Select a calendar:
Filter October Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for October
-
MIMO Broadcast Channels with Partial CSI: Throughput, Fairness, and Delay
Wed, Oct 06, 2004 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
SPEAKER: Mr. Masoud Sharif, CalTechABSTRACT: The downlink scheduling in a cellular system has to deal with two conflicting goals, namely maximizing the throughput yet at the same time maintaining fairness among the users and minimizing the delay. Information-theoretic results on broadcast channels yield schemes that only maximize the throughput, irrespective of the delay and fairness. In this talk, we first obtain the scaling laws of the sum-rate capacity (throughput) in terms of the number of receive/transmit antennas and the number of users n, under different channel state information (CSI) assumptions. It turns out that the sum-rate capacity heavily depends on the availability of CSI at the transmitter. In cellular systems, the number of users is typically large and obtaining full CSI from all users may not be practically feasible. Therefore, we propose a scheme that only requires partial CSI yet attains the same throughput scaling law as that of full CSI. We show that if the number of transmit antennas is large enough, our scheduling becomes fair, irrespective of the path-loss of users. We further show that the expected worst case delay when using throughput optimal scheduling is log n times worse than the minimum achievable delay. Finally, we look into the trade-offs between the throughput and the worst case delay.BIO: Masoud Sharif was born in 1977. He received the BS (with honors) and MS degrees in electrical engineering from Sharif University of Technology, in 1999 and 2001, respectively. He is currently working towards the Ph.D. degree in the Department of Electrical Engineering, California Institute of Technology, Pasadena, CA.HOST: Dr. Keith M. Chugg, x.07294, chugg@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - -248
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Massive Social Networks and Epidemiology
Wed, Oct 06, 2004 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Prof. Aravind Srinivasan, University of MarylandHost: Prof. David Kempe (CS)Abstract:Most mathematical models for the spread of disease use differential equations based on uniform mixing assumptions or ad hoc models for the contact process. We explore the use of dynamic bipartite graphs to model the physical contact patterns that result from movements of individuals between specific locations. The graphs are generated by large-scale individual-based urban traffic simulations built on actual census, land-use, and population-mobility data. We find that the contact network among people is a strongly connected small-world-like graph, and present provably-good algorithms and their empirical performance for outbreak detection by placing sensors. Within this large-scale simulation framework, we then analyze the relative merits of a number of proposed mitigation strategies for disease-spread.The talk will mostly be based on the following two papers, and will also briefly touch upon ongoing work:"Modelling Disease Outbreaks in Realistic Urban Social Networks", by S. Eubank, H. Guclu, V. S. A. Kumar, M. V. Marathe, A. Srinivasan, Z. Toroczkai and N. Wang. Nature, Vol. 429, 180-184, May 2004; and"Structural and Algorithmic Aspects of Massive Social Networks", by S. Eubank, V. S. A. Kumar, M. V. Marathe, A. Srinivasan, and N. Wang. Proc. ACM-SIAM Symposium on Discrete Algorithms (SODA), 711-720, 2004.
Location: Hedco Neurosciences Building (HNB) - 100 (Auditorium)
Audiences: Everyone Is Invited
Contact: Irina Strelnik
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Intrusion Detection in Ad hoc Networks
Mon, Oct 11, 2004 @ 02:30 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Saswati Sarkar, University of PennsylvaniaAbstract: We focus on detecting intrusions in ad hoc networks using misuse detection techniques. We allow for detection modules that have already failed or have been compromised by colluding intruders. Combining theories of hypothesis testing and approximation algorithms, we develop a framework to counter different threats while minimizing the resource consumption. We obtain computationally simple optimal decision rules to be followed by the nodes executing the intrusion detection software modules. But, we show that the selection of the optimal set of nodes for executing these modules is an NP-hard problem. We present polynomial complexity selection algorithms that attain guaranteeable approximation bounds. We also modify these algorithms for seamless operation in time-varying topologies. Using a combination of analysis and simulation, we identify the appropriate algorithms for different threat models, compromise levels, resource limitation and required detection rates.Bio: Saswati Sarkar received a Master of Engineering in Electrical Communication Engineering from the Indian Institute of Science in 1996 and a Ph.D. in Electrical and Computer Engineering from University of Maryland, College Park in 2000. She is currently an Assistant Professor in the department of Electrical and Systems Engineering in University of Pennsylvania. Her research interests are in resource allocation and performance analysis in communication networks. She received the Motorola gold medal for the best masters student in the division of electrical sciences at the Indian Institute of Science and a National Science Foundation (NSF) Faculty Early Career Development Award in 2003. She has been an associate editor of IEEE Transaction on Wireless Communications from 2001.Host: Dr. Urbashi Mitra, ubli@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Information Flow Decomposition for Network Coding
Tue, Oct 12, 2004 @ 03:15 PM - 04:15 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Christina Fragouli, School of Computer and Communication Sciences, EPFLAbstract: The famous min-cut, max-flow theorem states that a source node can send a commodity through a network to a sink node at the rate determined by the flow of the min-cut separating the source and the sink. Recently it has been shown that by linear re-encoding at nodes in communications networks, the min-cut rate can be also achieved in multicasting to several sinks. Constructing such coding schemes efficiently is the subject of current research.The main idea in this talk is a method to identify structural properties of multicast configurations, by decompositing the information flows into a minimal number of subtrees. This decomposition shows that very different networks are equivalent from the coding point of view, and offers a method to identify such equivalence classes. It also allows us to divide the network coding problem into two almost independent problems: one of graph theory and the other of classical channel coding theory. This approach to network coding enables us to derive tight bounds on the network code alphabet size, calculate the throughput improvement network coding can offer for different configurations. It also allows to develop algorithms to specify the coding operations at network nodes without the knowledge of the overall network topology. Such decentralized designs facilitate the construction of codes which can easily accommodate future changes in the network, e.g., addition of receivers and loss of links.This is joint work with Emina Soljanin.Bio: Christina Fragouli received her PhD from UCLA in Electrical Engineering in Fall 2000. Since then, she has worked at the Information Sciences Center at AT&T Labs (Florham Park, NJ) and at the National Capodistrian University of Athens, as a Research Associate. Currently she holds a postdoctoral position at the School of Computer and Communication Sciences at EPFL. She visited DIMACS (Rutgers University) and Bell Labs (Math. for Communication Dept., Murray Hill, NJ) in Spring 2003.Host: Dr. Urbashi Mitra, ubli@usc.edu
Location: Hedco Neurosciences Building (HNB) - -107
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
ARE THERE TURBO-CODES ON MARS
Wed, Oct 13, 2004 @ 03:00 PM - 04:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
PROF. BOB McELIECE (California Institute of Technology)Abstract:When the first close-up pictures of Mars were transmitted to Earth by NASA's Mariner 4 spacecraft in 1965, the data rate was less than 10 bits per second. By comparison, the current Mars Exploration Rover (Spirit and Opportunity) mission transmits images to earth at over 100,000 bits per second. What underlies this astonishing engineering improvement? Dr. Robert McEliece will argue that it is Newtonian physics and Shannon's information theory, in almost equal parts.Bio:Robert J. McEliece received B.S. and Ph.D. degrees in mathematics from the California Institute of Technology in 1964 and 1967, respectively. From 1963 to 1978, he worked at Caltech's Jet Propulsion Laboratory, and he has been a consultant at JPL since 1978. From 1978 to 1982, he was Professor of Mathematics and Research Professor at the Coordinated Science Laboratory, University of Illinois, Urbana-Champaign. Since 1982, he has been on the faculty at Caltech, where he is now the Allen E. Puckett Professor and Professor of Electrical Engineering. At JPL, Dr. McEliece and his students have contributed to the design and analysis of many coded interplanetary telecommunication systems, for example the Golay coded non-imaging system for the Voyager spacecraft, and the ``Big Viterbi Decoder'' which has been used on the Galileo, Mars Pathfinder, and Mars Exploration Rover Missions. Dr. McEliece is the author of three textbooks and more than 250 research articles, jointly with more than 100 coauthors. Among his research accomplishments are "McEliece's Theorem," on weight divisibility in cyclic codes, the "JPL Bound" (jointly with Rodemich, Rumsey, and Welch), which has been the world record-holder in a basic combinatorial problem of coding since 1977, the ``McEliece public-key cryptosystem,'' which has withstood the attacks of cryptanalysts for more than 25 years, and "repeat-accumulate codes" (jointly with Dariush Divsalar and Hui Jin), which bridge the gap between turbo-codes and LDPC codes. Dr. McEliece is a member of the American Mathematical Society, a Fellow of the IEEE and a member of the National Academy of Engineering. His Erdos number is One.Host: Prof. Solomon Golomb, x07333 ***A reception will follow the seminar at 4:00p.m.
Location: Ethel Percy Andrus Gerontology Center (GER) - ontology Auditorium (GER-124)
Audiences: Everyone Is Invited
Contact: Rosine Sarafian
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Boundary Control for Parabolic Partial Differential Equations and Turbulent Flows
Mon, Oct 18, 2004 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Miroslav KrsticUniversity of California, San DiegoAbstract: Classical methods for stabilization of PDEs require numerical solutions of operator Riccati equations. A new set of tools, based on infinite dimensional extensions of backstepping, and accessible to engineers with background not much beyond calculus, yields explicit formulae for boundary controllers for some classes of parabolic problems of practical interest. They employ integral transformations and boundary feedback to transform unstable PDEs into basic heat equations. This idea is similar to feedback linearization, where the starting system is nonlinear and unstable, whereas the target system is linear and stable. The new tools are particularly inspired by applications to fluid flows, either for turbulence suppression or for enhancement of fluid mixing in problems like combustion, and are implementable with MEMS/synthetic jet actuators and sensors. I will present designs for channels, pipes, jets, and bluff bodies, motivated by geometries in aerospace applications. Miroslav Krstic is Professor of Mechanical and Aerospace Egineering at University of California, San Diego.. He got his PhD in 1994 at University of California at Santa Barbara. Krstic is an IEEE Fellow and has received the NSF Career Award, the ONR Young Investigator Award, the Presidential PECASE Award, the George S. Axelby Outstanding Paper Award of IEEE Transactions on Automatic Control, and the O. Hugo Schuck Award for the best paper at American Control Conference. Krstic is a co-author of the books Nonlinear and Adaptive Control Design (Wiley, 1995), Stabilization of Nonlinear Uncertain Systems (Springer-Verlag, 1998), Flow Control by Feedback (Springer-Verlag, 2002), and Real Time Optimization by Extremum Seeking Control (Wiley, 2003). He has served as Associate Editor for the IEEE Transactions on Automatic Control, International Journal of Adaptive Control and Signal Processing, and Systems and Control Letters. Krstic is a Vice President for Technical Activities and a member of the Board of Governors of the IEEE Control Systems Society. Host: Petros Ioannou; ioannou@usc.edu
Location: Hedco Petroleum and Chemical Engineering Building (HED) - CO AUDITORIUM
Audiences: Everyone Is Invited
Contact: Irina Strelnik
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
CONTROL OF NONLINEAR DISTRIBUTED PROCESS SYSTEMS
Tue, Oct 19, 2004 @ 10:30 AM - 11:30 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Panagiotis D. ChristofidesDepartment of Chemical Engineering
University of California, Los AngelesABSTRACT:While most undergraduate control courses still deal with dynamics and control of lumped parameter systems using linear transfer function models, key technological needs in growth areas such as semiconductor manufacturing, nanotechnology and biotechnology have motivated extensive research on analysis and control of complex nonlinear distributed processes. Examples include the chemical vapor deposition of thin films for semiconductor manufacturing, as well as the crystallization of proteins and the aerosol-based production of nanoparticles. From a control point of view, the distinguishing feature of complex distributed processes is that they give rise to nonlinear control problems that involve the regulation of highly distributed control variables by using spatially-distributed control actuators and measurement sensors. Thus, complex distributed processes cannot be effectively controlled with control methods which assume that the state, manipulated and to-be-controlled variables exhibit lumped behavior or with linear control algorithms derived on the basis of linear/linearized distributed models.We have pioneered the development of a general and practical framework for the synthesis of practically implementable nonlinear feedback controllers for complex distributed processes based on fundamental models that accurately predict their behavior. The key difficulty in developing model-based control methods for distributed processes lies in the ``infinite-dimensional'' nature of the process models, which prohibits their direct use for control system synthesis. We have developed nonlinear order reduction techniques for deriving low-dimensional approximations that accurately reproduce the dynamics and solutions of distributed process models. We have used these approximate models for the synthesis of nonlinear feedback controllers via geometric methods and Lyapunov techniques. The controllers can be readily implemented in practice and enforce the desired control objectives in the infinite-dimensional closed-loop system. In addition to our order reduction and control work, we will present recent results on state estimation and control using multiscale (deterministic/stochastic) distributed models. We will present applications of the theoretical results to: a) temperature profile control in rapid thermal chemical vapor deposition and crystal growth, b) control of microstructure and composition in thin film growth including an experimental implementation, and c) control of size distribution in pharmaceutical crystallization. BIOGRAPHICAL SKETCH:Panagiotis D. Christofides was born in Athens, Greece, in 1970. He received the Diploma in Chemical Engineering degree in 1992, from the University of Patras, Greece, the M.S. degrees in Electrical Engineering and Mathematics in 1995 and 1996, respectively, and the Ph.D. degree in Chemical Engineering in 1996, all from the University of Minnesota.Since July 1996 he has been with the Department of Chemical Engineering at the University of California, Los Angeles, where he is currently Associate Professor. His theoretical research interests include nonlinear control, singular perturbations, and analysis and control of distributed parameter systems, multiscale systems and hybrid systems, with applications to advanced materials processing, particulate processes, biological systems and fluid flows. His research work has resulted in a large number of articles in leading scientific journals and conferences and two books entitled "Nonlinear and Robust Control of Partial Differential Equation Systems: Methods and Applications to Transport-Reaction Processes," Birkhäuser, Boston, 2001 and "Model-Based Control of Particulate Processes,'' Kluwer Academic Publishers, 2002. He has also co-authored the forthcoming book "Control of Nonlinear and Hybrid Process Systems: Designs for Uncertainty, Constraints and Time-Delays", Springer, 2005. A description of his research interests and a list of his publications can be found at http://www.chemeng.ucla.edu/pchristo/index.html.Professor Christofides has been a member of the Control Systems Society Conference Editorial Board and is an Associate Editor of "IEEE Transactions on Automatic Control", the 2004 Program Coordinator of the Applied Mathematics and Numerical Analysis Area of AIChE and the Program Vice-Chair for Invited Sessions for the 2004 American Control Conference.Professor Christofides has received the Teaching Award from the AIChE Student Chapter of UCLA in 1997, a Research Initiation Award from the ACS-Petroleum Research Fund in 1998, a CAREER award from the National Science Foundation in 1998, the Ted Peterson Student Paper Award from the Computing and Systems Technology Division of AIChE in 1999 and a Young Investigator Award from the Office of Naval Research in 2001. He has also received twice the O. Hugo Schuck Best Paper Award in 2000 (with Antonios Armaou) and 2004 (with D. Ni, Y. Lou, L. Sha, S. Lao and J. P. Chang), and the Donald P. Eckman Award in 2004, all from the American Automatic Control Council.Host: Petros Ioannou; ioannou@usc.eduLocation: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Irina Strelnik
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
Word-level Statistics Based High Level Crosstalk Estimation
Thu, Oct 21, 2004 @ 10:45 AM - 11:45 AM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Dr. Srinivas Katkoori
Associate Professor
Computer Science & Engineering Department, University of South Florida
katkoori@csee.usf.edu
http://vcapp.csee.usf.edu/~katkooriABSTRACT: Crosstalk estimation and optimization are critical issues in today's deep-sub-micron designs. We propose two word-level statistical techniques to estimate the probability of crosstalk events on the signal lines of a system bus. Given the word-level statistical parameters, namely mean, standard deviation, and lag-one temporal correlation coefficient, we analytically estimate the bit-level crosstalk probability. To reduce the complexity and efficiently scale the estimation technique for large bus-widths, we modify the first technique by using a procedure that maps disjoint values in a distribution to continuous values in a modified distribution. Experimental results for data streams from different data environments, compared against detailed HSPICE simulations, are presented. The statistical estimators yield average errors less than 7% and 12% respectively for bus-widths ranging from 8bits to 32bits. Compared to HSPICE, the execution times are reduced by factors of over 10x for the first technique and over two orders of magnitude for the second technique. The statistical approaches are shown to be compatible with existing bus re-ordering techniques. Ongoing work uses the proposed technique to estimate crosstalk between buses at the layout level and optimize crosstalk during high-level synthesis of designs. BIO: Dr. Srinivas Katkoori received his doctoral degree in computer engineering from the University of Cincinnati, Ohio, in 1998. He joined the CSE department at USF as an Assistant Professor in August of1997. Since 2004, he is working as an Associate Professor with tenure. His research interests are in the general area of VLSI CAD Algorithms and Design Methodologies. Specific research areas include: High level synthesis, Low power synthesis, VLSI CAD for DSM regime, and Radiation Hardness by Design. Dr. Katkoori is a recipient of 2001 National Science Foundation (NSF) CAREER award (Design Automation program) for a five-year duration on the topic "Interconnect-centric High Level Synthesis in DSM regime." Besides NSF, his research is being funded by Honeywell Inc., Clearwater (under their Academic Research Initiative program) and by the I4 High Tech Corridor Initiative Program. Currently, Dr. Katkoori's research group consists of 9 PhD students and several MS students. In 2003, he has received the USF Outstanding Faculty Research Award. Dr. Katkoori is serving as the faculty advisor of the student chapter of IEEE Computer Society. He is a member of ACM and IEEE professional societies. He served as the publicity and registration chair (2000) and as the general co-chair and registration chair (2001) for the Annual Workshop on VLSI held in Orlando, Florida. Since October 2003, he is serving as the editor of the Suncoast Signal, the newsletter of the IEEE Florida West Coast Section (FWCS). He is serving on the Technical Committees of 2004 ISVLSI and 2004 MAPLD. Host: Prof. Sandeep Gupta, sandeep@poisson.usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Lisa Inomata-O'Connell
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
A Single Chip, Low Power, Impulse-UWB Transceiver for Low-Rate, Indoor, Wireless Systems
Thu, Oct 21, 2004 @ 02:30 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Mr. Ian O'Donnell, UC BerkeleyAbstract: This talk describes the system architecture and circuit design constraints for a single-chip, impulse-based Ultra-Wideband transceiver. Targeting a sensor network application, the radio supports ranging and communication over a distance of 10m with a 1mW power budget at 100kb/s. Using simple digital switches to generate a narrow pulse, energy is transmitted in the low frequency band (DC-960MHz) allocated by the FCC for imaging systems and surveillance devices. Reception, after gain and filtering, occurs in a bank of 1-bit A/D converters that capture the pulse in an adjustable window of 16 to 64ns, sampled at 2 GSamples/s. This data is passed to the digital backend which performs matched filtering, acquisition, synchronization, and demodulation. The implementation issues of this system and simulation results will be discussed.Bio: Ian D. O'Donnell received the B.S. and M.S. degrees in Electrical Engineering and Computer Science from the University of California at Berkeley in 1993 and 1996 respectively. His master's topic was in the area of digital, low-power, CMOS circuit design for a wireless LAN receiver as part of the InfoPad project. From 1996 to 1999 he worked at Silicon Graphics, Inc. as a digital ASIC designer, and in 1999 he joined NVIDIA, Inc. where he worked on high-speed serial design. In 1998 he returned to Berkeley, joining the Berkeley Wireless Research Center to work in the area of low-power, integrated, picocellular radios. His Ph.D. research focus is the design and implementation of an impulse-based, low-power Ultra-Wideband transceiver in 0.13 micron CMOS suitable for sensor network applications.Host: Dr. Keith Chugg, chugg@usc.edu
Location: Hedco Neurosciences Building (HNB) - 107
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
