log in  |  register  |  feedback?  |  help  |  web accessibility
Logo
PhD Defense: Improving Wireless Efficiency: MAC and Above
Bo Han - University of Maryland, College Park
Thursday, July 26, 2012, 10:00-11:00 am Calendar
  • You are subscribed to this talk through .
  • You are watching this talk through .
  • You are subscribed to this talk. (unsubscribe, watch)
  • You are watching this talk. (unwatch, subscribe)
  • You are not subscribed to this talk. (watch, subscribe)
Abstract

THE DISSERTATION DEFENSE FOR THE DEGREE OF Ph.D. IN COMPUTER SCIENCE FOR

                                                Bo Han

Improving the efficiency of wireless networks becomes critical, as wireless capacity is limited by not only the shared spectrum but also regulations of the Federal Communications Commission (FCC). The main goal of this work is to increase the capacity of various wireless systems at different layers, through both experimental studies and theoretical analysis.

I first systematically study the sub-frame level 802.11 communication bit errors, which provides a deeper understanding of 802.11 data transmission for further improvement. The measurement results from a number of 802.11 wireless LAN testbeds demonstrate three bit error patterns: the slope-line, saw-line and finger patterns. Identifying these error patterns can potentially introduce new opportunities in several MAC layer mechanisms, such as forward error correction, network coding and frame combining.

Motivated by the observation from the aforementioned study that corrupted packets may have only a few localized errors, we propose Maranello, a practical partial packet recovery protocol that can significantly increase the throughput and reduce the latency of 802.11 networks. Maranello is compatible with off-the-shelf devices for incremental deployment. Moreover, our firmware-based implementation of Maranello offers a new approach to evaluate wireless protocols with hard timing constraints.

At the MAC layer, we also design algorithms with provable performance guarantee for simultaneous channel allocation of individual links and packet scheduling in multi-channel multi-radio wireless networks. With a new channel-access hash function (one of the key drivers of our algorithms), each radio can know the transmitters’ decisions for links in its interference set, without introducing any extra communication overhead between them. Further, by utilizing the inductive scheduling technique, each radio can also backoff appropriately to avoid collisions.

Finally, I present a lightweight and distributed protocol, called iWander, to identify influential users in mobile social networks through random-walk sampling. Influential users are individuals with high centrality in their social-contact graphs. iWander leverages fixed-length random walks initialized periodically by a small sample of users to find these critical mobile users. The most attractive feature of iWander is its extremely low computation and control-message overhead, which lends itself well to mobile applications. Through simulation studies using a real-world mobility trace, I demonstrate the effectiveness of iWander for two applications, targeted immunization of infectious diseases and target-set selection for mobile content delivery.

Examining Committee:

Committee Chair:                       Dr. Aravind Srinivasan

Co-Chair:                                  Dr. Bobby Bhattacharjee                       

Dean's Representative:              Dr. Richard J. La

Committee Members:                Dr. Amol Deshpande

                                                Dr. Jennifer Golbeck

This talk is organized by Jeff Foster