Â鶹ԼÅÄ

The aim was to see how well the system worked in a real-world environment and across different devices and to get feedback from the public on what they thought of it.

The trial was based on a prototype system developed by that adds to Wi-Fi multicast packets to recover those that aren’t received due to interference (multicast has no re-transmissions). We in turn developed a RAILS-based web application to allow us to monitor the performance of the system in real-time.
Three Access Points on a mixture of 2.4 and 5 GHz frequencies were used to cover , all fed from a single hardware multicast encoder. Since the Access Points were based on consumer-grade hardware, we had to go some lengths to make them watertight courtesy of some modified plastic crates from a DIY store.

On the three evenings of our trials, members of the public were invited to join our Wi-Fi network and to download a dedicated app for the trial. This then gave them access to two streams of content, one with a special Edinburgh comedy highlights package and the second showing footage from last year’s Edinburgh Tattoo. We also asked for user feedback in the form of a short questionnaire.

While we still have a mountain of performance data to process, the initial results are encouraging and the system worked well across the wide range and number of different devices we saw (around 130 in total) over the three evenings that we trialled the system.

We have a number of possible next steps that could include examining the effect of different Forward Error Correction parameters on the performance of the system, improving the synchronisation of the video across different devices and looking at how the system would work on enterprise wireless access points as well as further lab-based tests in a controlled environment.

-