Abstracts Submitted: 510
Number of Users: 710
View Abstracts Submitted
Back to home Page
In the proceedings, we treat the classification problem of surface and underwater targets using convolutional neural network (CNN) in the range-independent environment. The mode-space covariance matrix is used as the input data for the CNN. The mode function coefficients are estimated from the mode filtering of the vertical line array (VLA) data obtained from numerical propagation model. Our results are compared with those of two machine learning methods of random forests and support vector machine, and the classical methods in signal processing such as conventional matched-mode processing and other data-based techniques. In addition, we will present the results using the phone-space covariance matrix.
© Copyright 2017 All Rights Reserved