Performance comparison of super-resolution estimation algorithms used in real or complex LFMCW systems

Author

Yang, Li ; Liwan, Liang ; Weifeng, Pan ; Yanqin, Chen ; Zhenghe, Feng

Author_Institution

State Key Lab. on Microwave & Digital Commun., Tsinghua Univ., Beijing, China

fYear

2000

fDate

2000

Firstpage

658

Lastpage

661

Abstract

In real or complex linear FMCW (LFMCW) radar systems, employing algorithms with super-resolution capability and low SNR threshold can mitigate the system demands for power and bandwidth, and reduce system cost. In this paper, a comparative study among three typical algorithms is carried out on statistical SNR and resolution performance: MUSIC, AR, and neural network. Simulation results show that the neural network algorithm, having higher resolution and lower SNR threshold in both systems, is suitable for LFMCW applications

Keywords

CW radar; FM radar; Hopfield neural nets; Monte Carlo methods; autoregressive processes; eigenvalues and eigenfunctions; estimation theory; noise; radar resolution; statistical analysis; AR algorithm; LFMCW systems; MUSIC algorithm; linear FMCW systems; low SNR threshold; neural network algorithm; performance comparison; statistical SNR performance; statistical resolution performance; super-resolution estimation algorithms; Bandwidth; Chirp modulation; Costs; Frequency modulation; Multiple signal classification; Neural networks; Power system modeling; Signal processing algorithms; Signal resolution; Space technology;

fLanguage

English

Publisher

ieee

Conference_Titel

Microwave and Millimeter Wave Technology, 2000, 2nd International Conference on. ICMMT 2000

Conference_Location

Beijing

Print_ISBN

0-7803-5743-4

Type

conf

DOI

10.1109/ICMMT.2000.895773

Filename

895773