Non‐cooperative identification of civil aircraft using a generalised mutual subspace method

Abstract

Abstract:The subspace-based methods are effectively applied to classify sets of feature vectors by modelling them as subspaces. However, their application to the field of non-cooperative target identification of flying aircraft is barely seen in the literature. In these methods, setting the subspace dimensionality is always an issue. Here, it is demonstrated that a modified mutual subspace method, which uses softweights to set the importance of each subspace basis, is a promising classifier for identifying sets of range profiles coming from real in-flight targets with no need to set the subspace dimensionality in advance. The assembly of a recognition database is also a challenging task. In this study, this database comprises predicted range profiles coming from electromagnetic simulations. Even though the predicted and actual profiles differ, the high recognition rates achieved reveal that the algorithm might be a good candidate for its application in an operational target recognition system.

Radar HRRP Target Recognition Based on Stacked Autoencoder and Extreme Learning Machine

“…Radar target recognition based on high-resolution range profile (HRRP) has become a research hotspot due to the acquisition and processing of HRRP data being relatively easy [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ]. However, the non-cooperative recognition [ 8 , 9 ] with limited training samples is a challenging task. In the non-cooperative situation, such as at the battle with time, the amount of data under the test is usually huge but the training data is limited.…”

Radar HRRP recognition based on discriminant deep autoencoders with small training data size

“…In HRRP based target recognition, one of the most challenging task is the non-cooperative recognition [4,5] with limited training data size. Due to the poor measurement situation (e.g.…”