Skip to main content



Archived content

NOTE: this is an archived page and the content is likely to be out of date.

Cutting-Edge R&D

FSTJ 2014-1 Cover Image

2014-1 (Vol.50, No.1)

The mission of the Fujitsu Laboratories Group is to support the growth of the Fujitsu Group, by developing advanced technologies and driving innovation to enable the realization of a Human-Centric Intelligent Society envisioned by Fujitsu, as outlined in the Fujitsu Technology and Service Vision. With the aim of researching and developing technologies useful for people, and by contributing to business expansion, Fujitsu Laboratories conducts diverse R&D in a multitude of relevant fields, from solutions and services, to system/network/ubiquitous products, devices, and materials. This special issue highlights some of the leading-edge R&D we are undertaking, from a perspective of how such technologies can benefit and provide value to Fujitsu's customers.

Feature Papers

1. Preface (573 KB)
The world in which we live and environments around us are rapidly changing in diverse ways, giving rise to unprecedented new and challenging issues, many of which are complexly intertwined on a global scale. Conventional approaches of dealing with specific issues individually can no longer provide viable solutions—we are in an era in which closely inter-related problems must be treated systematically with a comprehensive approach and sense of urgency. Information and communication technology (ICT) is taking on an increasingly prominent role to help resolve such issues. In Fujitsu's vision of a Human-Centric Intelligent Society, the evolution of ICT and innovation enabled by people maximizing their full potential, and information that creates new value, will drive the sustainable growth of a prosperous and personally enriching society in which people can lead safe, secure and fulfilling lives: a world in which knowledge is continually harnessed to drive new value and support sustainable growth. ---[Tatsuo Tomita, President of Fujitsu Laboratories Ltd.]

Leveraging Big Data

2. Information Integration and Utilization Technology using Linked Data (1.44 MB )
Fujitsu Laboratories is conducting R&D on technology to integrate and utilize data by using Linked Data, which is a standard methodology for publishing data on the Web that is being promoted by the World Wide Web Consortium (W3C). W3C is the main international standards organization for Web technologies. Linked Data uses a machine-readable structured data format that can be processed mechanically. Much information that has public value has recently been published in Linked Data format, and a global data space on the Web called Linked Open Data (LOD) has been created. This paper introduces an LOD utilization platform technology that gathers and stores data published around the world in Linked Data format and that provides a centralized search function. It was developed by the authors in collaboration with the Digital Enterprise Research Institute at the National University of Ireland, Galway. ---[Nobuyuki Igata, Fumihito Nishino, Terunobu Kume, Takahide Matsutsuka]
3. Anomaly Detection Technology Using BigGraph (1.12 MB )
Many difficulties are encountered along all three axes of Big Data (volume, variety, and velocity), which limit the applicability of established technology. BigGraph is a research project and platform for realising the vision of an intelligent society, where requirements along all three axes can be accommodated. In the fraud detection scenario, working with Fujitsu UK & Ireland, BigGraph researchers consider different types of information, link external data sources as streams, and consolidate existing knowledge in order to identify undiscovered and fraudulent anomalies in Big Data. ---[Bo Hu, Aisha Naseer, Takahide Matsutsuka]
4. Development and Runtime Platform and High-speed Processing Technology for Data Utilization (1.20 MB )
Dramatic increases in computing power and network speed, along with advances in sensing technology, have broadened the range of devices that can connect to the Internet. Large quantities of diverse, time-sequenced data are flowing in from the Web and countless sensors, and there is a strong demand to rapidly and efficiently extract any valuable information from that data and to utilize that data for various types of navigation systems. Fujitsu Laboratories envisions a Human-Centric Intelligent Society and is building a cloud platform to support it. This article introduces an integrated development and runtime platform technology for utilizing large quantities of data, a technology for extracting parallelism from complex event processing as a base technology for high-speed processing, and a distributed-parallel technology for complex event processing and describes the goals, features, and effects of each. It also gives an overview of directions for future technology initiatives. ---[Hidetoshi Kurihara, Haruyasu Ueda, Yoshinori Sakamoto, Masazumi Matsubara]
5. StreamStorage: High-throughput and Scalable Storage Technology for Streaming Data (1.20 MB )
Real-time analytical processing (RTAP) of vast amounts of time-series data from sensors, server logs, and other sources has come to be widely used in recent years. To make thorough use of such streaming data, it is essential that the same data be analyzed iteratively from diverse viewpoints, which has increased the need for loss-free storage and data reconstruction functions for large volumes of data. Since individual items of data targeted for storage are generally small in size and repeatedly received from multiple sources, it is difficult for existing storage systems to simultaneously satisfy throughput and capacity demands. Additionally, high-speed access optimized for the chronologically ordered data characteristic of streaming data has not yet been achieved for such vast amounts of data. StreamStorage developed by Fujitsu Laboratories is a new storage technology that manages time-series data in units of streams, enabling high-speed storage and data reconstruction. It achieves high throughput and scalability by using distributed storage technology in which streaming data is partitioned into blocks and data is input and output in a parallel and asynchronous process. This paper presents an overview of the StreamStorage architecture and an application example. ---[Munenori Maeda, Toshihiro Ozawa]
6. Privacy-protection Technologies for Secure Utilization of Sensor Data (1.15 MB )
As the number of scenarios for utilizing sensor data such as household data from intelligent appliances and location information obtained from automobiles increases, social issues are arising, such as leakage of private information from sensor data. Fujitsu Laboratories has developed two privacy-protection technologies that provide total protection for such sensor data, from its collection to utilization of the results of its analysis. One, partial decryption technology, enables parts of the data to be masked while still encrypted, so other data can be substituted or the encryption key can be changed. The other, anonymous access technology, enables people to obtain the analysis results for their own data from a utilization service without providing their ID. These technologies give people appropriate control of their sensor-collected private information while also enabling them to use external analysis services anonymously. This article describes these technologies. ---[Tetsuya Izu, Koichi Ito, Hiroshi Tsuda, Kenichi Abiru, Takao Ogura]

Enabling a Human-Centric World

7. Touchless User Interface Utilizing Several Types of Sensing Technology (1.08 MB )
Recently users have come to demand a natural user interface (NUI) so that they can operate devices naturally, as an alternative to a graphical user interface (GUI). At Fujitsu Laboratories, with the aim of achieving an NUI, we are developing touchless user interfaces that make it possible to monitor users' behavior through intelligent sensing technologies, such as gesture recognition, eye tracking and speech recognition, and then understand users' intent to enable them to operate a device in a natural way. It is difficult to achieve this by using individual sensing technologies. Therefore, we have developed a way to ensure a device operation by integrating multiple sensing technologies for natural motion detection. Both gesture recognition and eye tracking technologies are combined in the developed interface, which gives the users the feeling that they are interacting more naturally and effectively with devices than they would if only individual sensing technologies were utilized. This paper describes an overview of the developed sensing technologies, the merits and problems regarding their combination, the developed interface, and future work. ---[Akihiro Minagawa, Junichi Odagiri, Yoshinobu Hotta, Satoshi Nakashima, Liu Wei, Fan Wei]
8. Data Transfer Technology to Enable Communication between Displays and Smart Devices (1.18 MB )
Recently, the chance to see videos in various places has increased due to the speed-up of networks and spread of digital signage. However, most of the videos are non-interactively broadcast at viewers, and currently viewers often input keywords related to the videos in a search site after they see the videos when they want to find out some information related to the videos. So that viewers can easily obtain information related to videos, Fujitsu Laboratories Ltd. has developed new data transfer technology that enables communication between videos and smart devices by embedding communication information into the videos invisibly and extracting it using the camera application of smart devices. Viewers are able to acquire information related to videos easily just by capturing them. In this paper, we introduce an outline and the usage scenarios of this data transfer technology via video data. ---[Kensuke Kuraki, Shohei Nakagata, Ryuta Tanaka, Taizo Anan]
9. Secure Managed Application and Execution Platform Technology for Smart Devices (1.52 MB )
With the rapid spread of smart devices such as smartphones and tablets, corporate IT systems are increasingly shifting away from PCs and towards such devices. In order for businesses to take full advantage of smart device capabilities, new security features are needed that are different from those conventionally designed for PCs. Accordingly, Fujitsu Laboratories Ltd. has developed platform technology for application management and execution that provides for flexible and secure use of business applications without sacrificing usability. In this paper, we introduce the concept and technical mechanism along with use cases. This technology provides the secure environment necessary for corporate use by distributing the applications associated with the users' contexts to their devices and properly protecting and controlling the execution of those distributed applications which include confidential information. ---[Hidenobu Ito, Kazuaki Nimura]

Realizing a Safe and Secure Society

10. Multi-layered Defense against Advanced Persistent Threats (APT) (1.13 MB )
Recently, crafty cyber-attacks such as advanced persistent threats (APT) have become a menace to systems in enterprises. The APT attack usually uses a variety of intelligent techniques to gain access to a specific target based on elaborate preliminary research on the target. Therefore, the risks involving the intrusion of such malicious unauthorized code (malware) have increased greatly. To confront such novel cyber-attacks, a new approach is needed. The new approach aims to reduce the damage of the cyber-attack by using technologies of early detection and avoidance. Fujitsu Laboratories is developing and commercializing technologies to enable comprehensive countermeasures against each stage of APT, from the initial spying stage of malware to the breakout stage via the system research stage, within an organization's network. This paper presents four research activities concerning cyber-security based on this approach. First, as an inlet measure to prevent APT at its initial spying stage, this paper describes the countermeasure implemented by a client against the targeted e-mail. Then, techniques to optimize security measures implemented by employees and organizations are described by giving consideration to human behavioral characteristics. Further, the techniques to appropriately detect intelligence activity conducted by the malware such as system research are described as an organization's network measure after malware has intruded. In the end, the paper covers the exit control measure after malware has intruded, and this detects the malware's attempt to communicate with an externally controlled server. ---[Satoru Torii, Masanobu Morinaga, Takashi Yoshioka, Takeaki Terada, Yuki Unno]
11. Fast Single-image Defogging (1.34 MB )
Bad weather conditions such as fog, haze, and dust often reduce the performance of outdoor cameras. To improve the effectiveness of surveillance and in-vehicle cameras under such conditions, we propose a method based on a dark channel prior for quickly defogging images. It first estimates the intensity of the atmospheric light by searching the sky area in the foggy image. Then it estimates the transmission map by refining a coarse map from a fine map. Finally, it produces a clearer image from the foggy image by using the estimated intensity and the transmission map. When implemented on a notebook PC with a graphics processing unit (GPU), it was able to process 50 images (720 x 480 pixels) per second. It can thus be used for real-time processing of surveillance and in-vehicle system images. ---[Zhiming Tan, Xianghui Bai, Bingrong Wang, Akihiro Higashi]
12. Online Failure Prediction in Cloud Datacenters (1.02 MB )
Once failures occur in a cloud datacenter accommodating a large number of virtual resources, they tend to spread rapidly and widely, impacting many cloud services and their users. One of the best ways to prevent a failure from spreading in the system is to identify signs of a failure before its occurrence and deal with it proactively before it causes serious problems. Although several approaches have been proposed to predict failures by analyzing past logs of system messages and identifying the relationship between the messages and the failures, it is still difficult to automatically predict the failure for several reasons such as variation of log message formats and frequent changes in their configurations. Based on this understanding, we propose a new failure prediction method that Fujitsu Laboratories has developed. The method automatically learns message patterns as signs of failure by classifying messages by their similarity regardless of their format and re-learning the message patterns in frequently changed configurations. We evaluated our method in an actual cloud datacenter. The experimental results showed that our approach predicted failures with 80% precision and 90% recall in the best case. ---[Yukihiro Watanabe, Yasuhide Matsumoto]

Contributing to Solutions for Societal Issues

13. Power Supply and Demand Control Technologies for Smart Cities (1.17 MB )
One of the missions of smart cities is to contribute to the environment and reduce social costs by using energy more efficiently and utilizing more renewable and distributed energy. This is reflected in the increasing use of renewable energy and in energy consumers becoming "prosumers," i.e., people who not only consume but also generate and sell power. However, several factors related to this trend are causing both the supply and demand sides of electrical power to fluctuate and complicate the control of the balance between the two sides. While supply and demand balance is controlled mainly with generators in conventional power systems, more effective and economical control is needed for future systems. We believe that it will be important to also optimally control the amount of energy that is needed for people to manage their daily lives at home and at work. Demand response (DR) is a technology that is receiving attention in light of this need. This article introduces R&D initiatives at Fujitsu Laboratories, including development of a Demand Response Automation Server (DRAS) conforming to OpenADR, which is advancing as a standard for DR, and a building-demand simulation technology that optimizes control of the balance between supply and demand for electric power using DR technology. ---[Tomoyoshi Takebayashi, Toshihiro Sonoda, Wei-Peng Chen]
14. Hassle-free Sensing Technologies for Monitoring Daily Health Changes (1.10 MB )
In an aging society beset with a multitude of social problems, there are growing expectations for technologies to support elderly people and patients in maintaining and increasing the "things that I can do for myself." Such expectations can be met by expanding the use of advanced mobile devices in diverse scenarios, by developing wearable sensors, and by using numerous sensors embedded in the environment. These actions will enable widespread extensive use of mechanisms for supporting human activities and health in everyday life. An important technological element for achieving such mechanisms is the ability to continually sense the state of the individual and the surrounding environment. This means first and foremost that sensors must be able to capture essential information as a performance requirement. It also means that sensing must be convenient for the user; i.e., the difficulty or bother of operating sensors must be eliminated, and the continuous collection of information must be hassle-free. Under the theme of "health support in daily life," this paper describes detection of a person's pulse through facial imaging and in-home monitoring using compact, light, and multi-functional wearable sensors as sensing technologies for hassle-free understanding of an individual's behavior and health. It also introduces joint research on assisted independent living for smart house residents. ---[Akihiro Inomata, Yoshinori Yaginuma]
15. A Mobile Platform for Real-time Continuous Monitoring (1.29 MB )
Advances in information and communications technology (ICT) are expected to help people lead healthier lives amidst the many stressors in today's fast-paced society. While smartphones and tablets continue to overtake desktops and laptops, the next computing revolution—Human-Centric Computing—is taking root. This shift in computing is enabled by the increasing ubiquity of sensors that are around us, on us, and even in us. Deploying intelligent, human-centric services using these connected sensors requires advancements in the underlying IT infrastructure itself. In this paper, we describe various novel services built atop a general-purpose mobile platform developed by Fujitsu Laboratories of America, Inc. for continuous mobile monitoring. Our platform was developed with next-generation healthcare services in mind but has broad applicability as an extensible platform for deploying real-time services that incorporate data from arbitrary sensors. We provide an overview of our platform, and highlight several services that act as new points of contact between a user and the IT infrastructure. In the domain of health and wellness, we show how continuous bio-monitoring enables us to measure stress and to provide stress management services. We describe how such services may be used in a typical day, contributing to a new and improved quality of life. ---[Ajay Chander, Albert Braun, Rajalakshmi Balakrishnan, Alex Gilman, Stergios Stergiou, Dave Marvit]
16. Energy Harvesting Technology for Maintenance-free Sensors (1.38 MB )
Energy harvesting is a technology that gathers energy such as sunlight, artificial light, and vibration and heat from machinery to obtain electric power. Applying this technology to wireless sensor modules in a machine-to-machine (M2M) wireless sensor network will eliminate the need for grid-based power and primary batteries and create new value in the form of maintenance-free, battery-free, and cable-free operation. Energy harvesting can also promote environment-friendly, clean technology that saves energy and reduces CO2 emissions, which makes it a useful technology for achieving the next-generation smart city and sustainable society. This paper introduces oxide-based thermoelectric material, all-solid-state secondary battery technology, and energy-harvesting tester technology now being researched and developed at Fujitsu Laboratories as elemental technologies for applying energy harvesting technology to M2M wireless sensor modules and achieving maintenance-free operation. ---[Tsutomu Tanaka, Takashi Suzuki, Kazuaki Kurihara]

Enabling High-Performance ICT Products

17. Photonic Network Defragmentation Technology Improving Resource Utilization during Operation (1.43 MB )
Recently, various new information services have emerged as a means for sustaining the significant growth of modern applications spearheaded by cloud and smartphones. These new services rely heavily on the underlying network and data center infrastructures. Thus, core networks are demanded to further expand the capacity to support the rapid increase in data traffic. Meanwhile, it is becoming increasingly important for core networks to be more flexible in order to accommodate feature-rich services such as bandwidth-on-demand. Additionally, for the realization of a sustainable human society and eco-friendly IT services, products capable of low energy consumption are strongly desired. In the near future, it is expected that conventional optical networks which operate on a rigid fixed channel basis will be replaced by flexible optical networks in which signals can be freely allocated on arbitrary frequency slots of the optical spectrum. This flexibility enables a more dynamic and efficient utilization of resources, which in turn leads to lowered energy consumption, heightened usable capacity, and superior agility, making it capable of providing adaptive networking services based on the dynamism of user requests. In flexible optical networks, due to the frequent setup and tear down of optical signals that occupy different spectrum slots, the utilization of such slots has the potential to become heavily fragmented. This so-called spectrum fragmentation phenomenon dramatically degrades resource utilization and reduces usable network capacity. Therefore, spectrum defragmentation technology that Fujitsu Laboratories has developed is needed to restore efficient resource utilization by reallocating the fragmented slots to more continuous ones. In this paper, we discuss a photonic network defragmentation technology that can improve resource utilization during network operation by continuous and in-sync reconfiguration of flexible optical nodes (transceivers and optical switches, etc.). We show the effectiveness of this technology through network simulations, as well as experimental results of hitless defragmentation. ---[Motoyoshi Sekiya, Wang Xi, Yasuhiko Aoki, Kyosuke Sone]
18. Ultra-high-speed Interconnect Technology for Processor Communication (1.25 MB )
In order to improve the performance of storage systems and servers that make up the cloud, it is essential to have high-bandwidth interconnects that connect systems. Fujitsu has already marketed a CMOS high-speed interconnect product that works between CPUs for the UNIX server (SPARC M10). Its data transfer rate per signal line is 14.5 Gb/s. Further, Fujitsu is currently engaged in studies of a CMOS interconnect that can operate at over 32 Gb/s per signal line to achieve a higher bandwidth interconnect. To realize higher speed, we developed a high-speed interconnect with a loss-compensation performance higher than 30 dB at 32 Gb/s by integrating a series of features: a communication method requiring no high-frequency operation in a transmitter circuit, a wideband loss-compensation equalizer circuit in the receiver circuit, and a data-receiving method that does not require generation of a highly accurate sample clock. When these technologies are implemented in a CPU chip, it is possible to improve the entire performance of server systems to a level more than twice to a level more than twice as high as the current one. This paper introduces the features of these ultra-high-speed interconnect technologies for 32-Gb/s serial interconnects implemented by using 28-nm CMOS technology. ---[Yoshiyasu Doi, Samir Parikh, Yuki Ogata, Yoichi Koyanagi]
19. Optical Interconnect Technology for High-bandwidth Data Connection in Next-generation Servers (1.44 MB )
In the near future, an improvement in the performance and integration of servers (for example blade servers) will cause an increase in the amount of data to be transferred in places where a data connection is required. Furthermore, the data rate of transmission is growing to 25 Gb/s. These will increase the interconnect bandwidth in servers to around the 10 Tb/s order. We propose "optical interconnect" that is a data connection using optical fibers so as to avoid a limitation of bandwidth caused by a degradation of waveform and interference in copper wiring. Optical interconnects can be used to transmit data signal at high speed and they excel in terms of wiring density and transmission distance. To apply such optical interconnects to servers, many optical channels must be integrated in the limited space of a server chassis. To do so requires low-cost and compact optical components. In this paper, we propose optical technology (an optical transceiver module, optical connector and optical mid-plane) that makes it possible to have low-cost and high-density optical interconnects, and we describe a verification of their practical use in trial manufacturing. ---[Tsuyoshi Yamamoto, Kazuhiro Tanaka, Satoshi Ide, Tsuyoshi Aoki]
20. Silicon Photonics Optical Transmitter Technology for Tb/s-class I/O Co-packaged with CPU (3.24 MB )
In the near future, Tb/s-class large-capacity input/output (I/O) circuits are considered to be necessary for high-end CPUs used in high-performance computing (HPC) systems and high-performance servers due to the progress in processing performance of the CPUs. A high-performance I/O using optical transmission technology is attracting public attention as a technology to overcome the limitations of a conventional electric I/O in terms of size and power consumption. Particularly, an integrated optical I/O chip based on silicon (Si) photonics technology is a very promising candidate to realize a Tb/s-class large-capacity I/O chip of a size that allows it to be co-packaged with a CPU. This paper describes the current status of progress in Fujitsu Laboratories' approaches to develop the Si optical transmitter for a large-capacity optical I/O chip. To realize the co-packaging of a large-capacity optical I/O chip in a CPU package, it is essential to realize a Si optical transmitter that ensures stable operation with low power even under drastic fluctuations of operation temperature. Therefore, the authors proposed a novel Si optical transmitter scheme using a highly energy-efficient Si ring modulator to operate stably without a complex wavelength-tuning procedure. A prototype integrated optical transmitter based on this scheme was fabricated and evaluated. The evaluation results showed that an integrated Si optical transmitter chip could operate at 10 Gb/s without any wavelength tuning over a temperature range of 25 to 60°C. Further, supported by successful developments of a compact-size, high-performance laser source using flip-chip bonding technology as well as a 4-ch integrated Si hybrid laser array for wavelength division multiplexing, there are certain prospects for application of a Si optical transmitter for large-capacity optical I/O chips. ---[Shinsuke Tanaka, Tomoyuki Akiyama, Shigeaki Sekiguchi, Ken Morito]
21. High-performance and Low-power Consumption Vector Processor for LTE Baseband LSI (1.08 MB )
Recently, the transmission rate for handheld devices has been increasing by Long Term Evolution (LTE), and baseband LSI has come to need a higher performance. In addition, handheld devices will use the second- and third-generation communication method, so a baseband LSI will need to handle multiple communication methods. Because implementing all communication circuits results in a large area, we have been developing Software Defined Radio (SDR), which switches each communication method with software. To implement SDR for handheld devices, a high-performance and low-power consumption digital signal processor (DSP) is needed. We have developed a DSP which inherits the architecture of vector supercomputers, and the architecture has advantages of a low power consumption and application developments. We have downsized the vector architecture for embedded systems. The peak performance is 12 giga operations per second (GOPS) at 250 MHz, and the power consumption is relatively low at 30 mW for 28 nm process technology on average. This paper presents the vector processor that we developed. ---[Yi Ge, Mitsuru Tomono, Makiko Ito, Yoshio Hirose]
22. Outlook for GaN HEMT Technology (1.27 MB )
It is expected that the high electron mobility transistor (HEMT) using gallium nitride (GaN) as its wide band gap semiconductor will be applied in diverse, green ICT systems because of its high efficiency. The GaN HEMT utilizes high-density two-dimensional electron gas (2DEG) accumulated in the boundary layer between GaN and AlGaN through their piezoelectric effect and natural polarization effect. This makes it possible to realize a low on-state resistance (Ron). Combined with a high breakdown voltage, the GaN HEMT indicates a superb performance as a power device. After the development of GaN HEMT technology started for power amplifiers of mobile base stations, it was expanded to a radar sensor application. Further expansion of its application is expected in the field of power conversion, in equipment such as server power systems. While the development of GaN HEMT technology has been promoted, focusing on "high output power" conventionally, further advantages such as high efficiency and low energy consumption have been attracting much attention in recent years. In this paper, the outlook for GaN HEMT technology is described in terms of its contribution to energy saving. ---[Kazukiyo Joshin, Toshihide Kikkawa, Satoshi Masuda, Keiji Watanabe ]