This paper introduces a data-driven model for generating synthetic satellite users based on population and geographical factors. The model applies the logarithmic opinion pool to combine complementary perspectives on user distribution, such as population density and rural predominance. This principled aggregation captures realistic spatial disparities in connectivity and improves the representativeness of satellite constellation simulations. The approach supports the development of adaptive non-terrestrial network (NTN) strategies aligned with both orbital topology and terrestrial demand.

We propose a fully decentralized, adaptive routing system that exploits the dynamic topology of Walker Delta Low Earth Orbit (LEO) constellations to achieve efficient traffic distribution and improved reliability without requiring global link-state information. The system addresses scalability challenges in mega-constellations, where congestion, limited ground infrastructure, and uneven user distribution degrade service quality in high-demand regions. Although inter-satellite links provide diverse routing paths, the lack of multibeam support in some terminals complicates routing and can increase path length during cross-plane communication. Our evaluation of a Hop-by-Hop based strategy in Walker Delta constellations shows that it achieves performance comparable to optimal congestion-minimization algorithms and outperforms source-routing methods, while maintaining near-minimal hop counts for low-latency communication.

The availability of abundant solar energy is driving renewed interest in orbital computing infrastructures hosted on Sun-Synchronous Orbit (SSO) constellations. However, these constellations suffer from limited instantaneous Earth coverage and constrained direct-to-ground, which is clearly incompatible with the connectivity and responsiveness requirements of datacenter clients. In this paper, we propose a hybrid architecture where SSO satellites offload traffic through Low Earth Orbit (LEO) Walker constellations—such as OneWeb and Starlink—which act as a dynamic relay layer toward the ground. We design and evaluate dynamic inter-orbital link topologies enabling SSO-to-Walker connectivity and analyze their performance in terms of link availability, relay opportunities, path continuity, and network resilience. Our results show that leveraging Walker constellations as relays significantly improves SSO systems, effectively compensating for their coverage limitations.

Low Earth Orbit (LEO) constellations are transforming today’s modern communication era by enabling highspeed, low-latency, and reliable broadband services at a global scale. Despite their huge potential, designing efficient mechanisms for satellite networks is challenging due to the dynamic nature of LEO satellites. In this paper, we focus on routing in LEO constellations including Inter-Satellite Links (ISLs) with several constraints like e.g. minimising the maximum link utilisation. We propose a static approach aiming to find optimal paths to transmit traffic demands in the network graph. We consider Linear Programming (LP) to formulate problems for different network constraints and show that we are able to provide solutions respecting the various constraints, including those related to quality of service (QoS).

International audience

Avionics Full-DupleX (AFDX) is a switched Ethernet-based network used in modern commercial airplanes for the transmission of command and control avionics flows. These critical flows require deterministic guarantees leading to a lightly loaded network. Aircraft manufacturers envision to carry additional non avionics flows (i.e. video, audio, service) to take advantage of the spare bandwidth. However, it is then compulsory to preserve the real-time guarantees of avionics flows in terms of bounded jitter at the transmitter output and of bounded end-to-end latency. Depending on the type of additional traffic, Quality of Service (QoS) end-to-end guarantees may be offered to the additional flows of lower criticality in terms of reduced delay or bounded jitter for instance. These guarantees can be ensured by scheduling policies at transmitter and switch level. However, an important safety-related constraint is the asynchronous design of avionics distributed systems that prohibits the use of a network-wide synchronization of end systems and switches. Thus, time-triggered networking such as emerging Time-Triggered Ethernet (TTEthernet) or Time-Sensitive Networking (TSN) cannot be leveraged. This paper underlines the benefits of only scheduling flows at the transmitter, which is compatible with the asynchronous safety constraint of avionics systems. We show that it is possible to build a table schedule that carries both critical avionics flows and additional video flows that meet their timing and bandwidth allocation constraints. Therefore, we design scheduling strategies for efficient distribution of flows in the table scheduling whose performance is compared to the optimal schedule minimizing the emission lag of additional flows. Proposed heuristics are constructed such as to favor the network responsiveness for avionics flows thanks to slot over-provisioning. Extensive results on an A350 AFDX industrial configuration show that for a transmitter load of up to 70% of the available bandwidth, the uniform allocation heuristic provides a performance close to optimal in terms of minimum emission lag for additional flows, so offering a practical configuration heuristic to industry.

A recent trend in embedded industry is to mix Integrated Modular Architectures (IMA) with Time Sensitive Networking (TSN). IMA principles allow resource sharing between numerous software functions in a deterministic way. On the other side TSN is considered a promising communication solution for distributed architectures. However, TSN has not been fully exploited with IMA, while it would meet the increasing communication needs due to the explosion of the number of on-board software functions. In order to better understand the challenges of designing IMA-TSN systems, this paper proposes a joint model which unifies task and message management and allows evaluation of end-to-end latency properties. The model is illustrated by simulation on an avionic case study.

International audience

Formal verification of SysML models contributes to detect design errors early in the life cycle of systems. Incremental modeling of systems leads to repeat verification of systems models parts that were already verified in previous versions of the SysML model. This paper proposes to optimize the verification process by generating first a dependency graph of the SysML model. The dependency generation algorithm is implemented by free SysML tool TTool. An Avionics Full DupleX network serves as case study.

This study investigates the importance of choosing the first (respectively last) hop to access (respectively to exit) a Low Earth Orbit (LEO) satellite constellation, which is of upmost importance for the LEO routing performance. Usually, basic routing strategies connect a ground station to its nearest satellite, and this strategy does not always lead to the optimal routing path. We propose to select this first/last satellites within a subset of k-nearest satellites. After performing routing simulations over one of the next-generation satellite constellations, preliminary results show that this in/out hop selection strategy leads to a better link capacity usage and a lower data loss rate, allowing a faster TCP bulk data transfer.

Nowadays, most of cyber-physical systems in avionics, automotive or recent Industry 4.0 domains require networked communication for mixed-critical applications. Ethernet-based networks such as AFDX, TTEthernet or TSN are capable to support transmission of both safety-critical and non-critical flows. This paper focuses on the TTEthernet network compliant with the avionics ARINC 664-P7 standard supporting time-triggered communication (TT) together with rate-constrained (RC) and best-effort (BE) traffic. Due to a global synchronization, TT communication with low latency and minimal jitter is ensured with static schedules computed offline. For event-triggered RC flows, bounded jitter at the source and end-to-end latency are guaranteed with worst-case analysis methods. With the increasing demands of applications, flows with Quality of Service (QoS) requirements such as video or audio may be transmitted as BE flows. However, on current configurations, no guarantees are offered to BE flows. In this paper, we aim at increasing the maximum RC utilization and improving the QoS of BE flows to allow the transmission of video or audio traffic with low jitter and end-to-end delay requirements. For this, we focus on the scheduling mechanisms and propose a scheduling approach based on a static slotted table that is applied at end systems. This table integrates the TT schedules usually obtained with Satisfiability Modulo Theories (SMT) approaches and establishes offsets of RC flows that reduce the end-to-end delay of BE flows. Several strategies for offset computations are proposed based on the distribution of flows locally at end system or globally at switch. We show that local strategies perform better than the global ones to reduce end-to-end delay of BE flows.

AFDX is the standard switched Ethernet solution for the transmission of avionics flows. Today's AFDX deployments in commercial aircrafts are lightly loaded to ensure the determinism of control and command operations. Manufacturers envision to take advantage of the remaining AFDX bandwidth to transmit additional non avionics flows (video, audio, service). These flows must not compromise the in-time transmission of avionics ones: constraints on jitter at source end system and end-to-end latency have to be insured for each avionics flow. In this paper, we investigate the scheduling of avionics and additional flows, mainly at the end system level. We show that an event-triggered strategy is better than a time-triggered one for additional flows at source level, but it might compromise the jitter constraint of avionics flows and increase the end-to-end latency of additional ones. We consider two time-triggered scheduling strategies, i.e. an optimal one and a simpler one based on a heuristic. We show that the later one performs nearly as well as the former one and that, for both of them, the difference with an event-triggered strategy at source level is limited and can be statically bounded.

The SImple Synchronization Protocol (SISP) has been designed for tiny sensors to offer a wireless synchronization service to the network. SISP is completely distributed with a flat architecture. Nodes broadcast a SYNC message periodically that contains the value of their view of a shared clock counter. Every time a SYNC message is received, nodes update their shared clock by averaging it with the clock value embedded in the message. This protocol converges in practice very well, and requires a small amount of energy as SYNC messages can be sent every second only. Moreover, computations are basic, perfectly fitting the tiny sensor platforms needed for the Internet of Things. Its distributed operations enable the network to adjust seamlessly to the appearance or disappearance of other nodes. This paper concentrates on the convergence analysis of this promising protocol. Convergence time and synchronization accuracy are determined analytically, by simulations and by experimenting a real sensor platform. All results show that this protocol offers an accuracy in the order of a few tens of microseconds. Moreover, our analytical derivations capture very well an upper bound on the synchronization accuracy.