Implementation of an Intelligent TargetClassifier with Bicoherence Feature Set

This paper examines the feasibility of bispectral analysing of acoustic signals emanated from underwater targets, for the purpose of classification. Higher order analysis, especially bispectral analysis has been widely used to analyse signals when non-Gaussianity and non-linearity are involved. Bicoherence, which is a normalized form of bispectrum, has been used to extract source specific features, which is finally fed to a neural network classifier. Vector quantization has been used to reduce the dimensionality of the feature set, thereby reducing computational costs. Simulations were carried out with linear, tan and log-sigmoid transfer functions and also with different code book sizes. It is found that the bicoherence feature set can provide acceptable levels of classification accuracy with a properly trained neural network classifier.

[1]  Mahmood R. Azimi-Sadjadi,et al.  Underwater target classification in changing environments using an adaptive feature mapping , 2002, IEEE Trans. Neural Networks.

[2]  G. R. Wilson,et al.  Nonlinear and non-Gaussian ocean noise , 1986 .

[3]  Zhishun Li,et al.  Ship recognition via its radiated sound: the fractal based approaches. , 2002, The Journal of the Acoustical Society of America.

[4]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.

[5]  M. Karimi,et al.  Ship noise classification using Probabilistic Neural Network and AR model coefficients , 2005, Europe Oceans 2005.

[6]  Leontios J. Hadjileontiadis,et al.  Analysis of Wheezes Using Wavelet Higher Order Spectral Features , 2010, IEEE Transactions on Biomedical Engineering.

[7]  Nafiz Arica,et al.  UNDERWATER ACOUSTIC SIGNAL RECOGNITION METHODS , 2009 .

[8]  Zhishun Li,et al.  Classification of ship-radiated signals via chaotic features , 2003 .

[9]  M.R. Raghuveer,et al.  Bispectrum estimation: A digital signal processing framework , 1987, Proceedings of the IEEE.

[10]  Qiang Huang,et al.  Underwater target classification using wavelet packets and neural networks , 2000, IEEE Trans. Neural Networks Learn. Syst..

[11]  Chrysostomos L. Nikias,et al.  Bispectrum estimation of transient signals , 1988, ICASSP-88., International Conference on Acoustics, Speech, and Signal Processing.

[12]  Jaakko Astola,et al.  Classification of ground moving targets using bicepstrum-based features extracted from Micro-Doppler radar signatures , 2013, EURASIP Journal on Advances in Signal Processing.

[13]  Stanley J. Wenndt,et al.  Bispectrum features for robust speaker identification , 1997, 1997 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[14]  Sohail Akhtar,et al.  Detection of helicopters using neural nets , 2001, IEEE Trans. Instrum. Meas..

[15]  Xinlong Wang,et al.  Ship classification using nonlinear features of radiated sound: an approach based on empirical mode decomposition. , 2010, The Journal of the Acoustical Society of America.

[16]  Nathan Intrator,et al.  Classification of underwater mammals using feature extraction based on time-frequency analysis and BCM theory , 1998, IEEE Trans. Signal Process..

[17]  Stanislaw Osowski,et al.  Higher order statistics and neural network for tremor recognition , 2002, IEEE Transactions on Biomedical Engineering.

[18]  M. Hinich,et al.  Bispectrum of ship‐radiated noise , 1989 .