Optical Performance Monitoring
The Optical Performance Monitoring (OPM) project aims to develop new ways to measure impairments within optical network that could affect performance.
December 11, 2007: We've signed our first customer. See media release
Other Media Releases
The Challenge
With the internet becoming an increasingly important tool for both businesses and consumers, ensuring the massive global network operates smoothly is a critical task. Every day billions of gigabytes of data flow through the complex web of optical cables and switches that span the globe. This rush of digital traffic is putting pressure on the infrastructure and requiring constant investment to keep it operational. Telecommunications companies are responding to the demand by increasing the capacity and flexibility of their backbone networks. Where 10 Gbit/s speeds were once sufficient, now they are rolling out reconfigurable 40 Gbit/s optical networks that combine fibre-optic cables and optical switches. While moving to such speeds greatly increases internet capacity, it brings with it management challenges. At speeds of 40 Gbit/s, the pulses of light moving through the cables are at such small timescales that a range of physical challenges come into play. Imperfections in the fibre can degrade performance, resulting in disrupted traffic. For network operators, the challenge is finding a way to efficiently monitor their high-speed fibre-optic networks, identify problems and predict where breakdowns might occur. They need a way of guaranteeing quality of service for their customers, as the future performance of the internet depends on it.
The NICTA Approach
Because monitoring techniques that work at slower network speeds no longer function as speeds increase, the NICTA team set about creating a different method of analysing the performance of high-speed optical networks. The Optical Performance Monitoring (OPM) project was established to develop new ways to detect and measure impairments within the network that could affect performance. The team took the novel approach of bringing together researchers with two very different sets of skills: optical networking and statistical signal processing. The rationale was that, if they could capture large amounts of data about network performance and then extract meaning from it, an accurate management tool could be created.
The Results
The team’s first success was the creation of a new-generation optical signal-to-noise ratio (OSNR) monitor that is compatible with optical switches. They discovered they could cancel out the signal being transmitted through a network. Their technique then accurately measures the remaining noise using inexpensive, off-the-shelf electronic components. The second success for the team has been the development of a multi-impairment monitor. It is this device which combines the optics and signal processing skills of the researchers. Essentially the approach taken was to look at network signals in a similar way a cardiologist looks at an ECG of a patient’s heart to determine and diagnose any problems. Impairments in a fibre cable will cause deterioration in signals passing through it, however it can be very difficult to determine which particular impairment is having what effect on a signal. Standard techniques can detect signal deterioration, however until now there has been no way to extract the effects of individual impairments. This is what the NICTA team has been able to achieve.
Through innovative use of signal processing techniques, the monitoring information is analysed and the effect of individual impairments extracted. So, for example, a network manager can determine whether it’s a stress in the fibre or a problem with an optical switch that is causing problems. This means proactive steps can be taken to remove or minimise any problems and improve the operating efficiency of the network. When that network is the global internet, the flow-on effects can be significant. In an interesting twist, the diagnostic techniques that we have developed for optical networks also have potential for analysing biological signals, and work in this direction has already commenced.
Commercialisation Opportunities
The prospects for these developments are significant. Network operators are looking for ways to automate the monitoring and management of their networks and this technology can assist them towards that goal. Commercialisation of the tools could take the form of licensing agreements with network equipment vendors. The technology would be embedded in switches and used by operators as part of their overall network management systems. Alternatively, NICTA could partner with a manufacturer and produce network monitoring equipment that could be attached to existing networks, providing significantly improved monitoring and management capabilities.
The Team
Trevor Anderson (Project leader), Masud Bakaul, Ken Clarke, Sarah Dods, Peter Farrell, Kerry Hinton, Adam Kowalcyzk, Thas Nirmalathas (Program Manager), Gavin Pearce, David Wright (Entrepreneur in Residence), Michael Yarrow.
Selected Publications
T. B. Anderson, K. Clarke, S. D. Dods, and M. Bakaul, "Robust, low cost, in-band optical signal to noise monitoring using polarization diversity," paper OMM3, presented at Optical Fiber Communications, Anaheim, USA, (2007). [slidepack] [paper]
S. D. Dods, T. B. Anderson, K. Clarke, M. Bakaul, and A. Kowalczyk, "Asynchronous sampling for optical performance monitoring (invited)," paper OMM5, presented at Optical Fiber Communications, Anaheim, USA, (2007). [slidepack] [paper]
T. B. Anderson, S. D. Dods, K. Clarke, J. Bedo and A. Kowalczyk, "Multi-impairment monitoring for photonic networks (invited)," presented at the 33rd European Conference on Optical Communications, Berlin, Germany, (2007). [slidepack] [paper]
Product Updates (these open in a new window)
Duration: 36 months
Participants: We are in discussion with a number of telecommunications carriers and network equipment manufacturers.
Links
updated SD 19 Dec 2007