Rice University logoGeorge R. Brown School of Engineering
Electrical and Computer Engineering

2013 ECE Affiliates Meeting - Poster/Demo Session - Graduate Research

In Progress 


Graduate Research &
Education Innovation
 Undergraduate Research &
Senior Designs
Startups by Rice Alums and Faculty 

Wednesday, April 3, 2013
1:00pm - Martel Hall, Duncan Hall

  1. Detailed mechanism for the orthogonal polarization switching of gold nanorod plasmons - Poster
  2. Designing "Good-enough" Parsimonious Hardware Systems through Inexact Computing- Poster
  3. Beyond Interference Avoidance: Distributed Sub-network Scheduling in Wireless Networks with Local Views - Poster
  4. Investigating Cognitive Side Effects of GPi Deep Brain Stimulation for Parkinson's Disease (pdf) - Poster
  5. Characterizing the Self-Interference Channel for Full-duplex Baseband Cancellation (pdf) - Poster
  6. Passive Self-Interference Suppression for Full-Duplex Wireless Communication - Poster
  7. Fundamental Limitations in Full-Duplex System Design (pdf) - Poster
  8. Algorithms and FPGA Implementation of Data Detection in the Large-Scale MIMO Uplink - Poster
  9. Parametrized Model Reduction (pdf) - Poster
  10. Detectors for Estimate-and-Forward Wireless Relay Networks (pdf) - Poster
  11. Functional Connectivity based Segmentaiton of Epileptic Activity (pdf) - Poster
  12. Routing and Power Allocation in Full-Duplex Networks - Poster
  13. Energy-Proportional Image Sensing (pdf)-  Poster
  14. Making I/O Virtualization Easy with Device Files - Demo with Poster
  15. Sulfur dioxide detecion using 7.4 μm DFB-QCL based cavity enhanced absorption spectroscopy (pdf) - Poster
  16. A compact high power CW DFB-QCL based QEPAS sensor for sensitive detection of carbon monoxide (pdf)- Demo with Poster
  17. A Software-Defined Radio Front-end for TV-Band Whitespace Research - Demo with Poster
  18. Argos: Practical Many-Antenna Basestations - Demo with Poster
  19. Distributed Full-duplex via Wireless Side Channels (pdf)- Poster
  20. Asthma Risk Estimator (pdf) - Demo with Poster
  21. Joint Sparse Factor Analysis and Topic Modeling for Learning and Content Analysis - Poster
  22. Damping of Ultra-Small Cobalt-HGNS - Poster
  23. Parallel Interleaver Design for High Throughput Configurable Turbo Decoder (pdf) - Poster
  24. Massive Manipulation (pdf) - Poster
  25. Authentication of implantable medical devices using biometric measurements - Poster
  26. Idetic: A High-level Synthesis Approach for Enabling Long Computations on Transiently-powered ASICs - Poster
  27. Nanoscale Device Metrology Utilizing Scanning Tunneling Microscopy (pdf) - Poster
  28. Large Population of Robots: Theory and Practice - Poster
  29. A Sound-to-Touch Sensory Substitution Prototype - Demo with Poster
  30. Adaptive Resource Allocation in Tiered Storage Systems - Poster
  31. A Single-Chip Dual-Mode CW/Pulse Electron Paramagnetic Resonance Spectrometer in 0.13μm SiGe BiCMOS  - Poster
  32. A Silicon-Based, Fully Integrated Pulse Electron Paramagnetic Resonance System for mm-Wave Spectroscopy - Poster
  33. Optical Waveguides and Photodiodes in 0.18μm CMOS SOI with No Post-processing (pdf) - Poster

Education Innovation
  1. Connexions and OpenStax College 

  1. Detailed mechanism for the orthogonal polarization switching of gold nanorod plasmons
    Authors: Jana Olson, Pattanawit Swanglap, Wei-Shun Chang, Saumyakanti Khatua, David Solis, and Stephan Link (Chemistry)

    Abstract: We describe an electro-optic material capable of orthogonally switching the polarization of the localized surface plasmon resonance scattering of single gold nanorods, independent of their orientation. Liquid crystal samples are prepared in a sandwich configuration with electrodes arranged so that an applied voltage induces alignment-switching of the liquid crystal molecules covering individual gold nanorods. Due to the birefringence of the nematic liquid crystal, the reorientation in the nematic director alignment causes a change in the output polarization of the scattered light. We propose the underlying mechanism to be based on a homogeneous nematic to twisted nematic phase transition and provide support for it via Jones calculus by modelling the effect of ideally aligned homogeneous nematic and twisted nematic phases on polarized light transmitted through the sample. In the model, we include the effects of sample thickness and surface plasmon resonance wavelength, expressed in terms of the phase retardation, χ, on the observed output polarization. We find four distinctively different trends for the output polarization as a function of the incident polarization as χ is varied. Two of these cases provide reproducible orthogonal polarization switching of the surface plasmon resonance while maintaining with a high degree of polarization. These results are verified experimentally with liquid crystal cells of different thicknesses. The deviation of the experimental samples from ideal behaviour can be explained by the inherent variations in the surface plasmon resonance maximum and local cell thickness.

  2. Designing "Good-enough" Parsimonious Hardware Systems through Inexact Computing
    Authors: Avinash Lingamneni, Krishna Palem (CS)

    Abstract: It is widely acknowledged that the exponentially improving benefits of sustained technology scaling prophesied by the Moore’s Law would end within the next decade or so, attributed primarily to an understanding that switching devices can no longer function deterministically as feature sizes are scaled down to the molecular levels. The benefits of Moore’s Law could, however, continue provided systems with probabilistic or “error-prone” elements could still process information usefully. We believe that this is indeed possible in contexts where the "quality" of the results of the computation is perceptually determined by our senses—audio and video information being significant examples. To demonstrate this principle, we will show how such ‘’inexact“ computing based devices, circuits and computing architectures can be used effectively to realize "good-enough" systems for many ubiquitous error-resilient applications. Further, we show that significant energy, performance and area gains can be achieved, while trading a perceptually tolerable level of error – that will be ultimately determined based on neurobiological models— applied in the context of video and audio data in digital signal processing. This design philosophy of inexact computing is of particular interest in the domain of embedded and (portable) multimedia applications and in application domains of budding interest such as recognition and data mining, all of which can tolerate inaccuracies to varying extents or can synthesize accurate (or sufficient) information even from inaccurate computations!

  3. Beyond Interference Avoidance: Distributed Sub-network Scheduling in Wireless Networks with Local Views
    Authors: Pedro Santacruz, Ashutosh Sabharwal

    Abstract: In most wireless networks, nodes have only limited local information about the network state, which includes connectivity and channel state information. With limited local information about the network, each node’s knowledge is mismatched, therefore they must make distributed decisions. We pose the following question - if every node has network state information only about a small neighborhood, then how and when should nodes choose to transmit? While scheduling answers the above question for point-to-point physical layers which assume interference must be avoided, we address this question assuming advanced physical-layer techniques that are able to manage interference beyond interference avoidance. We propose two constructive distributed algorithms which achieve rates higher than an interference-avoidance based link scheduling, especially if each node knows more than one hop of network state information.

  4. Investigating Cognitive Side Effects of GPi Deep Brain Stimulation for Parkinson's Disease
    Authors: Samantha R. Summerson, Caleb T. Kemere, and Behnaam Aazhang

    Abstract: The goal of this work is to characterize the side effects of electrical stimulation of the globus pallidus internus (GPi) as a treatment for Parkinson’s Disease (PD). This therapy is referred to as deep brain stimulation (DBS) and is used primarily to treat the motor symptoms of PD. Two significant side effects that are known to be associated with stimulation of the subthalamic nucleus (STN) are depression and impulsivity. While side effects of STN stimulation have been well investigated, the corresponding side effects for GPi stimulation have not. This work seeks to address the extent to which these are side effects of the alternative target, the GPi.
    Ten Long-Evans rats are used in this study to create a rodent model of PD. The rats are induced to be hemi-Parkinsonian via unilateral 6-OHDA lesions of the substantia nigra pars compacta (SNc) and are later treated with chronically implanted stereotrodes stimulating the entopeduncular nucleus (EP), which is the rat equivalent of the GPi. Sucrose preference and open field tests are used to assess anhedonia and depression. A lever-pressing reaction time task is used to test impulsivity. The rats’ performance is recorded prior to 6-OHDA lesions, after lesioning, and after DBS treatment. This work seeks to characterize levels of depression and impulsivity as a step towards addressing how the two possible DBS targets differentiate from each other.

  5. Characterizing the Self-Interference Channel for Full-duplex Baseband Cancellation
    Authors: Brett Kaufman, Jorma Lilleberg and Behnaam Aazhang

    Abstract: Full-duplex was once believed to be only a theoretical concept but has recently been demonstrated in wireless testbed experiments. In order to achieve the full potential of full-duplex, a large amount of self-interference cancellation is necessary. Our work focuses on baseband techniques as a final stage of cancellation before decoding. "

  6. Passive Self-Interference Suppression for Full-Duplex Wireless Communication
    Authors: Evan Everett, Achal Sahai, Ashu Sabharwal

    Abstract: Recent research results have demonstrated the feasibility of full-duplex wireless communication for short-range links. Although the focus of the previous works has been active cancellation of the self-interference signal, a majority of the overall self-interference suppression is often due to passive suppression, i.e., isolation of the transmit and receive antennas. We present a measurement-based study of the capabilities and limitations of three key mechanisms for passive self-interference suppression: directional isolation, absorptive shielding, and cross-polarization. The study demonstrates that more than 70 dB of passive suppression can be achieved in certain environments, but also establishes two results on the limitations of passive suppression: (1) environmental reflections limit the amount of passive suppression that can be achieved, and (2) passive suppression, in general, increases the frequency selectivity of the residual self-interference signal. These results suggest two design implications: (1) deployments of full-duplex infrastructure nodes should minimize near-antenna reflectors, and (2) active cancellation in concatenation with passive suppression should employ higher-order filters or per-subcarrier cancellation.

  7. Fundamental Limitations in Full-Duplex System Design
    Authors: Achaleshwar Sahai, Gaurav Patel, Chris Dick and Ashutosh Sabharwal

    Abstract: Recent experimental results have shown that full-duplex communication is possible for short-range communications. However, extending full-duplex to long-range communication remains a challenge, primarily due to residual self-interference even with a combination of passive suppression and active cancellation methods. In this poster, we investigate the root cause of performance bottlenecks in current full-duplex systems. The key bottleneck in current systems turns out to be the phase noise in the local oscillators in the transmit and receive chain of the full-duplex node. As a key by-product of our analysis, we propose signal models for wideband and MIMO full-duplex systems, capturing all the salient design parameters, and thus allowing future analytical development of advanced coding and signal design for full-duplex systems.

  8. Algorithms and FPGA Implementation of Data Detection in the Large-Scale MIMO Uplink
    Authors: Michael Wu, Bei Yin, Aida Vosoughi, Christoph Studer, and Joseph R. Cavallaro

    Abstract: The high processing complexity of data detection in multiple-input multiple-output (MIMO) systems having a large number of antennas at the base station (BS) necessitates efficint algorithms and high-throughput VLSI implementations. In this work, we propose a novel matrix inversion method suitable for data detection in systems having hundreds of antennas at the BS. We carry out the involved approximate matrix inversion using a small number of Neumann-series terms, which allows us to achieve near-optimal performance at low complexity. We propose a novel VLSI architecture to efficiently compute the approximate inverse using a systolic array and show reference FPGA implementation results for various system configurations. For a system where 128 BS antennas receive data from 8 single-antenna users, a single instance of our design processes 1.9 M matrices/s on a Xilinx Virtex-7 FPGA, while using less than 8% of the available resources.

  9. Parametrized Model Reduction
    Authors: Antonio Cosmin Ionita and Dr. Athanasios Antoulas

    Abstract: Dynamical systems are complex mathematical models used in a wide variety of fields, such as electrical circuits, vibration analysis of structures, simulation of fluid flows, molecular dynamics or weather prediction, to name just a few. Nowadays, there is an ever growing need for high accuracy of these mathematical models. This need leads to dynamical systems which are of large-scale; namely, they require huge amounts of data storage and take a very long time to simulate (usually in the order of days).
    Our research focuses on model reduction - finding simple, accurate models for large-scale dynamical systems. We extract reduced models that require minimal storage and have very fast simulation times (in the order of minutes), while still capturing the behavior of the large-scale system. Our approach uses rational interpolation and has a very desirable feature: it computes reduced models which match the frequency response of the large-scale system. We showcase our results through a series of plots illustrating diverse numerical examples such as finite element models resulting from discretization of partial differential equations.

  10. Detectors for Estimate-and-Forward Wireless Relay Networks
    Authors: Corina I. Ionita, Jorma Lilleberg(Renesas Mobile), Behnaam Aazhang

    Abstract: Relay radios have been introduced to improve the quality of the point-to-point wireless links with respect to achievable rates and symbol error rates. Under these metrics, the class of estimate-and-forward (EF) protocols has the best performance in a wide variety of cases. EF protocols are defined as relaying protocols in which the relay forwards an estimate of its received information. Optimal estimates have been identified, but an implementable detector at the receiver still remains to be found. In this paper, we derive an analytical form of the detector at the destination, by having the relay forward a piecewise linear approximation of the minimum mean square error (MMSE) estimate of the received symbol. We show that the symbol error rates of the proposed framework closely approximate the true performance of EF using a MMSE estimate of the received symbol at the relay for multiple types of modulations such as BPSK and 16QAM, in both the one-way and two-way relay channel topologies considered.

  11. Functional Connectivity based Segmentation of Epileptic Activity
    Authors: Rakesh Malladi, Behnaam Aazhang and Caleb Kemere

    Abstract: In this work, we develop a functional connectivity model for an epileptic brain using the ideas from probabilistic graphical models. This model will be generated using electroencephalographic (EEG) data from human patients. A patient-specific model is developed since there is high inter-patient variability in epileptic activity. This model will help in identifying the regions strongly connected to the seizure onset zone, which are the ideal locations for electrical stimulation.

  12. Routing and Power Allocation in Full-Duplex Networks
    Authors: David Ramirez and Behnaam Aazhang

    Abstract: We study a wireless full-duplex network with imperfect interference cancellation and solve the joint routing and power allocation problem. We use a simplified interference model that focuses on the effects of full-duplex by including residual self-interference and one hop interference while other interfering signals are assumed negligible. We first solve the optimal power allocation for a fixed route. We then propose a modification to Dijkstra's algorithm to find the joint route and power allocation to maximize throughput. Due to interference the algorithm proposed has a non decomposable objective function, however it is efficiently solved for every candidate route. We analyze the performance of our solution in a full interference model that does not neglect any interference. We derive a bound between the performance difference between both interference models. Through simulations we evaluate several scenarios and show the behavior of the solution in our model and then evaluate our solution in the full interference model.

  13. Energy-Proportional Imaging Sensing
    Authors: Robert LiKamWa, Steven Arroyo, Lin Zhong"

    Abstract: We see a new wave of innovations in Continuous Mobile Vision, based on capturing and processing a user’s visual experience in real-time. A major hurdle to this wave is the high power consumption of image sensing. We find that modern image sensors are not energy-proportional: Energy per pixel is inversely proportional to the spatiotemporal resolution, and thus fails to provide the opportunity to trade quality for energy efficiency.
    To solve this bottleneck, we use two energy-proportional mechanisms for current image sensors: (i) using an optimal clock frequency reduces the power by up to 50% or 30% for snapshot and video capturing, respectively; (ii) entering low-power standby mode between frames achieves almost constant energy per pixel for video, even at low frame rates. We also propose architectural modifications to the image sensor that would further improve the operational efficiency of image sensors. We use computer vision benchmarks to show the performance and efficiency tradeoffs that can be achieved with existing image sensors. For image registration, a key primitive for image mosaicking (e.g. panoramas) and depth estimation, an optimal clock frequency reduces power by 36% and aggressive standby mode reduces power by 95%. Our setup allows us to demonstrate the tradeoff in sacrificing spatiotemporal resolutions of video capture for reduced energy consumption."

  14. Making I/O Virtualization Easy with Device Files
    Authors: Ardalan Amiri Sani, Sreekumar Nair, Lin Zhong, Quinn Jacobson

    Abstract: We demonstrate devirtualization, a novel I/O virtualization technology that can easily virtualize various I/O devices such as GPU, input devices, camera, and audio devices, with interactive performance indistinguishable from that of native.

  15. Sulfur dioxide detecion using 7.4 μm DFB-QCL based cavity enhanced absorption spectroscopy
    Authors: J. Tarka, P. Stefański, R. Lewicki, S. So, M. Jahjah, and F. K. Tittel

    Abstract: A quantitative measurements of sulfur dioxide using CW, RT DFB-QCL based cavity-enhanced absorption spectroscopy (CEAS) and 2f wavelength modulation (WM) detection technique will be reported. Detection limits of 130 ppbv at 1σ were achieved with a 0.4-sec averaging time.

  16. A compact high power CW DFB-QCL based QEPAS sensor for sensitive detection of carbon monoxide
    Authors: P. Stefanski, R. Lewicki, J. Tarka, Y. Ma, M. Jahjah, F.K. Tittel

    Abstract: Development of an ultra-sensitive, selective, and compact QEPAS-based CO sensor platform employing a high power CW DFB-QCL will be reported. A minimum detectable concentration of 4 ppbv was achieved for the targeted 2169.2 cm-1 R(6) CO line using a 5 sec data acquisition time.

  17. A Software-Defined Radio Front-end for TV-Band Whitespace Research
    Authors: Ryan E. Guerra, Narendra Anand, and Dr. Edward Knightly

    Abstract: "Transmissions in TV UHF channels propagate further and through obstruction much better than similar ISM bands such as 2.4 and 5 GHz WiFi. Wireless spectrum sharing in the TV UHF band with frequency-agile radios promises to bring low-cost, efficient wireless internet service to communities and regions lacking last-mile broadband connectivity. In addition to broadband internet applications, such devices enable robust machine-to-machine communication and smart-grid data backhaul at a fraction of the cost of cellular systems.
    Until now, MAC and PHY research has been hindered by the lack of platforms that are both fully configurable and designed for real-time, full-power, long-distance operation in TV UHF channels. Today, we demonstrate a new radio front-end daughtercard developed at Rice University to enable up to 20 MHz, 1 Watt, full-duplex transmissions in UHF frequencies from 470-770 MHz, as well as 500 mW, full-duplex transmissions in ISM frequencies from 2400-2500 MHz.
    Built from the ground-up to be compatible with the WARP software-defined radio kit from Mango Comm as well as COTS FPGA development kits, this whitespace radio daughtercard enables new at-scale MIMO experiments in TV-whitespace frequencies with minimal RF programming and tuning required."

  18. Argos: Practical Many-Antenna Basestations
    Authors: Clayton Shepard, Lin Zhong

    Abstract: Many-antenna base stations promise to significantly improve the capacity of next-generation wireless networks, however these gains have only been demonstrated in theory and rudimentary prototypes. We will present our second generation base station, which is the first fully functional many-antenna implementation capable of real-time streaming wireless communication.

  19. Distributed Full-duplex via Wireless Side Channels
    Authors: Jingwen Bai, Ashutosh Sabharwal

    Abstract: We study a three-node full-duplex network, where the infrastructure node has simultaneous up- and downlink communication in the same frequency band with two half-duplex nodes. In addition to self-interference at the full-duplex infrastructure node, the three-node network has to contend with the inter-node interference between the two half-duplex nodes. The two forms of interferences differ in one important aspect that the self-interference is known at the interfered receiver. Therefore, we propose to leverage a wireless side-channel to manage the inter-node interference. We present one distributed full-duplex inter-node interference cancellation scheme called bin-and-cancel, which leverage the device-to-device wireless side-channel for improved interference cancellation. Bin-and-cancel is asymptotically optimal in high signal-to-noise ratio limit which uses Han-Kobayashi common-private message splitting and achieves within 1 bit/s/Hz of the capacity region for all values of channel parameters. We show analytically that leveraging the side channel can be highly beneficial in increasing the multiplexing gain of the system exactly in those regimes where inter-node interference has the highest impact.

  20. AsthmaGuru
    Authors: Rajoshi Biswas, Peter Chang, Hasitha Dharmasiri, Gaurav Patel, Ashutosh Sabharwal

    Abstract: One in 12 persons in the US suffer from Asthma. Although this disease cannot be completely cured, patients can manage and control their symptoms quite well using medication and spirometry. However, the reality of compliance in patients is far from ideal as 9 people die from asthma every day and the total costs amount to $56 billion per year. Our system is built to manage Asthma for the patients with minimal effort on their part. The low-power attachment brick built for the inhaler and spirometer wirelessly transmit usage, location, time stamp and lung function data to the smartphone through bluetooth. AsthmaGuru integrates information about air quality index (AQI) information through government websites, medication through inhaler and lung function through spirometry and records it in a smartphone application. This not only lets patients and doctors track their progress, control and manage Asthma better but also provides a metric that combines all the information to estimate the risk factor for each patient based on his/her compliance with Asthma.

  21. Joint Sparse Factor Analysis and Topic Modelingfor Learning and Content Analysis
    Authors: Andrew Lan, Andrew Waters, Christoph Studer and Richard Baraniuk

    Abstract: Personalized learning using machine learning algorithms is the key to achieving large scale personalized education catering to the different backgrounds of interests of learners. In this work, we propose the joint analysis of graded learner responses and question texts, in order to estimate i) Question--concept associations, ii) Learner concept knowledge and iii) the intrinsic difficulty of the questions, while also providing human-interpretable meaning to every estimated concept retrieved from the question text. We demonstrate the usefulness of our approach on real-world learner response and question text data. Our approach enable a personalized learning system to automatically generate human interpretable feedback to learners on their concept knowledge, and to course instructors on the underlying knowledge structure of a course/assessment.

  22. Damping of Ultra-Small Cobalt-HGNS
    Authors: Christyn A. Thibodeaux, Oara Neumann, Vikram Kulkarni, Wei-Shun Chang, Bruce Brinson, Chih-Wei Chen, Emilia Morosan, Stephan Link, Peter Nordlander, and Naomi J. Halas

    Abstract: We report the optical scattering profile of individual cobalt-HGNS using Total Internal Reflection Scattering (TIRS) Spectroscopy, which utilizes evanescent waves to excite nanostructures on a glass substrate. The signal to noise ratio is substantially increased rendering unprecedented scattering sensitivity for measuring ultra-small particles. The HGNS support localized surface plasmon resonances and contain cobalt impurities which are shown to dramatically broaden the plasmon line shape. This additional damping induced by cobalt atoms/ions can be utilized in the burgeoning field of hot electron generation.

  23. Parallel Interleaver Design for High Throughput Configurable Turbo Decoder
    Authors: Guohui Wang, Aida Vosoughi, Joseph R. Cavallaro

    Abstract: Parallel architecture is required for high throughput turbo decoder to meet the data rate requirements of the emerging wireless communication systems. However, due to the severe memory conflict problem caused by parallel architectures, the interleaver design has become a major challenge that limits the achievable throughput. Moreover, the high complexity of the interleaver algorithm makes the parallel interleaving address generation hardware very difficult to implement. In this paper, we propose a parallel interleaver architecture that can generate multiple interleaving addresses on-the-fly. We devised a novel scheduling scheme with which we can use more efficient buffer structures to eliminate memory contention. The synthesis results show that the proposed architecture with the new scheduling scheme can significantly reduce memory usage and hardware complexity. The proposed architecture also shows great flexibility and scalability compared to prior work.

  24. Massive Manipulation
    Authors: Aaron Becker, Golnaz Habibi, James McLurkin

    Abstract: Roboticists, biologists, and chemists are now producing large populations of simple robots. Controlling large populations of robots is still difficult, due to communication and onboard-computation constraints. Human control of large populations seems even more challenging.

    We investigate control of many mobile robots that move in a 2D workspace using three common system models. We focus on a model that uses broadcast control inputs specified in the global reference frame.

    This system model is uncontrollable, but we prove that adding a single obstacle makes the system controllable, for any number of robots. We provide a position control algorithm, but prove that the general form of this problem is hard.

    Nevertheless, we demonstrate through extensive testing with human subjects that many tasks are much easier using this model. Results are validated with extensive simulations and hardware experiments using hundreds of robots.

  25. Authentication of implantable medical devices using biometric measurements
    Authors: Masoud Rostami, Ari Juels, Farinaz Koushanfar

    Abstract: We present a system to authenticate external medical device controllers and programmers to Implantable Medical Devices (IMDs). IMDs often have built-in radio communication to facilitate non-invasive reprogramming and data readout. Many IMDs, though, lack well designed authentication protocols, exposing patients to the risks of over-the-air attack and physical harm. Our protocol uses biometric signals for authentication mechanism, ensuring access only by a medical instrument in physical contact with an IMD-bearing patient. Based on statistical analysis of real-world data, we propose and analyze new techniques for extracting time-varying randomness from biometric signals. We introduce a novel cryptographic device pairing protocol that uses this randomness to protect against attacks by active adversaries, while meeting the practical challenges of lightweight implementation and noise tolerance in biometric readings.

  26. Idetic: A High-level Synthesis Approach for Enabling Long Computations on Transiently-powered ASICs
    Authors: Azalia Mirhoseini, Ebrahim M. Songhori, and Farinaz Koushanfar

    Abstract: We develop Idetic, a set of mechanisms to enable long computations on ultra-low power Application Specific Integrated Circuits (ASICs) with energy harvesting sources. We address the power transiency and unpredictability problem by optimally inserting checkpoints. Idetic targets highlevel synthesis designs and automatically locates and embeds the checkpoints at the register-transfer level. We define an objective function that aims to find the checkpoints which incur minimum overhead and minimize recomputation energy cost. We develop and exploit a dynamic programming technique to solve the optimization problem. For real time operation, Idetic adaptively adjusts the checkpointing rate based on the available energy level in the system. Idetic is deployed and evaluated on cryptographic benchmark circuits. The test platform harvests RF power through an RFID-reader and stores the energy in a 3.3 F capacitor. For storage of checkpointed data, we evaluate and compare the effectiveness of various non-volatile memories including NAND Flash, PCM, and STTM. Extensive evaluations show that Idetic reliably enables execution of long computations under different source power patterns with low overhead. Our benchmark evaluations demonstrate that the area and energy overheads corresponding to the checkpoints are less than 5% and 11% respectively.

  27. Nanoscale Device Metrology Utilizing Scanning Tunneling Microscopy
    Authors: Corey Slavonic, Chaitra Rai, Kevin Kelly

    Abstract: We investigate the electronic properties of molecular machines, graphene, and other exotic material systems with atomic resolution using scanning tunneling microscopy (STM) and spectroscopy. Among these systems, we have studied the atomic structure of graphene on polycrystalline copper substrates. Using fast Fourier transform (FFT) we find the identities of the various metal facets and study their possible influence on defect formation in the graphene overlayer. Another system is composed of molecular machines called nanocars where we use the STM tip and variable surface temperatures to induce molecular motion. Our lab also uses STM as an electronic local probe to examine novel materials, such as topological insulators. Here, changes in the scattering of electrons can be seen in real-space and mapped out as a function of energy to reveal novel surface states. These projects represent part of an ongoing effort of our lab to use scanning probe techniques to discover the properties of nanoscale systems.

  28. Large Population of Robots: Theory and Practice
    Authors: Golnaz Habibi, SeoungKyou Lee, Aaron Becker, James McLurkin

    Abstract: Roboticists, biologists, and chemists are now producing large populations of simple robots. Multi-robot systems can perform many challenging applications, such as exploration, search-and-rescue, surveillance, and material transport, more effectively and efficiently than individual robots. Our research in Multi-Robot Systems Lab (MRSL) is focused on distributed algorithms for multi-robot systems. Our long-term goals are to understand the full potential of multi-robot systems in theory and in practice. We focus on transforming the underlying problem into one that is fundamentally different, and can be solved in an easier way, by using simpler robots. Here is a list of our current projects:
    1. Massive Multi-Robot Manipulation: Controlling large populations of robots is still difficult, due to communication and onboard-computation constraints. Human control of large populations seems even more challenging. We investigate control of mobile robots that move in a 2D workspace using three common system models. We focus on a model that uses broadcast control inputs specified in the global reference frame. This system model is uncontrollable, but we prove that adding a single obstacle makes the system controllable, for any number of robots.
    2. Cohesive Configuration Control: This algorithm produces three properties from a configuration of robots. It assumes some natural resting density, can be easily deformed into any shape by the environment, and has provable connectivity. Our approach builds upon algorithms for flocking, crystallization, boundary detection, broadcast tree navigation. It uses an estimate of the local vertex-cut to switch between two different motion controllers, essentially using the network topology as a distributed sensor.
    3. Max-Leaf Spanning Tree Clustering: The goal is to provide safe, efficient recovery of large populations of robots from complex environments. We use the robot’s local network geometry as a routing guide, but this can become disconnected in complex environments when all the robots are moving. We use a maximum-leaf spanning tree (MLST) for dynamic task allocation: internal robots remain stationary while leaf robots are mobile. This helps ensure connectivity during the recovery process and works well in practice – it is able to recover almost all of the robots, in contrast to a simple recovery algorithm that fails in all trials.
    4. Scale-Free Coordinates: We have proposed scale-free coordinates as an alternative coordinate system for multi-robot systems with large robot populations. Scale-free coordinates allow each robot to know, up to scaling, the relative position and orientation of other robots in the network. The proposed algorithm computes scale-free coordinates that is tailored to low-cost systems with limited communication bandwidth and sensor resolution.
    Our next project is to demonstrate a “grand challenge” experiment in Object Classification and Recovery that would integrate much of the above work. The goal is to field a large, heterogeneous system of small “scout” robots and larger “manipulator” robots to locate, classify, and recover objects in the environment.

  29. A Sound-to-Touch Sensory Substitution Prototype
    Authors: Scott Novich, David Eagleman

    Abstract: There are at least 2 million functionally deaf individuals in the United States alone and an estimated 53 million worldwide. The cochlear implant (CI) is an effective solution for regaining hearing capabilities for certain populations within this group, but not for all. First, CIs are expensive, ranging from $40,000 to $90,000 depending on the age of the recipient. This places CIs out of economic reach for many. Second, CIs require invasive surgery. Third, there is low efficacy of late CI implantation in early-onset deaf adults. Given this, there exists an important population of deaf individuals who would benefit from a hearing replacement that has low cost, does not involve an invasive procedure, and may have a higher efficacy for early-onset deaf adults. To address these problems, we are developing a low-cost, non-invasive, plasticity-based solution to deliver auditory information to the brain. Specifically, we are developing a ""vibratory vest"" by which auditory information is captured, digitally processed, and delivered to the skin of the torso using small vibratory motors. We call our device the Vibrotactile Acoustic Coder, or the ""VAC."" The term for such a technique is sensory substitution, and has previously proven successful in allowing those who are blind to have visual experience through the tongue or skin. In this demo and poster, we will present the current development status of the VAC prototype and preliminary results of several of our experiments that utilize it. Funding for this research is supported by a training fellowship from the Keck Center of Interdisciplinary Bioscience Training of the Gulf Coast Consortia (NIBIB Grant No. 5T32EB006350-05) and the Renz Foundation.

  30. Adaptive Resource Allocation in Tiered Storage Systems
    Authors: Hui Wang, Peter Varman

    Abstract: Increased consolidation in virtualized datacenters and public clouds has raised the importance of allocating shared server resources fairly among multiple tenants. In the storage domain, tiered storage made up of heterogeneous memory and storage devices are now the norm in high-end systems. In this paper we consider a two-tiered storage system made up of SSDs and hard disks (HDs), and address some of the challenges in achieving both high utilization and fair allocation. Our work is complementary to fairness versus efficiency tradeoffs studied in the context of sequential versus random IOs.

  31. A Single-Chip Dual-Mode CW/Pulse Electron Paramagnetic Resonance Spectrometer in 0.13μm SiGe BiCMOS
    Authors: Xuebei Yang, Payam Seifi, and Aydin Babakhani

    Abstract: Microwave circuitry for electron paramagnetic resonance (EPR) spectroscopy is implemented in a 0.13μm SiGe BiCMOS process. The chip can operate in both continuous wave (CW) and pulse modes. The frequency is tunable from 770MHz to 970MHz, corresponding to Zeeman magnetic fields from 28mT to 35mT for a free electron. The CW-EPR absorption line of a DPPH powder sample is acquired. The chip consists of a VCO, a PA, an LNA, a down-conversion mixer, baseband amplifiers, and a pulse generation block.

  32. A Silicon-Based, Fully Integrated Pulse Electron Paramagnetic Resonance System for mm-Wave Spectroscopy
    Authors: Charles Chen, Payam Seifi, Aydin Babakhani

    Abstract: An integrated transceiver for time-domain EPR spectroscopy is implemented using a 0.13μm SiGe BiCMOS technology. The system utilizes an on-chip resonator to study time domain relaxation behavior of paramagnetic samples, i.e. materials with unpaired electron spins. The single-chip EPR spectrometer consists of an EPR resonator, 22-26GHz tunable VCO, a programmable pulse generation block, RF buffer and power amplifier, a multi-stage LNA, and down-conversion mixer all in a 2mm2-size chip area. Index Terms — Electron paramagnetic resonance, mm-wave, silicon, fully integrated, spectroscopy.

  33. A Silicon-Based, Fully Integrated Pulse Electron Paramagnetic Resonance System for mm-Wave Spectroscopy
    Authors: Xuebei Yang and Aydin Babakhani

    Abstract: We demonstrate the first optical waveguide and fastest photodiode implemented in a commercial CMOS technology without performing any post-processing. The measured waveguide loss at 1.55μm is 37dB/cm and bandwidth of photodiode exceeds 9.2GHz at 850nm.


 Education Innovation
  1. Connexions and OpenStax College
    Authors: Richard Baraniuk, Daniel Williamson, Denver Greene

    Abstract: Connexions, an initiative of Rice University, is a dynamic digital educational ecosystem consisting of an educational content repository and a content management system optimized for the delivery of educational content. Connexions is one of the most popular open education sites in the world. Its more than 21,000 learning objects or modules in its repository and over 1200 collections (textbooks, journal articles, etc.) are used by millions of people per month. Its content services the educational needs of learners of all ages, in nearly every discipline, from math and science to history and English to psychology and sociology.

    OpenStax College is the peer-reviewed publishing arm of Connexions. OpenStax College's free textbooks are developed and peer-reviewed by educators to ensure they are readable, accurate, and meet the scope and sequence requirements of the most common high-enrollment courses. Through our partnerships with companies and foundations committed to reducing costs for students, OpenStax College is working to improve access to higher education for all.