A FULLY COMPLEX-VALUED RADIAL BASIS FUNCTION NETWORK AND ITS LEARNING ALGORITHM
Authors:
- 1 August 2009
- journal article
- research article
- Published by World Scientific Pub Co Pte Ltd in International Journal of Neural Systems
- Vol. 19 (04) , 253-267
Abstract
In this paper, a fully complex-valued radial basis function (FC-RBF) network with a fully complex-valued activation function has been proposed, and its complex-valued gradient descent learning algorithm has been developed. The fully complex activation function, sech(.) of the proposed network, satisfies all the properties needed for a complex-valued activation function and has Gaussian-like characteristics. It maps Cn → C, unlike the existing activation functions of complex-valued RBF network that maps Cn → R. Since the performance of the complex-RBF network depends on the number of neurons and initialization of network parameters, we propose a K-means clustering based neuron selection and center initialization scheme. First, we present a study on convergence using complex XOR problem. Next, we present a synthetic function approximation problem and the two-spiral classification problem. Finally, we present the results for two practical applications, viz., a non-minimum phase equalization and an adaptive beam-forming problem. The performance of the network was compared with other well-known complex-valued RBF networks available in literature, viz., split-complex CRBF, CMRAN and the CELM. The results indicate that the proposed fully complex-valued network has better convergence, approximation and classification ability.Keywords
This publication has 33 references indexed in Scilit:
- ONLINE DETECTION OF THE MODALITY OF COMPLEX-VALUED REAL WORLD SIGNALSInternational Journal of Neural Systems, 2008
- FREQUENCY-MULTIPLEXING ABILITY OF COMPLEX-VALUED HEBBIAN LEARNING IN LOGIC GATESInternational Journal of Neural Systems, 2008
- CHAOTIC SYNCHRONIZATION USING A NETWORK OF NEURAL OSCILLATORSInternational Journal of Neural Systems, 2008
- PSEUDO-RELAXATION LEARNING ALGORITHM FOR COMPLEX-VALUED ASSOCIATIVE MEMORYInternational Journal of Neural Systems, 2008
- ASSOCIATIVE MEMORY IN QUATERNIONIC HOPFIELD NEURAL NETWORKInternational Journal of Neural Systems, 2008
- THE UNIQUENESS THEOREM FOR COMPLEX-VALUED NEURAL NETWORKS WITH THRESHOLD PARAMETERS AND THE REDUNDANCY OF THE PARAMETERSInternational Journal of Neural Systems, 2008
- FLEXIBLE NONLINEAR BLIND SIGNAL SEPARATION IN THE COMPLEX DOMAINInternational Journal of Neural Systems, 2008
- LEARNING BY CRITERION OPTIMIZATION ON A UNITARY UNIMODULAR MATRIX GROUPInternational Journal of Neural Systems, 2008
- POLARIZED SIGNAL CLASSIFICATION BY COMPLEX AND QUATERNIONIC MULTI-LAYER PERCEPTRONSInternational Journal of Neural Systems, 2008
- The complex backpropagation algorithmIEEE Transactions on Signal Processing, 1991