Taking the multi-track route to all-you-can-eat bandwidth

Are we running out of bandwidth? Our data transmission needs increase 40% every year, and our current communications system is nearing its limits. With more and more businesses and infrastructure running online, this is a problem that could throttle the digital economy. To keep the information flowing, we need to wring every last bit of capacity from our traditional modes of use, and think of even more ways to use what we have, more efficiently and with less expense. Dr Benn Thomsen of Electronic and Electrical Engineering is leading a project to address this problem, funded by a £2.3million EPSRC grant and involving top researchers from the universities of Oxford, Cambridge, Southampton and UCL.

The best method we have of transmitting data is optically, where the data can be encoded in certain properties of an electromagnetic wave travelling down an optical fibre.  Currently, several messages can be sent at once down the same length of cable, as long as they are transmitted on different frequencies or with different polarisations, so they can be unscrambled and separated out at the other end.

However, there is room to squeeze yet more signals in simultaneously. Within the fibre, each beam could follow several different paths. This can be undesirable if they interfere with each other and blur the signal, but it can also be engineered as a feature that lets us get more bandwidth out of the same size cable. If you could send different signals down different paths or ‘modes’ in the cable, and confidently unscramble them at the other end, you would multiply the bandwidth that is achievable many times.

Although this allows greater amounts of data to be sent, it brings its own problems in keeping all the signals separate.  This project will attack these challenges on four fronts:

  • Physically engineering the optical cable to maximise the number of signals it can carry cleanly.
  • Dynamic control of the signal processing at the receiver, to use the method which will extract the best signal and minimize the complexity
  • Amplifying the signal using energy efficient methods, to allow signals to travel further.
  • Measuring not only the amplitude, but the phase, of the wave at the receiver, to give information about the path it took to get there.  This will not only make multi-mode transmission possible, but also make existing methods for sending simultaneous signals (on different frequencies, say) more practicable.

The technologies and systems developed within this project will find applications that range from upgrading the capacity of existing multi-modal fibre networks in data and computer processing centres, through to the installation of new high capacity metropolitan and long haul fibre transmission systems using the optimised fibres and technologies developed in this project.

It is a collaborative project involving Cambridge, Oxford and Southampton and UCL, funded by the EPSRC. The four year project started in January 2012 and will run to 2016. The overall project lead is Dr Benn Thomsen, at UCL, where the work will be focused on the transmitter and receiver development, receiver-based digital signal processing and systems demonstration. Dr Tim Wilkinson of Cambridge leads the work on efficient spatial multiplexing using Spatial light modulators. Dr Frank Payne, at Oxford, takes the lead on multimode fibre modelling, design and characterisation, while Prof David Richardson, at Southampton, leads the work in multi-fibre fabrication and development of multimode optical amplification.

Read more about the research on the group’s site.

Photo from Humuku’s Flickr stream, licensed under Creative Commons.