Home / Information / Laboratories and Academic Staff / Advanced Signal Processing

Advanced Signal Processing (Sato Lab)

Advanced signal processing is a technology that supports the digital infrastructure of our increasingly information-oriented society, by dealing with data compression in communications, object recognition by robots, and many other things. However, the fundamental definition of what is classified as "signal" and what is considered "noise" depends on the goal at hand.

This lab pursues research on the signals handled in applications such as electromagnetic radiation measurements and optical communications, as typified by lasers, with a focus on determining the essence of "signals", for realizing advanced processing techniques that transcend the limitation of conventional processing methods.

Academic Staff

Toru SATO

Toru SATOProfessor (Graduate School of Informatics)

Research Interests

Atmospheric radar, Remote sensing for observation of precipitation, middle- and upper atomosphere, and orbital objects, Indoor environmental measurement with radar signal processing

Contacts

Room 223, Engineering Bld. 2, Yoshida Campus
TEL: +81-75-753-3362
FAX: +81-75-753-3342
E-mail: tsato (at) kuee.kyoto-u.ac.jp

Seiji NORIMATSU

Seiji NORIMATSUAssociate Professor (Graduate School of Informatics)

Research Topics

Optical fiber communication systems, Optical modulation/demodulation schemes, Optical fiber nonlinearities, Optical receiver performances

Contacts

Room 214, Engineering Bld. 2, Yoshida Campus
TEL: +81-75-753-3363
FAX: +81-75-753-3342
E-mail: norimatu (at) kuee.kyoto-u.ac.jp

Introduction to R&D Topics

Radar signal processing [leader: Prof. Sato]

  • Atmosphere observation radar signal processing
    R&D on new technologies for improving large atmosphere observation radars such as Kyoto University's MU radar and the Equatorial Atmosphere Radar (EAR). Work is focused particularly on adaptive signal processing for clutter suppression utilizing digital receiver arrays, and on three-dimensional high-resolution observations using multistatic methods.
  • Space debris observation
    The presence of space waste (space debris) left in orbit around the earth is becoming a space environmental problem. We are developing techniques to improve the sensitivity of debris observation using radars such as the KSGC radar constructed in Okayama prefecture.
  • Using radars for indoor environment measurements
    We are engaged in research focused on household robots, to develop techniques that use ultra-wide-band (UWB) pulse radar for observation and high-speed 3D imaging of indoor environments.
  • Measurement of object shapes using ultrasound
    We are doing studies on the use of pulse ultrasonic sensors to image obstacles to assist the visually impaired to walk safely, and on the use of medical ultrasonography for imaging the interior of the body.

pic1

Optical fiber communication systems [leader: Assoc. Prof. Norimatsu]

Almost all backbone networks are realized by utilizing optical fiber communication systems, and the expanding deployment of fiber-optic lines to subscriber networks means that capacity of backbone networks need to be increased even further. In addition, to continue utilizing existing infrastructure it is also important that possible transmission distances do not decrease as capacity is increased.

Since wavelength-division-multiplexing (WDM) holds considerable promise as data capacity is increased, we are pursuing research aimed at improving frequency efficiency and transmission distances of WDM systems. Most important issue is to overcome various fiber-induced obstacles such as fiber nonlinearities, chromatic dispersion, and polarization-mode dispersion.

UWB radar imaging technology [leader: Asst. Prof. Sakamoto]

UWB signals were standardized by the U.S. FCC in 2002. Radar systems with UWB signals can realize extremely high range resolutions, down to as little as several tens of mm. However, radar imaging for near-field targets is a type of inverse problem, and all conventional techniques for this are based on iterative solution algorithms, which makes real-time processing difficult. Our research has successfully clarified the reversible transform relationship between the target shape and received signal, based on simple conditions, to obtain exact solutions to the inverse problem without relying on iteration. We are now applying this principle to develop the SEABED method—a UWB radar imaging technique—and conducting research to improve the precision and anti-noise performance of the method.

pic2