林宗男教授主要研究領域為網路通訊、分散式多媒體系統及視訊/影像處理。目前積極進行中之研究計畫包括：(一)可適性階層多解析彩色影像處理與分析(國科會)；(二)教育部通訊科技改進計畫(教育部)。在此計畫中，結合了電機系及電信所多位專精網路通訊教授之教學與研究之能量，重點領域包括網路運算、無線網路內容分布，以及行動計算資料管理；(三)叢集運算系統架構之設計與實作，計畫包括以叢集式之架構設計高效能之運系統及對此系統之評估與分析(廣達電腦)；(四)數位學習系統架構之設計及效能評估(國科會)，包括數位學習系統之規畫與設計以及與各教學機構合作之效能測試與回饋評量；(五)對等式網路搜尋傳遞及安全議題之研究(國科會)。此研究針對等式網路中之資料搜尋、分布、傳遞及網路之延展性、可靠性、安全性考量，設計高效能之對等式網路系統。

The major research focus of Professor Tsung-Nan Lin’s research team is in the technical developments related to network communication, distributed multimedia systems, and video/image processing.  The projects under active investigations at the time of writing include: (1) Adaptive hierarchical multi-resolution color image processing algorithms (NSC); (2) The project for improvement of communication technology of ministry of education (Ministry of Education). This project combines the strength on both research and education of several professors excelling in network communication, focusing on web computing, data management in mobile computing, and content distribution over wireless technology; (3) Design and implementation of cluster computing systems (Quanta Computer). The goals of this project are to design a high performance computing system under the cluster architecture and analyze and evaluate the performance of this system; (4) Platform Design and Simulation of E-Learning System Architecture, including the coordination and design of the e-learning platform and cooperation of educational institutes for performance measurement and feedback evaluation. (NSC); and (5) Routing Algorithms and Security Issues in P2P Content Networks (NSC), considering issues of data distribution, search, and delivering and scalability, reliability, and security in the networks in order to design a high performance P2P network systems.

• BS: Electrical Engineering, National Taiwan University

• PhD: Electric Engineering, Princeton University

• Address: Room 536, 
                  Department of Electrical Engineering,
                  National Taiwan University,
                  Taipei 106, Taiwan

• Phone: +886-2-23635251ext.536

• Fax: +886-2-23656149

• E-mail: tsungnan@cc.ee.ntu.edu.tw

• Personal homepage: http://140.112.41.70

1. Tsungnan Lin, B. G. Horne, P. Tino and C. L. Giles, Learning long-term dependencies in NARX recurrent networks.  IEEE Transactions on Neural Networks, Vol. 7, No. 6, p. 1329-1338, 1996. This paper won the 1997 Outstanding Paper Award of IEEE Transactions on Neural Networks.  

2. Tsungnan. Lin, C. L. Giles, B. G. Horne, S. Y. Kung, A Delay Damage Model Selection Algorithm for NARX Neural Networks. IEEE Transactions on Signal Processing, "Special Issue on Neural Networks," 45(11), p. 2719, 1997. This paper won the IEEE Signal Processing Society 1998 Young Author Best Paper Award.

3. C. L. Giles, B. G. Horne and Tsungnan Lin Learning a Class of Large Finite State Machines with a Recurrent Neural Network. Neural Networks, Vol. 8, No. 9, p. 1359-1365. 1995.

4. Tsungnan Lin, B. G. Horne, C. L. Giles. How Embedded Memory in Recurrent Neural Network Architectures Helps Learning Long-term Temporal Dependencies, Neural Networks, Vol. 11, p.861, 1998.

5. Tsungnan Lin, Joseph Shu, Adaptive-Hierarchical-Filtering Technique for Image Enhancement, IEEE Pacific-Rim Conference on Multimedia 2002.

6. Tsung-Nan Lin and Joseph Shu, A Color Consistency Algorithm Between Different Printers, Image Processing, Image Quality, Image Capture, System Conference, p409-411, The Society for Imaging Science and Technology, 1998.