TOP Page >  Faculty List by Area >  Profile

Lab
情報研究棟IS Building III 4F
TEL:0761-51-1265
To Lab's Site
 

English

Full text / JAIST Repository

 

 

Tadashi Matsumoto Professor
School of Information Science、Security and Networks Area

■Degrees

B.S. from Keio University(1978)、M.S.from Keio University(1980)、Ph.D from Keio University(1991)

■Professional Career

Nippon Telegram and Telephone Public Corporation(NTT)(1980), NTTDoCoMo(1992), Professor, Wireless Communications, at Oulu University (2002)、Guest professor, Ilmenau University of Technology (2006)

■Specialties

Wireless Communications, Information Theory, Coding Theory, Iterative (Turbo) Algorithm, Network Information Theory, Information Theoretic Analysis and Coding Techniques for Relay and Sensor Network, Multi-Dimensional Channel nalysis

■Research Keywords

Turbo Coding, Turbo Equalization, Network Information Theory, Mutual Information Transfer Chart

■Research Interests

Joint Decoding of Source and Channel Codes using Message Techniques
Jount decoding of source and channel codes using the Turrbo principle is sought for. Convergence property analysis using extrinsc information transfer chart provides us with the information about the matching optimality of the codes, and hence the EXIT curve matching techniques will be used as a tool for the optimization.
Optimal Activation Control of Multiple Turbo Loops
To detect signals via detector-decoder chains having multiple Turbo loops, the optimal path in the extrinsic information transfer plain has to be found to minimize the decoding/detection complexity. The primary goal of this research is to develop algorithms that can achieve the optimality in activation control of the multiple Turbo loops to minimize the decoding/detection complexity.
A Unified Apprach to the MAC and Slepian-Wolf Region and its Applications
The primary goal of this research is to establish methodologies allowing us to calculate the multiple access (MAC) and Slepian Wolf regions for correlated sources. Major applications of the outcomes of this research include joint optimization of cooperative source and channel coding in sensor and/or relaying networks.
Compression Techniques for Sensor Network
The purpose of this research work is to fullfill the battery longevity requirement in sensor network by significantly reducing the information bit rate of the signal transmitted from sensors. To achieve this goal, Turbo decoding techniques will be used, where the correlation between the multiple sources is modeled as a hidden Markov source, and message passing takes place over the trellis diagrams representiong the source correlation.
Cooperative Coding for Multi-Hop Netwroks
In wireless multi-hop networks, cooperative coding techniques allow us to achieve diversity and coding gains, while also improving the throuput efficiency. This research work aims to develop signal relaying algorithms where account is taken of the fact that the signals received from the primary sender's and relayed terminals are correlated; The correlation is first estimated by the receiver, and then decoding of the codes used for relaying is performed using Turbo techniques.
Cross Layer Optimization in Wireless Communications Network towards Autonomous Resource Allocation and Adaptive Coding
Cross layer optimization is one of the crucial issues that have to be solved when designing spectrum- and power-efficient ubiquitous wireless networks. This research category includes a lot of issues, all related cross layer optimization , such as optimal resource allocation, adaptive coding and modulation, and scheduling. The major aim of this research category is to create algorithms that can bring autonomously the wireless networks to the optimal operation points using the message passing algorithm over the network nodes.
Semantic Language Analysis using Turbo Algorithms
Language recognition systems can be seen as a system having distributed multiple local decision making nodes. By utilizing the message passing technique over the multiple decision making nodes at different understanding level, the language literacy, as a whole, is expecetd to be significantly enhanced, especialy in the noisy environments. This research aims to establish algoerithms that can solve the semantic level language analysis using the Turbo techniques.
Turbo Estimation Techniques for Channel Prediction, Filtering, and Synchronization.
This research aims to apply the Turbo techniques to solving several estimation problems in wireless communications, including channel estimation, prediction, and synchronization. The factor graph-based detection and estimation technique using message passing algorithm will be used.

■Publications

◇Books

  • Chapter: "Equalization" in "Mobile Broadband Multimedia Networks: Techniques, Models and Tools for 4th Generation Communication Networks",Tadashi Matsumoto,Elsevier,2006,pp. 51-65
  • Chapter: "Iterative (Turbo) Signal Processing Techniques for MIMO Signal Detection and Equalization" in "Smart Antenna; State-Of-the-Art",Tadashi Matsumoto,EURASIP Book Series on Signal Processing and Communications: Hindawi,2005,pp.119-146

◇Published Papers

  • APDOA-DRSS Hybrid Factor Graph-based Unknown Radop Wave Geolocation,Shofiyai Nur Karimah , MuhammadReza Kahar, and Tad Matsumoto,The 2nd International Conference on Signal and System, 2018(ICSigSys 2018), Published. DOI 10.1109/ICSIGSYS.2018.8372773
  • LDPC-based Joint Source Channel and NetworkCoding for the MARC,Marwa Ben Abdessalem, Amin Zribiy, Tadashi Matsumoto, and Ammar Bouall`egue,The IEEE 6th International Conference on Wireless Networks and Mobile Communications WINCOM'18, October 16-19, 2018, Marrakesh, Morocco. Accepted
  • Joint Source-Channel Decoding for MDC-encoded Sources Transmitted over Relay Systems,Amin Zribi and Tad Matsumoto,International Conference on Antenna Measurement and Applications, 2018 IEEE CAMA, Sweden, from 3 to 6 September 2018. Accepted, to be published

Display All

◇Lectures and Presentations

  • Lossy Communications for IoT: from multi-terminal source coding viewpoint (Invited),Tadashi Matsumoto,The Eleventh China Wireless Sensor Network ConferenceTianjin, Oct 13-15, 2017,Tianjin University, China
  • Cooperative Communications: from the correlated source coding theorem viewpoint (Invited),Tadashi Matsumoto,The 7th International Conference on Electronics, Communications and Networks, Nov. 24-27, 2017, National Dong Hwa University, Hualien, Taiwan,Hualien, Taiwan
  • Lossy Communications for IoT: from multi-terminal source coding viewpoint (Invited),Tadashi Matsumoto,"Wenjin Forum" (中国Anhui 師範大学における最高レベルの技術討論フォーラム),中国Anhui 師範大学,2017/10/16

Display All

■Extramural Activities

◇Academic Society Affiliations

  • IEEE,Vehicular Technology Society Board of Governer, Fellow,2002-
  • IEICE,Member,1978-

◇Other Activities

  • 2nd International Conference on Telematics and Future Generation Networks to be held,ADVISORY COMMITTEE,2018/04/01 - 2018/07/26
  • 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC),Technical Program Committee Track Co-Chair,2017/10/01 - 2018/09/12
  • EU FP7 RESCUR Projec Summer School, Technically Supported by School of Information Science, JAIST,Lecturer,2015/08/24 - 2015/08/17

Display All

■Academic Awards Received

  • Recognition of Outstanding Contribution as an IEEE VTS Distinguished Speaker,IEEE Vehicular Technology Society,201807
  • IEEE Communications Society 2017 Exemplary Reviewers Recognition,IEEE Communications Society,2018
  • Recognition of Outstanding Distinguished Lecturer for 2011-2015, by IEEE Vehicular Technology Society,IEEE

Display All