![](bg62.jpg)
97
0.2
0.4
0.6
0.8
1
scales
(a)
1
4
8
12
16
0
0.05
0.1
scales
(b)
1
4
8
12
16
0
0.005
0.01
0.015
scales
(c)
time (s)
100 200 300 400 500 600
1
4
8
12
16
Figure 9 – (a) CWT of the signal; (b) CWT of the signal after finite
differentiation; (c) Teager-Kaiser energy of the CWT after finite
differentiation
7. CONCLUSIONS
Radar based helicopter classification is not an easy task spe-
cially under noisy environment, and this paper shows that
Continuous Wavelet Transform can be a very useful tool
when trying to determine one of the most important features
for classifying helicopters, the time between successive
peaks.
In [2], a way of finding the blade tip velocity from a coher-
ently integrated radar signal is presented. This paper presents
a way of measuring time between peaks also using a coher-
ently integrated radar signal.
A conclusion achieved from this work is the expression for
the minimum radar pulse repetition frequency (PRF) neces-
sary for avoiding aliasing in the helicopter echo presented in
equation (5).
It is important to mention that in the case where the radar
PRF does not obey the equation (5) and the aliasing occurs, it
is almost impossible to measure blade tip velocity and, there-
fore, in this case the time between peaks can be the only fea-
ture available when trying to perform helicopter classifica-
tion. In these cases, the method presented in this paper for
measuring time between peaks also achieved good results.
Finally, the use of other techniques, as the Wavelet transform
modulus maxima and the Teager-Kaiser energy, shows that
many other tools already existent in signal processing can
turn out to be very useful in the determination of features for
this task.
0.01
0.02
0.03
0.04
0.05
scales
(b)
1
4
8
12
16
100 200 300 400 500 600
1
4
8
12
16
time (s)
scales
(c)
0.2
0.4
0.6
0.8
1
amplitude
(a)
Figure 10 – (a) Signal in time domain; (b) Teager-Kaiser energy of
the CWT after finite differentiation; (c) Wavelet transform modulus
maxima lines (WTMML).
REFERENCES
[1] C. E. Rotander and H. von Sydow, “Classification of Helicopters
by the L/N-quotient”, in Proc. IEE Radar 97 (Conf. Publ. No. 449),
Oct. 1997, pp. 629-633.
[2] J. M. Tikkinen, E. E. Helander and A. Visa (2005), “Joint Utili-
zation of Incoherently and Coherently Integrated Radar Signal in
Helicopter Categorization”, in Proc. IEEE International Radar
Conference, May 2005, pp. 540-545.
[3] S. Yoon, B. Kim and Y. Kim, “Helicopter Classification Using
Time-Frequency Analysis”, IEE Electronics Letters Online, vol. 36,
pp. 1871-1872, Oct. 2000.
[4] J. Misiurewicz, K. Kulpa and Z. Czekala, “Analysis of Recorded
Helicopter Echo”, in Proc. IEE Radar 97 (Conf. Publ. No. 449),
Oct. 1997, pp. 449-453.
[5] J. Ding, Z. Yu, D. Yan and C. Ruan, “Modeling and Validation
of Modulated Characteristics for Aircraft Rotating Structure”, in
Proc. IEEE International Radar Conference, May 2005, pp. xiii-
xxiii.
[6] A. V. Oppenheim, R. W. Schafer and J. R. Buck, Discrete-Time
Signal Processing, Prentice Hall, 1999.
[7] M. S. Diallo, M. Kulesh, M. Holschneider, F. Scherbaum and F.
Adler, “Characterization of polarization attributes of seismic waves
using continuous wavelet transforms”, Geophysics, vol. 71, pp.
V67–V77, May-June 2006.
[8] J. Sadowsky, “Investigation of Signal Characteristics Using the
Continuous Wavelet Transform”, Johns Hopkins APL Technical
Digest, vol. 17, pp. 258-269, 1996.
[9] J. F. Kaiser, “On a simple algorithm to calculate the ‘energy’ of
a signal”, in Proc. IEEE ICASSP, 1990, pp. 381-384.
[10] M. C. de Matos and P. R. S. Johann, “Revealing geological
features through seismic attributes extracted from the wavelet trans-
form Teager-Kaiser energy”, 77th Annual International Meeting,
SEG, Expanded Abstracts, pp. P1442-P1446.
[11] M. C. de Matos, P. L. M. Osorio and P. R. S. Johann, “Unsu-
pervised seismic facies analysis using wavelet transform and self-
organizing maps”, Geophysics, vol. 72, pp. P9-P21, Jan.-Feb. 2007.