In predictable mobile networks, network nodes move in a predictable way and therefore have dynamically changing but predictable connectivity. We have developed a model that formalizes predictable dynamic topologies as sequences of static snapshots. We use this model to design and evaluate a predictable mobile-routing protocol based on link-state routing, whose performance is superior to its static and ad hoc counterparts. Our routing protocol accounts for occurrences of additional, unpredictable changes, as well as their interaction with predictable changes. We evaluate our protocol using simulations based on randomly generated topologies and spacecraft-network scenarios. In both cases, we show that our protocol outperforms traditional routing protocols and is well suited for routing in next-generation space networks.

In recent times, the importance of information security in civilian space communication systems has increased significantly among space agencies. Those systems often require a high degree of flexibility, interoperability and scalability. Therefore the usage of traditional, physical layer based security solutions is not a desirable option for the agencies. In this paper, we investigate the application of security services to the CCSDS Packet TM/TC space communication protocol stack and propose a transparent and open solution that provides an optimal trade-off between security, interoperability and technical implementation feasibility. We take into consideration the special restrictions that are superimposed by the space environment and existing legacy systems.

Predictable mobile networks are dynamic in terms of the mobility and connectivity of the network nodes. However, in contrast to mobile ad-hoc networks, their dynamics are mostly predictable, as the name suggests. Currently, no adequate topology model exists for such networks. Moreover, routing protocols based on either static or mobile ad-hoc topology models do not exploit this predictability and thus are too inefficient for use in some application areas. We present a model that formalizes predictable dynamic topologies as sequences of static snapshots. We use this model to design and prove the correctness of a protocol based on link-state routing, whose performance is superior to its static and ad-hoc counterparts. Our routing protocol accounts for occurrences of additional, unpredictable changes, as well as their interaction with predictable changes. As an application area, we focus on routing in spacecraft networks and explain the suitability of our protocol for this application domain.