 10 cm, in the 3 dimensions (length, width, hight). The shown room measures 5x9x3 meter.

The life analytics page is shown here and can be tracking a beacon on request.