PATH AWARENESS TECHNIQUES FOR TRANSPORT PROTOCOLS OVER SATELLITE (ARTES AT 3A.157)

Description

The objective of the activity is to develop and test Path Aware Networking (PAN) techniques for LEO, MEO or GEO systems with dynamic bandwidth allocation and oversubscription and for hybrid satellite-terrestrial networks with multipath. The new protocols will be implemented in software and verified in a laboratory environment first, and then they will be validated over the air. Targeted Improvements: Enabling transport layer protocol performance optimisation over encrypted links and/or multiple paths. Description: Path Aware Networking (PAN) is a new networking paradigm that aims to respond to the challenges posed by unpredictable characteristics ofmodern networks in terms of available bandwidth and transmissions latency. PAN enables endpoints to be aware of the path characteristics and use this knowledge to direct end-to-end transport-layer decisions. Important information that can be exploited by the end hosts includes the knowledge that the path supports certain services, or has specific characteristics (e.g., some form of QoS support),that alternative routes to the destination exist (for multipath extensions of transport protocols), and of the level of congestion on the paths. While PAN can offer substantial benefits in satellite contexts, satcom systems use instead specific appliances, such as Performance-enhancing Proxies (PEPs) with TCP accelerators and caches, to adapt transport protocols to the characteristics of satellite networks, which is a strategy that cannot be applied when traffic is encrypted at network layer, e.g., with IPSec (VPN traffic).To enable optimised transport layer protocol performance even when using VPNs or traffic being encrypted at network layer, e.g., with IPSec, this activity will develop path aware networking for LEO, MEO or GEO systems with dynamic bandwidth allocation and oversubscription and for hybrid satellite-terrestrial networks with multipath transport support. Security aspects related to the trust and risk models associated with path information will be considered. The new protocols will be first tested by simulation, then in a laboratory environment, emulating in real-time different satellite scenarios with several network paths and, finally, the protocols will be tested over the air.

Tender Specifics