Ultra-Low-Voltage Frequency Synthesizer and Successive-Approximation Analog-to-Digital Converter for Biomedical Applications

Table of Contents

1. Frontmatter

2. Chapter 1. Introduction to Biomedical Signals and Their Applications

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   Autonomous biomedical signals from the central nervous system and functional tissues of the brain are transmitted through corresponding nerve fibers to complete the functions of life actions and activities, such as electroencephalogram. Non-autonomous biomedical control signals are sent by organs to maintain the basic operations of human survival, such as the electrocardiogram. Since biomedical signals are in charge of human life and all activities, it is necessary for us to understand the origin of biomedical signals, the operating principles behind them, and their transmission behavior, and to enumerate common biological signal characteristics, including their amplitude and frequency.

   Collecting the corresponding biomedical signals can not only understand the biomedical status of the human body in time, but it is also very important to send out early warning signals of symptoms before the onset of the patient’s disease. However, measurement of biological signals may introduce some noise during the process. Therefore, noise must be eliminated first when processing biomedical signals. Only by identifying the characteristics of biomedical signals through signal processing can they become meaningful information and be used for special purposes or applications. For example, rehabilitation aids or equipment can be manufactured through a human–machine interface to improve the quality of life. Implantable devices can also allow patients to immediately apply appropriate feedback, such as implanting a cochlear implant to stimulate the auditory nerve or implanting a pacemaker to stimulate the heartbeat. These signals can also be transmitted wirelessly to a medical center or emergency center to notify relevant units of immediate response and preparation so that patients can receive treatment and immediate care.

3. Chapter 2. Low-Power and Low-Voltage VLSI Circuit Design Techniques for Biomedical Applications

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   Since wearable devices, sensor devices, or implanted devices are powered by batteries, they need to maintain long working time, especially implantable devices, because the battery needs to be replaced by surgery. This chapter introduces how to implement low-power, low-voltage VLSI circuit design.

   In addition to realizing more transistors in the same area to comply with Moore’s Law, advances in semiconductor technology have also brought many benefits, such as smaller areas, faster speed, more functions, and lower power consumption. From the perspective of technological evolution, there are many compromises in consideration of transistor characteristics. To produce good transistor characteristics, designers must weigh these transistor characteristics to achieve the best design performance or operating point. However, these advanced process nodes have also brought some problems, such as various leakage currents, and battery-powered devices cannot tolerate the waste of electricity. Therefore, we will discuss how to achieve low voltage and low power consumption in digital circuit design, including possible solutions and recommendations, and the trade-offs of reducing dynamic power, static power, and leakage current. Next, we will present the problems faced by low-voltage analog circuit design, and then discuss traditional design methods and g_m/I_D design methods. Finally, we will discuss some considerations for the design and implementation of nano-analog circuits.

4. Chapter 3. Introduction of Frequency Synthesizer

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   For either implanted or most in vitro medical signal detectors, it is necessary to wirelessly transmit the measured biomedical signal information to the relevant receiving unit. The most crucial component in a wireless transceiver is a frequency synthesizer, which must provide a wide range of bands of frequency and stable oscillation frequency.

   This chapter first introduces the basic principle of the integer N frequency synthesizer and the stability analysis of its linear model. Then, some key trade-offs when designing an integer-N frequency synthesizer are discussed, including the trade-offs among settling time, phase noise, and reference spurs. Next, each building block in the frequency synthesizer, including phase detector, charge pump, loop filter, oscillator, and frequency divider, are presented. In addition to the integer-N frequency synthesizer, we also briefly introduce the concept of non-integer frequency synthesizer, direct synthesizer, and all-digital phase-locked loop (ADPLL). Finally we discuss in depth the two important units in ADPLL, digital-controlled oscillation (DCO) and time-to-digital conversion (TDC).

5. Chapter 4. A 0.35-V 240-μW Fast-Lock and Low-Phase-Noise Frequency Synthesizer for Implantable Biomedical Applications

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   For implantable frequency synthesizers, realizing ultra-low voltage (ULV) and low power in addition to meeting PLL targets, fast lock and low phase noise, poses a difficult challenge. This chapter presents techniques to achieve PLL targets as well as ULV and low power in the same chip through the use of a regular CMOS technology node. A curvature-PFD technique achieves both faster locking and lower jitter compared with conventional techniques. A two-step switching technique substantially reduces the power consumption in current mirrors and reduce noise when switching from a charge pump. Leakage analysis and subthreshold-leakage-reduction technique reduce reference spur and jitter to the voltage-controlled oscillator (VCO). A dither technique randomizes and averages reference spurs. The proposed chip was implemented in 90-nm CMOS technology; the 0.35-V medical-band frequency synthesizer consumes 238-μW power while generating output clock of 401.8–431.31-MHz and exhibiting a phase noise of −105.7 dBc/Hz at 1-MHz frequency offset with 20 μs locking time.

6. Chapter 5. Introduction of ADC

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   The detection of biomedical signals needs to be converted into digital signals by an analog-to-digital converter, and then relevant signal processing is performed to identify the characteristics of the biomedical signal so as to become meaningful information for subsequent processing. This chapter introduces the basic working principles of several commonly used analog-to-digital converters (ADC) and simulation details based on digital signal processing. Because different applications have different requirements for ADC speed, accuracy, and power consumption, a suitable ADC architecture needs to be selected for a specific application. We also present the basic operating steps and principles of ADCs, as well as some of its important static and dynamic performance indicators. In addition, the limitations imposed by the working environment will be discussed, including timing jitter and thermal noise limitations. This chapter also introduces common ADC architectures, including FLASH, pipeline, successive approximation, and Sigma-Delta data converters. More detailed discussion of the successive approximation ADC algorithm, calibration technology, and latest development trend are illustrated. In addition, ADC simulation and testing are discussed. Because ADC performance needs to be obtained through digital signal processing approaches, it is necessary to understand some basic knowledge of sampling frequency and input frequency selection, as well as FFT leakage problems. Finally, the commonly used test methods of Nyquist ADCs are presented.

7. Chapter 6. A 0.3 V 10b 3 MS/s SAR ADC with Comparator Calibration and Kickback Noise Reduction for Biomedical Applications

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   This chapter presents a 10-bit successive approximation analog-to-digital converter (ADC) that operates at an ultralow voltage of 0.3 V and can be applied to biomedical implants. The study proposes several techniques to improve the ADC performance. A pipeline comparator was utilized to maintain the advantages of dynamic comparators and reduce the kickback noise. Weight biasing calibration was used to correct the offset voltage without degrading the operating speed of the comparator. The incorporation of a unity-gain buffer improved the bootstrap switch leakage problem during the hold period and reduced the effect of parasitic capacitances on the digital-to-analog converter. The chip was fabricated using 90-nm CMOS technology. The data measured at a supply voltage of 0.3 V and sampling rate of 3 MSps for differential nonlinearity and integral nonlinearity were +0.83/−0.54 and +0.84/−0.89, respectively, and the signal-to-noise plus distortion ratio and effective number of bits were 56.42 dB and 9.08 b, respectively. The measured total power consumption was 6.6 μW at a figure of merit of 4.065 fJ/conv.-step.

8. Chapter 7. Summary

   Chung-Chih Hung, Shih-Hsing Wang

   Abstract

   This book combines expertise in different fields, including biotechnology and electrical and electronic engineering. The biological signals are usually dedicated to a specific frequency range or a specific voltage range. As engineers, we need to think about how to capture disturbed or noisy biomedical signals and convert them into meaningful biosignals. This intermediate process includes signal amplification (amplifier design), noise removal (filter design), analog-to-digital conversion (analog-to-digital converter design), digital signal processing (algorithm design), and data transmission through human–machine interface (interface circuit design) or wirelessly (frequency synthesizer design, radio frequency circuit design). All these topics belong to electronic engineering. Information brought by biological signals can be used for many applications. For example, rehabilitation aids or equipment can be manufactured through a human–machine interface to improve the quality of life. The implanted devices also allow the patient to immediately apply appropriate feedback, such as cochlear implant to stimulate the auditory nerve or pacemaker to stimulate the heartbeat. These signals can also be wirelessly transmitted to a medical center or emergency center to notify relevant units of immediate responses and preparation so that patients can receive treatment and immediate care.

9. Backmatter