Optimizing link reinforcements to improve Internet protocol networks

September 25, 2014, University of Electro Communications
Researchers at the University of Electro-Communications, Tokyo, have developed a new optimization scheme capable of predicting the weak links in a network in order to pinpoint where reinforcements are required. Their work could help lessen network congestion on the internet.

Internet protocols are a set of communication 'rules' which are followed in order to deliver data packets between computers. Nowadays, the amount of data that the internet carries is so vast that significant network congestion can occur.

At any given moment in time, a can be represented by a series of nodes and links. Each link is given a 'weight' - a measure of the link's quality of service. The aim of an internet engineer is to keep the links as uncongested as possible, allowing as much data as possible to flow freely at any one time.

In order to transfer in the fastest and cheapest way, 'start-time optimization' procedures are used to find the best set of link weights to minimize congestion. Stephane Kaptchouang and co-workers at the University of Electro-Communications in Tokyo have successfully built a new 'preventative start-time link-weight optimization' scheme (PSO) capable of predicting the 'weaker links' in any given chain - singling out the ones which need reinforcing in order to create a less congested network.

Applying reinforcements to every link in a network is a costly process in terms of resources, and so predicting which links need to be reinforced is very useful. Previous start-time optimizations considered all possible failure scenarios at any given time on a network, disregarding reinforcements already in place. Kaptchouang's team built their PSO to allow for existing reinforcements, therefore reducing the number of failures the network operators need to address, and speeding up the process considerably.

Explore further: What metadata does the government want about you?

More information: Kaptchouang, S., Ouedraogo, I.A., & Oki, E. Preventative start-time optimization of link weights with link reinforcement. IEEE Communications Letters, 18 (7) (2014) DOI: 10.1109/LCOMM.2014.2325814

Related Stories

What metadata does the government want about you?

August 28, 2014

With the leaking of a discussion paper on telecommunications data retention, we are at last starting to get some clarity as to just what metadata the Abbott government is likely to ask telecommunications companies, internet ...

Cutting congestion on the data network highway

September 12, 2014

Perhaps no other consumer-driven technology has made such incredible advances in such a relatively short space of time as the mobile phone. Today's smartphones are used to stream videos, access social media and perform all ...

Recommended for you

Fish-inspired material changes color using nanocolumns

March 20, 2019

Inspired by the flashing colors of the neon tetra fish, researchers have developed a technique for changing the color of a material by manipulating the orientation of nanostructured columns in the material.


Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.