Stanford IEEE Student Chapter Events 2011-2012

Follow us on Twitter and/or Facebook

Connecting People to Products using Big Fast and Heterogenous Data

Please join us for a talk by Abhishek Gattani, Senior Director/Distinguished Architect from Walmart Labs.
Abstract: With over 200 million people walking into stores each week, 43+ million unique visitors per month online, and access to the full twitter fire-hose (i.e. 200 million tweets per day), we at WalmartLabs are processing large amounts of data to help connect people to products and also provide unexpected insights to merchants. How do we turn this data into insights? How do we combine data internal to an enterprise with social data such as tweets and data from the "traditional" Web such as Wikipedia? When is social and web data useful? These are key questions when building any Big Data system. In this talk I will discuss our work in this area and give examples of where social and web data was used to solve some key challenges faced when building the search engine for the second largest eCommerce retailer in the world.
Presentation will start at 6:00 PM, pizza served afterwards

Feedback Communication Systems: Fundamental Limits and Control-Theoretic Approach

Please join us for a talk by Ehsan Ardestanizadeh from ASSIA, Inc.
Abstract: We combine techniques from information theory, estimation and control, and optimization theory to investigate the benefits of feedback in improving fundamental limits on information flow in communication networks. We focus on three network models. First, combining the Lagrange duality technique and tools from information theory including the dependence balance bound, we derive an upper bound on the sum rate achievable by linear codes for the k-sender Gaussian multiple access channel with feedback. This upper bound is further shown to coincide with the known lower bound by Kramer, hence establishing the linear sum capacity. Next, we study the application of tools from classic linear quadratic Gaussian (LQG) control in designing codes for feedback communications. For the k-receiver Gaussian broadcast channel with feedback, we construct a linear code based on a mapping to the LQG control problem, which achieves the best known lower bound on the sum rate. In addition, depending on the correlation among the receivers' noises, it is shown that in the high signal-to-noise ratio regime this code can achieve k degrees of freedom, i.e., the sum rate increases linearly with the number of receivers. Finally, we consider the wiretap channel with rate-limited feedback and discuss the benefits of feedback in improving the secrecy capacity.
Food will be available at 5:30 PM in front of the room 202. Presentation will start at 6:00 PM.

Feedback Communication Systems: Fundamental Limits and Control-Theoretic Approach

Please join us for a talk by Ehsan Ardestanizadeh from ASSIA, Inc.
Abstract: We combine techniques from information theory, estimation and control, and optimization theory to investigate the benefits of feedback in improving fundamental limits on information flow in communication networks. We focus on three network models. First, combining the Lagrange duality technique and tools from information theory including the dependence balance bound, we derive an upper bound on the sum rate achievable by linear codes for the k-sender Gaussian multiple access channel with feedback. This upper bound is further shown to coincide with the known lower bound by Kramer, hence establishing the linear sum capacity. Next, we study the application of tools from classic linear quadratic Gaussian (LQG) control in designing codes for feedback communications. For the k-receiver Gaussian broadcast channel with feedback, we construct a linear code based on a mapping to the LQG control problem, which achieves the best known lower bound on the sum rate. In addition, depending on the correlation among the receivers' noises, it is shown that in the high signal-to-noise ratio regime this code can achieve k degrees of freedom, i.e., the sum rate increases linearly with the number of receivers. Finally, we consider the wiretap channel with rate-limited feedback and discuss the benefits of feedback in improving the secrecy capacity.
Food will be available at 5:30 PM in front of the room 202. Presentation will start at 6:00 PM.

From Embedded DSP to Embedded AI: towards low-power biomedical sensors with intelligence

Please join us for a talk by Professor Naveen Verma from Princeton.
Abstract: By 'intelligent biomedical devices', I mean systems that can interpret physiologically-complex patient signals. These can take advantage of the unprecedented sensing and stimulations modalities that have recently emerged, making possible closed-loop implants that have the potential for very high clinical impact. Or, they can take advantage of advanced networking technologies to enable distributed clinical decision support, allowing healthcare to scale to extremely large patient populations outside a hospital (a possibility we are particularly focused on). The ability to accurately detect specific physiological processes of clinical importance, however, poses some notable challenges. The physiology that is expressed in the patient signals that are available through chronic sensors is extremely complex and, as seen across clinical applications, highly variable both from patient-to-patient and over time. Modeling the signal correlations thus becomes extremely difficult, especially under the tight energy constraints (i.e., 1-10mW for wearable devices, 10-100 microW for implantable devices). Data-driven modeling, particularly through the use of machine learning, offers some powerful benefits. In this talk, I describe how these algorithmic frameworks can be exploited at multiple levels in a low-power sensing system. The benefits range from the hardware platform itself up to the applications. Clinical applications (from the domain of cardiac and neurological monitoring) are used to illustrate the range of challenges posed by the signal models involved. I discuss hardware-level approaches that take advantage of algorithmic structure as well as opportunities to reformulate the kernel functions for substantial energy reduction. The embedded inference capabilities that result lead to new algorithmic possibilities at the application level but also new approaches for hardware resilience at the platform level.
Biography:
Naveen Verma received the B.A.Sc. degree in Electrical and Computer Engineering from the University of British Columbia, Vancouver, Canada in 2003 and the M.S. and Ph.D. degrees in Electrical Engineering from Massachusetts Institute of Technology in 2005 and 2009 respectively. Since July 2009 he has been an Assistant Professor of Electrical Engineering at Princeton University. His research focuses on ultra-low-power integrated circuits including low-voltage digital logic and SRAMs, low-noise analog instrumentation and data-conversion, and low-energy processing algorithms especially for biomedical applications. He is co-recipient of 2008 ISSCC Jack Kilby Award for Outstanding Student Paper, and 2006 DAC/ISSCC Student Design Contest Award.

Groupon Tech Talk

Please join us for a technical talk by Groupon

Lunch with CS/EE Women Faculty

Engineering women: Please join  for a lunch with women faculty in EE/CS departments on Thursday Nov 10. Prof. Fei Fei Li, Prof. Audrey Ellerbee, Prof. Monica Lam, Prof. Julie Zelenski and Prof. Ada Poon will be joining for this lunch event. Hear them talk about their backgrounds and their exciting path to Stanford. Ask them questions in an informal setting.

Talk on Body Area Networks (IEEE 802.15.6)

Presenter: Arthur Astrin PhD, Astrin Radio
Body Area Networks (BAN) devices operate in close vicinity to, on, or inside body and can enable a wide range of applications, including medical support, healthcare monitoring and consumer electronics with increased convenience or comfort. Due to strong demands of medical, healthcare and information technology industries, IEEE was requested to standardize the Body Area Network. IEEE 802.15 task group 6 (TG6) was set up to develop an IEEE international standard for BAN in January 2008. This talk reviews major issues, history and current status of TG6. Early on, the TG6 invited representatives from industry to present applications which require body area networks. We then developed an application matrix, and summarized it into a single document, which was issued to proposers. The proposers were asked to propose a communication protocol that would accommodate this application summary document. The other issue facing TG6 was to have a detail understanding of available spectrum for BANs. And finally we needed at an accurate model of the channel; in this case the human body. This channel is much more difficult than free space/air to measure and to correctly model. The draft of the standard is in Sponsor Ballot stage and is being worked on by the team. We expect comments resolution in the next IEEE 802 meeting and completion of the standard in 2012.
Food will be available at 5:30 PM in front of the room 202. Presentation will start at 6:00 PM.

From millibits to Terabits per Second and Beyond: Over 60 years of Innovation

As described in Moore's Law, circuit complexity (e.g. in transistors per square cm) has been doubling every 24 months. Communication bandwidth has doubled every 18 months. This meteoric increase was enabled by three key developments: (a) the demonstration of the point-contact bipolar transistor in 1947 by Bardeen, Brattain and Shockley, which started the solid-state revolution, (b) the development of Information Theory as enunciated by Claude Shannon in 1948, and (c) the invention of laser by Schawlow and Townes in 1958. Serious efforts to develop high-performance lightwave systems incorporating integrated electronics with MOS technology began around 1980. However, initial attempts at boosting receiver sensitivity and data-rates was seriously hampered by a lack of understanding of the noise performance of the MOS device. Prof. Jindal's contributions in this area not only led to a deeper understanding of the noise behavior of MOS devices but also produced an order of magnitude improvement in their performance. This set the stage for MOS to become the technology of choice for lightwave and now low-cost wireless terminal applications. The ubiquitous nature of cell phones is a testimony to these key developments in the early 80?s. In this talk, starting from smoke signals at millibits per second, we will trace these events from a historical perspective to see how these key technologies led to the development of modern wireless and optical networks of terabit capacity, with petabits looming in sight.

BIOGRAPHIC SKETCH
Professor Renuka P. Jindal,
President, IEEE Electron Devices Society
IEEE Electron Devices Society Distinguished Lecturer
William and Mary Hansen Hall Chair of Electrical and Computer Engineering
University of Louisiana at Lafayette

IEEE and You: The EDS Connection

Pizza lunch follows in the CIS Courtyard.

In this discussion session with students and associated faculty, Prof. Jindal will discuss the benefits of a professional involvement that starts at school and carries throughout an entire professional career. Associated faculty and student members of IEEE will share their perspectives into the benefits of such professional engagement and networking.

Big Sister/Little Sister Mentorship Event

First Big/Little Sister mentorship program event this school year with a photo scavenger hunt & dinner! The event supported by Women in Electrical Engineering (WEE)and IEEE Women in Engineering (WiE). In addition to our big sisters who are current graduate students, we also have big sisters from industry - Cisco, CoolIris, Lookheed Martin and someone earlier in US Navy.
Photo Scavenger Hunt at 5.30 pm and Dinner at 7pm at Packard 036.

Twitter Tech Talk: "Saying more with less"

At Twitter, we are building a novel advertising platform that gives advertisers the power to shape conversations. Come learn about how our platform shows relevant ads and maximizes engagement, all while scaling to billions of requests. Walk away with a basic understanding of internet advertising and how it involves solving challenging problems in almost every computer science discipline ranging from distributed systems and machine learning to algorithms, analytics, UX design, and visualization.
Also bring your resumes, as we'll be raffling off t-shirts at the end of the Tech Talk.

Welcome Picnic for Women Engineers

This year Stanford IEEE student chapter founded a Women in Engineering (WiE) chapter. IEEE WiE group and Women in Electrical Engineering group like to welcome new and returning EE women with a picnic dinner on Monday October 10th.
You do not want to miss this!! There are amazon gift cards as gift prizes....Come and meet your fellow women engineers...

Ethics in Engineering-Beyond Good Intentions

Join us for a discussion on ethics and integrity in engineering led by Rick Baily, Vice President of Engineering & Mission Assurance for Boeing Defense, Space & Security, and John Iwasaki, Director of Ethics and Business Conduct for Engineering, Operations & Technology/Shared Services Group.

Co-hosted by AIAA & IEEE. Food will be provided.

IEEE BBQ Night

Our first event this year is our quarterly barbecue on Oct 6 (IEEE day) from 6-8 pm. Invite your friends, especially those new to Stanford. We have raffle prizes like Amazon Kindle, ipods, and gift cards in the BBQ Night Please RSVP below to help us gauge attendance. If you don't, you're still welcome to come. If you have trouble viewing or submitting this form, you can fill it out online: https://docs.google.com/spreadsheet/viewform?hl=en_US&formkey=dHZzbE5uSjMyWjFKaERkdU9hdkdGWXc6MQ#gid=0

The Role of Information Theory in Public Key Cryptography

Presenter: Martin E. Hellman, Professor Emeritus of Electrical Engineering, Stanford University
Abstract: People are often surprised that public key cryptography was invented within Stanford's EE department, not Computer Science. Yet my background in Information Theory played a key role in that advance. This talk reviews my professional odyssey, from research in more traditional information theory to cryptography to estimating the odds that civilization will survive another century. The unifying theme is the role that my training in information theory played in guiding that journey.

Mountains of Data: A Hikers Guide

Sir Edmund Hillary said, "It is not the mountain we conquer, but ourselves". Today, networking and storage technology presents us with incredible capabilities. However, the data in isolation, like the mountain, doesn't do much. Its true value emerges in the context of planning, access, and distribution. It's been said that humans are creating an exabyte of new data every day. Moving it, storing it, and making good use of it is becoming a challenge. Tom Coughlin, Mark Day, and Gordon Brebner discuss their perspectives for handling the data explosion.

New Frontiers in Computing 2011: Emerging Automotive Computing: Engineering in Overdrive

On board computers are now the major cost of any new car. This year's NFIC will explore where innovations in automotive computing will go in the next generation. Starting with an introduction and overview of the expanding field of automotive computing, this conference will examine exciting new developments taking place in both university and corporate research facilities. Key to much of this innovation, both theoretical and practical, the conference will cover the development of computer based autonomous vehicles. Technical presentations on inter-vehicle communication and hardware/software design challenges will follow this discussion of growing vehicular computerization and autonomy. Finally, the technical issues of Universal and Human computer Interface in this changing automotive environment will receive attention. This July 16th at Stanford University's Braun Auditorium, NFIC will over the challenges of developing the next generation auto that will push computing and engineering into technical overdrive.
Registration and more information at http://www.nfic-us.org/
Speakers : Dr. Steven Shladover, UC Berkeley; Marcial Hernandez, Volkswagen; Paolo Giusto, General Motors; Fuliang Weng, Robert Bosch LLC; Rama Vuyyuru, Toyota; Dr. Christopher Pribe, OneHandRunning
The NFIC conference is a collaboration between Stanford IEEE, Santa Clara Valley IEEE Computer Society, and NATEA Silicon Valley. We look forward to seeing you at the conference!