Machine-Learning-assisted Optical Network Planning on NDFF

Network researchers used NDFF for Machine-Learning-assisted and SDN-based Optical Network Planning with Network-Scale Monitoring Database

From August to September 2017, researchers from the University of Bristol and Hong Kong Polytechnic University used a 436 km long link of the NDFF network for optical field trials of a Software Defined Network (SDN) based framework, which utilised machine learning to predict link performance and maximise the capacity. Adaption of the spectral efficiency utilising probabilistic-shaping BVT based on link performance prediction is demonstrated.

This experimental work was presented as a post-deadline paper at the European Conference on Optical Communications (ECOC 2017), considered as a key update in the rapidly growing field of optical communications and networks.


Published: 26 September 2017