One and Two Dimensional Self Organized Learning Applied To Global Positioning System (GPS) Data

摘要

In this paper, we applied self organizing feature map learning to Global Positioning System Data with multiple inputs and multiple outputs. We used a new visualization method to display several one dimensional data versus connection weights. Also we studied the effect of several learning rates on cluster separation errors; furthermore we extended our research on self Organizing feature map to simulated navigation data in two dimensions and obtain promising results. In the near future, we plan to extend our approach to three and higher dimensional networks related to navigation and communication systems.