I am currently working with a team of researchers on a new optical access network to accelerate the evolution of information-communication services.
Our goal here is achieve and deploy a new optical access network by researching network architecture and elemental technologies that can dramatically improve system performance and flexibility while engaging in collaborative activities on a global basis. In the present optical access network, communications traffic is transferred to a relay network at an exchange building much like a bus passenger might change to a train at a station. In contrast, our goal for the future is a network that can eliminate such traffic transfers and transmit optical signals directly to the location needed by merging the access network and relay network.
Up to now, the optical access network has been supporting the expansion of broadband services called Fiber To The Home (FTTH). Today, however, with the coming of the mobile Internet era, it might be thought "the wired optical network is no longer needed," but the facilities within an exchange building, fifth-generation mobile communications system (5G) antennas, and next-generation wireless LAN antennas are all interconnected by the optical fiber network. Furthermore, considering that all kinds of things such as factory machines, sensors, transportation systems, and power systems will one day be connected over the network, we can expect requirements for bandwidth, latency, etc. to become significantly broader in scope. Against this background, we expect the optical access network to migrate from a FTTH platform to a common access platform for diverse services and systems, and to this end, we pursue research and development that can drive the evolution of the optical access network into the future.
To begin with, we are working to dramatically improving the transmission performance of the optical access network to meet the ever-increasing demand for broader bandwidths and lower latency. For example, our team has achieved a real-time digital coherent optical transceiver circuit for the optical access network, a world's first. Although a digital coherent receiver system has been used for large-capacity transmission in the backbone network, applying it to the access network requires that it receive intermittent signals (burst signals) having greatly different levels of power from different optical network units (ONUs). We therefore devised and developed a real-time signal processing circuit in addition to a burst-supporting coherent receiving circuit and achieved errorless transmission of 20 Gbit/s with power differences greater than 20 dB (100 times). This development received a top score in the field of access networks at European Conference On Optical Communication (ECOC) 2016, a major international conference in the optical communications industry.
Next, we are converting transmission functions to software to achieve a highly flexible network. Our reasoning here is that, if transmission functions can be made to run on general-purpose (commodity) equipment such as general-purpose servers and personal computers, it will become significantly easier to replace, combine, and fine-tune transmission functions according to bandwidth and distance needs. We consider that a conversion to software is the key to achieving a new type of network that can access a desired location solely by optical means as I discussed earlier. We continue to research possibilities in this area. For example, we have been conducting joint research with The University of Tokyo on the application of graphics processing units (GPUs) commonly used in image processing and machine learning to software-based processing and on the development of new algorithms. As a result of these efforts, we achieved a processing speed of 10 Gbit/s in the present access system in relation to computationally intensive error-correction processing. Our paper describing this achievement received the Transmission, Access, and Optical Systems (TAOS) Technical Committee Best Paper Award at IEEE Globecom 2016, a core international conference in the field of communications. We also achieved the world's first implementation of digital coherent optical transmission by software through high-speed software-based digital signal processing (DSP). For this achievement, we received a top score in the field of access networks at Optical Fiber Communications Conference (OFC) 2018, a major international conference in the field of optical communications.
The future society that Japan should aspire to is laid out in the government's Science and Technology Basic Plan under the name "Society 5.0." In simple terms, we can think of Society 5.0 as a society that merges cyberspace and physical space. To achieve such a convergence, much attention will be focused on artificial intelligence (AI) and virtual reality (VR), but on a more basic level, the smooth transmission of information through optical communications in the infrastructure supporting those technologies will be very important. In the sense that optical communications is a technology supporting the base foundation of the network, it has a background role, but we can also see it as an important and socially valuable technology that accelerates the evolution of information-communication services. As part of this future vision, we would like to continue our efforts in developing groundbreaking optical access technologies.
NTT has announced the concept of an Innovative Optical and Wireless Network (IOWN) as its future vision of an optics-based innovative network and information-processing platform. With an eye to making this vision a reality, our team will take up the challenge of achieving new forms of networking and creating an altogether new ICT world.
Research on novel optical access networks to accelerate the evolution of information and communication services
Senior Distinguished Researcher
Jun-ichi Kani
Currently, he is Senior Research Fellow and Group Leader of Optical Access Infrastructure SE Group, Optical Access Infrastructure Project, NTT Access Network Service Systems Laboratories. He is also Visiting Professor in the Collaboration Field of Hokkaido University Graduate School of Information Sciences.