MULTIDIMENSIONAL WAVELET FRAMES

Pati and Krishnaprasad [1] first studied the connection between neural networks and wavelet transforms. Zhang and Benveniste [2] gave a different treatment of this connection. The problem of constructing multidimensional wavelet frames for use in neural networks has not been satisfactorily studied, however. In this report, one-dimensional wavelet frame is generalized to the multidimensional Ease by using single scaling and multiscaling parameters. The construction of multi-dimensional wavelet frames is also discussed. These results provide more insight on the use of wavelets in neural networks.