The SVM-based smart antenna for estimation of the directions of arrival of electromagnetic waves

被引:30
作者
Pastorino, Matteo [1 ]
Randazzo, Andrea [1 ]
机构
[1] Univ Genoa, Dept Biophys & Elect Engn, I-16145 Genoa, Italy
关键词
array antennas; direction of arrival estimation; multiple antennas; regression; support vector machines (SVMs);
D O I
10.1109/TIM.2006.884295
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Microwave systems for object detection and imaging usually illuminate a given scene and measure the reflected signals. One of the most important tasks is a rapid and efficient detection of the angular positions of the scatterers present in the scene. This task can be accomplished by designing a smart antenna composed of multiple receiving elements, which is able to estimate the angles of incidence of the incoming electromagnetic waves. In this paper, a new efficient method based on a support vector regression is proposed for the detection of arriving electromagnetic waves scattered by objects located both in the far-field and near-field regions of the smart antenna. The effectiveness of the proposed approach is evaluated by means of several numerical simulations.
引用
收藏
页码:1918 / 1925
页数:8
相关论文
共 20 条
[1]  
ABOUKHOUSA MA, 2006, P MED MICR S MMS 200, P185
[2]  
[Anonymous], 1998, PROCEEDINGS OF THE 8TH INTERNATIONAL CONFERENCE ON ARTIFICIAL NEURAL NETWORKS, PERSPECTIVES IN NEURAL COMPUTING
[3]   Multiple antenna systems: Their role and impact in future wireless access [J].
Blostein, SD ;
Leib, H .
IEEE COMMUNICATIONS MAGAZINE, 2003, 41 (07) :94-101
[4]   Practical selection of SVM parameters and noise estimation for SVM regression [J].
Cherkassky, V ;
Ma, YQ .
NEURAL NETWORKS, 2004, 17 (01) :113-126
[5]  
Choi MS, 2003, IEEE VTS VEH TECHNOL, P1041
[6]  
Collin R.E., 1985, ANTENNAS RADIOWAVE P
[7]   Content-based audio classification and retrieval by support vector machines [J].
Guo, GD ;
Li, SZ .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (01) :209-215
[8]   Direction of arrival estimation based on temporal and spatial processing using a direct data domain (D3) approach [J].
Kim, K ;
Sarkar, TK ;
Wang, H ;
Salazar-Palma, M .
IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, 2004, 52 (02) :533-541
[9]   Linear dependency between ε and-the input noise in ε-support vector regression [J].
Kwok, JT ;
Tsang, IW .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 2003, 14 (03) :544-553
[10]  
Liberti J.C., 1999, PR H COMMUN ENGN EM