3D Time-of-flight distance measurement with custom - Universität ...

More documents

Recommendations

Info

DEMODULATION PIXELS IN CMOS/CCD 115 Thermal diffusion drift E-field drift (1 V) Transport distance Transport time Transport distance Transport time 30 µm 270 ns 30 µm (1 V) 7 ns 15 µµµµm 68 ns 15 µm (1 V) 1.8 ns 10 µm 30 ns 10 µm (1 V) 0.8 ns 5 µm 7.5 ns 5 µm (1 V) 0.2 ns 3⋅10 µm 90 ns 3⋅10 µm (1 V=3·0.33 V) 7 ns 3⋅⋅⋅⋅5 µµµµm 23 ns 3⋅⋅⋅⋅5 µµµµm (1 V=3·0.33 V) 1.8 ns Table 5.2 Transfer time estimation with pure diffusion and pure E-field model. (p-substrate, NA=4⋅10 14 cm -3 , µn=1300 cm 2 /Vs, Dn=33.6 cm 2 /s). The 1-tap structure benefits from stronger fringing fields, since the pixel architecture allows the realization of shorter CCD gates. Also the transfer does not necessarily require high charge transfer efficiency, as in the multitap approach. According to Table 5.2 we can expect transfer times between 2 and 10 ns (depending on the actual fringing field influence) corresponding to demodulation frequencies up to 100 MHz or even 500 MHz. The subdivision of the photogate into three separate gates results in a speed improvement of at least a factor of three (considering only diffusion), more probably a factor of 10 to 20, since with (1) smaller gate lengths and (2) more gate junctions (three instead of one) the fringing field influence increases. We demonstrate the 1-tap pixel with demodulation frequencies up to 20 MHz (actual operation frequency of the 3D cameras introduced in Chapter 6). For higher frequencies it would be even more difficult to realize the highly accurate control and driving electronics. But another fact also reduces the demodulation performance at high modulation frequencies, the penetration depth of the modulated light. So far we have only considered the transfer times of electrons that are already captured by the electric field of the CCD gates. Generally, however, depending on the wavelength (c.f. Figure 3.4), the electron hole pairs are generated (far) away from the semiconductor surface. They will first have to diffuse to the top, before being captured by the CCD-field. This random diffusion process takes time and additionally limits the pixel-bandwidth, especially for modulation light of long wavelength.
116 CHAPTER 5 5.2 Characterization of 1-tap pixel performance In this section, we present the characterization measurements for the 1-tap pixel. In the first part we characterize the on-chip output stage (source follower) in terms of sensitivity, linearity and saturation. Then we outline the motivation for the measurements, describe the measurement setup and discuss the results to be expected. With this knowledge, we present the measurements of demodulation contrast and phase accuracy as a function of the control voltage amplitudes, the modulation speed, the wavelength and the power of the modulated light. We also present the results of noise measurements and compare the measured phase accuracy with the prediction we presented in Section 4.2. We try to specify all measurement parameters completely so that the measurements may be reproduced. In order not to confuse the reader and to keep the measured data clearly presented, we summarize some measurement details in the appendix (coded as measurement conditions: MCDxyS for SCCD measurements and MCDxyB for BCCD measurements). The 1-tap pixel has been realized (1) as a linear array containing 108 pixels and (2) as a 2D-array with 64 by 25 pixels. The line-sensor has been fabricated in two technologies: as BCCD and as SCCD. The pixel characterizations have been carried out with the line sensors. This offers the possibility to compare the buried channel and surface channel CCD performance. More details of the overall sensor architectures can be found in Chapter 6, where the camera implementations are described. 5.2.1 Charge to voltage conversion Sensitivity and Responsivity The sensitivity is defined as the ratio of output voltage to input charge (unit: Volt/electron). It is characterized by the output stage’s conversion capacitance and amplification. Since, in our implementation, the input charge can only be generated optically or thermally, it cannot be measured directly. So the sensitivity cannot really be distinguished in isolation from the responsivity, which is the ratio of output voltage to input light (unit: V/photon). The responsivity additionally contains the quantum efficiency and therefore depends on the wavelength. None of the
Page 1 and 2:
3D Time-of-flight distance measurem
Page 3 and 4:
To my parents
Page 5 and 6:
Meinen Eltern
Page 7 and 8:
II 4. Power budget and resolution l
Page 10 and 11:
Abstract Since we are living in a t
Page 12:
centimeter accuracy. With the singl
Page 15 and 16:
X Eine elegantere Methode zur Entfe
Page 18 and 19:
INTRODUCTION 1 1. Introduction One
Page 20 and 21:
INTRODUCTION 3 light source observe
Page 22 and 23:
INTRODUCTION 5 Propagation time, ho
Page 24 and 25:
INTRODUCTION 7 Chapter 3 gives a sh
Page 26 and 27:
OPTICAL TOF RANGE MEASUREMENT 9 2.
Page 28 and 29:
Page 30 and 31:
OPTICAL TOF RANGE MEASUREMENT 13 pr
Page 32 and 33:
OPTICAL TOF RANGE MEASUREMENT 15 hi
Page 34 and 35:
OPTICAL TOF RANGE MEASUREMENT 17 Pu
Page 36 and 37:
OPTICAL TOF RANGE MEASUREMENT 19 Ad
Page 38 and 39:
OPTICAL TOF RANGE MEASUREMENT 21 pr
Page 40 and 41:
OPTICAL TOF RANGE MEASUREMENT 23 wh
Page 42 and 43:
OPTICAL TOF RANGE MEASUREMENT 25 δ
Page 44 and 45:
OPTICAL TOF RANGE MEASUREMENT 27 c(
Page 46 and 47:
OPTICAL TOF RANGE MEASUREMENT 29 A
Page 48 and 49:
OPTICAL TOF RANGE MEASUREMENT 31 ph
Page 50 and 51:
OPTICAL TOF RANGE MEASUREMENT 33 Ts
Page 52 and 53:
OPTICAL TOF RANGE MEASUREMENT 35 f1
Page 54 and 55:
OPTICAL TOF RANGE MEASUREMENT 37 an
Page 56 and 57:
OPTICAL TOF RANGE MEASUREMENT 39 in
Page 58 and 59:
OPTICAL TOF RANGE MEASUREMENT 41 Ss
Page 60 and 61:
OPTICAL TOF RANGE MEASUREMENT 43 Fr
Page 62 and 63:
OPTICAL TOF RANGE MEASUREMENT 45 1
Page 64:
Page 67 and 68:
50 CHAPTER 3 the smearing effect, r
Page 69 and 70:
52 CHAPTER 3 3.1 Silicon properties
Page 71 and 72:
54 CHAPTER 3 (a) Figure 3.3 Optical
Page 73 and 74:
56 CHAPTER 3 Quantum efficiency in
Page 75 and 76:
58 CHAPTER 3 The different operatio
Page 77 and 78:
60 CHAPTER 3 (a) (b) Depletion widt
Page 79 and 80:
62 CHAPTER 3 Very special processes
Page 81 and 82: 64 CHAPTER 3 1 kT Dn = ⋅ vth ⋅
Page 83 and 84: 66 CHAPTER 3 The proportionality fa
Page 85 and 86: 68 CHAPTER 3 difference of only 1 V
Page 87 and 88: 70 CHAPTER 3 Flicker noise is cause
Page 89 and 90: 72 CHAPTER 3 A s = sf ⋅ q C ⎡ V
Page 91 and 92: 74 CHAPTER 3 amount of the charge p
Page 93 and 94: 76 CHAPTER 3 substrate (Equation 3.
Page 95 and 96: 78 CHAPTER 3 charge carriers the CT
Page 97 and 98: 80 CHAPTER 3 without causing short
Page 99 and 100: 82 CHAPTER 3 Over an address decode
Page 101 and 102: 84 CHAPTER 3 Orbit’s 2.0µm CMOS/
Page 103 and 104: 86 CHAPTER 4 Figure 4.1 P lens Lamb
Page 105 and 106: 88 CHAPTER 4 Required optical power
Page 107 and 108: 90 CHAPTER 4 4.2 Noise limitation o
Page 109 and 110: 92 CHAPTER 4 number of pseudo-backg
Page 111 and 112: 94 CHAPTER 4 Cmod = 1 QE(λ) = 0.65
Page 113 and 114: 96 CHAPTER 4 Range accuracy (std) i
Page 116 and 117: DEMODULATION PIXELS IN CMOS/CCD 99
Page 168 and 169: IMAGING TOF RANGE CAMERAS 151 6. Im
Page 170 and 171: IMAGING TOF RANGE CAMERAS 153 contr
Page 172 and 173: IMAGING TOF RANGE CAMERAS 155 6.1.2
Page 174 and 175: IMAGING TOF RANGE CAMERAS 157 gate
Page 176 and 177: IMAGING TOF RANGE CAMERAS 159 6.2 2
Page 178 and 179: IMAGING TOF RANGE CAMERAS 161 n+ di
Page 180 and 181: IMAGING TOF RANGE CAMERAS 163 pixel
Page 182 and 183:
IMAGING TOF RANGE CAMERAS 165 Power
Page 184 and 185:
IMAGING TOF RANGE CAMERAS 167 Delay
Page 186 and 187:
IMAGING TOF RANGE CAMERAS 169 6.3 3
Page 188 and 189:
IMAGING TOF RANGE CAMERAS 171 Gate
Page 190 and 191:
IMAGING TOF RANGE CAMERAS 173 Bound
Page 192 and 193:
IMAGING TOF RANGE CAMERAS 175 Figur
Page 194 and 195:
IMAGING TOF RANGE CAMERAS 177 Final
Page 196 and 197:
IMAGING TOF RANGE CAMERAS 179 #5 #6
Page 198 and 199:
SUMMARY AND PERSPECTIVE 181 7. Summ
Page 200 and 201:
SUMMARY AND PERSPECTIVE 183 into th
Page 202:
SUMMARY AND PERSPECTIVE 185 or phas
Page 205 and 206:
188 CHAPTER 8 8.2 Typical parameter
Page 207 and 208:
190 CHAPTER 8 Spectral luminous int
Page 209 and 210:
192 CHAPTER 8 MCD03S: Conditions SC
Page 211 and 212:
194 CHAPTER 8 MCD06S: Measurement c
Page 213 and 214:
196 [BRK] Brockhaus, “Naturwissen
Page 215 and 216:
198 [KAI] I. Kaisto, et al., “Opt
Page 217 and 218:
200 [SP1] T. Spirig et al., “The
Page 220 and 221:
ACKNOWLEDGMENTS 203 Acknowledgments
Page 222:
206 Publication list 1. R. Lange, P
show all

3D Time-of-flight distance measurement with custom - Universität ...

Create successful ePaper yourself

Delete template?

Save as template?