Section: New Results

Wireless & Wired Networks

Participants : Thomas Begin, Anthony Busson, Isabelle Guérin Lassous.

Conflict graph-based model for IEEE 802.11 networks: A Divide-and-Conquer approach

WLANs (Wireless Local Area Networks) based on the IEEE 802.11 standard have become ubiquitous in our daily lives. We typically augment the number of APs (Access Points) within a WLAN to extend its coverage and transmission capacity. This leads to network densification, which in turn demands some form of coordination between APs so as to avoid potential misconfigurations. In our article [20], we describe a performance modeling method that can provide guidance for configuring WLANs and be used as a decision-support tool by a network architect or as an algorithm embedded within a WLAN controller. The proposed approach estimates the attained throughput of each AP, as a function of the WLAN's conflict graph, the AP loads, the frame sizes, and the link transmission rates. Our modeling approach employs a Divide-and-Conquer strategy which breaks down the original problem into multiple sub-problems, whose solutions are then combined to provide the solution to the original problem. We conducted extensive simulation experiments using the ns-3 simulator that show the model's accuracy is generally good with relative errors typically less than 10%. We then explore two issues of WLAN configuration: choosing a channel allocation for the APs and enabling frame aggregation on APs.

Video on Demand in IEEE 802.11p-based Vehicular Networks: Analysis and Dimensioning

This is a joint work with A. Boukerche. In [31], we consider a VoD (Video on-Demand) platform designed for vehicles traveling on a highway or other major roadway. Typically, cars or buses would subscribe to this delivery service so that their passengers get access to a catalog of movies and series stored on a back-end server. Videos are delivered through IEEE 802.11p Road Side Units deployed along the highway. In this paper, we propose a simple analytical and yet accurate solution to estimate (at the speed of a click) two key performance parameters for a VoD platform: (i) the total amount of data down-loaded by a vehicle over its journey and (ii) the total "interruption time" , which corresponds to the time a vehicle spends with the playback of its video interrupted because of an empty buffer. After validating its accuracy against a set of simulations run with ns-3, we show an example of application of our analytical solution for the sizing of an IEEE 802.11p-based VoD platform.

An accurate and efficient modeling framework for the performance evaluation of DPDK-based virtual switches

This is a joint work with B. Baynat, G. Artero Gallardo and V. Jardin [4]. Data plane development kit (DPDK) works as a specialized library that enables virtual switches to accelerate the processing of incoming packets by, among other things, balancing the incoming flow of packets over all the CPU cores and processing packets by batches to make a better use of the CPU cache. Although DPDK has become a de facto standard, the performance modeling of a DPDK-based vSwitch remains a challenging problem. In this paper, we present an analytical queueing model to evaluate the performance of a DPDK-based vSwitch. Such a virtual equipment is represented by a complex polling system in which packets are processed by batches, i.e., a given CPU core processes several packets of one of its attached input queues before switching to the next one. To reduce the complexity of the associated model, we develop a general framework that consists in decoupling the polling system into several queueing subsystems, each one corresponding to a given CPU core. We resort to servers with vacation to capture the interactions between subsystems. Our proposed solution is conceptually simple, easy to implement and computationally efficient. Tens of comparisons against a discrete-event simulator show that our models typically deliver accurate estimates of the performance parameters of interest (e.g., attained throughput, packet latency or loss rate). We illustrate how our models can help in determining an adequate setting of the vSwitch parameters using several real-life case studies.

Association optimization in Wi-Fi networks

Densification of Wi-Fi networks has led to the possibility for a wireless station to choose between several access points (APs), improving coverage, wireless link quality and mobility. But densification of APs may generate interference, contention and decrease the global throughput as these APs have to share a limited number of channels. The recent trend in which Wi-Fi networks are managed in a centralized way offers the opportunity to alleviate this problem through a global optimization of the resource usage. In particular, optimizing the association step between APs and stations can increase the overall throughput and fairness between stations. In this work, we propose an original solution to this optimization problem. First, we propose a mathematical model to evaluate and forecast the throughput achieved for each station for a given association. The best association is then defined as the one that maximizes a logarithmic utility function using the stations' throughputs predicted by the model. The use of a logarithmic utility function allows to achieve a good trade-off between overall throughput and fairness. A heuristic based on a local search algorithm is used to propose approximate solutions to this optimization problem. This approach has the benefit to be tuned according to the CPU and time constraints of the WLAN controller. A comparison between different heuristic versions and the optimum solution shows that the proposed heuristic offers solutions very close to the optimum with a significant gain of time.

In the first place, we consider a saturated network. Even if such traffic conditions are rare, the optimization of the association step under this assumption has the benefit to fairly share the bandwidth between stations. Nevertheless, traffic demands may be very different from one station to another and it may be more useful to optimize associations according to the stations' demands. In a second step, we propose an optimization of the association step based on the stations' throughputs and the channel busy time fraction (BTF). The latter is defined as the proportion of time the channel is sensed busy by an AP. We propose an analytical model that predicts BTF for any configuration. Associations are optimized in order to minimize the greatest BTF in the network. This original approach allows the Wi-Fi manager to unload the most congested AP, increase the throughput for most of the stations, and offer more bandwidth to stations that need it. We present a local search technique that finds local optima to this optimization problem. This heuristic relies on an analytical model that predicts BTF for any configuration. The model is based on a Markov network and a Wi-Fi conflict graph. NS-3 simulations including a large set of scenarios highlight the benefits of our approach and its ability to improve the performance in congested and non-congested Wi-Fi networks.

Lastly, we consider the latest amendments of the IEEE 802.11 standard. The main challenges are to propose models that take into account recent enhancements such as spatial multiplexing (MIMO) at the physical layer and frame aggregation mechanism at the MAC layer. To assess these new features, we derive an association optimization approach based on a new metric, named Hypothetical Busy Time Fraction (H-BTF), that combines the classical Busy Time Fraction (BTF) and the frame aggregation mechanism [3].

Transient analysis of idle time in VANETs using Markov-reward models

The development of analytical models to analyze the behavior of vehicular ad hoc networks (VANETs) is a challenging aim. Adaptive methods are suitable for many algorithms (e.g. choice of forwarding paths, dynamic resource allocation, channel control congestion) and services (e.g. provision of multimedia services, message dissemination). These adaptive algorithms help the network to maintain a desired performance level. However, this is a difficult goal to achieve, especially in VANETs due to fast position changes of the VANET nodes. Adaptive decisions should be taken according to the current conditions of the VANET. Therefore, evaluation of transient measures is required for the characterization of VANETs. In the literature, different works address the characterization and measurement of the idle (or busy) time to be used in different proposals to attain a more efficient usage of wireless network. We focus on the idle time of the link between two VANET nodes. Specifically, we have developed an analytical model based on a straightforward Markov reward chain (MRC) to obtain transient measurements of this idle time. Numerical results from the analytical model fit well with simulation results [12].