We study topology aware and proportionally fair medium access control in a Poisson network using spatial Aloha. We show that there exists a continuum of adaptive controls based on local stopping sets, with the extreme cases being (1) the case without any topology information and (2) the case with full network topology information. We give fixed point equation for the medium access probability in the general case, and we also derive the associated performance measures. As local information increases, the performance levels of these schemes are shown to get arbitrarily close to that of the full information based scheme. I will also briefly discuss another work where we develop LP relaxation based distributed algorithms for scheduling in ad hoc networks. As a part of our solution, we propose an algorithm for solving the LP relaxation of Maximum Weighted Independent Set (MWIS) problem, which exhibits faster convergence compared to the earlier proposed algorithms.
Chandramani Singh is a postdoctoral research associate at Coordinated Science Laboratory, University of Illinois at Urbana-Champaign. He received ME and PhD degrees in the Electrical Communication Engineering department at the Indian Institute of Science, Bangalore in 2005 and 2012, respectively. He worked as a member of technical staff at ESQUBE Communication Solutions Pvt. Ltd. from 2005 to 2006. From Feb. 2012 to Feb. 2013, he was a postdoctoral researcher at TREC, a joint research team between INRIA Rocquencourt and ENS de Paris.