HTML5 Webook

6/84

exceeds the Shannon limit. However, the team showed only the theorem of such existence, and concrete means to ex-ceed the Shannon limit and to reach the Holevo limit were still unknown.We started the study by extracting important points to construct a model to perform experiments. In 1996, the author met Prof. Holevo for the rst time at an interna-tional conference held in Hakone and spent a week with him aer the conference, observing him study the gener-alization of the capacity theorem. Discussion with him gave the author an idea about a mechanism enabling exceeding the Shannon limit. e principle is to generate quantum interference between the states of code words by quantum computing during the process of decoding in order to improve the distinguishability of signals. e eect can be expressed easily (when the amount of communication re-source doubles, the capacity of transmitted data increases more than twice: super-additive coding gain). By the conventional theorem, the capacity of transmitted informa-tion increases twice at maximum. A principle demonstra-tion experiment of super-additive coding gain succeeded in 2003. However, it became clearer that the practical ap-plication of the theorem was more dicult than expected.On the other hand, since 2000, full-edged experiments in QKD started in various countries. In 2005, a project funded by the Defense Advanced Research Project Agency (DARPA) of the Department of Defense (DoD) in the USA succeeded in a eld experiment by constructing a QKD network between three points in the Boston area. In Europe, a project of the European Union, SECOQC, was established in 2004 and research by a team consisting of 41 teams from 12 countries started. At the same time, NICT was developing basic technologies for a prototype of QKD by outsourcing to Mitsubishi Electric Corporation, NEC and the University of Tokyo.In the eld of quantum metrology standards, the tech-nology to improve the assurance of frequency standards and technology for controlling the generation of single photons were developed by controlling a single ion freely by enclosing it in a cavity.2The Second Medium-Term Plan (FY2006-2010)In the experiment to prove super additive coding gain performed in 2003, a specic signal format called polariza-tion of single photons and path modulation coding was used. However, it is necessary to use quantum computing to treat states of laser light (coherent state) for practical application. In the Second Medium-Term Plan, full-edged development of technology to control quantum bits con-sisting of a coherent state started. A quantum bit in a co-herent state is known as the paradox of Schrodinger’s cat, and the generation of it had been a dream in quantum physics. In NICT, we had been trying to realize it. In 2004, a group of Laboratoire Charles Fabry in France published a paper on basic technology, from which we knew for the rst time the existence of a competitor for the same goal. In fall of 2005, we received news that the state of Schrodinger’s cat was generated in the Niels Bohr Institute in Denmark. And in December, we came to know that Laboratoire Charles Fabry submitted a paper on generation of Schrodinger’s cat to Science.We stood up again from the discouragement of falling behind the competition and made progress in improving our unique experiment apparatus. In the summer of 2006, we succeeded in generating a high-purity Schrodinger’s cat state that was greatly advanced from the previous experi-ments in quality. Aer that, new technologies such as for amplifying the cat’s state (amplitude of wave) and for freely controlling the weight of odd and even photons in the cat’s state have been developed one aer another by using the technology and foundations for new ICT have been established by exploiting new aspects in quantum optics. e achievements were accepted by the most famous international journal in the eld of physics and optics.e photon number resolving detector that correctly identies the photon number in a pulse is indispensable to realize quantum ICT as well as Schrodinger’s cat state. e photon number resolving detector must be low noise and its eciency to transform input photons to electric signals (quantum eciency) must be almost 100%. In order to develop such a high performance photon number resolving detector that meets these requirements, we commissioned research and development of a superconductive transition edge sensor to the National Institute of Advanced Industrial Science and Technology, Nippon University and the National Institute for Materials Science, and they have developed the highest level photon number resolving de-tector in the world.e eld of QKD entered into eld experiments and NTT also joined the collaboration following Mitsubishi Electric Corporation and NEC in Japan. e Second Medium-Term Plan started in NICT in 2006. In Europe, the SECOQC project started, establishing a QKD network between multi-stations. On October 8th in 2008, the eld 1 Intoroduction2　　　Journal of the National Institute of Information and Communications Technology Vol. 64 No. 1 (2017)