The underwater acoustic noise radiated from the propellers of surface and underwater vehicles is characterized as high-frequency broadband noise modulated by low-frequency narrowband noise. Since the modulation is affected by propeller rotation speed, blade rate, and inception of cavitation, the measured propeller noise can be used to extract the information of propeller system and also to monitor its operating condition. In this research, we apply the cyclostationary signal processing technique to monitor vehicle propeller operation. In particular, we show that the rotation speed of a propeller can be estimated from cyclostationary characteristics such as the cyclic autocorrelation function and its Fourier counterpart - the cyclic spectrum. The cyclic spectrum shows the spectral support of the broadband modulating noise in addition to the modulation frequencies of the envelope signal. The effectiveness of the proposed technique is examined using two experimental data sets which are the scaled propeller model test data from a large cavitation tunnel and the ROV self-noise measurement data. The data analysis results show that the propeller rotation speed as well as the frequency band of the broadband noise can be identified using the cyclostationary statistics of acoustic noise data.
[1]
W. Gardner.
Exploitation of spectral redundancy in cyclostationary signals
,
1991,
IEEE Signal Processing Magazine.
[2]
Richard O. Nielsen,et al.
Sonar Signal Processing
,
1991
.
[3]
Antonio Napolitano,et al.
Cyclostationarity: Half a century of research
,
2006,
Signal Process..
[4]
Woojae Seong,et al.
Noise Localization Method for Model Tests in a Large Cavitation Tunnel Using a Hydrophone Array
,
2016,
Remote. Sens..
[5]
J. Antoni.
Cyclostationarity by examples
,
2009
.
[6]
Seung Woo Lee,et al.
Multiband Enhancement for DEMON Processing Algorithms
,
2013
.
[7]
J. Antoni,et al.
Detection of Surface Ships From Interception of Cyclostationary Signature With the Cyclic Modulation Coherence
,
2012,
IEEE Journal of Oceanic Engineering.