This is the highest speed ever achieved with a standard diameter optical fiber, compatible with the current infrastructure. Technology full of promise.
On the occasion ofInternational Conference on Optica Fiber and Communications, Japanese researchers from National Institute of Information and Communications Technology (NIICT) presented stunning work: they managed to transfer data at a staggering rate of 319 terabits per second for more than 3 km. Above all, their technology is compatible as it is with modern fiber optic networks; we could well be at the dawn of a small revolution.
This figure would almost look like an entry error, but it is not. The previous record for a standard diameter optical fiber, established in 2020 at 178 Tb / s, was simply shattered. By way of comparison, the fiber optic packages offered by operators to the general public reach a maximum of 10 Gb / s. To achieve the speed proposed by the Japanese team, it would thus be necessary to accumulate nearly 32,000 high-end fiber boxes. This would allow downloading more than 300 movies in 4K… every second.
An already known technique
However, these defying performances are not conditioned by a new revolutionary material; they are based on the same fiber optic technology already present in our infrastructure. Rather than using a single fiber optic channel, the research team relied on a technique called “Wavelength-division multiplexing“, or WMD.
The idea is based on a combination of several “cores” of optical fiber. Thanks to this device, the signal can be distributed over different bands simultaneously. This avoids having all the load bear on a single fiber, which generates interferences in particular. It’s a bit like an exceptional convoy of several vehicles, each carrying part of a gigantic machine; each vehicle will generate fewer traffic jams (interference) and arrive at its destination faster.
To achieve this record speed, the Japanese researchers used an optical fiber with four “cores”, and made use of a third “highway” for the signal; in addition to the two usually used in this type of applications, they also used the S-band. They also subjected the signal to two new types of amplification, before undergoing the normal amplification process (we speak ofamplification de Raman).
A technology applicable in the “short term”
As mentioned below, what makes this work so disturbing is its practical feasibility. The external diameter of the fibers proposed by the team is strictly identical to that of a standard fiber optic cable. This has a major implication. To move from current networks to this type of technology, there would be no need to replace the entire infrastructure; the new one is already compatible with the old one.
Certainly, there are technologies capable of climbing much higher. In 2017, the Rennes start-up CAILabs supplied a Japanese operator with technology which made it possible to achieve a speed of… 10 Pbit / s. That is several dozen times the global internet speed. But these are gigantic cables, with 19 cores and absolutely not suitable for existing networks. This kind of technology can only be used in niche cases. That of NIICT, on the other hand, is directly applicable on the current network.
For researchers, this very important feature “demonstrates the potential of standard compatibility fibers in the short-term implementation of ultra-high speed fiber networks”. As unbelievable as it may sound, the insane speeds we have access to today could seem ridiculous a lot faster than you think!