Revista: IEEE Transactions on Aerospace and Electronic Systems. Vol. 45. Nº 4. Pp. 1373-1384. Octubre 2009.

Autores: J. M. Munoz-Ferreras, F. Pérez-Martínez, y M. Burgos-García

Helicopter classification using a high resolution linear frequency modulated continuous wave (LFMCW) radar is addressed. A blade echo modelling for this radar system is derived, taking into account that the stop-and-go assumption is not longer applicable. The standard classification schemes based on the L/N-quotient and the evenness of the number of blades are easily reproduced in this high resolution context, although two main advantages are added: a high pulse repetition frequency (PRF) to acquire the Doppler bandwidth is no longer necessary, and the time for which we receive returned signal from the target may be maximized. On the other hand, a characteristic critical classification problem, for which the standard approaches do not properly work, is presented. Two classification schemes are proposed for this situation, the first one based on coherent matched masks and the second one based on incoherent masks where a robustness parameter has also been defined. Identification probabilities as a function of the signal-to-noise ratio (SNR) are obtained in order to characterize the performance of both techniques. Real data validate both the derived model and the proposed classification algorithms.