UC Berkeley EECS Technical ReportsThe UC Berkeley EECS Technical Memorandum Series provides a dated archive of EECS research. It includes Ph.D. theses and master's reports as well as technical documents that complement traditional publication media such as journals. For example, technical reports may document work in progress, early versions of results that are eventually published in more traditional media, and supplemental information such as long proofs, software documentation, code listings, or elaborated examples.http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019-02-22T02:37:04Z2019-02-22T02:37:04ZenLow Power ISM band receiver front endAli NiknejadHari VEMURIhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-4.html2019-02-21T08:00:00Z2019-02-21T08:00:00Z<p>Low Power ISM band receiver front end</p>
<p>
Ali Niknejad and Hari VEMURI</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2019-4<br>
February 21, 2019</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-4.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-4.pdf</a></p>
<p>This report discusses the design and simulation results of an ultra low power receiver front end circuit for the ISM band(2.4GHz). The design has been done using a 65nm CMOS process. The direct conversion receiver architecture employs synchronous detection using a local oscillator whose frequency is same as the carrier frequency of the signal. As a result, the circuit complexity is significantly reduced, enabling integration with the baseband circuitry. Using simultaneous Inphase(I) and Quadrature(Q) mixing, the image problem is eliminated. This project investigates both active and passive downcoversion techniques. Besides, driver circuits have also been designed to amplify and buffer the VCO output to drive the Local Oscillator(LO) ports of the mixer. The receiver based on the active mixer has a simulated noise figure of 4.8dB, IIP3 of -19dBm and a power consumption of 1.95mW including the LO drivers. The receiver based on the passive mixer has a simulated noise figure of 4.8dB, IIP3 of -15dBm and a power consumption of 1.92mW, with the mixers and LNA alone consuming 1.6mW. The active mixer is based on a single balanced topology whereas the passive mixer is based on a fully differential Transimpedance Amplifier(TIA) to convert the mixer current to voltage output. A CMOS Transimpedance amplifier along with Common Mode Feedback(CMFB) circuit have also been designed for implementing the passive mixer. The LO driver-buffer stage comprises of an amplifier and series of invertors for achieving the requisite fanout to drive the LO input ports of the mixer. All circuits have been implemented at transistor level and have realistic passives with a quality factor of 10 for inductors and 50 for capacitors. The maximum supply voltage is 1V. This power constrained design is a tradeoff between noise, linearity and power dissipation.</p>2019-02-21T08:00:00ZCloud Programming Simplified: A Berkeley View on Serverless ComputingEric JonasJohann Schleier-SmithVikram SreekantiChia-Che TsaiAnurag KhandelwalQifan PuVaishaal ShankarJoao Menezes CarreiraKarl KrauthNeeraja YadwadkarJoseph GonzalezRaluca Ada PopaIon StoicaDavid A. Pattersonhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-3.html2019-02-10T08:00:00Z2019-02-10T08:00:00Z<p>Cloud Programming Simplified: A Berkeley View on Serverless Computing</p>
<p>
Eric Jonas, Johann Schleier-Smith, Vikram Sreekanti, Chia-Che Tsai, Anurag Khandelwal, Qifan Pu, Vaishaal Shankar, Joao Menezes Carreira, Karl Krauth, Neeraja Yadwadkar, Joseph Gonzalez, Raluca Ada Popa, Ion Stoica and David A. Patterson</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2019-3<br>
February 10, 2019</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-3.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-3.pdf</a></p>
<p>Serverless cloud computing handles virtually all the system administration operations needed to make it easier for programmers to use the cloud. It provides an interface that greatly simplifies cloud programming, and represents an evolution that parallels the transition from assembly language to high-level programming languages. This paper gives a quick history of cloud computing, including an accounting of the predictions of the 2009 Berkeley View of Cloud Computing paper, explains the motivation for serverless computing, describes applications that stretch the current limits of serverless, and then lists obstacles and research opportunities required for serverless computing to fulfill its full potential. Just as the 2009 paper identified challenges for the cloud and predicted they would be addressed and that cloud use would accelerate, we predict these issues are solvable and that serverless computing will grow to dominate the future of cloud computing.</p>2019-02-10T08:00:00ZTripAware: Emotional and Informational Approaches to Encourage Sustainable Transportation via Mobile ApplicationsJesse ZhangJohn SullivanVasudev Venkatesh P. B.Kyle TseAndy YanJohn LeydenKalyanaraman ShankariRandy H. Katzhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-2.html2019-01-11T08:00:00Z2019-01-11T08:00:00Z<p>TripAware: Emotional and Informational Approaches to Encourage Sustainable Transportation via Mobile Applications</p>
<p>
Jesse Zhang, John Sullivan, Vasudev Venkatesh P. B., Kyle Tse, Andy Yan, John Leyden, Kalyanaraman Shankari and Randy H. Katz</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2019-2<br>
January 11, 2019</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-2.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-2.pdf</a></p>
<p>This exploratory study investigates how mobile applications assist commuters with sustainable transportation choices. Our goal is to determine persuasiveness of two broad categories of features: emotional or informational. A controlled trial randomly assigned 41 users to three mobile applications: Emotion, Information, and Control. During the ten week study, we recorded user interactions and changes in transportation habits.
<p>Several features distinguish this study from prior work, the most salient of which are: (1) automatic trip recording, segmentation, and classification; (2) statistical assessment of metrics that reflect a user's interactions and behaviors; (3) larger and more diverse samples. </p>
<p>Using hypothesis testing, we found that Emotion resulted in greater engagement with the application (p=0.006, 0.035, 0.031, 0.040) while Information improved the sustainability of travel behavior (p = 0.043). This suggests a combination of both approaches is required in order to both maintain user engagement and have an effect on carbon emissions.</p></p>2019-01-11T08:00:00ZGordian: Formal Reasoning Based Outlier Detection for Secure LocalizationMatthew WeberBaihong JinGil LedermanYasser ShoukryEdward A. LeeSanjit A. SeshiaAlberto L. Sangiovanni-Vincentellihttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-1.html2019-01-11T08:00:00Z2019-01-11T08:00:00Z<p>Gordian: Formal Reasoning Based Outlier Detection for Secure Localization</p>
<p>
Matthew Weber, Baihong Jin, Gil Lederman, Yasser Shoukry, Edward A. Lee, Sanjit A. Seshia and Alberto L. Sangiovanni-Vincentelli</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2019-1<br>
January 11, 2019</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-1.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2019/EECS-2019-1.pdf</a></p>
<p>Accurate localization from Cyber-Physical Systems (CPS) is a critical enabling technology for context aware applications and CPS control. As localization plays an increasingly safety-critical role, location systems must be able to identify and eliminate faulty measurements to prevent dangerously inaccurate localization. In this paper we consider the range-based localization problem and propose a method to detect coordinated adversarial corruption on anchor positions and distance measurements. Our algorithm, Gordian, rapidly finds attacks by identifying geometric inconsistencies at the graph level without requiring assumptions about hardware, ranging mechanisms or cryptographic protocols. We give necessary conditions for which attack detection is guaranteed to be successful in the noiseless case, and use that intuition to extend Gordian to the noisy case where fewer guarantees are possible. In simulations generated from real-world sensor noise, we empirically show Gordian’s trilateration counterexample generation procedure enables rapid attack detection even for combinatorially difficult problems.</p>2019-01-11T08:00:00ZSculpture Designs Based on Borromean Soap FilmsCarlo H. Séquinhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-192.html2018-12-31T08:00:00Z2018-12-31T08:00:00Z<p>Sculpture Designs Based on Borromean Soap Films</p>
<p>
Carlo H. Séquin</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-192<br>
December 31, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-192.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-192.pdf</a></p>
<p>The Borromean rings are used as the border curves of aesthetically pleasing 2-manifolds, approximating minimal-area surfaces. Two- and three-level deep recursive configurations are then constructed and the border curves of the various levels are gracefully joined to obtain a single 2-manifold covering all the nested levels. The results are intricate, but highly symmetrical, abstract geometrical sculptures.</p>2018-12-31T08:00:00ZScaling Interactive Data Science Transparently with ModinDevin Petersohnhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-191.html2018-12-19T08:00:00Z2018-12-19T08:00:00Z<p>Scaling Interactive Data Science Transparently with Modin</p>
<p>
Devin Petersohn</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-191<br>
December 19, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-191.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-191.pdf</a></p>
<p>The combined growth of data science and big datasets has increased the performance requirements for running data analysis experiments and workflows. However, popular data science toolkits such as Pandas have not adapted to the technical demands of modern multicore, parallel hardware. As such, data scientists aiming to work with large quantities of data find themselves either suffering from libraries that under-utilize modern hardware or being forced to use big data processing tools that do not adapt well to the interactive nature of exploratory data analyses.
<p>In this report we present the foundations of Modin, a library for large scale data analysis. Modin emphasizes performant, parallel execution on big datasets previously deemed unwieldy for existing popular toolkits, all while importantly maintaining an interface and set of semantics similar to existing interactive data science tools. The experiments presented in this report demonstrate promising results towards developing a new generation of performant data science tools built for parallel and distributed modern hardware.</p></p>
<p><strong>Advisor:</strong> Anthony D. Joseph</p>2018-12-19T08:00:00ZHindSight: Enhancing Spatial Awareness by Sonifying Detected Objects in Real-Time 360-Degree VideoEldon SchoopJames SmithBjörn Hartmannhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-190.html2018-12-19T08:00:00Z2018-12-19T08:00:00Z<p>HindSight: Enhancing Spatial Awareness by Sonifying Detected Objects in Real-Time 360-Degree Video</p>
<p>
Eldon Schoop, James Smith and Björn Hartmann</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-190<br>
December 19, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-190.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-190.pdf</a></p>
<p>Our perception of our surrounding environment is limited by the constraints of human biology. The field of augmented perception asks how our sensory capabilities can be usefully extended through computational means. We argue that spatial awareness can be enhanced by exploiting recent advances in computer vision which make high-accuracy, real-time object detection feasible in everyday settings. We introduce HindSight, a wearable system that increases spatial awareness by detecting relevant objects in live 360-degree video and sonifying their position and class through bone conduction headphones. HindSight uses a deep neural network to locate and attribute semantic information to objects surrounding a user through a head-worn panoramic camera. It then uses bone conduction headphones, which preserve natural auditory acuity, to transmit audio notifications for detected objects of interest. We develop an application using HindSight to warn cyclists of approaching vehicles outside their field of view. To evaluate HindSight, we first conduct an exploratory study with 15 users. We next create a VR platform to simulate realistic traffic scenarios and use it to evaluate HindSight in a controlled user study with 21 participants. Participants using HindSight had fewer collisions, increased their space to other vehicles, experienced reduced cognitive load, and reported a perceived increase in awareness.</p>
<p><strong>Advisor:</strong> Björn Hartmann</p>2018-12-19T08:00:00ZReducing Actuation Voltage in RF MEMS Switches and the Impact of Scaling on Performance and ReliabilityJaime Castrohttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-189.html2018-12-19T08:00:00Z2018-12-19T08:00:00Z<p>Reducing Actuation Voltage in RF MEMS Switches and the Impact of Scaling on Performance and Reliability</p>
<p>
Jaime Castro</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-189<br>
December 19, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-189.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-189.pdf</a></p>
<p>For over two decades, researchers have written about microelectromechanical switches and their remarkable performance in terms of low insertion loss, high linearity, high isolation, and extremely low power consumption. Although these characteristics are highly desired in RF applications, the high actuation voltage currently required to operate these switches (typically in the 20 to 80 volts range), presents a challenge for incorporating MEMS switches into portable wireless, low-power, and battery-operated systems. Continuing to push for yet smaller dimensions can help in reducing actuation voltage requirements and provides additional benefits such as higher integration and speed. Despite these advantages, scaling down can also emphasize reliability concerns that reduce the lifetime of the switch. The work presented here touches on the fundamentals of electrostatically actuated RF MEMS switches and the impact of scaling to both reliability and performance.</p>
<p><strong>Advisor:</strong> Vladimir Stojanovic</p>2018-12-19T08:00:00ZReducing Actuation Voltage in RF MEMS Switches and the Impact of Scaling on Performance and ReliabilityJaime Castrohttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-188.html2018-12-18T08:00:00Z2018-12-18T08:00:00Z<p>Reducing Actuation Voltage in RF MEMS Switches and the Impact of Scaling on Performance and Reliability</p>
<p>
Jaime Castro</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-188<br>
December 18, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-188.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-188.pdf</a></p>
<p>For over two decades, researchers have written about microelectromechanical switches and their remarkable performance in terms of low insertion loss, high linearity, high isolation, and extremely low power consumption. Although these characteristics are highly desired in RF applications, the high actuation voltage currently required to operate these switches—typically in the 20 to 80 volts range, presents a challenge for incorporating MEMS switches into portable wireless, low-power, and battery-operated systems. Continuing to push for yet smaller dimensions can help in reducing actuation voltage requirements and provides additional benefits such as higher integration and speed. Despite these advantages, scaling down can also emphasize reliability concerns that reduce the lifetime of the switch. The work presented here touches on the fundamentals of electrostatically actuated RF MEMS switches and the impact of scaling to both reliability and performance.</p>2018-12-18T08:00:00ZNegative capacitance and hyperdimensional computing for unconventional low-power computingJustin Wonghttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-187.html2018-12-17T08:00:00Z2018-12-17T08:00:00Z<p>Negative capacitance and hyperdimensional computing for unconventional low-power computing</p>
<p>
Justin Wong</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-187<br>
December 17, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-187.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-187.pdf</a></p>
<p>Properties that emerge from the collective behavior of constituents at different length scales can be exploited to reduce power consumption below conventional limits in computing. At the device level, ferroelectric-dielectric coupling ("negative capacitance") can reduce energy consumption below 1 / 2 CV 2 in capacitors. However, this effect is still not well understood. We construct a microscopic model and analyze energy flow from the perspective of Poynting's theorem to clear up these misunderstandings. At the circuit level, high-dimensional distributed representations relax requirements on signal-to-noise ratio and supply voltage, and enable new architecture designs. Computing with these representations ("hyperdimensional computing") is natural for performing energy efficient cognitive computing at the application level. However, data in practice is always measured in some sort of representation, which may not be natural for hyperdimensional computing. We bridge this gap by proposing to use an approximation of the bispectrum to map data measured in practice into high-dimensional distributed representations for use with hyperdimensional computing.</p>
<p><strong>Advisor:</strong> Sayeef Salahuddin</p>2018-12-17T08:00:00ZScaling Effect on RF MEMS SwitchXiao Chenhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-186.html2018-12-16T08:00:00Z2018-12-16T08:00:00Z<p>Scaling Effect on RF MEMS Switch</p>
<p>
Xiao Chen</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-186<br>
December 16, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-186.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-186.pdf</a></p>
<p>The purpose of this project is to study the scaling effect on (radio frequency) RF micro electromechanical systems (MEMS) Switch. Many MEMS devices’ performance could be enhanced tremendously by scaling. This capstone report talks about the benefits and drawbacks of scaling. A detailed analysis on reliability will be introduced. Scaling effects on other parameters such as switching speed and actuation voltage will be discussed. This paper will point out a suggestion of how RF MEMS switch could be optimized and designed in the future.</p>
<p><strong>Advisor:</strong> Jan M. Rabaey</p>2018-12-16T08:00:00ZOptimization Everywhere: Convex, Combinatorial, and EconomicSam Wonghttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-185.html2018-12-14T08:00:00Z2018-12-14T08:00:00Z<p>Optimization Everywhere: Convex, Combinatorial, and Economic</p>
<p>
Sam Wong</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-185<br>
December 14, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-185.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-185.pdf</a></p>
<p>In this thesis we study fundamental problems that arise in optimization and its applications. We present provably efficient algorithms that achieve better running times or approximation guarantees than previously known. Our method draws on the toolkit from convex and combinatorial optimization as well as economics. By intertwining techniques from these disciplines, we are able to make progress on multiple old and new problems, some of which have stood open for many years. Main results of this thesis include the following:
<p>Convex Programming: We show how to solve convex programming with an expected O(n log(nR/\epsilon)) evaluations of the separation oracle and additional time O(n^3\log^{O(1)}(nR/\epsilon)). This matches the oracle complexity and improves upon the O(n^{\omega+1}\log(nR/\epsilon)) additional time of the previous fastest algorithm achieved over 25 years ago for the current value of the matrix multiplication constant when R/\epsilon=O(\poly(n)). </p>
<p>Submodular Function Minimization: We provide new weakly and strongly polynomial time algorithms with a running time of O(n^{2}\log nM\cdot\text{EO}+n^{3}\log^{O(1)}nM) and O(n^{3}\log^{2}n\cdot\text{EO}+n^{4}\log^{O(1)}n), improving upon the previous best of O((n^{4}\cdot\text{EO}+n^{5})\log M) and O(n^{5}\cdot\text{EO}+n^{6}) respectively. We also provide the first subquadratic time algorithm for computing an approximately optimal solution. </p>
<p>Matroid Intersection: We provide new algorithms with a running time of O(nr\mathcal{T_{\text{rank}}}\log n\log(nM)+n^{3}\log^{O(1)}nM) and O(n^{2}\mathcal{T_{\text{ind}}}\log(nM)+n^{3}\log^{O(1)}nM , achieving the first quadratic bound on the query complexity for the independence and rank oracles. In the unweighted case, this is the first improvement since 1986 for independence oracle. </p>
<p>Submodular Flow: We obtain a faster weakly polynomial running time of O(n^{2}\log(nCU)\cdot\EO+n^{3}\log^{O(1)}(nCU)), improving upon the previous best of O(mn^{5}\log(nU)\cdot\EO) and O\left(n^{4}h\min\left\{ \log C,\log U\right\} \right) from 15 years ago by a factor of \tilde{O}(n^{4}). </p>
<p>Semidefinite Programming: We obtain a running time of \tilde{O}(n(n^{2}+m^{\omega}+S)), improving upon the previous best of \tilde{O}(n(n^{\omega}+m^{\omega}+S)) for the regime S is small. </p>
<p>Market Equilibrium: We present the first polynomial time algorithm for computing market equilibrium in an economy with indivisible goods and general buyer valuations having only access to an aggregate demand oracle. </p>
<p>Vertex Cover with Hard Capacity: We give a f-approximation algorithm for the minimum unweighted Vertex Cover problem with Hard Capacity constraints on f-hypergraphs This improves over the previous 2f-approximation and is the best possible assuming the unique game conjecture. </p>
<p>Network Design for Effective Resistance: We initiate the study of network design for s-t effective resistance. Among other results we present a constant factor approximation by applying classic techniques to a convex quadratic programming relaxation.</p></p>
<p><strong>Advisor:</strong> Christos Papadimitriou</p>2018-12-14T08:00:00ZDomain-Specific Techniques for High-Performance Computational Image ReconstructionMichael Driscollhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-184.html2018-12-14T08:00:00Z2018-12-14T08:00:00Z<p>Domain-Specific Techniques for High-Performance Computational Image Reconstruction</p>
<p>
Michael Driscoll</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-184<br>
December 14, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-184.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-184.pdf</a></p>
<p>The widespread emergence of parallel computers in the last decade has created a substantial programming challenge for application developers who wish to attain peak performance for their applications. Parallel programming requires significant expertise, and programming tools---general-purpose languages, compilers, libraries, etc.---have had limited success in hiding the complexity of parallel architectures. Furthermore, the parallel programming burden is likely to increase as processor core counts grow and memory hierarchies become deeper and more complex.
<p>The challenge of delivering productive high-performance computing is especially relevant to computational imaging. One technique in particular, iterative image reconstruction, has emerged as a prominent technique in medical and scientific imaging because it offers enticing application benefits. However, it often demands high-performance implementations that can meet tight application deadlines, and the ongoing development of the iterative reconstruction techniques discourages ad-hoc performance optimization efforts. </p>
<p>This work explores productive techniques for implementing fast image reconstruction codes. We present a domain-specific programming language that is expressive enough to represent a variety of important reconstruction problems, but restrictive enough that its programs can be analyzed and transformed to attain good performance on modern multi-core, many-core and GPU platforms. We present case studies from magnetic resonance imaging (MRI), ptychography, magnetic particle imaging, and microscopy that achieve up to 90% of peak performance. We extend our work to the distributed-memory setting for an MRI reconstruction task. There, our approach gets perfect strong scaling for reasonable machine sizes, and sets the best-known reconstruction time for our particular reconstruction task. The results indicate that a domain-specific language can be successful in hiding much of the complexity of implementing fast reconstruction codes.</p></p>
<p><strong>Advisor:</strong> Katherine A. Yelick and Armando Fox</p>2018-12-14T08:00:00ZNon-Linear Stiffness Extraction & Modeling of Wineglass Disk ResonatorsAlain Antonhttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-183.html2018-12-14T08:00:00Z2018-12-14T08:00:00Z<p>Non-Linear Stiffness Extraction & Modeling of Wineglass Disk Resonators</p>
<p>
Alain Anton</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-183<br>
December 14, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-183.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-183.pdf</a></p>
<p><strong>Advisor:</strong> Clark Nguyen</p>2018-12-14T08:00:00ZLow Dimensional Methods for High Dimensional Magnetic Resonance ImagingFrank Onghttp://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-181.html2018-12-14T08:00:00Z2018-12-14T08:00:00Z<p>Low Dimensional Methods for High Dimensional Magnetic Resonance Imaging</p>
<p>
Frank Ong</p>
<p>
EECS Department<br>
University of California, Berkeley<br>
Technical Report No. UCB/EECS-2018-181<br>
December 14, 2018</p>
<p>
<a href="http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-181.pdf">http://www2.eecs.berkeley.edu/Pubs/TechRpts/2018/EECS-2018-181.pdf</a></p>
<p>Magnetic Resonance Imaging (MRI) is an amazing imaging modality in many aspects. It offers one of the best imaging contrast for visualizing soft issues. It has no ionizing radiation at all. Its flexibility has also enabled many applications, including assessing blood flow, imaging brain activity via oxygenation contrast, and measuring tissue stiffness. Since MRI was invented, this imaging technology has saved numerous lives, and has been the frontier of biomedical and engineering research.
<p>On the other hand, imaging speed remains a main limitation of MRI. Inherently, MRI takes time to collect measurements, and often requires minutes to complete a scan. In this regard, MRI is quite similar to early cameras: Subjects have to be motionless for minutes to obtain an image, which is uncomfortable to patients. This often leads to motion and motion artifacts. When severe motion artifacts occur, scans have to be repeated. </p>
<p>This dissertation aims to change that by developing techniques to reconstruct three-dimensional (3D) dynamic MRI from continuous acquisitions. An ideal 3D dynamic scan would be able to resolve all dynamics at a high spatiotemporal resolution. Subjects would not have to be motionless. The comprehensive information in the single scan would also greatly simplify clinical workflow. While this dissertation has not achieved this ideal scan yet, it proposes several innovations toward this goal. In particular, www.doi.org/10.6084/m9.figshare.7464485 shows a 3D rendering of a reconstruction result from this dissertation. Arbitrary slices at different orientation can be selected over time. Respiratory motion, contrast enhancements, and even slight bulk motion can be seen. </p>
<p>The main challenge in high resolution 3D dynamic MRI is that the reconstruction problem is inherently underdetermined and demanding of computation and memory. To overcome these challenges, this dissertation builds on top of many fundamental methods, including non-Cartesian imaging, parallel imaging and compressed sensing. In particular, this dissertation heavily relies on the compressed sensing framework, which has three components: 1) the image of interest has a compressed signal representation. 2) MRI can acquire (pseudo)-randomized samples in k-space, which provides incoherent encoding of the underlying image. 3) sparsity/compressibility can be efficiently enforced in reconstruction to recover the compressed representation from the undersampled measurements. </p>
<p>In this dissertation, I propose a multiscale low rank model that can compactly represent dynamic image sequences. The resulting representation can be applied beyond MRI, and is useful for other applications, such as motion separation in surveillance video. With the multiscale low rank representation, I propose a technique incorporating stochastic optimization to efficiently reconstruct 3D dynamic MRI. This makes it feasible to run such large-scale reconstructions on local workstations. To further speed up the reconstruction time, I propose accelerating the convergence of non-Cartesian reconstruction using a specially designed preconditioner. Finally, I leverage external undersampled datasets to further improve reconstruction quality using convolutional sparse coding.</p></p>
<p><strong>Advisor:</strong> Michael Lustig</p>2018-12-14T08:00:00Z