Meta-heuristic Solution for Dynamic Association Control in Virtualized Multi-rate WLANs Dawood Sajjadi, Maryam Tanha and Jianping Pan Department of Computer Science, University of Victoria, Victoria, BC, Canada Email: {sajjadi, tanha, pan}@uvic.ca Abstract—Chaotic deployment of Wireless Local Area Net- works (WLANs) in dense urban areas is one of the common issues of many Internet Service Providers (ISPs) and Wi-Fi users. It results in a substantial reduction of the throughput and impedes the balanced distribution of bandwidth among the users. Most of these networks are managed independently and there is no cooperation among them. Moreover, the conventional association mechanism that selects the Access Points (APs) with the strongest Received Signal Strength Indicator (RSSI) aggravates this situa- tion. In this paper, we present a versatile near-optimal solution for the fair bandwidth distribution over virtualized WLANs through dynamic association control. The proposed scheme is called ACO- PF, which is developed on top of Ant Colony Optimization (ACO) as a meta-heuristic technique to provide Proportional Fairness (PF) among the greedy clients. In fact, it presents a generic and centralized solution for ISPs that are using a common, virtualized or overlapped WLAN infrastructure for serving their customers. We have evaluated the efficacy of ACO-PF through numerical analysis versus popular existing schemes for both downlink and uplink scenarios. Our proposed technique has less complexity in terms of the implementation and running time for large- scale WLANs and it can be easily developed and customized for different objective functions. In addition, it is implemented in a testbed environment to investigate the key challenges of real deployment scenarios. Index Terms- Dynamic Association Control, Optimization, Virtual Multi-rate WLAN, Ant Colony, Proportional Fairness. I. I NTRODUCTION According to Cisco visual networking index [1], the amount of cellular traffic that will be offloaded to WLANs is expected to increase from 33% in 2012 to 47% by 2017. This means Wi- Fi networks are becoming drastically dense and chaotic, which is one of the consequences of bandwidth provisioning for highly growing user demands through adding more APs. By increasing the number of APs, theoretically we can promote the Quality of Experience (QoE) for the users through reducing the number of associated users to each AP. Nonetheless, since the number of non-overlapping channels is limited, neighboring APs within the dense areas must operate on the same channel. This fact exacerbates the design complexity of WLANs. Moreover, due to the contention based nature of 802.11 networks and the backoff procedure, there is a considerable throughput degradation in such areas. Also, it is important to note that AP selection based on the strongest RSSI has remained as the most common approach for asso- ciating the Wi-Fi users to WLANs; however it cannot reach the maximum network throughput. In addition, the existing 802.11 MAC protocol attempts to give the same chance to all stations that are associated to the same WLAN and it reduces overall network throughput in multi-rate WLANs, significantly. The main reason behind this phenomenon is the unbalanced channel occupancy among the stations with different data rates [2]. Thus, regular fairness provisioning techniques for wired networks such as max-min fairness can not be directly applied to multi-rate WLANs. Also, it should be noted that although there are a lot of challenges for the management of dense WLANs, there is an opportunity for Wi-Fi stations to associate with the APs which provide them with the highest end-to-end throughput. In enterprise and community networks of ultra-dense WLANs interconnected by high-speed wired links, the con- solidation of distributed APs that belong to different ISPs through a centralized association control not only promotes the customer satisfaction, but also improves the network performance. Since finding the optimal association map of Wi-Fi stations in large-scale WLANs is NP-hard, utilizing centralized solutions on top of the novel technologies, e.g., SDN, is one of the most efficient options. Although the decentralized systems may seem more realistic, the growth of centralized solutions has attracted a lot of attentions due to the feasibility of using efficient algorithms at the controllers. Moreover, distributed schemes impose noticeable convergence latency in large networks and they might miss the near-optimal configuration due to the lack of a holistic view of the system. Hence, by applying cooperative association control schemes, it would be possible to alleviate the impact of drastic interference and throughput degradation in such environments. Furthermore, sharing a common or overlapped WLAN in- frastructure among several ISPs extends the network coverage that facilitates the user connectivity as well as reducing the operational costs. In this situation, since the network capacity is shared within a WLAN infrastructure which is serving the customers of different ISPs, the growth of traffic in one virtual network can lead to the traffic decline in another one. Thus, using an efficient resource sharing scheme through a centralized controller is a necessity for the management of virtualized WLAN infrastructures. Fig. 1 illustrates an example, in which each AP of the virtual WLAN infrastructure broadcasts the SSIDs of three different ISPs. So, the customers of each ISP can be associated to any AP that belongs to the illustrated infrastructure and they will be served based on their service agreements and the total airtime share of their ISPs. All the resource allocation process will be handled by a controller. 2016 IEEE 41st Conference on Local Computer Networks © 2016, Dawood Sajjadi. Under license to IEEE. 253 2016 IEEE 41st Conference on Local Computer Networks © 2016, Dawood Sajjadi. Under license to IEEE. DOI 10.1109/LCN.2016.53 253 2016 IEEE 41st Conference on Local Computer Networks © 2016, Dawood Sajjadi. Under license to IEEE. DOI 10.1109/LCN.2016.53 253