Breakout Session Abstracts

Thursday, February 28

Energy-Efficient Electronics, Corwin Pavillion

Innovations for Increased Energy Efficiency

Global power consumption is approximately 14 Terawatts (TW) and rising rapidly, as more and more countries improve their quality of life. Eighty percent of this power is generated by burning fossil fuels, and that sends nearly nine billion tons of carbon into our atmosphere every year in the form of greenhouse gasses. This is having a major impact on our environment, our economy, and our quality of life.
UCSB is establishing an Institute on Energy Efficiency. Our focus is on reducing demand, because the cleanest power plant is the one you don’t build. The Institute is developing key technologies to make our use of energy more efficient. These technologies will be able to drive economic growth without increasing energy consumption. Improvements in energy efficiency can have a far bigger economic and environmental impact, realized far sooner, than the development of alternative energy sources. Our research is directed toward lighting, computing, communication, transportation, buildings and solar power. Several of these areas will be highlighted in this talk, and other areas will be covered in the subsequent talks.

Engineered Materials for Thermoelectric Power Generation

We are developing new materials for the direct thermoelectric generation of electricity from heat. The performance of thermoelectric materials can be enhanced by increasing their electrical conductivity and decreasing their thermal conductivity. We are producing these characteristics by incorporation of crystalline metallic nanoparticles (erbium arsenide) in compound semiconductors (InGaAs) We have achieved record figures of merit at temperatures above 225C and have incorporated the materials into power generation modules.

Energy-Efficient Computing: From Low-Power Devices to Energy-Aware Data Centers

Energy-Efficient Solid State Lighting

Toward Terabit Access Networks, State Street Room

MultiGigabit Millimeter Wave Communication: New Architectures and System Concepts

Integrated Circuits for High Capacity Communications

The bandwidths of both Si and InP transistors will soon approach 0.5 to 1 THz. InP transistors have low noise, high power, and low distortion, while Si CMOS devices offer large integration scales. These offer complementary benefits to communications links. With InP transistors, we have developed (with our partners) many ICs, including 300+ GHz amplifiers for sub-mm-wave imaging, and mm-wave op-amps for low-distortion microwave amplification. Massively complex high speed CMOS ICs will allow massive parallelism in mm-wave communications links for both highly directional communications and for massive numbers of parallel communications links, while the high available bandwidth and large # of parallel of signal processing channels will permit comprehensive compensation of channel imperfections, allowing transmission at near SNR limits.

Integrated Photonic Technologies for Optical Networks

Digitally Enhanced Microwave Circuits for Next Generation Wireless Systems

Bio-Inspired Engineering, SB Harbor Room

Biotechnology Opens New Paths to Semiconductors for Energy Applications

Biological systems fabricate multifunctional, high-performance materials at low temperatures and near-neutral pH with a precision of three-dimensional nanostructural control that exceeds the capabilities of present human engineering. We discovered the mechanism governing the nanofabrication of silica in a marine sponge, and translated this mechanism to develop a generic new, “biologically inspired” low-temperature route for the kinetically controlled synthesis of a wide range of nanostructured semiconductor thin films and nanoparticles without the use of organic templates. We learned that the silicateins - proteins we found occluded in the glass skeletal elements of a marine sponge - self assemble to form macroscopic, crystallographically order filaments, and that these filaments are capable of enzymatically catalyzing and templating the synthesis of silica and a wide variety of silsesquioxanes and metal oxide semiconductors from the corresponding molecular precursors at low temperature and near-neutral pH. These were the first reported examples of enzyme-catalyzed, nanostructure-directed synthesis of semiconductors. Interaction with the template-like protein surface stabilizes polymorphs of these materials (e.g., the anatase form of titanium dioxide and the spinel polymorph of gallium oxide) otherwise not formed at low temperatures. This observation and the preferential alignment of the Ga2O3 nanocrystallites suggested a pseudo-epitaxial relationship between the mineral crystallites and specific functional groups on the templating protein surface. The novel mechanisms of self-assembly and catalytic nanofabrication revealed through detailed genetic and molecular analyses of the silicateins enabled the development of a novel and highly generic strategy for the low-temperature synthesis of a wide range of nanostructured inorganic materials.
From the mechanism of synthesis revealed by the silicateins, and confirmed with a series of biomimetic, synthetic analogs and catalytic self-assembled monolayers that mimic the catalytic and templating features of the natural protein, we developed a generic new, biologically inspired low-temperature route for the kinetically controlled synthesis of a wide range of nanostructured metal oxide, -hydroxide, -phosphate and bimetallic perovskite semiconductor thin films and nanoparticles without the use of organic templates. Post-synthesis conversion to the nitrides and sulfides has been demonstrated. Because no organics are used, this new biologically inspired synthesis method yields high purity inorganic semiconductors, and thus is potentially integrable with MOCVD, CMOS and other conventional manufacturing methods.
Employing gentle catalysis at low temperature, this method can preserve the intermetallic organization of bimetallic precursors that are thus incorporated into crystalline solids as bimetallic molecular units without phase segregation. This has led to the first low-temperature synthesis of 6 nm barium titanate nanoparticles with low polydispersity, good electronic properties and no organic contaminants. We also have used this process for the low-temperature synthesis of a wide range of supported (substrate-grown) and unsupported (free-standing) nanostructured thin films. One such material is strongly photoconductive cobalt hydroxide-based thin film material that exhibits high dopant density, high surface area of single-crystal domains and strong absorption in the visible, making it potentially attractive for photovoltaic applications. Conversion to the oxide yields a nanostructured material now under investigation for its advantages in high power-density 3-dimensional batteries. A wide range of other materials made by this low-temperature process offers unique combinations of structures and properties not readily attainable by conventional high-temperature processes; these exhibit potential advantages now under investigation for improved energy conversion and storage, ferroelectric random access memory (FeRAM), infrared and piezoelectric detectors, optoelectronics and flexible displays.

Swarming and Teaming: Biologically Inspired Distributed Estimation and Control

Trauma, Implants and Inorganic Surfaces

Friday, February 29

Energy-Efficient Building Systems, Corwin Pavillion

Integrated Building Systems

We discuss potential savings enabled by energy efficiency improvements in modern buildings, and the fundamental and applied research that is leading to those savings. A building is a complex system, where the physical effects due to complicated airflow patterns are coupled to communications and control layers in building’s HVAC, elevatoring and security systems. Thus it is a proper, but earthly example of a cyber-physical system. The design of such a system is ridden with uncertainty, not only in physical properties, but also in types of devices used to implement the cyber layer. We discuss necessary improvements in design procedures and use of an integrated, dynamical systems, airflow dynamics and control perspective, coupled to platform-based design. These improvements are projected to lead to saving of 100’s of billions of dollars a year in energy use, and half the CO2 emissions from buildings by 2030.

Curbing the Energy Demand in Buildings

Reducing energy consumption is perhaps the simplest and quickest approach to reducing greenhouse gas emissions and our dependence on external sources of fossil fuel. In the USA, roughly one third of the energy consumption occurs in buildings, which also consume about 70 percent of the electricity. Most of the building energy is consumed for heating and cooling, lighting, and appliances. To reduce energy consumption, traditional approaches have relied on improving the performance of individual components, e.g. converting incandescent lights to fluorescent ones or increasing the compressor efficiency in an HVAC unit. While this is necessary, it is not sufficient to meet the California goals, which are 90 percent reduction in new buildings and 50 percent in existing ones, with reasonable market averaged payback times. Buildings have rarely been studied as a system or a system of systems. Such an approach can identify opportunities to save energy at the interface between components, which traditional approaches may have ignored. In the future, this integrated approach needs to followed in both design of buildings as well as operation. For the latter, it is critical to develop a building operating platform, which will integrate feedback control with sensor communication, as well as energy and exergy analysis with real time optimization to reduce either cost, carbon footprint, or total energy consumption.

Challenges in Control of Buildings

We will start with experiences from an old project in control of buildings. The main experience is that it is possible to achieve significant long lasting results in a project of moderate size provided the funding is long range and that good cooperation with industry is established at an early stage. Mechanisms for creating such cooperations are also discussed. Some dramatic advances in control which set the scene for novel approaches in model based control. While much of control theory has focused on design there are challenges opportunities in the areas of modeling, commissioning and operation. These issues will be discussed together with applications to air conditioning systems.

Reduced Order Models for Airflows in Buildings

Energy efficiency promises to be one of the key issues facing society over the next generation. The economic and environmental motivations for improving energy efficiency are well documented; the challenge lies in developing specific approaches and technologies to accomplish this. In this talk, we will describe a methodology which promises to aid in the design and control of indoor airflows to improve ventilation and efficiency in buildings. This involves developing low-dimensional models of the airflows which capture enough physics to be meaningful, but which are computationally inexpensive and allow the development of control strategies. After illustrating this approach for the academic example of plane Couette flow, we will show how it can be applied to the design and control of airflows in buildings.

Design and Computing with Emerging Technologies, State Street Room

Planning for the Technology Pipeline

Computer and communication systems design has been driven by rapidly developing technologies. In this talk, I present an overview of our strategy at UCSB to position Computer Engineering to most nimbly adapt to emerging technologies, and how this strategy fits in to the initiatives of the campus and our collaborations with industry. Finally, I give an example of this strategy in my own research in quantum computing architectures.

Advancing Computer Hardware and Performance with Nanoelectronics

My talk will highlight power and interconnect related challenges in nanometer scale VLSI circuits and discuss a few alternate emerging technologies that can overcome these challenges. More specifically, I will discuss the potential for hybrid NEMS-CMOS circuits and carbon-nantube based interconnects.

Emerging Technologies for Introspection and Security

Emerging Wireless and Wireline Technologies for Chip-to-Chip Interface

The integration of antenna on chip with the transceiver can enable a wide range of new applications requiring very-low- cost, short-range wireless data link. The ability to perform wireless testing at the wafer level can reduce the testing cost of integrated circuits by avoiding expensive high-frequency probe cards. Wireless sensors and biomedical implants will also greatly benefit from the ultra small form factor of a fully integrated radio. On-chip folded dipole and loop antennas operating at different frequencies ranging from 1 to 24 GHz have been characterized to evaluate their efficiency. The reception of a FM signal sent from an on-chip loop antenna at 5.8 GHz is achieved at 10 cm. As the data rate and wire density continue to increase for chip-to-chip communication, equalization circuits have become indispensable to high-speed transceiver. The major challenge is to compensate high-frequency chip-to-chip channel losses with low power consumption. A very-low-power continuous-time adaptive equalizer has been designed utilizing an integrated passive tunable filter to overcome this challenge. The proposed equalizer does not require a recovered clock and hence it can be inserted at the front-end of existing high-speed receivers to boost the maximum data throughput. As data rate approaches 10 Gb/s and beyond, passive equalization filters offer substantial power saving compared to conventional active filters.

Large Scale Image/Video Databases, SB Harbor Room

Scaling Database Technology to Large Image Collections

Rapid advances in personal and scientific imaging technologies have led to rapidly growing collections of images and videos. The amount of metadata and the need for analysis over these datasets varies greatly based on the application domain. In this talk, I will present sensitive and yet scalable distance measures and algorithms for searching over large collections of images. Subsequently, I will discuss the increasingly important aspect of uncertainty in image databases. How to model uncertainty and how to extract features from uncertain data will be presented. Finally, I will discuss new techniques for analyzing and accessing probabilistic data.

Effective Statistical Modeling Tools for Pattern Recognition and Dynamics Analysis of Visual Data

Effective statistical modeling tools for pattern recognition and dynamics analysis of visual data Professor Rose will talk about statistical modeling tools for solving challenging pattern analysis problems. Emphasis will be given to hidden Markov models: one-dimensional to capture time correlations, and two-dimensional to capture space correlations. Main examples and results will cover extremes ranging from the bio-imaging problem of microtubial tracking and dynamics analysis, to face recognition applications, so as to illustrate the broad applicability and effectiveness of such tools.

Scalable Interactive Visualization (When Neither a Picture nor a Thousand Words Will Do)

Prof. Hollerer will talk about visualization problems, scalable interactive graph visualization, the tie-in to bio imaging, and the Allosphere.

BISQUE: Building a Web-Based Scientific Image Database and Analysis System.

Scientific imaging is a fundamental technology for many disciplines. While our capacity to generate images has grown extraordinarily over the past few years, our ability to organize, track and analyze experimental images has lagged. In this talk I will present BISQUE, a web-based image database and analysis system for scientific images. Bisque supports distributed image collections, managing associated image metadata, and large-scale image analysis.

Invite a Friend
© 2007-2008, The Regents of the University of California. All Rights Reserved.
UC Santa Barbara, College of Engineering • Terms of Use