A Novel Fast Kolmogorov's Spline Complex Network for Pattern Detection

In this paper, we present a new fast specific complex-valued neural network, the fast Kolmogorov’s Spline Complex Network (FKSCN), which might be advantageous especially in various tasks of pattern recognition. The proposed FKSCN uses cross correlation in the frequency domain between the input data and the input weights of neural networks. It is proved mathematically and practically that the number of computation steps required for the FKSCN is less than that needed by conventional Kolmogorov’s Spline Complex Network (CKSCN). Simulation results using MATLAB confirm the theoretical computations. Keywords— Fast Kolmogorov’s Spline Complex Network, Cross Correlation, Frequency Domian, Pattern Detection, Neural Networks, Modeling of Time-Variant Multidimensional Data.