DocumentCode :
71239
Title :
Application of Linear Predictive Coding for Doppler Through-Wall Radar Target Tracking
Author :
Yipeng Ding ; Jingtian Tang ; Xuemei Xu ; Jiliang Zhang
Author_Institution :
Sch. of Phys. & Electron., Central South Univ., Changsha, China
Volume :
12
Issue :
6
fYear :
2015
fDate :
Jun-15
Firstpage :
1317
Lastpage :
1321
Abstract :
In this letter, a target tracking approach, which combines short-time Fourier transform (STFT) and linear predictive coding (LPC), is proposed for a Doppler through-wall radar. The LPC is applied to extend the known echo data in each STFT sliding window, thus helping in improving the estimation accuracy of target instantaneous frequency. Compared with the traditional LPC process which determines the prediction data size empirically, the proposed approach takes advantage of the fitting error array as a control parameter to intelligently adjust the data size and reduce the prediction error. Moreover, the proposed approach can also enhance the radar processing efficiency by preventing the unqualified prediction data, which is of great importance for real-time detecting applications. Series of experimental measurements are presented as a preliminary assessment of the proposed approach.
Keywords :
Doppler radar; Fourier transforms; error statistics; frequency estimation; image coding; linear predictive coding; radar cross-sections; target tracking; Doppler through-wall radar target tracking; LPC; STFT sliding window; data size adjustment; fitting error array; linear predictive coding; prediction error reduction; radar echo; radar processing efficiency enhancement; short-time Fourier transform; target instantaneous frequency estimation accuracy; Accuracy; Arrays; Doppler effect; Estimation; Frequency estimation; Prediction algorithms; Radar; Doppler through-wall radar (TWR); fitting error array; linear predictive coding (LPC); short-time Fourier transform (STFT); target instantaneous frequency estimation;
fLanguage :
English
Journal_Title :
Geoscience and Remote Sensing Letters, IEEE
Publisher :
ieee
ISSN :
1545-598X
Type :
jour
DOI :
10.1109/LGRS.2015.2397451
Filename :
7045445
Link To Document :
بازگشت