Abstract
Satellite communications are characterized by long delays, packet losses, and sometimes intermittent connectivity and link disruptions. The TCP/IP stack is ineffective against these impairments and even dedicated solutions, such as performance enhancing proxies (PEPs), can hardly tackle the most challenging environments, and create compatibility issues with current security protocols. An alternative solution arises from the delay- and disruption-tolerant networking (DTN) architecture, which specifies an overlay protocol, called bundle protocol (BP), on top of either transport protocols (TCP, UDP, etc.), or of lower layer protocols (Bluetooth, Ethernet, etc.). The DTN architecture provides long-term information storage on intermediate nodes, suitable for coping with disrupted links, long delays, and intermittent connectivity. By dividing the end-to-end path into multiple DTN hops, in a way that actually extends the TCP-splitting concept exploited in most PEPs, DTN allows the use of specialized protocols on the satellite (or space) links. This paper discusses the prospects for use of DTN in future satellite networks. We present a broad DTN overview, to make the reader familiar with the characteristics that differentiate DTN from ordinary TCP/IP networking, compare the DTN and PEP architectures and stacks, as a preliminary step for the subsequent DTN performance assessment carried out in practical LEO/GEO satellite scenarios. DTN security is studied next, examining the advantages over present satellite architectures, the threats faced in satellite scenarios, and also open issues. Finally, the relation between DTN and quality of service (QoS) is investigated, by focusing on QoS architectures and QoS tools and by discussing the state of the art of DTN research activity in modeling, routing, and congestion control.