Experiments with NetLogo for Distributed Channel Assignment in Dense WLAN Networks Vangelis Gazis, Konstantinos Sasloglou, Andreas Merentitis, Kostas Mathioudakis AGT Group (R&D) GmbH Darmstadt, Germany e-mail: {vgazis,ksasloglou,amerentitis,kmathioudakis}@agtinternational.com Abstract—Dense IEEE 802.11 networks require thorough radio network planning to minimize the detrimental effects of channel interference and achieve good performance. When multiple IEEE 802.11 networks operating under different administrations reside in the same geographical area (as defined by the bounds of radio signal transmission and reception), the problem of efficient radio resource allocation arises. Particularly, in highly populated urban environments, efficient radio resource allocation can be quite a challenge, due to the finite number of interference- free channels available in the IEEE 802.11 family of standards. To address this shortcoming, we have investigated how the radio resource management capabilities provided by the IEEE 802.11k standard can support a distributed approach in channel assignment. In this paper, we present our family of distributed channel assignment algorithms and elaborate on their pros and cons. Furthermore, we present our simulation environment for simulating distributed channel assignment scenarios based on the NetLogo agent modeling environment. Index Terms—Channel assignment; Channel Allocation; WLAN; Distributed Systems; Simulation; NetLogo I. I NTRODUCTION The establishment of wireless communication as a tech- nological commodity in modern society and the insatiable demand for Internet connectivity paved the way to the tremen- dous popularity and commercial success of the IEEE 802.11 family of standards. With IEEE 802.11b/g as their flag- ship and—currently—most widespread standard, IEEE 802.11 Wireless LAN (WLAN) type systems feature an ever in- creasing footprint worldwide. Standardisation work is feverish and the IEEE 802.11 working groups regularly publishes improvements to existing standards and introduces additional features to address emerging use cases. To address the complexity associated with the manage- ment of large scale deployments in the Future Internet (FI), the vision of self-managing systems has been proposed and adopted. Under the umbrella self-management concept, self- awareness, self-configuration, self-protection, self-healing and self-optimization stand as prominent properties. The funda- mental premise of the self-management paradigm is to con- strain the operational expenses resulting by the current network management practices in the Internet. To this end, three thematic areas have converged [1] under the umbrella of the Future Internet architecture: autonomic computing, cognitive networking and self-organization. A. Radio interference In a wireless infrastructure, network management and, in particular radio network planning, becomes a challenging task due to the volatile and unpredictable nature of the wireless medium and the mobility patterns of terminal devices. The increasing footprint and diversification of IEEE 802.11 type components in ICT equipment suggests that a large portion of FI devices will employ one or more types of wireless access technology. Due to the lack of coordination that characterizes applications in the mass consumer market segments, static and centralized solutions to wireless network planning prove inefficient in this chaotic environment [2]. Minimizing radio interference is essential to realizing good system performance in IEEE 802.11 networks due to the finite number of interference-free channels available [3]. However, the volatile nature of the wireless medium, combined with varying patterns of traffic demand, render any radio network planning exercise a challenging task. The latter is in iteself a sophisticated task, which, besides expert knowledge, may also require a solid understanding of the propagation profile at each radio site [4]. In cases where IEEE 802.11 networks operating under different administrations reside in the same geographical area (as defined by the bounds of radio signal transmission and reception, efficient radio resource allocation becomes quickly problematic. This is due to available technology solutions for radio resource management of IEEE 802.11 installations supporting only a centralized model of administration (e.g., Cisco Unified Wireless Access [5], HP MultiService Mobility Controller [6]). In addition, they lack support for peer-to- peer communication between IEEE 802.11 access points in regard to radio resource management procedures. Thereupon, as IEEE 802.11 systems continue to increase their footprint in the residential and enterprise market segments, efficient radio resource allocation in a dense urban environment rapidly devolves to a chaotic situation [2]. B. Channel assignment Channel assignment in IEEE 802.11 systems may result in a conflict where more than one adjacent (in terms of radio coverage) wireless access points use the same channel, thus causing a substantial drop in performance. In addition, as Fig. 2 illustrates for the case of IEEE 802.11b/g/n, adjacent access points may use different channels but still experience a substantial spectrum overlap [3], thus still suffering from 44 Copyright (c) IARIA, 2014. ISBN: 978-1-61208-331-5 ICAS 2014 : The Tenth International Conference on Autonomic and Autonomous Systems