Radar High Resolution Range Profile Recognition with Small Sample Set Based on Linear Gaussian Mixture Dynamic Model

2016 ◽  
Vol 13 (10) ◽  
pp. 6549-6554
Author(s):  
Wang Penghui ◽  
Xia Shuangzhi ◽  
Zhang Xuefeng
Keyword(s):  
High Resolution ◽  
Dynamic Model ◽  
Gaussian Mixture ◽  
Small Sample ◽  
Statistical Characteristics ◽  
Recognition Method ◽  
Range Profile ◽  
Training Samples ◽  
Sample Set ◽  
High Resolution Range Profile

A novel recognition method is proposed to relieve the heavy requirement of training samples in the radar High Resolution Range Profile (HRRP) target recognition. Firstly, the statistical characteristics of HRRP’s frequency spectrum amplitude (FSA) are analyzed. Then a Linear Gaussian Mixture Dynamic Model (LGMDM) is proposed to describe the stationarity and multi-modality of the FSA. Afterwards, the Expectation Maximization (EM) algorithm is derived for model parameter estimation. Finally, experimental results show the proposed method can achieve satisfactory recognition accuracy and rejection performance with only a few training samples.

scholarly journals A Novel Radar HRRP Recognition Method with Accelerated T-Distributed Stochastic Neighbor Embedding and Density-Based Clustering

Sensors ◽  
2019 ◽  
Vol 19 (23) ◽  
pp. 5112 ◽  
Author(s):  
Hao Wu ◽  
Dahai Dai ◽  
Xuesong Wang
Keyword(s):  
Prior Information ◽  
Signal To Noise Ratio ◽  
Recognition Performance ◽  
Principal Component ◽  
High Dimensional ◽  
Recognition Method ◽  
Range Profile ◽  
Density Based Clustering ◽  
Computation Speed ◽  
High Resolution Range Profile

High-resolution range profile (HRRP) has attracted intensive attention from radar community because it is easy to acquire and analyze. However, most of the conventional algorithms require the prior information of targets, and they cannot process a large number of samples in real time. In this paper, a novel HRRP recognition method is proposed to classify unlabeled samples automatically where the number of categories is unknown. Firstly, with the preprocessing of HRRPs, we adopt principal component analysis (PCA) for dimensionality reduction of data. Afterwards, t-distributed stochastic neighbor embedding (t-SNE) with Barnes–Hut approximation is conducted for the visualization of high-dimensional data. It proves to reduce the dimensionality, which has significantly improved the computation speed. Finally, it is exhibited that the recognition performance with density-based clustering is superior to conventional algorithms under the condition of large azimuth angle ranges and low signal-to-noise ratio (SNR).

High resolution range profile formation based on LFM signal fusion of multiple radars

2007 ◽  
Vol 24 (1) ◽  
pp. 75-82 ◽  
Author(s):  
Cheng Wang ◽  
Weidong Hu ◽  
Xiaoyong Du ◽  
Wenxian Yu
Keyword(s):  
High Resolution ◽  
Signal Fusion ◽  
Range Profile ◽  
High Resolution Range Profile
Sign in / Sign up

Export Citation Format

Share Document