techniques for pixel level analog to digital...
TRANSCRIPT
![Page 1: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/1.jpg)
Techniques for Pixel Level Analog to Digital Conversion
Boyd Fowler, David Yang, and Abbas El Gamal
Stanford University
Aerosense 98 3360-1 1
![Page 2: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/2.jpg)
Output MuxOutput
Digital
Image Sensor
Array
ADC
ADC
ADC
ADC
ADC
ADC
ADC
ADC
ADC
ADC
ADC
ADC
Array Core
Image Sensor
ADC
Multiplxer outputDigital
Array
ADC
Approaches to Integrating ADC with Image Sensor
• Chip Level
• Column Level
• Pixel Level
Aerosense 98 3360-1 2
![Page 3: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/3.jpg)
Pixel Level ADC
• Advantages
– Continuous observation of the pixel
– Low noise
– Low power
• Disadvantages
– Limited area at the pixel level – this becomes
less of a problem as technology scales
Aerosense 98 3360-1 3
![Page 4: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/4.jpg)
Outline
⇒ • Architecture
– Σ∆ ADC– MCBS ADC– Readout circuitry
• Circuits
– Σ∆ pixel– MCBS pixel
• Results
• Comparision
Aerosense 98 3360-1 4
![Page 5: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/5.jpg)
Pixel Level ADC
• Pixel size limits ADC architecture
• Cannot use conventional bit-parallel ADC
techniques, because they require memory at the
pixel
• Bit serial ADC is required but conventional
techniques are too complicated
• To overcome this problem we developed two bit
serial techniques
– Oversampled
– Nyquist rate
Aerosense 98 3360-1 5
![Page 6: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/6.jpg)
Oversampled ADC
• Robust operation over process variation and noise
• Simple imprecise circuits can be used
• Use first order 1 bit Σ∆ modulation
– Requires small silicon area
Aerosense 98 3360-1 6
![Page 7: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/7.jpg)
dTInput Output
First Order 1 Bit Σ∆ Modulation
Output decimated using LPF to obtain pixel values -performed off chip.
Aerosense 98 3360-1 7
![Page 8: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/8.jpg)
Oversampling Ratio L as a function of SNR
Number of bit planes L per frame for a desired average
SNR
log2(L) =SNR + 5.2dB
9.0
Examples:• SNR = 48dB (8 bits) ⇒ L = 60
• SNR = 20dB (3.3 bits) ⇒ L = 7
Aerosense 98 3360-1 8
![Page 9: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/9.jpg)
Σ∆ ADC Limitations
• Increased output bandwidth caused by
oversampling
• Decimation/DSP is required to resample the data
at the Nyquist frequency
• Poor low light response due to nonuniform
quantization
Aerosense 98 3360-1 9
![Page 10: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/10.jpg)
Pixel Level Nyquist Rate ADC
Recently developed a new multi-channel bit-serial(MCBS) ADC technique (Yang, Fowler, El GamalCICC’98)
• Low output data rate
• Requires very few transistors at the pixel
• Uses very simple circuits
• Complex circuits shared by all pixels
• No external DSP is required to consturct image
Aerosense 98 3360-1 10
![Page 11: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/11.jpg)
MCBS ADC
ADC assigns binary codewords to quantization intervals3-bit example: signal S ∈ [0, 1))
Quantization table
S Gray Code
0 − 18 0 0 0
18 − 2
8 0 0 128 − 3
8 0 1 138 − 4
8 0 1 048 − 5
8 1 1 058 − 6
8 1 1 168 − 7
8 1 0 178 − 1 1 0 0
Key observation: Each output bit can be viewed as abinary-valued function over intervals (independent ofother bits!)
Aerosense 98 3360-1 11
![Page 12: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/12.jpg)
Operation of the 1-bit Comparator/Latch
3 Bit LSB Example
Aerosense 98 3360-1 12
![Page 13: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/13.jpg)
MCBS ADC – Block Diagram
Aerosense 98 3360-1 13
![Page 14: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/14.jpg)
MCBS ADC Limitation
The gain bandwidth product must exceed 2Fd × 4m
• To perform m bit quantization at least 2m − 1
comparison must be performed
• Assuming uniform quantization, the gain of the
comparator must be proportional to 2m.
Aerosense 98 3360-1 14
![Page 15: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/15.jpg)
Sense Amplifiers and Latches
Row
Addr
ess D
ecod
er
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
PixelBlock
WORD
BIT
Block Diagram for Image Sensor with Pixel Level ADC
Aerosense 98 3360-1 15
![Page 16: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/16.jpg)
1
2
34
Bit Plane
1
0
1
1
1
1
0
L1
2
Frame
Next Frame
Time
Output of Image Sensor with Pixel Level ADC
Aerosense 98 3360-1 16
![Page 17: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/17.jpg)
Outline
• Architecture
– Σ∆ ADC– MCBS ADC– Readout circuitry
⇒ • Circuits
– Σ∆ pixel– MCBS pixel
• Results
• Comparision
Aerosense 98 3360-1 17
![Page 18: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/18.jpg)
Multiplexed Σ∆ Pixel Block Diagram - 17 Transistors
Aerosense 98 3360-1 18
![Page 19: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/19.jpg)
Multiplexed Σ∆ Pixel Block Circuit Schematic - 17
Transistors
Aerosense 98 3360-1 19
![Page 20: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/20.jpg)
Σ∆ ADC Circuit
• Advantages
– Small size
– Low sensitivity to transistor noise, and no
offset FPN
– Large charge handling capacity
• Disadvantages
– Moderate gain FPN
– Poor low light response
– High output data rate
Aerosense 98 3360-1 20
![Page 21: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/21.jpg)
MCBS ADC Pixel Block Circuit Schematic - 19
Transistors
Aerosense 98 3360-1 21
![Page 22: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/22.jpg)
MCBS ADC
• Advantages
– Small size
– Low gain and offset FPN
– Complete testability
• Disadvantages
– Moderate comparator gain-bandwidth
– Reduced sensitivity caused by sample capacitor
Aerosense 98 3360-1 22
![Page 23: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/23.jpg)
Outline
• Architecture
– Σ∆ ADC– MCBS ADC– Readout circuitry
• Circuits
– Σ∆ pixel– MCBS pixel
⇒ • Results
• Comparision
Aerosense 98 3360-1 23
![Page 24: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/24.jpg)
Chip Characteristics
Σ∆ ADC MCBS ADC
Technology 0.8µm 3M1P 0.35µm 4M1P
Pixel Area 20.8 µm × 19.8 µm 8.9µm × 8.9µm
Transistors per pixel 4.25 (17 per four pixels) 4.5 (18 per four pixels)
Fill Factor 30% 25%
Array Size 128×128 320×240
Supply Voltage 3.3 v 3.3 v
Aerosense 98 3360-1 24
![Page 25: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/25.jpg)
0 0.2 0.4 0.6 0.8 1
x 10−3
−1
0
1
2
3
Time (ms)
!CK
(vo
lts)
0 0.2 0.4 0.6 0.8 1
x 10−3
−2
−1
0
1
2
3
4
Time (ms)
Pix
el O
utpu
t (vo
lts)
Σ∆ ADC Output Waveforms
CLKB
PIXEL
OUTPUT
SNR = 43dB with OSR = 64, Gain FPN = 1%
Aerosense 98 3360-1 25
![Page 26: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/26.jpg)
1 1.5 2 2.50
50
100
150
200
250
300
Input Voltage
Out
put C
odeADC Transfer Characteristic (10us)
Measured MCBS ADC Transfer Function
INL = 2.3 LSB, DNL = 1.2 LSB, Gain FPN = 0.24%,and Offset FPN = 0.2%Aerosense 98 3360-1 26
![Page 27: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/27.jpg)
Outline
• Architecture
– Σ∆ ADC– MCBS ADC– Readout circuitry
• Circuits
– Σ∆ pixel– MCBS pixel
• Results
⇒ • Comparision
Aerosense 98 3360-1 27
![Page 28: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/28.jpg)
Pixel Level ADC Comparison
Σ∆ ADC MCBS ADC
Size 4.25 transistors per pixel 4.5 transistor per pixel
Conversion mode Charge to bits Voltage to bits
Charge handling Capacity Large Small
FPN Moderate Small
Transistor noise sensitivity Small Moderate
Date rate Large Small
Quantization Nonuniform – fixed by L Programmable
External processing Decimation Filtering None
Required memory Large Small
Power Dissipation Moderate Small
Aerosense 98 3360-1 28
![Page 29: Techniques for Pixel Level Analog to Digital Conversionabbas/group/papers_and_pub/aerosense98_slide.pdfTechniques for Pixel Level Analog to Digital Conversion Boyd Fowler, David Yang,](https://reader033.vdocument.in/reader033/viewer/2022050601/5fa91156384f016bca782b23/html5/thumbnails/29.jpg)
Conclusions
• Σ∆ ADC is better suited to IR sensors
– Large charge handling capacity
– Fine quantization near the center of the range
(i.e. small difference between large charge
value can be determined)
• MCBS ADC is better suited to visible range
sensors
– Low output data rate
– Programmable quantization
– High sensitivity
Aerosense 98 3360-1 29