Design and Analysis of a CMOS Ratio-Memory Cellular Nonlinear Network (RMCNN) Requiring No Elapsed Time

A CMOS ratio-memory cellular nonlinear network (RMCNN) requiring no elapsed time is proposed. The correlations between any two neighboring cells are stored in the memories. The ratio weights of each cell are generated through a comparison of the four correlations around one cell with the mean value of these four correlations. With this method, the elapsed time required by the previously existing RMCNN algorithm is no longer required and, therefore, the ratio weights can be generated individually. Moreover, the use of multi-dividers can be avoided to make the circuit simple. Based on the proposed algorithm, a CMOS RMCNN chip requiring no elapsed time has been designed and fabricated using TSMC 0.35-μm 2P4M mixed-signal technology. In the fabricated chip, three test patterns can be learned and recognized.

[1]  L. Chua.  Cnn: A Paradigm for Complexity , 1998 .

[2]  Franco Bagnoli,et al.  Cellular Automata , 2002, Lecture Notes in Computer Science.

[3]  Chung-Yu Wu,et al.  The design of CMOS nonself-feedback ratio memory cellular nonlinear network without elapsed operation for pattern learning and recognition , 2005, 2005 9th International Workshop on Cellular Neural Networks and Their Applications.

[4]  L. Goras,et al.  On the Feature Extraction Performances of CNN Gabor-Type Filters in Texture Recognition Applications , 2006, 2006 10th International Workshop on Cellular Neural Networks and Their Applications.

[5]  Leon O. Chua,et al.  Cellular neural networks: applications , 1988 .

[6]  Tommaso Toffoli.  Cellular automata , 1998 .

[7]  Ángel Rodríguez-Vázquez,et al.  A 0.8-μm CMOS two-dimensional programmable mixed-signal focal-plane array processor with on-chip binary imaging and instructions storage , 1997, IEEE J. Solid State Circuits.

[8]  L.O. Chua,et al.  Cellular neural networks , 1993, 1988., IEEE International Symposium on Circuits and Systems.

[9]  Jerome A. Feldman,et al.  Connectionist Models and Their Properties , 1982, Cogn. Sci..

[10]  Ricardo Carmona-Galán,et al.  A CNN Universal Chip in CMOS Technology , 1996, Int. J. Circuit Theory Appl..

[11]  Chung-Yu Wu,et al.  A learnable cellular neural network structure with ratio memory for image processing , 2002 .

[12]  Lin-Bao Yang,et al.  Cellular neural networks: theory , 1988 .

[13]  S. Grossberg.  Nonlinear difference-differential equations in prediction and learning theory. , 1967, Proceedings of the National Academy of Sciences of the United States of America.

[14]  Tommaso Toffoli,et al.  Cellular automata machines - a new environment for modeling , 1987, MIT Press series in scientific computation.

[15]  Kah Phooi Seng,et al.  Lyapunov Theory-Based Multilayered Neural Network , 2009, IEEE Transactions on Circuits and Systems II: Express Briefs.

[16]  K. Nakamura,et al.  Recognition of object orientation and shape by a rotation spreading associative neural network , 2001, IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222).

[17]  Chung-Yu Wu,et al.  CMOS current-mode neural associative memory design with on-chip learning , 1996, IEEE Trans. Neural Networks.

[18]  Tamás Roska,et al.  The CNN universal machine: an analogic array computer , 1993 .

[19]  Tamás Roska.  Analogic algorithms running on the CNN Universal Machine , 1994, Proceedings of the Third IEEE International Workshop on Cellular Neural Networks and their Applications (CNNA-94).

[20]  Zhong Zhang,et al.  Cellular Neural Network for Associative Memory and Its Application to Braille Image Recognition , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.

[21]  Chung-Yu Wu,et al.  CMOS current-mode outstar neural networks with long-period analog ratio memory , 1995, Proceedings of ISCAS'95 - International Symposium on Circuits and Systems.