A single layer fast learning fuzzy controller/filter

This paper describes an orthonormal neural network (ONN) designed to function as a fuzzy controller/filter. This single layer network is tested in simulating a third order function with two input variables. The simulation results are described. In this application ONN used one output and two input nodes, Each input node is connected to seven /spl pi/ functions to perform fuzzification resulting in 7/sup 2/=49 fuzzy rules. The conclusions of these rules are the system output and they are given in numerical form. As a result of using a new fast learning algorithm, the training process requires only 100 cycles. The fuzzy controller with the help of some switching can function as a bandpass and/or in-band filter.<<ETX>>

[1]  M. E. Ulug,et al.  A signal processing neural network resembling the simple cells of the visual cortex , 1992, [Proceedings] 1992 RNNS/IEEE Symposium on Neuroinformatics and Neurocomputers.

[2]  D. Pollen,et al.  Phase relationships between adjacent simple cells in the visual cortex. , 1981, Science.

[3]  M. E. Ulug.  A rapid learning orthonormal neural network for signal processing , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[4]  M. E. Ulug.  ANN bandpass filters for electro-optical implementation , 1992, [Proceedings 1992] IJCNN International Joint Conference on Neural Networks.

[5]  Sankar K. Pal,et al.  Fuzzy measures in determining seed points in clustering , 1986, Pattern Recognit. Lett..