Eur. Phys. J. B (2015) 88: 211
https://doi.org/10.1140/epjb/e2015-60234-y

Regular Article

Uncovering transportation networks from traffic flux by compressed sensing

¹ School of Systems Science, Beijing Normal University, Beijing 100875, P.R. China
² State Information Center, Beijing 100045, P.R. China

^a e-mail: shenzhesi@mail.bnu.edu.cn

Received: 23 March 2015
Received in final form: 8 July 2015
Published online: 12 August 2015

Abstract

Transportation and communication networks are ubiquitous in nature and society. Uncovering the underlying topology as well as link weights, is fundamental to understanding traffic dynamics and designing effective control strategies to facilitate transmission efficiency. We develop a general method for reconstructing transportation networks from detectable traffic flux data using the aid of a compressed sensing algorithm. Our approach enables full reconstruction of network topology and link weights for both directed and undirected networks from relatively small amounts of data compared to the network size. The limited data requirement and certain resistance to noise allows our method to achieve real-time network reconstruction. We substantiate the effectiveness of our method through systematic numerical tests with respect to several different network structures and transmission strategies. We expect our approach to be widely applicable in a variety of transportation and communication systems.