第五章 結論
近年來生物學家、醫學家對於研究人腦的運作以及分析不遺餘力，目前各領 域學家也提出了許多針對腦波訊號分析以及處理的相關研究。本論文以 FPGA 為 基礎實現了一套棘波分類系統，透過管線化的方式將運算速度大幅縮短，因此就 算於低時脈運作下本系統仍然能擁有著卓越的輸出產能。同時，本系統在合理的 資源消耗下，對於運算時間、準確率與功率消耗取得了一個較佳的平衡點。因此 本論文兼具著低資源消耗與即時運算的優點。
根據前一章比較的結果，可以知道本論文所提出的架構不論在分類正確率或 計算速度上與現有提出的設計[4,5,6]相較都有著不錯的表現，雖然採用 FPGA 來 實現本系統可能會使功率消耗比 ASIC 來得大，但整個系統的功率密度仍舊在安 全範圍內，因此雖然功率消耗不如 ASIC 低，但對於植入體內的需求與安全考量 亦能滿足。
總結來說，本論文所提出的電路架構在棘波分類的應用上確是有其助益，而 且也具有充分的延伸性與可調整性，同時與現有的架構相比更有著不少的優勢。
因此，本論文所設計的棘波分類系統確是有其需求和效率的電路架構。
參考文獻
[1] M.S. Lewicki,“A review of methods for spike sorting: the detection and classifi-cation of neural action potentials,” Network Computer Neural System, Vol. 9, pp.
R53R78, 1998.
[2] S. Haykin, Neural Networks and Learning Machines, 3rd ed.; Pearson: Upper Saddle River, NJ, USA, 2009.
[3] T.D. Sanger, “Optimal unsupervised learning in a single-layer linear feedforward neural network,” Neural Network, Vol. 12, pp.459-473, 1989.
[4] Y. Sun, S. Huang, J. J. Oresko, and A. C. Cheng, “Programmable Neural Pro-cessing on a SmartDust for Brain-Computer Interfaces,” IEEE Trans. Biomedical Circuits and Systems, Vol. 4, pp.265-273, 2010.
[5] T.-C. Chen, W. Liu and L.-G. Chen,“VLSI Architecture of Leading Eigenvector Generation for On-chip Principal Component Analysis Spike Sorting System,”
Proc. 30th Annual International IEEE EMBS Conference Vancouver, British Co-lumbia, Canada, pp.3192-3195, 2008.
[6] B. Yu, T. Mak, X. Li, F. Xia, A. Yakovlev, Y. Sun, C.-S. Poon,“A Reconfigurable Hebbian Eigenfilter for Neurophysiological Spike Train Analysis,” Proc. Interna-tional Conference on Field Programmable Logic and Applications, pp.556-561, 2010.
參考文獻
[7] M. Aghagolzadeh, F. Zhang, and K. Oweiss,“An Implantable VLSI Architecture for Real Time Spike Sorting In Cortically Controlled Brain Machine Interfaces,”
Proc. 32nd Annual International Conference of the IEEE EMBS Conference, Buenos Aires, Argentina, pp.1569-1572, 2010.
[8] A. M. Kamboth, M. Raetz, K. G. Oweiss, and A. Mason, “Area-power efficient VLSI implementation of multichannel DWT for data compression in implantable neuroprosthetics,” IEEE Trans. Biomedical Circuits and Systems, Vol. 1,
pp.128-135, 2007.
[9] A. M. Kamboh, and A. J. Mason, “On-Chip Feature Extraction for Spike Sorting in High Density Implantable Neural Recording Systems,” Proc. IEEE Biomedical Circuits and Systems Conference, pp.13-16, 2010.
[10] Y. Yang, and A. J. Mason,“On-Chip Spike Clustering & Classification using Self Organizing Map for Neural Recording Implants,” Proc. IEEE Biomedical Circuits and Systems Conference, pp.145-148, 2011.
[11] S.-J. Lin, Y.-T. Hung, and W.-J. Hwang, “Efficient hardware architecture based on generalized Hebbian algorithm for texture classification,” Neurocomputing, pp.3248-3256, 2011.
[12] N. Sudha, A.R. Mohan, P.K. Meher, “A self-configurable systolic architecture for face recognition system based on principal component neural network,” IEEE Trans. Circuits Syst. Video Technol, Vol. 21, pp.1071-1084, 2011.
[13] S. -J. Lin, W. -J. Hwang, andW. -H. Lee, “FPGA Implementation of Generalized Hebbian Algorithm for Texture Classification,” Vol. 12, pp.6244-6268, Sensors, 2012.
modified fuzzy c-means clustering algorithm for realtime applications,” Micro-processor and Microsystems, Vol. 29, pp.375-380, 2005.
[15] H. -Y. Li, W. -J. Hwang, and C. -Y. Chang, “Efficient Fuzzy C-Means Architec-ture for Image Segmentation, Sensors, Vol. 11, pp.6697-6718, 2011.
[16] Y.-J. Yeh, H.-Y. Li, C.-Y. Yang, and W.-J. Hwang, ”Fast Fuzzy C-Means Cluster-ing Based on Low-Cost High-Performance VLSI Architecture in Reconfigurable Hardware,” pp.112-118, IEEE International Conference on Computational Sci-ence and Engineering, 2010.
[17] L. S. Smith and N. Mtetwa, “A tool for synthesizing spike trains with realistic in-terference,” Vol. 159, pp.170-180, Journal of Neuroscience Methods, 2007.
[18] Altera Corporation. NIOS II Processor Reference Handbook ver 10.0. 2010.
Available online: http://www.altera.com/literature/lit-nio2.jsp (accessed on 3 May 2012).
[19] Altera Corporation. SOPC Builder User Guide. 2011. Available
online:http://www.altera.com/literature/lit-sop.jsp (accessed on 3 May 2012).
[20] W. Reichert, “Indwelling neural implants: Stategies for contending with the in vivo environment,” in BMI-Related Thermal Studies. Boca Raton, FL: CRC, 2007.