# Modeling Information Propagation in General V2V-enabled Transportation Networks

## TL;DR

This paper presents a continuous-time Markov chain model to analyze how information propagates among vehicles in V2V-enabled transportation networks, accounting for traffic density and communication conditions, with applications to real-world scenarios.

## Contribution

The paper introduces a novel Markov chain-based model for information spread in V2V networks that converges to epidemiological differential equations for efficient computation.

## Key findings

- Models accurately match real trajectory data with minimal error.
- Information propagation varies with traffic density and communication conditions.
- Applicable to real-world scenarios and system perturbations.

## Abstract

V2V technologies bridge two infrastructures: the communications infrastructure and the transportation infrastructure. These infrastructures are interconnected and interdependent. On the one hand, the communications network enables V2V interactions, while, on the other hand, the density of vehicles on the roadway enabled with V2V and the level of congestion on the roadway determine the speed and quality of communications between vehicles and infrastructure. The V2V technology is expected to contribute significantly to the growth of shared mobility, in turn, receives a significant boost from the deployment of a large number of connected vehicles in shared mobility services, provided challenges towards the deployment can be overcome. Vehicle mobility patterns and communication conditions are not only heterogeneous, but they also evolve constantly, leading to dynamic coupling between the communication and the transportation infrastructure. We consider the communication of messages amongst the vehicles in a transportation network, and estimate how quickly messages spread under different conditions of traffic density (traffic congestion, the presence of an accident, and time of day such as morning and evening rush hour) and communication conditions. We developed a continuous-time Markov chain to describe the information propagation process through enabled vehicles. Our models converge to a solution of a set of clustered epidemiological differential equations which lend itself to fast computation. We then demonstrate the applicability of this model in various scenarios: both real-world scenarios and hypothesized scenarios of outages and system perturbations. We find that our models match actual trajectory data with very little error, demonstrating the applicability of our models to study the spread of information through a network of connected vehicles.