It has been nearly 60 years since the launch of Intelsat-I, the world’s first commercial satellite communications system, and satellite communications have evolved to offer broadcast, fixed satellite, and mobile satellite services. As 5G base stations are widely deployed, cellular networks still cover less than 40% of the terrestrial surface, making mobile satellite Internet a key complement for 6G’s seamless global coverage. Low-Earth-orbit (LEO) satellites have emerged as promising candidates due to low launch costs, transmission latency, and path loss, with examples such as SpaceX’s Starlink aiming to deploy over 42000 satellites and completing its first direct-to-cell (DTC) constellation in December 2024.

Satellite-to-ground wireless transmission faces fundamental limitations. With link distances ranging from 300 to 36000 km, path loss is significantly larger than in terrestrial cellular communications (typically under 10 km). The downlink transmission rate is governed by a link budget formula involving payload effective isotropic radiated power (EIRP,) user terminal (UT) noise–temperature ratio (G/T) ratio, carrier frequency, satellite-to-ground distance, and demodulation threshold. Payload EIRP, tied to satellite power supply and antenna aperture, is constrained by manufacturing costs and ITU power-flux density limits. UT (G/T) ratio is limited by portability demands, as miniaturized antennas restrict aperture size despite advances in low-noise amplifiers.

Frequency band selection presents tradeoffs: Ka/Ku bands offer large bandwidth but require costly, less portable terminals, while L/S bands support compact terminals and terrestrial-satellite integration but face spectrum scarcity. Orbital altitude balances coverage, constellation scale, and handover frequency, with LEO satellites needing ultra-dense constellations. Waveform design must address dynamic channel fading, Doppler shifts, and long propagation delays, which challenge conventional cellular technologies such as orthogonal frequency-division multiplexing (OFDM) and hybrid automatic repeat request (HARQ).

Key enabling technologies address these limitations. Extremely large antenna arrays (ELAAs) use beamforming to boost EIRP and spatial multiplexing, with beam pattern-oriented and user-specific approaches balancing complexity and performance. Mobility management, critical for LEO’s rapid movement, employs conditional handover and multi-connectivity to handle frequent beam shifts. Multi-satellite cooperative transmission (MSCT) in spatial, time-frequency, or multi-domain modes aggregates resources, with carrier aggregation and multi-connectivity leveraging existing cellular technologies. Advanced UT antennas, either dedicated or reusing cellular multiple-input multiple-output (MIMO) arrays, enhance gain while maintaining compactness.

Quantitative analyses show doubling ELAA aperture increases transmission rates by 3–4 Mbps for leading systems, while MSCT schemes like multi-connectivity and distributed beamforming boost rates by 44%–76%. As the industry advances, balancing UT optimization, ELAA efficiency, and MSCT maturity will be pivotal for realizing mobile satellite Internet’s potential, aligning with the vision of unified terrestrial and non-terrestrial networks in 6G.

The paper “Toward Mobile Satellite Internet: The Fundamental Limitation of Wireless Transmission and Enabling Technologies,” is authored by Wenjin Wang, Yiming Zhu, Yafei Wang, Rui Ding, Symeon Chatzinotas. Full text of the open access paper: https://doi.org/10.1016/j.eng.2025.07.007. For more information about Engineering, visit the website at https://www.sciencedirect.com/journal/engineering.