In this article we use real traffic data to confirm that vehicle velocities follow Gaussian distribution in steady state traffic regimes (free-flow, and congestion). We also show that in the transition between free-flow and congestion, the velocity distribution is better modeled by generalized extreme value distribution (GEV). We study the effect of the different models on estimating the probability distribution of connectivity duration between vehicles in vehicular ad-hoc networks.
1 May 2015 in AEU - International Journal of Electronics and Communications
Random matrix theory is finding an increasing number of applications in the context of information theory and communication systems, especially in studying the properties of complex networks. Such properties include short-term and long-term correlation. We study the spectral fluctuations of the adjacency of networks using random-matrix theory. We consider the influence of the spectral unfolding, which is a necessary procedure to remove the secular properties of the spectrum, on different spectral statistics. We find that, while the spacing distribution of the eigenvalues shows little sensitivity to the unfolding method used, the spectral rigidity has greater sensitivity to unfolding
28 January 2021 in IEEE Transactions on Intelligent Transportation Systems
In vehicular networks, the instantaneous connectivity between any two consecutive vehicles is a function of both the headway distance between them and the communication channel characteristics. We derive a generalized analytic expression for vehicle-to-vehicle link connectivity probability in highway vehicular networks that considers the change in vehicle headway distribution under different traffic conditions. The introduced formulation is derived under the Nakagami-m small-scale fading model which captures a wide range of small-scale fading conditions in freeway vehicular networks and was shown to well fit the empirical data. This formulation is of interest in analyzing and simulating vehicular communication systems.
1 February 2018 in Physica A: Statistical Mechanics and its Applications
To help with the planning of intervehicular communication networks, an accurate understanding of traffic behavior and traffic phase transition is required. We calculate intervehicle spacings from empirical data collected in a multilane highway in California, USA. We calculate the correlation coefficients for spacings between vehicles in individual lanes to show that the flows are independent. We determine the first four moments for individual lanes at regular time intervals, namely, the mean, variance, skewness, and kurtosis. We follow the evolution of these moments as the traffic condition changes from the low-density free flow to high-density congestion. We find that the higher moments of intervehicle spacings have a well-defined dependence on the mean value. The variance of the spacing distribution monotonously increases with the mean vehicle spacing. In contrast, our analysis suggests that the skewness and kurtosis provide one of the most sensitive probes toward the search for the critical points. We find two significant results. First, the kurtosis calculated in different time intervals for different lanes smoothly varies with the skewness. They share the same behavior with the skewness and kurtosis calculated for probability density functions that depend on a single parameter. Second, the skewness and kurtosis as functions of the mean intervehicle spacing show sharp peaks at critical densities expected for transitions between different traffic phases. The data show a considerable scatter near the peak positions, which suggests that the critical behavior may depend on other parameters in addition to the traffic density.