Distributed and centralized algorithms for large-scale IEEE 802.11b infrastructure planning

Nonlinear Optimization of IEEE 802.11 Mesh Networks

“…In [3], we simulated the distributed deployment algorithm on a realistic scenario with a significant number of basic nodes (≈ 50), corresponding to Vigo, Spain. We observed that most WLCs got connected and the resulting mesh had link diversity.…”

“…From them, in a previous paper [3] we derived two mesh deployment algorithms: a distributed one, to be executed by infrastructure nodes themselves, and a centralized one via a mixed integer linear optimization model. In this paper we enrich the centralized version, by adding co-channel and inter-channel interference estimates that yield a mixed integer non-linear optimization model.…”

“…It is straightforward to extend the results of this work to other substandards like IEEE 802.11a or 802.11g. This paper is organized as follows: section 2 reviews the work in [3]. Section 3 describes the new proposal, a mixed integer non-linear optimization model satisfying the deployment rules in [3].…”

“…This paper is organized as follows: section 2 reviews the work in [3]. Section 3 describes the new proposal, a mixed integer non-linear optimization model satisfying the deployment rules in [3]. Section 4 presents numerical tests on a realistic scenario -a sector in Vigo, Spain-.…”

“…From a formal point of view, it is a mixed integer non-linear optimization model that considers both co-channel and inter-channel interference in the same compact formulation. It may serve as a planning tool by itself or to provide a performance bound to validate simpler planning models such as those in [3]. …”