Thesis (M. A.)--University of Hong Kong, 2006. / Title proper from title frame. Also available in printed format.

This thesis strives to enhance the performance of delta-sigma modulators in two areas: increasing their speed and enabling their operation in a low voltage environment. Parallelism based on time-interleaving can be used to increase the speed of delta-sigma modulators. A novel single-path time-interleaved architecture is derived and analyzed. Finite opamp gain and bandwidth result in a mismatch between the noise transfer functions of the internal quantizers which degrades the performance of the new modulator. Two techniques are presented to mitigate the mismatch problem: a hybrid topology where the first stage uses multiple integrators while the rest of the modulator uses a single path of integrators and a digital calibration method. The input-feedforward technique removes the input-signal component from the internal nodes of delta-sigma modulators. The removal of the signal component reduces the signal swing and distortion requirements for the opamps. These characteristics enable the reliable implementation of delta-sigma modulators in modern CMOS technology. Two implementation issues for modulators with input-feedforward are considered. First, the drawback of the analog adder at the quantizer input is identified and the capacitive input feedforward technique is introduced to eliminate the adder. Second, the double sampled input technique is proposed to remove the critical path generate by the input feedforward path. Novel input-feedforward delta-sigma architecture is proposed. The new digital input feedforward (DIFF) modulator maintains the low swing and low distortion requirements of the input feedforward technique, it eliminates the analog adder at the quantizer input, and it improves the achievable resolution. To demonstrate these advantages, a configurable delta-sigma modulator which can operate as a feedback topology or in DIFF mode is implemented in 0.18μm CMOS technology. Both modulators operate at 20MHz clock with an oversampling ratio of 8. The power consumption in the DIFF mode is 22mW and in feedback mode is 19mW. However, the DIFF mode achieves a peak SNDR of 73.7dB (77.1dB peak SNR) while the feedback mode achieves a peak SNDR of 64.3dB (65.9dB peak SNR). Therefore, the energy required per conversion step for the DIFF architecture (2.2 pJ/step) is less than half of that required by the feedback architecture (5.7 pJ/step).

One of the world’s largest freshwater deltas (~4000 km2), the Peace-Athabasca Delta (PAD), is located at the convergence of the Peace and Athabasca rivers and Lake Athabasca in northern Alberta, Canada. Since the early 1970s, there has been increasing concern regarding the ecological impacts on the PAD after flow regulation of the Peace River began in 1968, decreased discharge in the Peace and Athabasca rivers as a result of hydroclimatic changes in Western Canada, and increased Athabasca River water usage by oil sands development to the south. This thesis is part of an ongoing, multi-disciplinary project assessing current and past hydrological and ecological conditions in the PAD. Research conducted in this thesis aims to better understand the processes controlling water balance of lakes in the PAD using mainly stable water isotope data collected from lakes and their input sources. Isotope data are used to describe and quantify hydrological processes for individual lakes (seasonal and annual) and across the delta and are supported by other chemical and hydrometric data. An isotopic framework in d18O-d2H-space is developed for the PAD using evaporation-flux-weighted local climate data, and isotopic data collected from a reference basin, lakes throughout the PAD, and lake input sources (i.e., snowmelt, rainfall, and river water). The framework is comprised of two reference lines, the Local Meteoric Water Line, which is based on measured isotopic composition of precipitation, and the Local Evaporation Line, which is based on modelled isotopic composition of reference points. Evaporation pan data is used to assess short-term variations in key isotopic reference values, which are important for addressing short-term changes in the isotopic signature of shallow basins. This framework is used in subsequent chapters including assessment of seasonal and annual water balance of two hydrologically-contrasting shallow lakes, and to quantify the impacts of flood water and snowmelt on a set of 45 lakes in spring 2003. Five years of isotope data using time-series analysis and the isotopic framework suggested that a perched (isolated) lake and its catchment (forest and bedrock) in the northern, relict Peace sector captured sufficient rain, snow, and runoff to maintain a relatively stable water balance, and also that a low-lying lake in the southern, active Athabasca sector was regularly replenished with river water in both spring and summer. Snowmelt and rainfall were found to have diluted the perched basin by an average of 16% and 28 % respectively, while spring and summer floods were found to almost completely flush the low-lying lake. Using the spring 2003 regional dataset, flooded lakes were separated from snowmelt-dominated lakes through use of suspended sediment concentrations, isotope data, and field observations. Application of an isotope mixing model translated d18O values into a range of replenishment amount by either river water or snowmelt, which compared well with hydrological conditions at the time of sampling and previously classified drainage types of the lakes. Spatial mapping of replenishment amounts illustrated flooding of much of the Athabasca sector due to ice-jams, except for two sub-regions isolated from flooding by artificial and natural northern diversion of flow from the Athabasca River. It is also shown that most of the relict landscape of the Peace sector was replenished by snowmelt except for a few low-lying lakes close to the Peace River and its tributaries. Overall, improved understanding of lake and regional hydrology in the PAD, especially the ability to quantify the affects of various lake inputs, will improve the ability to develop effective guidelines and management practices in the PAD as lakes respond to future changes in climate and river discharge.

This thesis strives to enhance the performance of delta-sigma modulators in two areas: increasing their speed and enabling their operation in a low voltage environment. Parallelism based on time-interleaving can be used to increase the speed of delta-sigma modulators. A novel single-path time-interleaved architecture is derived and analyzed. Finite opamp gain and bandwidth result in a mismatch between the noise transfer functions of the internal quantizers which degrades the performance of the new modulator. Two techniques are presented to mitigate the mismatch problem: a hybrid topology where the first stage uses multiple integrators while the rest of the modulator uses a single path of integrators and a digital calibration method. The input-feedforward technique removes the input-signal component from the internal nodes of delta-sigma modulators. The removal of the signal component reduces the signal swing and distortion requirements for the opamps. These characteristics enable the reliable implementation of delta-sigma modulators in modern CMOS technology. Two implementation issues for modulators with input-feedforward are considered. First, the drawback of the analog adder at the quantizer input is identified and the capacitive input feedforward technique is introduced to eliminate the adder. Second, the double sampled input technique is proposed to remove the critical path generate by the input feedforward path. Novel input-feedforward delta-sigma architecture is proposed. The new digital input feedforward (DIFF) modulator maintains the low swing and low distortion requirements of the input feedforward technique, it eliminates the analog adder at the quantizer input, and it improves the achievable resolution. To demonstrate these advantages, a configurable delta-sigma modulator which can operate as a feedback topology or in DIFF mode is implemented in 0.18μm CMOS technology. Both modulators operate at 20MHz clock with an oversampling ratio of 8. The power consumption in the DIFF mode is 22mW and in feedback mode is 19mW. However, the DIFF mode achieves a peak SNDR of 73.7dB (77.1dB peak SNR) while the feedback mode achieves a peak SNDR of 64.3dB (65.9dB peak SNR). Therefore, the energy required per conversion step for the DIFF architecture (2.2 pJ/step) is less than half of that required by the feedback architecture (5.7 pJ/step).

One of the world’s largest freshwater deltas (~4000 km2), the Peace-Athabasca Delta (PAD), is located at the convergence of the Peace and Athabasca rivers and Lake Athabasca in northern Alberta, Canada. Since the early 1970s, there has been increasing concern regarding the ecological impacts on the PAD after flow regulation of the Peace River began in 1968, decreased discharge in the Peace and Athabasca rivers as a result of hydroclimatic changes in Western Canada, and increased Athabasca River water usage by oil sands development to the south. This thesis is part of an ongoing, multi-disciplinary project assessing current and past hydrological and ecological conditions in the PAD. Research conducted in this thesis aims to better understand the processes controlling water balance of lakes in the PAD using mainly stable water isotope data collected from lakes and their input sources. Isotope data are used to describe and quantify hydrological processes for individual lakes (seasonal and annual) and across the delta and are supported by other chemical and hydrometric data. An isotopic framework in d18O-d2H-space is developed for the PAD using evaporation-flux-weighted local climate data, and isotopic data collected from a reference basin, lakes throughout the PAD, and lake input sources (i.e., snowmelt, rainfall, and river water). The framework is comprised of two reference lines, the Local Meteoric Water Line, which is based on measured isotopic composition of precipitation, and the Local Evaporation Line, which is based on modelled isotopic composition of reference points. Evaporation pan data is used to assess short-term variations in key isotopic reference values, which are important for addressing short-term changes in the isotopic signature of shallow basins. This framework is used in subsequent chapters including assessment of seasonal and annual water balance of two hydrologically-contrasting shallow lakes, and to quantify the impacts of flood water and snowmelt on a set of 45 lakes in spring 2003. Five years of isotope data using time-series analysis and the isotopic framework suggested that a perched (isolated) lake and its catchment (forest and bedrock) in the northern, relict Peace sector captured sufficient rain, snow, and runoff to maintain a relatively stable water balance, and also that a low-lying lake in the southern, active Athabasca sector was regularly replenished with river water in both spring and summer. Snowmelt and rainfall were found to have diluted the perched basin by an average of 16% and 28 % respectively, while spring and summer floods were found to almost completely flush the low-lying lake. Using the spring 2003 regional dataset, flooded lakes were separated from snowmelt-dominated lakes through use of suspended sediment concentrations, isotope data, and field observations. Application of an isotope mixing model translated d18O values into a range of replenishment amount by either river water or snowmelt, which compared well with hydrological conditions at the time of sampling and previously classified drainage types of the lakes. Spatial mapping of replenishment amounts illustrated flooding of much of the Athabasca sector due to ice-jams, except for two sub-regions isolated from flooding by artificial and natural northern diversion of flow from the Athabasca River. It is also shown that most of the relict landscape of the Peace sector was replenished by snowmelt except for a few low-lying lakes close to the Peace River and its tributaries. Overall, improved understanding of lake and regional hydrology in the PAD, especially the ability to quantify the affects of various lake inputs, will improve the ability to develop effective guidelines and management practices in the PAD as lakes respond to future changes in climate and river discharge.

Water isotope tracers and multi-proxy paleolimnological approaches are used to characterize the present and past hydrology of the Slave River Delta (SRD), NWT. This research addresses crucial gaps in knowledge about the role of major hydrological processes on the water balances of northern freshwater lakes, and responds to concerns expressed by local land users about declining flood frequency in the delta following upstream river regulation. Contemporary hydrological studies were conducted using multiple lakewater sampling campaigns from a suite of 41 delta lakes situated in three previously recognized biogeographical zones - outer delta, mid-delta and apex – that were initially sampled in fall 2002, and again immediately following the spring melt, during summer, and in the fall of 2003-2005. Paleolimnological studies aimed at reconstructing flood frequency in the Slave River focus on a sediment core obtained from a flood-susceptible lake in the active delta. Together, contemporary and past studies of SRD hydrology provide a detailed picture of environmental change and variability in an important northern freshwater ecosystem.

This dissertation presents a comprehensive design and analysis framework for system-on-a-chip analog-to-digital conversion design. The design encompasses a broad class of systems, which take advantage of system-on-a-chip complexity. This class is exemplified by an interferometric photodetector array based bio-optoelectronic sensor that is built and tested as part of the reported work. While there have been many discussions of the technical details of individual analog-to-digital converter (ADC) schemes in the literature, the importance of the analog front-end as a pre-processor for a data converter and the generalized analysis including converter encoding and decoding functions have not previously been investigated thoroughly, and these are key elements in the choice of converter designs for low-noise systems such as bio-optoelectronic sensors. Frequency domain analog front-end models of ADCs are developed to enable the architectural modeling of ADCs. The proposed models can be used for ADC statistically worst-case performance estimation, with stationary random process assumptions on input signals. These models prove able to reveal the architectural advantages of a specific analog-to-digital converter schemes quantitatively, allowing meaningful comparisons between converter designs. The modeling of analog-to-digital converters as communication channels and the ADC functional analysis as encoders and decoders are developed. This work shows that analog-to-digital converters can be categorized as either a decoder-centered design or an encoder-centered design. This perspective helps to show the advantages of nonlinear decoding schemes for oversampling noise-shaping data converters, and a new nonlinear decoding algorithm is suggested to explore the optimum solution of the decoding problem. A case study of decoder-centered and encoder-centered data converter designs is presented by applying the proposed theoretical framework. The robustness and flexibility of the resulting analog-to-digital converters are demonstrated and compared. The electrical and optical sensitivity measurements of a fabricated oversampling noise shaping analog-to-digital converter circuit are provided, and a sensor system-on-a-chip using these ADCs with integrated interferometric waveguides for bio-optoelectronic sensing is demonstrated.

The thesis proposes an integrator with an OPAMP in the feedback loop to fulfill 1MHz bandwidth SI Sigma Delta modulator. The OPAMP is used to pull down the input impedance and get high speed and high resolution. Oversampling and noise shaping are the two keys of Sigma Delta modulator. In structure, multistage is helpful for depressing noises and we use three stages to fulfill this 4-order proposed Sigma Delta modulator. The proposed Sigma Delta modulator uses TSMC 0.18£gm CMOS process and it is a 4-order and three stages SI Sigma Delta modulator. The sampling rate is 32MHz, bandwidth is 1MHz, and oversampling ratio is 16.

none

The Best PhD Thesis in the Faculties of Architecture, Arts, Business &Economics, Education, Law and Social Sciences (University of HongKong), Li Ka Shing Prize, 2009-2010 / published_or_final_version / Sociology / Doctoral / Doctor of Philosophy