Through-Wall Imaging With UWB Radar System - KEMT FEI TUKE

More documents

Recommendations

Info

4.1 Through-Wall TOA Estimation 48 Z Antenna � a Wall �w �a A d E T 1 H X W 1 T H Z Fig. 4.1.3: Two layer through-wall model. By a simple geometrical manipulations as we described in [137] the dinit can by estimated as: � εa HXDW dinit ≈ εw HZ W di D w X . (4.1.12) In [137] we used (4.1.12) as a final estimation of d for TOA computation, here it is used only to estimate the initial condition - dinit. 4.1.4 Simulation Results Several simulation experiments are shown in order to demonstrate efficiency and precision of the proposed algorithm. The simulated scenario of an object behind the wall is shown in Fig. 4.1.4. Firstly, the investigated parameter d is computed with initial condition given by (4.1.10). The results dependant the on number of iterations are compared with the true value of d and shown in Table 4.1. In next, Table 4.1: Dependence of d on number of iterations. dinit ≈ HX 2 . Iterations 0 1 2 3 4 5 6 7 d[cm] 150 63.636 26.6 18.797 17.7 17.673 17.673 17.673 Errd[%] 748.752 260.075 50.512 6.360 0.153 0.000 0.000 0.000 d is computed with initial condition given by (4.1.12). The results are shown in Table 4.2. Errd[%] represents a relative error of d in percentage. It is obvious, that only a few iterations lead to precise results. Only 5 iterations are required to obtain Errd < 0.0005% in case the dinit is computed according to the (4.1.10). In case the dinit is computed by (4.1.12), only four iterations are required.
4.1 Through-Wall TOA Estimation 49 Z Wall � rw = 8 Antenna d = ? H X = 3 m W di = 0.5 m D w = 0.5 m X H Z = 1.5 m Fig. 4.1.4: Simulated scenario of object behind the wall. Table 4.2: Dependence of d on number of iterations. dinit ≈ � εa HXDW εw HZ Iterations 0 1 2 3 4 5 6 7 d[cm] 35.355 20.828 17.857 17.674 17.673 17.673 17.673 17.673 Errd[%] 100.051 17.852 1.041 0.006 0.000 0.000 0.000 0.000 The computation complexity of TOA estimation is a very important aspect, e.g. in order to compute migration from SAR scanning consisting from N = 300 scans and with final image resolution Nx = 500 × Nz = 500 pixels, the parameter d has to be computed Nc = 150 millions times. Nc = NxNzN. (4.1.13) The computation method of dinit should be therefore chosen with sophisticated trade-off between precision and complexity, and with regard to the hardware performance. In Table 4.3 precisions of several TOA methods are compared. Firstly, the TOA is computed as a straight line between the antenna and the target according to the (2.5.1) (one way antenna - target only) with no wall compensation. Secondly, a simple wall compensation method (Section 2.5.3) is used to compute TOA. Finally, the TOA is computed according to the proposed precise method. ErrT OA[%] represents a relative error of TOA estimation in percentage. The shown precision is obtained only in case of homogeneous walls. In addition, the velocity of the wave propagation depends also on conductivity, permeability,
Page 1 and 2:
TECHNICAL UNIVERSITY OF KOˇSICE Fa
Page 3 and 4:
Acknowledgement I give a sincere gr
Page 5 and 6:
CONTENTS v 3 Goals of Dissertation
Page 7 and 8:
LIST OF FIGURES vii List of Figures
Page 9 and 10:
LIST OF FIGURES ix 4.4.3 Aquarium f
Page 11 and 12: LIST OF FIGURES xi List of Symbols
Page 13 and 14: LIST OF FIGURES xiii kx - Horizonta
Page 15 and 16: Chapter 1 Introduction 1.1 Motivati
Page 17 and 18: 1.3 Thesis Organization 3 In additi
Page 19 and 20: Chapter 2 State Of The Art 2.1 UWB
Page 21 and 22: 2.2 Through-Wall Radar Basic Model
Page 23 and 24: 2.3 SAR Scanning 9 and shall remain
Page 25 and 26: 2.4 Calibration and Preprocessing 1
Page 27 and 28: 2.4 Calibration and Preprocessing 1
Page 29 and 30: 2.5 Radar Imaging Methods Overview
Page 55 and 56: Chapter 4 Selected Research Methods
Page 57 and 58: 4.1 Through-Wall TOA Estimation 43
Page 61: 4.1 Through-Wall TOA Estimation 47
Page 71 and 72: 4.2 Measurements of the Wall Parame
Page 81 and 82: 4.3 Highlighting of a Building Cont
Page 91 and 92: 4.4 Measurements of the Practical S
Page 99 and 100: 5.1 Conclusion and Future Work 85 5
Page 101 and 102: 50 100 150 200 250 300 350 400 Z 45
Page 103 and 104: 50 100 150 200 250 300 350 400 Z 45
Page 105 and 106: 50 100 150 200 250 300 350 400 Z 45
Page 107 and 108: 50 100 150 200 250 300 350 400 Z 45
Page 109 and 110: Z [cm] Z [cm] 50 100 150 200 250 30
Page 111 and 112: Z [cm] 50 100 150 200 250 300 b) Em
Page 113 and 114:
BIBLIOGRAPHY 99 [13] T. W. Barrett,
Page 115 and 116:
BIBLIOGRAPHY 101 [46] G. Derveaux,
Page 117 and 118:
BIBLIOGRAPHY 103 [81] S. Kidera, T.
Page 119 and 120:
BIBLIOGRAPHY 105 [115] J. Sachs, M.
Page 121:
BIBLIOGRAPHY 107 [150] O. Yilmaz, S
show all

Through-Wall Imaging With UWB Radar System - KEMT FEI TUKE

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?