Machine-to-machine (M2M) communications are increasingly critical in supporting new applications such as environmental sensing (e.g. weather, flood or volcanic activity tracking, climate or wildlife monitoring), sustaining human activity over large areas (smart farming, smart transportation), and generally enabling the operation of connected devices in remote areas not or poorly covered by terrestrial networks. While terrestrial radio access networks (RAN) typically suffer from blind spots and imperfect coverage in remote or rural areas, non-terrestrial networks (NTN) have the potential to provide uniform and seamless coverage for such applications. This domain is a very hot topic and we can already observe a blossom of new start-ups, new constellations of satellites and new services.
In M2M communications, the information to be transmitted is not data flows (video, music) or data files (photos, files); instead, the goal is to report sporadically – periodically or randomly – on sensing actions to transfer small size packets. Furthermore, M2M wireless links are becoming an element of a large interconnected system, with terminals spread over a wide area, possibly lacking awareness of their geographical position, dynamic and with little learning capabilities. Thus, the specific traffic generated in M2M scenarios requires to profoundly modify the needs in terms of RAN design.
The prototypical set-up considered in WARM-M2M consists in a large-scale system communication system encompassing many (potentially millions of) Internet of things (IoT) devices capable of two-way communication but with limited processing capabilities, several low-earth orbit (LEO) satellites with regenerative payloads and high-throughput inter-satellite links, as well as a ground segment including one or several terrestrial gateways providing overall coordination and cloud-like computing capabilities (Fig. 1). In this scenario, the number of terminals within reception range of a satellite is orders of magnitude higher than ground-based IoT systems, resulting in a significant over-the-air
contention that cannot be resolved through classical methods. For these reasons, NTN would benefit from new dedicated protocols and waveforms while the first commercial generation of unsupervised M2M like LORA or Sigfox are not able to fully exploit the available channel capacity.
The objective of theWARM-M2M project is to develop novel physical layer (PHY) and medium access control (MAC) layers approaches, radio access protocols and distributed coordination mechanisms for massive M2M scenarios allowing multiple LEO satellites to jointly serve a massive number of nodes with sporadic traffic, under controlled reliability and/or latency constraints, achieving a high area spectral efficiency at the network scale, with limited IoT device complexity and protocol overhead.