Belangrijkste concepten
Estimating traffic demands and paths in road networks using Dynamic Programming.
Samenvatting
This article discusses a Dynamic Programming approach for estimating traffic demands and paths in road networks. It introduces a method based on higher-order cumulants to estimate these quantities. Theoretical properties and simulation results on synthetic data from NSFnet and Sioux Falls networks are presented.
Introduction
- Discusses a Dynamic Programming approach for traffic estimation.
- Focuses on estimating traffic demands and paths in road networks.
Literature Review
- Mentions the broader field of Network Tomography.
- Discusses the estimation of traffic demands from traffic flows.
Statement of Contribution
- Introduces a novel approach for estimating traffic demands and road usage.
- Presents a Dynamic Programming procedure for efficient estimation.
Notation
- Defines notation for the mathematical aspects of the problem.
Problem Formulation
- Formulates the problem of estimating traffic demands and paths in road networks.
Simulation Results
- Tests the algorithm on synthetic data from NSFnet and Sioux Falls networks.
- Compares the estimated demands with the actual demands.
Conclusions
- Summarizes the key points discussed in the article.
- Mentions future developments for the algorithm.
Statistieken
We consider a road network represented by a directed graph.
The flows of all user groups are modeled as independent Poisson processes.
The algorithm relies on the knowledge of high order cumulants.
The method is tested on synthetic data from NSFnet and Sioux Falls networks.
Citaten
"We consider a road network represented by a directed graph."
"Our focus is estimating the paths followed by each user group."