Menu Close

Research

Nishi Laboratory has proposed a network transparent edge computing infrastructure that can add all functions without modifying the terminal communication system, aiming at the realization of the next-generation Internet, and the transparent add-on that is the core of it.・ We have aimed to build transparent load balancing technology.

Then, as a realization of a smart community as an application of the next-generation Internet, we built 150 EVs and a smart island for electric power and tourism in Nagasaki Prefecture and implemented a smart city centered on information and electric power in Kurihara City and Kawasaki City. rice field. Currently, we are promoting information smart city initiatives in collaboration with Saitama City and Tsushima Smart Sustainable Town, and have experience in demonstrating infrastructure in collaboration with abundant local governments. Recently, we are also building a smart agricultural infrastructure in collaboration with the National Agriculture and Food Research Organization.

Please see the following items for details.

Research Topics

  • Smart City/Smart Community Information Infrastructure

He is conducting research on smart cities, particularly on the construction of information infrastructure to activate information distribution. He is a leader in this field as he is the chairman of an incorporated association for smart city in Saitama City and also the chairman of an incorporated association for ICT in tourism. The information infrastructure that he has built is being tested in actual operation through FeliCa Pocket Marketing, Softbank, and venture companies. The elemental technologies are listed below.

  • Data Anonymization

In a smart city, information from residents and other information concerning privacy issues are highly valuable, and if used correctly, attractive services can be developed. Contrastingly, while information anonymization is necessary to solve privacy issues, service quality will also decline if data quality declines due to anonymization. In other words, the value of the information will decrease. Therefore, it is necessary to construct a method that captures the various features of data and anonymizes them correctly. So far, we have proposed a new method for anonymizing location history information from smartphones and other devices and using the latent space of deep learning. In addition, we have also proposed a mechanism that can control the trade-off between quality of service and anonymization.

  • Edge Computing

In the provision of smart city services, the location of their provision has not been discussed until now. For example, health checks and device control, which are closely related to personal information and cause communication latency problems, are naturally provided in the IoT domain, while services that handle large amounts of information to grasp the overall situation and are provided over a wide area are provided in the cloud, and existing systems are constructed in such a manner. From this perspective, the edge is somewhere in between and is assumed to be a form of service that uses information from a specific region, loosely anonymizes it, and provides services to that region. Its location could also change over time as the service changes over time. We are building a new edge construction technology based on this new edge concept.

  • Application-specific hardware and its migration

Cloud has high communication latency but high computing power. The edge has low communication latency but low computational capacity. In other words, it is not simple to determine which has the shorter processing time. Smart city services may be critical in processing latency; from this perspective, one cannot simply choose to go to the cloud or the edge to meet that latency. Low latency requires low latency by zero-copy processing, which is tied to the communication module since it is at the edge and multi-parallel pipeline processing using hardware design and implementation. In addition, as indicated earlier, FPGA Logic State Migration is also necessary to follow changes in service provision locations, and we are building the technology to achieve this.

  • Smart Agriculture

We collaborate with NARO on this project. The primary industry has only seen an increase in greenhouse gas emissions since 2010, especially in horticulture. The project is building technologies such as spatial temperature distribution monitoring using deep learning, all of which are designed based on the same infrastructure as the smart city described above.

  • Technology Standardization

We are a member of the technical standardization of IEEE-SA regarding the technologies above, such as information linkage in smart agriculture and IoT. Nishi is the only Japanese leading the standardization as the chair of two standardization working groups.

Past Researches

See below for past researches. The contents of each website are current as of the end of the project.

  • Network type energy-saving system / smart grid
    • Green Society ICT Life Infrastructure Research Center
      In recent years, there has been increasing recognition that adaptation measures to prepare for the adverse effects of climate change, which cannot be dealt with by mitigation measures alone, are important. In this task, two local governments were used as fields, and first, the impact of climate change on local governments was estimated using mesh data, and regional vulnerability analysis was conducted. On top of that, we will develop a “green society ICT life infrastructure” that utilizes sensor networks to measure and optimize household energy consumption information, as well as areas associated with climate change, such as adverse effects on health, medical care, and agriculture. We have formulated adaptation measures to address the vulnerabilities of the above and demonstrated their effectiveness. The purpose of this task was to reform the social system to realize a new social vision of aiming to form a resilient community by increasing social capital.
    • Nagasaki EV & ITS Project
      This project aims to build an EV / PHV town by centrally investing electric vehicles and their infrastructure, centering on the Goto Islands in Nagasaki Prefecture, and to experience and investigate future towns and lifestyles in advance. It was also featured in the EV City Casebook published by the IEA.
    • FUSION project
      We conducted a demonstration experiment of a futuristic energy-saving system using ICT in the Fukueko Terminal area of Goto City, Nagasaki Prefecture. In this project, we mainly aimed to achieve the following two goals. (1) Global standardization of technical standards for information and communication networks to reduce CO2 emissions (2) Build an information processing system that reduces energy consumption by 10% For energy reduction, we use information and communication technology and information processing technology to reduce energy consumption by 10% while maintaining the comfort of facility users, instead of reducing energy consumption by enduring heat and cold. I aimed. We also installed solar panels in the parking lot to contribute to CO2 reduction.
    • Koevo House (Keio type co-evolved house)
      Keio-type co-evolutionary housing “Coevo House” is a new home with advanced technology under the theme of “House in 2030” in collaboration with the laboratories of multiple faculties and graduate schools of Keio University and many companies. It is a model house built on the occasion of “Enemane House 2014”, which proposes, builds, and exhibits people. Nishi Laboratory has implemented a control system that links various equipment and sensors of Koevo House. In order to realize a control system to create a comfortable living environment, we first sense the environment in order to grasp the current situation, and not only detailed information on power consumption, gas, and water volume, but also temperature, humidity, illuminance, wind speed, and carbon dioxide. We measured carbon concentration in real time and estimated and controlled comfort and number of people. Various control systems not only implement and develop new sensors, but also analyze existing board circuits, design new circuits, controllers that control them, software that operates on them, and higher-level services that connect to databases via the Internet. And, it is a fusion of various technical fields. By collaborating with companies in industry and academia and implementing such a control system for students, we were able to gain learning and joy that cannot be obtained in class.
  • Information-based Open Innovation Platform An innovative network virtualization platform and a new generation network R & D program for its applications and research. This project was supported by NICT.
    • Research on new-generation networks using service-oriented routers
      In order to introduce a new mechanism to the modern Internet environment where information is inexhaustibly increasing and traffic is bloated, traffic is reduced and the soil for new services is laid, we said, “Routers actively collect information from higher layers. I researched for the realization of the provision of service-oriented routers that are managed by the database. We carried out unique activities such as the release of open source, the development and deployment of the world’s first service-oriented router.
Scroll Up