本論文針對目前現有的棘波分類系統設計架構，並使用Network on chip技術於硬體中實現此架構。本論文採用Generalized Hebbian Algorithm (GHA) 來擷取棘波的特徵值，搭配Fuzzy C-Means (FCM) 演算法將擷取到的棘波特徵值進行分類。且對GHA電路稍作修改使的原本在高雜訊干擾下無法正確分類的問題成功解決，GHA演算法可高速計算主成分特徵值供後續分群演算法進行運算，同時利用FCM演算法對於初始質心選取好壞不敏感的特性可獲得較佳的分類結果。為了減少硬體資源的消耗，GHA架構中在計算調整不同組權重值時皆共享相同一塊計算電路，而FCM採用逐步增量計算權重係數與質量中心點，這可以避免原本需要大量儲存空間儲存權重係數矩陣所造成的空間消耗。因此，本論文所提出的架構同時擁有低area cost與高輸出產量的優點。加上採用Network on chip(NOC)技術，使本論文之棘波分類系統執行速度大為提升。為了驗證本論文所提出的架構有效性，我們於現場可程式邏輯閘陣列 (Field Programmable Gate Array , FPGA) 中實作出本架構並進行實際效能量測。實驗結果證明針對棘波分類本論文所提出的架構同時具有低判斷錯誤率、低area cost與高速計算的優點。

1.M.S. Lewicki,“A review of methods for spike sorting: the detection and classification of neural action potentials,” Network Computer Neural System, Vol. 9, pp. R53R78, 1998.
2.M.A. Lebedev and M.A.L. Nicolelis, “Brainmachine interfaces: past, present and future,” Trends in Neurosciences, Vol.29, pp.536-546, 2006.
3. E.E. Fetz, “Real-time control of a robotic arm by neuronal ensembles,” Natural Neural Science,Vol. 2, 00.583-584, 1999.
4. S. Hauck and A. Dehon, Reconfigurable Computing: The Theory and Practice of FPGA-Based Computing, Morgan Kaufmann: San Fransisco, CA, USA, 2008.
5. S. -J. Lin, W. -J. Hwang, and W. -H. Lee, “FPGA Implementation of Generalized Hebbian Algorithm for Texture Classification,” Vol. 12, pp.6244-6268, Sensors, 2012.
6. S. Haykin, Neural Networks and Learning Machines, 3rd ed.; Pearson: Upper Saddle River, NJ,USA, 2009.
7. T.D. Sanger, “Optimal unsupervised learning in a single-layer linear feedforward neural network,Neural Network, Vol. 12, pp.459-473, 1989.
8. S. Miyamoto, H. Ichihashi, and K. Honda, Algorithms for Fuzzy Clustering, Springer: Berlin,Heidelberg, Germany, 2010.
9. J. C. Bezdek, Pattern Recognition with Fuzzy Objective Function Algoritms, Plenum Press: NewYork, 1981.
10.T. Bjerregaard and S. Mahadevan, A Survey of Research and Practices of Network-on-Chip, ACM Computing Surveys, Vol. 38, 2006.
11. 李偉豪,應用於棘波分類之硬體架構實現,國立台灣師範大學資訊工程研究所,2012
12 Altera Corporation. SOPC Builder User Guide. online:http://www.altera.com/literature/lit-sop.jsp
13. Holger Blume, Thorsten von Sydow, Jochen Schleifer and Tobias G. Noll, Petri Net Based Modelling of Communication in Systems on Chip, Petri Net - Theory and Applications (book), Vienna, Austria,2008 ,
14. 周凱楓,晶片網路多核心處理器之模擬器設計與實現,國立成功大學資訊工程學系,2007
15. Altera Corporation. Quartus II Handbook Version 13.0. 2013. Available online:
http://www.altera.com/literature/lit-qts.jsp (accessed on 26 June 2013).
16. Altera Corporation. NIOS II Processor Reference Handbook ver 11.0. 2011. Available online: http://www.altera.com/literature/lit-nio2.jsp (accessed on 26 June 2013).
17. T. Bjerregaard and S. Mahadevan, A Survey of Research and Practices of Network-on-Chip, ACM Computing Surveys, Vol. 38, 2006.
18. Y. Sun, S. Huang, J. J. Oresko, and A. C. Cheng, “Programmable Neural Processing on a Smartdust for Brain-Computer Interfaces,” IEEE Trans. Biomedical Circuits and Systems, Vol. 4, pp.265-273,2010.
19. B. Yu, T. Mak, X. Li, F. Xia, A. Yakovlev, Y. Sun, and C.-S. Poon,“A Reconfigurable Hebbian Eigenfilter for Neurophysiological Spike Train Analysis,” Proc. International Conference on Field Programmable Logic and Applications, pp.556-561, 2010.
20. A. Oliynyk, C. Bonifazzi1, F. Montani and L. Fadiga1, ”Automatic online spike sorting with singular value decomposition and fuzzy C-mean clustering,” BMC Neural Science, Vol. 13, 2012.
21. W.-J. Hwang and H. Chen, Efficient VLSI Architecture for Spike Sorting Based on Generalized Hebbian Algorithm, Proc. European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, pp. 71-76, 2013.
22. J. Lazaro, J. Arias, J.L. Martin, C. Cuadrado, A. Astarloa, “Implementation of a modified fuzzy c-means clustering algorithm for realtime applications,” Microprocessor and Microsystems, Vol.29, pp.375-380, 2005.