TY - JOUR
T1 - Radar target recognition by machine learning of k-nearest neighbors regression on angular diversity RCS
AU - Lee, Kun Chou
N1 - Funding Information:
The authors would like to acknowledge (1) the financial support of the Ministry of Science and Technology, Taiwan, under Grant MOST 107-2221-E-006-098, (2) the National Center for High Performance Computing, Taiwan, for computer time and facilities.
Publisher Copyright:
© ACES.
PY - 2019/1
Y1 - 2019/1
N2 - In this paper, the radar target recognition is given by machine learning of K-NN (K-nearest neighbors) regression on angular diversity RCS (radar cross section). The bistatic RCS of a target at a fixed elevation angle and different azimuth angles are collected to constitute an angular diversity RCS vector. Such angular diversity RCS vectors are chosen as features to identify the target. Different RCS vectors are collected and processed by the K-NN regression. The machine learning belongs to the scope of artificial intelligence, which has attracted the attention of researchers all over the world. In this study, the K-NN rule is extended to achieve regression and is then applied to radar target recognition. With the use of K-NN regression, the radar target recognition is very simple, efficient, and accurate. Numerical simulation results show that our target recognition scheme is not only accurate, but also has good ability to tolerate random fluctuations.
AB - In this paper, the radar target recognition is given by machine learning of K-NN (K-nearest neighbors) regression on angular diversity RCS (radar cross section). The bistatic RCS of a target at a fixed elevation angle and different azimuth angles are collected to constitute an angular diversity RCS vector. Such angular diversity RCS vectors are chosen as features to identify the target. Different RCS vectors are collected and processed by the K-NN regression. The machine learning belongs to the scope of artificial intelligence, which has attracted the attention of researchers all over the world. In this study, the K-NN rule is extended to achieve regression and is then applied to radar target recognition. With the use of K-NN regression, the radar target recognition is very simple, efficient, and accurate. Numerical simulation results show that our target recognition scheme is not only accurate, but also has good ability to tolerate random fluctuations.
UR - http://www.scopus.com/inward/record.url?scp=85063165238&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85063165238&partnerID=8YFLogxK
M3 - Article
AN - SCOPUS:85063165238
SN - 1054-4887
VL - 34
SP - 75
EP - 81
JO - Applied Computational Electromagnetics Society Journal
JF - Applied Computational Electromagnetics Society Journal
IS - 1
ER -