Title :
BVQ neural network for non-connected service areas satellite coverage plans optimization
Author :
Burrascano, Pietro
Author_Institution :
INFOCOM Dept., Rome Univ., Italy
Abstract :
The use of the backpropagation vector quantization neural network, proposed by the author (1992) for the solution of the constrained optimization problem of satellite antenna coverage plans project, is extended to the case of non-connected service areas. The network´s structure and characteristics are described in order to clarify the proposed procedure. The effectiveness of the approach is demonstrated by means of examples.
Keywords :
backpropagation; neural nets; optimisation; satellite antennas; vector quantisation; backpropagation vector quantization neural network; constrained optimization problem; nonconnected service areas; satellite antenna; satellite coverage plans optimization; Backpropagation; Constraint optimization; Cost function; Design optimization; Earth; Neural networks; Probability density function; Satellite antennas; Training data; Vector quantization;
Conference_Titel :
Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
Print_ISBN :
0-7803-1421-2
DOI :
10.1109/IJCNN.1993.714083
