Skip to main content



Archived content

NOTE: this is an archived page and the content is likely to be out of date.

Fujitsu Develops Spatiotemporal Data-Processing Technology to Quickly and Precisely Search Areas of Interest

Opens door to new services sensitive to the locations of people and motor vehicles

Fujitsu Laboratories Ltd.

Kawasaki, Japan, October 17, 2011

Fujitsu Laboratories Limited today announced the development of a new spatiotemporal data-processing technology that can quickly and accurately search areas in which events of interest are occurring, based on latitude and longitude positional data.

As GPS and RFID sensing have made it possible in recent years to capture massive amounts of positional data from people and motor vehicles, there has been growing interest in founding new services that make good use of that data. Taking advantage of the massive volume of positional data in a timely fashion requires exceptional data processing speeds so that it can be quickly reflected by the service.

With existing technologies, however, problems have come about in processing this kind of data as they require the setting of either an advance mesh suitable to the characteristics of the data or tailored to the objectives for analysis; or without that, lengthy processing times would be needed to thoroughly check an innumerable amount of areas. Given these issues, Fujitsu developed this new technology which dispenses with the need to supply a tailored mesh, and is capable of searching precise areas of complex shapes and various sizes while operating roughly 60 times faster than previous methods.

This technology makes it possible to quickly and precisely discover areas of interest in response to the disposition of people and motor vehicles, which, in turn, promises new services based on positional information.

Figure 1: Areas where taxis are expected to be used most (rainy evening example, Tokyo)(Background map data distributed by the Geospatial Information Authority of Japan's "Digital Japan" web system)

Larger View (382 KB)


Together with the advance of ICT and sensor technologies, GPS sensors deployed to mobile phones and motor vehicles, and RFID tags on goods, now make it possible to capture massive amounts of sensor data. This puts a spotlight on services that use positional information captured from people and motor vehicles. Time and location data are important in any business or field, and great promise is seen for the creation of new businesses that employ this information.

Technological Issues

It is vital to have precise and current data for services that offer timely information by analyzing the movements of people and motor vehicles. For example, a taxi-dispatch support service would need to know precisely what areas are seeing larger demand at any given moment. The method of employing an administrative boundary or rectilinear grid, however, requires that a mesh of appropriate granularity be set beforehand. A mesh that is set too coarsely will only obtain results that are too general, and conversely, a mesh that is too fine will have its borders severely affected. This is why an appropriate mesh somewhere in between these extremes is necessary. A mesh's adequacy depends upon the data's characteristics and analytical objectives, and for this reason, determining granularity requires knowledge about the data and expertise in its analysis. One particular challenge is determining the ever-changing mesh requirements when handling something such as sensor data that has constantly fluctuating characteristics. General data mining techniques can be used to discover optimum areas based on individual data, relying as little as possible on data knowledge or analytical expertise. In data mining, every possible candidate area is searched to discover the optimal value. However, the existence of numerous candidate areas to search makes it impossible to conduct a realistic computation. This meant, for example, that there was an issue of lengthy times needed for processing, even when the search was limited to rectangular shapes.

About the Technology

The technology that Fujitsu Laboratories has developed can precisely identify areas where events of interest occur with high probability, achieving a significant boost to speed of roughly 60 times compared to off-the-shelf data-mining techniques.

Features of the technology are as follows.

1. Precisely defined areas of interest

This technology eliminates the need to define a tailored mesh and enables a high degree of flexibility when searching for areas of complex shapes and different sizes. A total area is automatically divided into sufficiently precise zones in accordance with data characteristics and method of aggregation. These areas are then combined into a candidate area for an optimum search. In this way, areas of differing shapes and sizes are combined—dispensing with the need for rectangles or other set shapes—so that areas of interest with complex shapes and irregularities can be searched with precision.

2. Faster searching

This technology adopts a newly developed algorithm to enable an extremely fast search of an area of interest. Rather than enumerate the combined areas one by one, the technology repeatedly processes to filter out areas that should not be included in the areas of interest based on the likelihood that something of interest will occur in a candidate area being searched. In this way candidate areas on the list are significantly reduced in number and only those areas that show the most promise are searched. Compared to typical data-mining techniques, this technique speeds up the search process by a factor of 60 or more, making it possible to search dynamically changing areas in a timely manner.

Figure 2: Locations covered by core-optimized area search technology and usage scenario.

Larger View (133 KB)


This technology makes it possible to precisely and quickly search areas of interest that change minute by minute. This could be used to benefit the public to efficiently control electricity distribution based on differences in supply and demand between areas, or to produce highly detailed traffic and high-crime danger maps. It could also be used to help improve business efficiency and develop new services, such as real-time and accurate trading area analysis, by making use of the positional data already at hand.

Future Plans

In June 2011, Fujitsu announced its new SPATIOWL service, which uses location data. SPATIOWL provides services for using and managing a variety of location-related data types statistically, based on a massive volume of location data. The spatiotemporal data-processing technology being announced today is set to be rolled into SPATIOWL during fiscal 2011, bringing its customers the ability to analyze their own data more effectively and to integrate and manage it with the layer of location information provided by SPATIOWL. The technology will also be introduced at the 18th World Congress on Intelligent Transport Systems held in Orlando, Florida in the United States, in October 2011.

About Fujitsu

Fujitsu is the leading Japanese information and communication technology (ICT) company offering a full range of technology products, solutions and services. Over 170,000 Fujitsu people support customers in more than 100 countries. We use our experience and the power of ICT to shape the future of society with our customers. Fujitsu Limited (TSE:6702) reported consolidated revenues of 4.5 trillion yen (US$55 billion) for the fiscal year ended March 31, 2011. For more information, please see

About Fujitsu Laboratories

Founded in 1968 as a wholly owned subsidiary of Fujitsu Limited, Fujitsu Laboratories Limited is one of the premier research centers in the world. With a global network of laboratories in Japan, China, the United States and Europe, the organization conducts a wide range of basic and applied research in the areas of Next-generation Services, Computer Servers, Networks, Electronic Devices and Advanced Materials. For more information, please see:

Press Contacts

Public and Investor Relations Division

Company:Fujitsu Limited

All other company or product names mentioned herein are trademarks or registered trademarks of their respective owners. Information provided in this press release is accurate at time of publication and is subject to change without advance notice.

This press release has been revised as of December 17, 2018.

Date: 17 October, 2011
City: Kawasaki, Japan
Company: Fujitsu Laboratories Ltd., , , , , , , , , ,