Research and Development Projects

Transport technology for the all-photonics network

To transport a huge volume of many types of traffic with ultra-low latency, we are developing a transport network technology for the all-photonics network (APN), which overcomes the limits of the existing network.
We are mainly working on three technologies: high-capacity transmission technology in order to provide direct optical links between data centers and users at [rates of] 1 Tb/s or higher; technology to increase the number of wavelengths multiplexed with a target of multiplexing more than 1,000 wavelengths, which will be achieved by fully exploiting both wavelength multiplexing and space multiplexing; and optical cross-connect technology that flexibly controls the destinations of optical signals. ^*1*2
In addition, with a view to further enhancing the reliability of transport networks, we are engaged in the development of technologies that can quickly identify the locations of optical fiber disconnections or optical repeater failures, and can also predict disconnections and failures.

*1　NTT Holding Company Press Release
"Succeeded: World’s fastest 600Gbps per Lambda Optical Transmission with 587Gbps Data Transfer," December 2018

*2　NTT Technical Review
"Optical Full-mesh Network Technologies Supporting the All-Photonics Network," Vol. 18 No. 5 May 2020
https://www.ntt-Review.jp/archive/ntttechnical.php?contents=ntr202005fa4.html

Software for controlling white box switches and high-precision network monitoring technology

To provide an easily customizable network in which new functions can be added quickly in order to adapt to advances in technology and evolving customer demands, we are developing white box switches in which hardware and software, which conventionally have been integrated, are separated. We are mainly developing Beluganos, a software program for controlling white box switches, and working to promote its adoption through joint experiments with telecommunications carriers in other countries^*1 and by providing the program as open-source software^*2.
In addition, to provide a network that can offer the quality and performance best suited to individual customers, we are developing HANMOC, a system that monitors the network state with a high degree of precision. By simply connecting this system at any point in the target network, the delay time of the entire network can be measured with a high degree of precision, in the order of microseconds. Through feeding back measurement results to our study, we aim to develop an ultra-low-latency network that will be required in Beyond 5G.

*1　NTT Technical Review
"NTT and Telefónica Start PoC on Open Optical and Packet Transport Technologies within Telecom Infra Project"
https://www.ntt-review.jp/archive/ntttechnical.php?contents=ntr202003sr1.pdf&mode=show_pdf

*2　Beluganos
https://github.com/beluganos

Self-evolving zero-touch operations

We are studying self-evolving zero-touch operations, which autonomously adapt to changes in the environment without the need for human intervention, from the following three perspectives:

• Technology for AI to learn new operations through dialogue with operators (AI teaching technology)
• Technology for AI to learn from changes in the digital twin space (self-evolving mechanism of AI)
• Technology for autonomously creating workflows that are adapted to changes in the environment (self-evolving operations technology)