HRRP target recognition method based on attentional depth bidirectional loop neural network

Abstract

The invention discloses a HRRP target recognition method based on attentional depth bidirectional circulating neural network, firstly, the time-domain features of the data are extracted and the extracted time-domain features are segmented and non-uniformly quantized to obtain the coding of the local structure, and then the co-occurrence matrix is obtained by using the relationship between the local structure and several surrounding local structures, and then the structural embedding characteristics of the data are obtained by the co-occurrence matrix, and then the extracted embedded features are sent to a deep neural network composed of a full connection layer and a bi-directional loop neural network based on attention LSTM for training, at the same time, according to the output of the hidden layer of the loop network, the softmax layer is used to obtain the weight parameters of the attention model. Finally, the HRRP is identified by the softmax layer and the weight of the attention model, and the recognition results are obtained.

Description

technical field [0001] The invention belongs to the field of radar target recognition, and in particular relates to a radar high-resolution range image target recognition method based on deep attention cycle neural network. Background technique [0002] The radar high-resolution range profile (HRRP) can reflect the geometric structure information of the scattered point target in the direction of the radar line of sight, and compared with the synthetic aperture radar (SAR) and inverse synthetic aperture radar (ISAR) images, it contains the size of the target Important structural information such as the distribution of scattering points and scattering points has the advantages of easy acquisition and small storage capacity, so it has been widely used in radar target recognition technology, so the use of HRRP for target recognition has become a current research hotspot. At present, the classic models of HRRP recognition mainly include the template matching classifier (MCC-TMM) ...

AI Technical Summary

Problems solved by technology

At present, many literatures have applied the method based on the internal structure characteristics of HRRP samples to conduct experiments on the time domain characteristics of radar HRRP and the overall radar HRRP sequence, and achieved good recognition results. However, in practical applications, there are still the following: The main problems: (1) In the process of building the HMM model, it is necessary to assume that the signal conforms to the first-order Markov property, that is, the signal at the current time point is only related to the signal at the previous time point. , Spectral feature recognition, the current local feature structure only depends on the previous local feature structure, and the hidden overall physical structure correlation between the local feature structures of the sample cannot be further explored, and there is still a lot of room for improvement; (2 ) The cyclic neural network model based on the attention mechanism is a one-way cyclic neural network model, which can only establish dependencies in one direction, and fails to make good use of the overall structural information of HRRP; (3) The cyclic neural network model based on the attention mechanism The cyclic neural network used in the neural network model is too simple, without the use of long short-term memory (LSTM) units and deep neural networks, the output of the cyclic neural network will be excessively dependent on the local structure, and it is difficult to take into account the overall information

Images