Optimization-Based Wireless Network Architectures: Complexity, Decentralization and Performance Guarantees
Abstract
The goal of the project was to use optimization-based techniques to develop new architectures, algorithms and performance analysis tools for wireless networks. Our main results are as follows: (i) we developed a novel architecture using virtual backlog queues to accommodate real-time and elastic flows in the same network, (ii) we developed a distributed CSMA algorithm , using a statistical physics idea called Glauber dynamics, which achieved 100% throughput in wireless networks with low complexity, and (iii) we used techniques used to bound Markov chain mixing times to understand the performance of our algorithms. The sum total of our effort led to new algorithms for wireless networks which significantly improved the state-of-the-art.
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 09, 2011
- Accession Number
- ADA564105
Entities
People
- R. Srikant
Organizations
- University of Illinois Urbana–Champaign