The performance of intelligent and unintelligent approaches on aircraft identification tasks

This paper is a report of our research progress in the area of pattern recognition in endeavouring finding a novel method for aircraft identification. In this paper we address the performance comparison between intelligent and unintelligent approaches in performing aircraft identification tasks in a generic system called generic aircraft identification system (G-AIS) especially in accessing the knowledge stored in database. For this purpose, we select two types of neural networks namely, back propagation network (BPN) for the supervised exemplar and adaptive resonance theory (ART) for the unsupervised one for intelligent identification approach. For unintelligent approach we select standard approach in database technology namely linked list. As for previous research, we use two kinds of input namely aircraft radar cross section (RCS) and average speed. Their performance will be validated by using already-learnt and never-learnt patterns.