Model T: A model for user registration patterns based on campus WLAN data

Abstract

We discuss the derivation of an empirical model for spatial registration patterns of mobile users in a campus wireless local area network (WLAN). Such a model can be very useful in a variety of simulation studies of the performance of mobile wireless systems, such as that of resource management and mobility management protocols. We base the model on extensive experimental data from a campus WiFi LAN installation. We divide the empirical data available to us into training and test data sets, develop the model based on the training set, and evaluate it against the test set.The model shows that user registration patterns exhibit a distinct hierarchy, and that WLAN access points (APs) can be clustered based on registration patterns. Cluster size distributions are highly skewed, as are intra-cluster transition probabilities and trace lengths, which can all be modeled well by the heavy-tailed Weibull distribution. The fraction of popular APs in a cluster, as a function of cluster size, can be modeled by exponential distributions. There is general similarity across hierarchies, in that inter-cluster registration patterns tend to have the same characteristics and distributions as intra-cluster patterns. In this context, we also introduce and discuss the modeling of the disconnected state as an integral part of real traffic characteristics.We generate synthetic traffic traces based on the model we derive. We then compare these traces against the real traces from the test set using a set of metrics we define. We find that the synthetic traces agree very well with the test set in terms of the metrics. We compare the derived model to a simple modified random waypoint model, and show that the latter is not at all representative of the real data. We also show how the model parameters can be varied to allow designers to consider 'what-if' scenarios easily. Finally we develop an extended version of Model T that uses an alternative modeling of relative popularity of APs and clusters, with certain generalization advantages, and evaluate its fidelity to the real data also, with positive results.

WiQoSM: An Integrated QoS-Aware Mobility and User Behavior Model for Wireless Data Networks

“…More recently, some authors have derived synthetic models of user behavior which try to mimic the observed user behavior in a certain environment. This is the case, for instance, of the ModelT model proposed in [13], which is based on the traces collected from the Dartmouth College campus over a period of 2 years. This model has been recently extended to account for spatio-temporal correlation in the user registration patterns [21].…”

“…The fact that APs in a wireless data network display different popularity degrees is well documented in the literature [1,4,18]. In our model, we initially assign each AP a with a popularity degree p(a) ∈ [p min , p max ] according to a probability distribution resembling a bounded power law (as suggested in [13]). …”

“…Each deployed AP a is then assigned with a degree of popularity p(a) ∈ [p min , p max ], which reflects its potential of attracting network users. The degree of popularity of APs is computed according to a probability distribution resembling a bounded power law (see [13]). Note that other probability distributions for AP popularity can be easily included in our model.…”

“…In case of a mobility model generating user and traffic traces such as WiQoSM, model validation consists in comparing certain features of the generated traces with real-world traces. However, to this purpose we cannot use features which are directly accounted for in the model [17], such as the degree of AP popularity in case of WiQoSM: it is clear that such features can easily be reproduced by embedding the properties observed in the real-world trace at hand directly in the model (and, in fact, the way AP popularity is set in WiQoSM is based on the observations reported in [13]). Hence, for WiQoSM assessment we use features of the generated traces that cannot be directly controlled.…”

“…On one hand, we have synthetic models aimed at resembling user registration patterns observed in a certain realworld deployment (e.g., the Dartmouth college [13,17,20,21], or the ETH Zurich campus [27]). While these models allow some flexibility in the choice of some parameters, such as number of users and APs, they do not consider user traffic patterns.…”