Existing Internet protocols assume persistent end-to-end connectivity, which cannot be guaranteed in disruptive and high-latency space environments. To operate over these challenging networks, a store-carry-and-forward communication architecture called Delay/Disruption Tolerant Networking (DTN) has been proposed. This work provides the first examination of the performance and robustness of Contact Graph Routing (CGR) algorithm, the state-of-the-art routing scheme for space-based DTNs. To this end, after a thorough description of CGR, two appealing satellite constellations are proposed and evaluated by means of simulations. Indeed, the DtnSim simulator is introduced as another relevant contribution of this work. Results enabled the authors to identify existing CGR weaknesses and enhancement opportunities.

Virtual Prototyping has been widely adopted as a costeffective solution for early hardware and software co-validation. However, as systems grow in complexity and scale, both the time required to get to a correct virtual prototype, and the time required to run real software on it can quickly become unmanageable. This paper introduces a feature-rich integrated virtual prototyping solution, designed to meet industrial needs not only in terms of performance, but also in terms of ease, rapidity and automation of modelling and exploration. It introduces novel methods to leverage the QEMU dynamic binary translator and the abstraction levels offered by SystemC/TLM 2.0 to provide the best possible trade-offs between accuracy and performance at all steps of the design. The solution also ships with a dynamic platform composition infrastructure that makes it possible to model and explore a myriad of architectures using a compact high-level description. Results obtained simulating a RISC-V SMP architecture running the PARSEC benchmark suite reveal that simulation speed can range from 30 MIPS in accurate simulation mode to 220 MIPS in fast functional validation mode.

a b s t r a c tDelay-Tolerant Networking (DTN) has been proposed for satellite networks with no expectation of continuous or instantaneous end-to-end connectivity, which are known as Delay-Tolerant Satellite Networks (DTSNs). Path computation over large and highly-dynamic yet predictable topologies of such networks requires complex algorithms such as Contact Graph Routing (CGR) to calculate route tables, which can become extremely large and limit forwarding performance if all possible routes are considered. In this work, we discuss these issues in the context of CGR and propose alternatives to the existing route computation scheme: first-ending, first-depleted, one-route , and per-neighbor strategies. Simulation results over realistic DTSN constellation scenarios show that network flow metrics and overall calculation effort can be significantly improved by adopting these novel route table computation strategies. .ar (J.A. Fraire).DTN can be precisely computed in advance based on orbital elements. Also, power-conserving spacecrafts may communicate on infrequent, fixed intervals established by configuration. In any case, forthcoming episodes of communications (a.k.a. contacts ) are typically scheduled weeks or months before they occur and can be imprinted in a contact plan . The resulting contact plan can be either distributed in advance to DTN nodes, or used by a centralized node (i.e., mission control) to execute route determination procedures.A DTN paradigm can indeed be used to forward data on near-Earth satellite networks with sporadic satellite-to-satellite and satellite-to-ground communication opportunities. If so, we define them as Delay-Tolerant Satellite Networks (DTSNs). DTSNs differ from other space DTNs in the size of the topology and the speed at which it changes. In particular, interplanetary networks are rather scarce in terms of spacecrafts as a few rovers on a remote planet plus some orbiters are typically assumed in the literature [7] . While this density of deep-space nodes is unlikely to change in the near future, DTSNs topologies are expected to be promptly based on dozens or even hundreds of satellites [1] . Furthermore, while in interplanetary DTNs the topological changes are dictated by planetary dynamics, communication opportunities in DTSNs typically occur much more frequently between satellites in Low-Earth Orbit (LEO). As a result, the scalability limits of current DTN protocols and algorithms are likely to be met sooner in DTSN than in deep-space applications. Thus, DTSNs become an immediate object of study for evaluating efficient routing strategies which will also, in the long term, be valuable in the interplanetary domain.

No abstract