AI-Driven Integration of Terrestrial and Non-Terrestrial Networks 

The evolution of communication networks is entering a pivotal era, marked by 3GPP Release 17, which has empowered 5G operators to extend their services beyond terrestrial boundaries. This expansion reaches far beyond traditional communication networks, impacting not only remote communities but also maritime, airborne, and other isolated environments. The introduction of Non-Terrestrial Networks (NTN), which includes Uncrewed Aerial Vehicles (UAVs), High Altitude Platform Stations (HAPS), and satellites, brings unprecedented coverage capabilities to the telecommunications industry. This enables a wide array of applications—such as Machine-to-Machine (M2M) communication, emergency response, and enhanced connectivity for high-speed platforms (airplanes, trains, ships)—across various sectors like agriculture, transportation, environmental monitoring, and asset tracking. However, the integration of NTN into Terrestrial Networks (TN) presents unique technical and regulatory challenges. Unlike traditional TN base stations, satellites, especially those in Low Earth Orbit (LEO), move at high velocities, introducing complexities like Doppler shift and frequency variations. Addressing these issues requires compensating for satellite mobility and ensuring the reliability of user devices. Moreover, NTN systems face challenges related to higher path loss due to extended signal paths through the atmosphere, which affects latency and network capacity. These networks must carefully balance resource allocation based on real-time system dynamics and user demand, and spectrum-sharing between NTN and TN demands efficient dynamic spectrum access strategies. Collaboration among satellite operators, mobile network providers, government agencies, and standards bodies will be essential to overcoming these regulatory and technical hurdles. 

One of the most exciting developments in this space is the direct-to-device communication paradigm, especially the prospect of direct-to-smartphone connectivity. Historically, the NTN narrative focused on providing coverage for rural and remote areas, but recent advancements in satellite mega constellations and HAPS networks have expanded its potential dramatically. By the 2030s, these systems could be powerful enough to communicate directly with 6G smartphones, smart glasses, and other consumer devices, enabling high-end use cases like video streaming across a global $10 trillion market. With an estimated 10 billion smartphones and 100 billion Internet of Things (IoT) devices globally, this shift will impact everyone, far beyond rural and remote areas. The direct-to-device approach has transformative potential and poses a significant challenge to traditional Mobile Network Operators (MNOs), potentially disrupting the current telecom business model.

Achieving seamless integration of NTN into TN for such advanced use cases requires more than just overcoming physical and regulatory challenges. Wireless devices that were initially designed for TN will require new architectures to establish stable and efficient links with satellite systems. Here, Artificial Intelligence (AI) becomes a critical enabler. AI-supported methods can optimize future 6G networks, allowing them to function effectively in dynamic, unpredictable environments. By analyzing large datasets and making real-time decisions, AI can help manage the complexities of TN-NTN integration, ensuring the networks adapt to changing conditions and user demands. The effectiveness of these AI models, however, relies on the availability of high-quality training data. 

This Special Interest Group (SIG) serves as a platform for researchers, engineers, and professionals to explore the role of AI in bridging TN and NTN, overcoming the technical challenges of integration, and examining the disruptive potential of direct-to-device satellite communications. Together, we aim to shape the future of global telecommunications by facilitating discussions, research, and innovation in these transformative technologies.

Motivated by current trends in the field of integrated TN and NTN using latest AI algorithms,  the topics of interest include the following areas, but not limited to:

• Al enabled Beamforming techniques for energy efficient Integrated TN-NTN networks.

• Multiple access schemes, e.g., RSMA, NOMA for integrated TN-NTN systems.

• Al enabled Backscatter communications for Integrated TN-NTN networks.

• Energy harvesting, wireless information, and power transfer for integrated TN-NTN networks.

• Al enabled intelligent reflecting surface design for Integrated TN-NTN networks.

• Holographic MIMO for Integrated TN-NTN networks.

• PHY algorithms and protocol designs for integrated TN-NTN.

• AI enabled optimization-oriented designs for Integrated TN-NTN.

• Satellite communications and networking, e.g., LEOs, ground-to-space, space-to-ground links.

• AI enabled cost and power efficient design of Integrated TN-NTN networks.

• AI backed security, privacy, and interference exploitation challenges in integrated TN-NTN based systems.

• AI algorithms for joint TN and NTN resource allocations.

• Generative AI for TN and NTN.

Link to SSC TC SIG :AI-Driven Integration of Terrestrial and Non-Terrestrial Networks (AITNTN)