论文信息 - Radar Signals Sorting with Kohonen Neural Net

Radar Signals Sorting with Kohonen Neural Net

Kohonen neural network is capable of selforganizing and recognizing clustering center, which is used in many artificial Intelligence (AI) fields. A radar electronic support measures (ESM) system must sort the received radar pulse signal to cells with the same features according to the single pulse parameters, such as radio frequency (RF), angle of arrival (AOA), pulse width (PW), etc. Considering the variety character of pulse parameters, a new definition of distance is proposed in this paper, which decreases the effect of large variety range of special parameter among them. And in order to increase the convergence speed, this paper revised the SOFM algorithm according to the effect of pulse special parameter. The computer simulation shows the validity of these improvements.

Yongjun Zhao | Jianqi Lu | Chuang Zhao

[1] Nasser M. Nasrabadi,et al. Vector quantization of images based upon the Kohonen self-organizing feature maps , 1988, ICNN.

[2] H. Ritter,et al. Convergence properties of Kohonen's topology conserving maps: fluctuations, stability, and dimension selection , 1988, Biological Cybernetics.

[3] P. S. Ray. A novel pulse TOA analysis technique for radar identification , 1998 .

[4] Donald F. Specht,et al. Generation of Polynomial Discriminant Functions for Pattern Recognition , 1967, IEEE Trans. Electron. Comput..

[5] Teuvo Kohonen,et al. Self-Organization and Associative Memory, Third Edition , 1989, Springer Series in Information Sciences.

[6] H. E. Hassan. A new algorithm for radar emitter recognition , 2003, 3rd International Symposium on Image and Signal Processing and Analysis, 2003. ISPA 2003. Proceedings of the.

[7] S. Carrato. Image vector quantization using ordered codebooks: Properties and applications , 1994, Signal Process..

[8] D. S. Bradburn. Reducing transmission error effects using a self-organizing network , 1989, International 1989 Joint Conference on Neural Networks.