Share Email Print

Spie Press Book

Digital Converters for Image Sensors
Author(s): Kenton T. Veeder
Format Member Price Non-Member Price

Book Description

This book is intended for image sensor professionals and those interested in the boundary between sensor systems and analog and mixed-signal integrated circuit design. It provides in-depth tips and techniques necessary to understand and implement these two types of complex circuit systems together for a wide variety of architectures or trade off one against another. The tutorial begins with a brief introduction to the history and definition of a digital image sensor, as well as converter characteristics, before addressing DAC and ADC architectures. Later chapters cover pipeline ADC designs, digital correction, calibration, and testing according to IEEE standards.

Book Details

Date Published: 12 January 2015
Pages: 160
ISBN: 9781628413892
Volume: TT97

Table of Contents
SHOW Table of Contents | HIDE Table of Contents

1 Introduction to Digital Image Sensors
1.1 Data Converters and their History
1.2 Digital Image Sensors and Closely Related Integrated Circuits
1.3 Reasons To Use and Not Use On-Chip Analog-to-Digital Conversion
1.4 Parallelism
     1.4.1 Serial architecture
     1.4.2 Column-parallel architecture
     1.4.3 Pixel-parallel architecture
1.5 Pseudo-differential Architectures

2 Converter Characteristics
2.1 Basics of Data Conversion
     2.1.1 Sampling, Nyquist sampling, and oversampling
     2.1.2 Resolution
     2.1.3 Quantization and full-scale range
     2.1.4 Quantization error
     2.1.5 Converter coding
2.2 Converter Static Characteristics
     2.2.1 Introduction to the transfer function
     2.2.2 Accuracy and precision
     2.2.3 Gain and offset errors
     2.2.4 Integral nonlinearity
     2.2.5 Differential nonlinearity
2.3 Converter Dynamic Characteristics
     2.3.1 Settling and full-scale step response
     2.3.2 Noise
     2.3.3 Effects of static nonlinearity on noise
     2.3.4 Distortion
     2.3.5 Signal-to-noise ratio
     2.3.6 Signal-to-noise and distortion ratio, and spurious free dynamic range
     2.3.7 Effective number of bits
     2.3.8 Dynamic range for Nyquist and oversampled converters
2.4 Figures of Merit for ADCs on Imagers
     2.4.1 Energy figures of merit
     2.4.2 Data density

3 DACs Used in ADC Architectures and Read-in ICs
3.1 Resistor DACs
     3.1.1 Resistor dividers
     3.1.2 R-2R ladder DACs
3.2 Current-Steering DACs
     3.2.1 Unary DACs
     3.2.2 Binary DACs
     3.2.3 Dynamic calibration of DAC current elements for high resolution
     3.2.4 Switching current elements for improved speed
3.3 Switched-Capacitor DACs
     3.3.1 Capacitive divider
     3.3.2 Charge redistribution 2-cap DACs
3.4 The Specialized Multiplying DAC
3.5 Combining Architectures
3.6 Characteristics Unique to Digital-to-Analog Converters
     3.6.1 Pedestal error and droop
     3.6.2 Glitches

4 ADC Architectures for Image Sensors
4.1 Flash ADC (Serial)
     4.1.1 Interpolation
4.2 Folding Technique
4.3 Integrating and Sloping Architectures (Pixel and Column Parallel)
     4.3.1 Single-slope ADC (pixel and column parallel)
     4.3.2 Multislope ADCs (column parallel)
4.4 Successive Approximation ADCs (Column Parallel and Serial)
4.5 Sub-ranging and Two-Step ADCs (Serial, and Column and Pixel Parallel)
4.6 Algorithmic or Cyclic ADCs (Column Parallel and Serial)
4.7 Pipeline ADCs (Column Parallel and Serial)
4.8 Time-to-Digital Converters (Serial, and Column and Pixel Parallel)
4.9 Voltage-Controlled Oscillator ADCs (Serial, and Column and Pixel Parallel)
4.10 Time-Interleaved ADCs (Column Parallel and Serial)
4.11 Oversampling Architectures
     4.11.1 Oversampling with conventional ADCs (pixel and column parallel, and serial)
     4.11.2 Pulse frequency modulation ADCs (pixel parallel)
     4.11.3 Delta-sigma modulation ADCs (pixel and column parallel)
 Delta-sigma ADCs for digital pixel sensors
4.12 Surveys of State-of-the-Art ADCs and Their Design Evolution Convergence
     4.12.1 What can be learned from state-of-the-art academic publications
     4.12.2 What can be learned from commercially available products

5 Case Study: Pipeline ADCs
5.1 Why the Pipeline ADC is Worth a Special Case Study
5.2 Reasons for Image Sensor Professionals to Understand the Pipeline ADC
5.3 Pipeline ADC Architecture
     5.3.1 The ideal nine-bit structure and transfer function
     5.3.2 Pipeline delay
5.4 Pipeline Pieces
5.5 Pipeline Errors and the Need for Redundancy
     5.5.1 Offset and gain errors
     5.5.2 Introduction to redundancy
     5.5.3 An ideal nine-bit ADC with redundancy
     5.5.4 Redundancy in the 1.5-bit-per-stage ADC and the stage transfer function
5.6 The Real-World Transfer Function
5.7 Pipeline ADC Noise Calculations
     5.7.1 The ADSC surprise
     5.7.2 Total pipeline ADC noise
5.8 Pipeline Stage Optimization
5.9 Resource Sharing in Pipeline ADCs

6 Automatic Calibration and Error Correction
6.1 Analog Error Correction for the Pipeline ADC
     6.1.1 MDAC gain trimming
6.2 Digitally Calibrating the Pipeline ADC
     6.2.1 Digital foreground calibration
 Digital self-calibration algorithm
 16-bit digital self-calibration example
 Look-up table values
 Digital self-calibration for the 1.5-bit-per-stage ADC
 Popular digital foreground calibration variants
     6.2.2 ADC background calibration
6.3 ADC Calibration on Image Sensors

7 Testing ADCs on Image Sensors
7.1 Test Overview
     7.1.1 IEEE Standard 1241-2010
     7.1.2 ADC and image sensor test similarities
7.2 Steady State Input Tests
     7.2.1 Gain error and offset
     7.2.2 Servo or code edge tests
     7.2.3 Noise contribution curve
     7.2.4 Steady state input and linearity measurements
7.3 Dynamic Input Tests
     7.3.1 Histogram testing
     7.3.2 Linearity from histogram data
     7.3.3 Noise for image sensor ADCs from dynamic signals
     7.3.4 ADC and image sensor FSR matching
     7.3.5 Full-scale step response
     7.3.6 Out-of-range recovery
7.4 In Situ Test and Measurement
     7.4.1 Power
     7.4.2 Notes on testing and data acquisition
7.5 Built-In Test and Self-Test Circuits for Digital Image Sensors
     7.5.1 Test taps and signal monitors
     7.5.2 Self-calibrated ramp for linearity and SNDR



This book is intended for image sensor professionals and those interested in the boundary between sensor systems and analog and mixed-signal integrated circuit design. If you have at least a basic electrical engineering background and are technically interested in how digital image sensors and digital readout integrated circuits are constructed, you might find this book useful. My goal was to provide a broad tutorial for image sensor professionals without getting bogged down in the transistor-level design aspects associated with most converter design texts. It is my hope that this approach will give you a new understanding of the topic and escape the common pitfalls associated with these complex on-chip imaging systems. If you are a designer, I hope that this text will spark your creativity to develop new and useful architectures of your own and point you to more in-depth resources as needed.

To my knowledge, there are currently no textbooks that focus on the details and issues of integrating converters into image sensors. There are many textbooks and tutorial papers available for digital converter design; there are also many books and tutorial papers available for image sensor design for both infrared and visible sensors. These are focused more broadly on the entire imaging system or narrowly on one particular type of digital converter signal chain. Unfortunately, these references do not provide in-depth tips and techniques necessary to understand and implement these two types of complex circuit systems together for a wide variety of architectures or trade off one style versus another. This situation has generated some confusion in the published literature between image-sensor-focused and converter-focused publications. Bridging this gap is useful because much of the visible image sensor industry has adopted digital image sensors, and the infrared industry is rapidly moving in the same direction.

Chapter 1 provides a brief introduction to the history and definition of a digital image sensor. Chapter 2 covers converter characteristics for readers not already familiar with the basics of data conversion. Each of these characteristics is discussed with respect to image sensor needs and requirements in preparation for the remainder of the book, which discusses converter architectures and evaluation. The architecture portion of the book begins in Chapter 3 with a brief discussion of digital-to-analog converters (DACs) used for displays and read-in integrated circuits, and as analog-to-digital converter (ADC) subcomponents. DACs are important subcomponents for many of the ADC architectures described in Chapter 4. Each ADC architecture is classified as suitable for serial, column-parallel, or pixel-parallel conversion and examples of each type are provided from the published literature. At the end of Chapter 4, a survey of academic and industrial state-of-the-art architectures is presented with a discussion of the concept of industrial design evolution convergence down to the most flexible architectures for the widest number of applications. This sets the stage for an in-depth dive into pipeline ADCs in Chapter 5, which also illustrates real-world issues that designers face when working with all types of ADCs with resolutions suitable for image sensing. Chapter 6 covers digital correction and calibration, which is an integral part of many pipeline ADC architectures. Even though Chapter 6 focuses on the pipeline ADC, the techniques gleaned from this chapter can and should be applied to a wide variety of ADC architectures. The book wraps up with Chapter 7: testing ADCs on image sensors according to IEEE standards. This final chapter addresses specific misunderstandings in the published image sensor literature and covers test issues that are of particular concern to digital image sensor designers.

Writing this book in parallel with running a young business turned out to be more of an undertaking than I first realized. I want to thank my wife, Tricia Veeder, for putting up with my mental absence during evenings and weekends over the last couple years and for listening to some of my ideas. It is a wonderful thing to have a technically savvy life partner who works in the same industry and can critically assess the occasional idea. I want to thank Eric Kurth from FLIR, who originally asked me to put together an in-depth course on ADCs for image sensors from which this book is derived. I will always remember talking all day for four days straight to his integrated circuit design team while sucking on cough drops so that I didn't sound too much like Marlon Brando in The Godfather by the end of the course. I also want to thank John Caulfield from Cyan Systems for telling the team over at SPIE that I had this course and might be interested in converting it into a short, half-day course for SPIE Defense, Security, and Sensing. Finally, I would like to thank the SPIE team (my editors, Scott McNeill and Dara Burrows, as well as Jim Harrington and Tim Lamkins) for asking me to write this book and for providing the resources and valuable feedback to make it happen.

Kenton Veeder
December 2014

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?