盲蔽訊號分離之可適性學習率設計與優化

本論文提出基於區間第二型模糊小腦模型 (interval type-2 fuzzy cerebellar model articulation controller, T2FCMAC) 調整盲蔽訊號分離 (blind source separation, BSS) 之學習率，並利用粒子群最佳化 (particle swarm optimization, PSO) 演算法優化T2FCMAC提高系統之性能。近年來獨立成份分析 (independent component analysis, ICA) 演算法被提出用以解決BSS問題，而基於學習率的梯度演算法為其中一種。為了平衡系統的失調性及收斂速率，基於輸出訊號的二階和高階相關係數，利用T2FCMAC調整BSS之學習率。T2FCMAC為一種擁有較好學習能力的網路系統用以調整BSS之學習率，而為了提高系統性能，本論文利用粒子群最佳化演算法將T2FCMAC優化。此外在應用方面，基於T2FCMAC調整BSS之學習率方法可實現在影像加密 (image encryption) 系統，根據BSS的欠定問題 (underdetermined problem) ，將原始影像(original images)加密時加入密鑰影像(key images)可確保加密系統的安全性。

關鍵字

盲蔽訊號分離；獨立成份分析；小腦模型；區間型第二型模糊系統；粒子群最佳化；影像加密

並列摘要

This thesis proposes the interval type-2 fuzzy cerebellar model articulation controller (T2FCMAC)-based learning rate adjustment for the blind source separation (BSS). To enhance the performance of the T2FCMAC-based learning rate, the T2FCMAC system is optimized by PSO algorithm. Recently, independent component analysis (ICA) algorithms have been proposed to solve the BSS problems. The gradient algorithm is a popular method deals with separating independent signal step by step with learning rate. In order to balance the mis-adjustment and the speed of convergence, the learning rate will be computed by T2FCMAC with input of the second-order and higher order correlation coefficients of output components. The T2FCMAC system is a more generalized network with better learning ability to provide the adaptive learning rate of the BSS. Furthermore, the method we proposed can be implemented in the image encryption. According to the underdetermined BSS problem, the original images are encrypted by key images to ensure the security of the cryptosystem. Finally, we present the T2FCMAC-based learning rate for the BSS and implement it in the image encryption system.

並列關鍵字

blind source separation ； independent component analysis ； cerebellar model articulation controller ； interval type-2 fuzzy system ； particle swarm optimization ； image encryption

參考文獻

[52] H. H. Yang, S. Amari, and A. Cichocki, “Information-theoretic Approach to Blind

[1] J. S. Albus, “A new approach to manipulator control: The cerebellar model articulation controller (CMAC),” J. Dynamic Syst. Meas. Control, vol. 97, no. 3, pp. 220–227, 1975.

[2] J. S. Albus, “Data storage in the cerebellar model articulation controller (CMAC),” J. Dynamic Syst. Meas. Control, vol. 97, no. 3, pp. 228–233, 1975.

[3] S. Amari, A. Cichocki, and H. H. Yang, “A new learning algorithm for blind signal separation,” Advanced in Neural Information Processing Systems. Cambridge, MA: MIT Press, vol. 8, pp. 752–763, 1996.

[4] S. Amari and A. Cichocki, “Adaptive blind signal processing—Neural network approaches,” Proc. IEEE, vol. 86, pp. 2026–2048, Nov. 1998.

國際替代計量

盲蔽訊號分離之可適性學習率設計與優化

全文下載

主題瀏覽