News

ESE Seminar Series: Fall 2008

"Fairness and Load Balancing In Wireless LANs"

Yigal Bejerano Dr. Yigal Bejerano - Bell Labs
Wednesday, October 15, 2008

2:00pm - 3:00pm

337 Towne Building

Abstract:

In recent years, IEEE 802.11 wireless LANs (WLANs) have been rapidly deployed all over the world, in enterprises, public areas and residential environment. Recent studies on operational WLANs have shown that the traffic load is often unevenly distributed among the access points (APs). In WLANs, by default, a user scans all available channels and associates itself with an AP that has the strongest received signal strength, while being oblivious to the load of the APs. As users are, typically, not evenly distributed, some APs tend to suffer from heavy load, while their adjacent APs may carry only light load. Such load imbalance among APs is undesirable as it hampers the network from fully utilizing the network capacity and providing fair services to the users. In this talk, I will address the challenge of providing fair service as well as high throughput in large WLANs without changing the standard. I will show that this objective can be greatly achieved by balancing the load on the APs. To this end, I will present two efficient approaches for achieving such load balancing. The first approach, termed association control, is based on intelligent control of the user-AP associations. To this NP-hard problem I will present a 2-approximation algorithm. The second approach, termed cell-breathing, balances the load on the APs by controlling the size of their cells. The latter is obtained by tuning the transmission power of APs' beacon messages. To this power control problem I will describe the optimal solution. I will conclude with a performance comparison of the two approaches.

 

 

"Superconductivity and Biomagnetism"

Harold WeinstockDr. Harold Weinstock
Friday, October 3, 2008
1:30pm - 2:30pm
Heilmeier Hall - Towne Building

Abstract: 

This presentation will review recent advances in Magnetic Resonance Imaging (MRI). Some advances are related to the use of higher magnetic fields, i.e., increasing the standard field strength from 1.5T to 3.0T. Others involve using non-superconducting magnets at lower fields, but with magnetization of the patient measured by a superconducting quantum interference device (SQUID) gradiometer. Yet another improvement to some MRI measurements is the utilization of high-temperature superconductive (HTS) pick-up coils operating at about liquid nitrogen temperature. The major part of this seminar will be devoted to biomedical SQUID applications, with SQUID gradiometers being used to detect magnetic signals in the brain, the liver and the heart. Magnetic anomalies in the brain can be triggered by auditory, visual or mechanical stimuli or by electrical malfunctions such as uncorrelated noise due to epileptic events. Another application that should replace surgery with a non-invasive SQUID measurement, relates to excess iron in the liver, symptomatic of potentially fatal tumors. By far, the most promising SQUID biomagnetic application relates to monitoring magnetic signals produced by surface currents in the heart. Dynamic SQUID-generated magnetic field maps provide the most precise non-invasive determination of ischemic heart disease in comparison to all other non-invasive, non-contact methodologies. These are measurements taken without the need for generating stress, and generally they can be made in an ordinary hospital environment without the need for ambient magnetic field shielding. With a full array of 6 x 6 SQUID axial gradiometers, the entire measurement takes no more than 90 seconds, with a PC-driven analysis available a few seconds later.

 

NEW FACULTY

Rahul Mangharam, Assistant Professor of Electrical and Systems Engineering, PhD in Electrical & Computer Engineering (2007), Carnegie Mellon University. Dr. Mangharam's interests are in sensing, communication and actuation between the physical and virtual worlds.  His focus includes scheduling algorithms for wireless and embedded systems for time-critical and safety-critical applications
« read more »

SWARMS on CNET

SWARMCNET news blog features Scalable sWarms of Autonomous Robots and Mobile Sensors (SWARMS) project.
« read article »

 

Engheta

Engheta. EE TimesProf. Nader Engheta has been featured in the 2/27/08 edition of EE Times for his work in nanotechnology which explores circuit theory in a new regime in which 'current' is "...no longer defined as the movement of electrons and holes, but instead as an electromagnetic wave".
« read article »

 

Prof. Nader Engheta was selected as the 2008 recipient of the George H. Heilmeier Faculty Award for Excellence in Research. This prestigious award recognizes Prof. Engheta's extraordinary research career and his leadership in technical innovation and public service.
« read article »

MURI Award

Prof. George Pappas has lead a multi-disciplinary research team to a MURI Award.  The team includes faculty representatives from Georgia Institute of Technology, University of CA (Berkeley), Arizona State University, University of Washington, Michigan Technological University, Yale University, and M.I.T. The highly-competitive MURI program, sponsored by the Air Force Office of Scientific Research (AFOSR) supports multi-disciplinary research in areas of DoD relevance.  The team received the award for the MURI topic:  Biologically-Inspired Approaches for Team and Coalition Adaptation of Heterogeneous Unmanned Systems for Surveillance over Large and Complex Areas.  
Click here to read Penn's media release. Additional coverage of this project may be found here.

Hugo Schuck Award

Prof. Ali Jadbabaie and his student, Nader Motee, have won the Hugo Schuck Award, given to the best overall American Control Conference paper of the previous year.  Their paper, « read paper » , introduces a new formalism to study optimal control of heterogeneous spatially distributed systems.

New robot-oriented undergraduate classes announced

EduBotThe six-legged mobile robot, EduBot, will be used as the working platform for the new ESE classes:  ESE 313 and ESE 112 are the new robot-oriented undergraduate classes.

 

DARPA Urban Challenge 2007

Little Ben completed the DARPA Urban Challenge 2007 course in 6 hours!!!  Penn's Autonomous Car "Little Ben" Advances to the Finals of the DARPA Urban Challenge -- Under the team leadership of Professor Daniel D. Lee, "Little Ben", the autonomous vehicle engineered by the University of Pennsylvania and Lehigh University faculty and students to drive itself, has advanced to the finals of the 2007 DARPA Urban Challenge. The finals will be broadcast live on Saturday, November 3, 2007 at 11 a.m. EST at www.grandchallenge.org.
For further information on "Little Ben", « read article »