Previously associated with space programs or emergency communication, satellites now play a key role in the expansion of telecommunications networks. Their ability to connect remote or inaccessible regions, such as mountains, deserts or even oceans, makes them a strategic asset. Use of the Starlink constellation to compensate for the destruction of communication infrastructures in Ukraine or, just recently, in Réunion, is another conclusive example. Satellites are therefore opening up new prospects for extending connectivity to blank signal reception spots that have yet to be covered by standard networks, and to regions that have suffered devastation. However, this ambitious idea raises a key question: how can 5G by satellite be rolled out, in particular using millimeter frequencies that have barely been exploited, while accommodating constraints relating to the movement of satellites and of mobile devices on the ground?
This is the challenge partly taken up by Philippe Martins, a Télécom Paris professor in mobile phone networks, through a project supported by BPI and led by Constellation Technologies & Operations. Relying on 5G NTN (Non-Terrestrial Network), this project, named simply 5G NTN mmWave, aims to design an all-French satellite infrastructure allowing two-way communication between terminals and satellites. “The goal is to obtain a sovereign constellation that meets connectivity needs while ensuring technological self-sufficiency in an increasingly strategic field”, he adds, emphasizing that many countries, including China, have already launched their own constellations. But to achieve such deployment, the means to integrate this satellite infrastructure into existing networks remain to be found.
Strengthened connectivity between earth and sky
Unlike traditional infrastructures based on permanent base stations, 5G NTN uses a satellite constellation to send signals into zones where terrestrial installations are non-existent or inconceivable. The approach of the 5G NTN mmWave project, however, aims to complement existing infrastructures, rather than compete with them. “What we intend to do is extend the coverage of terrestrial operators by using standardized millimetric bands for 5G. To our knowledge, this has not yet been done”, argues Philippe Martins. This makes the envisaged system different from the existing, so-called ‘proprietary’ systems, such as Starlink, which exploit their own millimetric frequency bands (Ku, Ka) and are not designed to be directly integrated into terrestrial 5G networks.
The project also relies on a multi-orbit architecture, which combines satellites traveling at different altitudes, in this case, at Very Low Earth Orbit (VLEO) and Middle Earth Orbit (MEO). To design these satellites, the consortium brings together several fields of expertise. For example, the Office National d’Etudes et de Recherches Aérospatiales (ONERA) and the company Greenerwave contribute their expertise in space and antenna engineering (RIS). “As for me, I don’t know how to make satellites fly or design RIS antennas!”, jokes Philippe Martins. “But François Baccelli [mathematician and director of research at Inria] and I were initially consulted by the company Constellation for our technical opinion, before joining the consortium to bring our expertise on networks to the project.” Indeed, to deploy a constellation, it isn’t ‘enough’ to simply send satellites into orbit. You also have to coordinate them and optimize the movement of data between space and Earth.
Organizing a moving network
One of the project’s major technical challenges lies in the movement of the relay satellites. Unlike terrestrial base stations, which are permanently fixed, satellites are in perpetual movement and may follow different orbits. This movement results in time lag and frequency shift, which makes it necessary to optimize the management of signal synchronization between satellites and users on the ground.
The 5G NTN mmWave research teams are therefore working actively to develop the timing advance algorithms required for this synchronization. It is a complex task because, to compensate for these issues caused by satellite movement, constant adjustments in real time are necessary to avoid desynchronization. “This is the first step because, without perfect management of time lag and frequency shift, nothing can function”, Philippe Martins points out.
Indeed, synchronization is only part of the equation. The Télécom Paris teams are also working on the design of routing and handover algorithms. Routing determines the path taken by data between satellites and users, while handover enables a terminal to switch from one satellite to another without any service interruption. Again, both these aspects are particularly complex in a context of dynamic mobility, in which satellites and terminals are constantly moving.
Optimal distribution of satellites
The resulting algorithms will first be tested in a simulator developed by Télécom Paris, before eventually being incorporated into the Constellation Technologies simulator. The latter, which is a digital twin simulating the behavior of a satellite constellation, will hence confirm the algorithms’ performance in a realistic environment.
Alongside the definition of these three algorithms, which are essential for the system to work, the team led by Philippe Martins and François Baccelli are working on performance models to identify the system’s limitations and optimize its efficiency in real-life conditions. The ultimate aim is to develop a global model to assess the availability of satellites at all times in order to ensure continuous coverage. By determining the number of satellites able to communicate with a terminal at a given moment, the model will make it possible to optimize the distribution of satellites and anticipate the risk of signal loss. This coordination task is all the more crucial that, on its launch, the constellation may be composed of only a limited number of satellites.
“Your space is saturated”
Besides the technical challenges relating to telecommunication issues, the project also requires consideration of the design of the satellites themselves, especially in terms of their energy performance. Unlike terrestrial stations, satellites are powered by solar panels and do not benefit from unlimited resources. It is therefore vital to optimize their software component for the best possible management of their on-board resources. This energy issue is nevertheless a mere facet of the environmental challenges raised by the deployment of satellite constellations.
Space is far from empty, as Philippe Martins laments. In low orbits of 300 to 1,000 km above the Earth, a whole ocean of space debris can be found, with some 900,000 fragments exceeding 1 centimeter and traveling at very high speeds. This debris presents a major risk to active satellites and future constellations. To avoid exacerbating this pollution of space, satellites must therefore be provided with collision-avoiding mechanisms, as well as solutions to guarantee their total disintegration at the end of their life cycle. “When a satellite has completed its mission, it must not leave debris in orbit”, insists Philippe Martins. An out-of-service satellite potentially remains in space for several months before falling back into the atmosphere, with the risk of fragments reaching the earth’s surface. The aim is therefore to control its destruction from the moment it reenters the atmosphere, which leads us to think about the choice of materials used to create it.
“These coexistence issues arising from the multiplication of objects in orbit have not always been taken into account in the past”, the researcher observes. “But whether it’s pollution from debris, electromagnetic pollution or visual pollution, these issues can no longer be ignored when a constellation is deployed.” Energy optimization, space debris management and preservation of the orbital environment are hence becoming criteria that must be met to ensure the responsible implementation of space technologies.
