Optimization of the Mobile Router and Traffic Sources in Vehicular Delay-Tolerant Network

• Published in: IEEE transactions on vehicular technology 2009-11, Vol.58 (9), p.5095-5104
• Main Authors: Niyato, D., Ping Wang
• Format: Article
• Language: English
• Subjects: Applied sciences; Buffer storage; Constraint optimization; Delay-tolerant network (DTN); Disruption tolerant networking; Electrical engineering. Electrical power engineering; Electrical power engineering; Electronics; Exact sciences and technology; Formulations; Game theory; Games; Hardware; Input-output equipment; Joints; Mobile ad hoc networks; Mobile communication; Nash equilibrium; Networks; Operation. Load control. Reliability; Optimization; Power networks and lines; Quality of service; Road vehicles; Routers; Studies; Systems, networks and services of telecommunications; Telecommunication traffic; Telecommunications; Telecommunications and information theory; Teletraffic; Traffic control; Traffic engineering; Traffic flow; Users connections and in door installation; vehicular communications
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2009.2025379

Vehicular delay-tolerant networks (VDTNs) are introduced for communications between traffic sources and sinks, in which direct end-to-end connection is not available. In this VDTN, mobile routers installed on the vehicles are used as carriers to carry data from traffic sources to sinks. We first present the optimization formulation based on the constrained Markov decision process (CMDP) to obtain an optimal decision for the mobile router on whether to accept packets from a traffic source. This decision is made to maximize the reward of data delivery while the quality-of-service (QoS) performance is guaranteed. Then, the noncooperative game and optimization formulations are presented for the cases when traffic sources have a self-interest to maximize their own benefit and a global interest to maximize the total benefit of the network, respectively. From the performance evaluation, a mobile router with CMDP optimization can achieve the highest reward while the maximum packet-blocking probability requirement is met. With noncooperative game formulation for the traffic sources, the Nash equilibrium can be obtained. Alternatively, if the traffic sources are cooperative, the largest total benefit of the network can be achieved.