Enhanced ART1-based Self-Organizing Supervised Learning Algorithm for Channel Optimization in Mobile Cellular Networks
Kwang-Baek Kim, Cheul-Woo Ro, Kwang-eui Lee, Kyung-min Kim · 2006
Summary In this paper, we propose a novel approach for evolving the architecture of a multi-layer neural network. Our method uses combined ART1 algorithm and Max-Min neural network to selfgenerate nodes in the hidden layer. We have applied the proposed method to the optimal channel allocation problem in mobile cellular networks. Experimental results show that the proposed method has better performance than conventional neural networks and the resulting neural network computes the optimal guard channel number g within ignorable error bound for GoS.