A Blind Architecture for Signal Combining in Distributed Satellite Communications

To meet the increasing demand for high data rates in satellite communications (SATCOM), the requirements in terms of link quality and reliability are also increasing. Reception of SATCOM signals with multiple ground stations and combining them to increase the signal-to-noise ratio (SNR) provides a flexible and robust alternative to larger antenna dish sizes and higher transmit powers. In practical scenarios, the signals received at different ground stations may exhibit different time delays, phase/frequency offsets and noise powers. This might be due to movement of the satellite in low-earth orbit (LEO) and ground stations as well as temperature differences on the ground causing different thermal noise levels. When the transmitting satellite does not cooperate with the ground stations, the central signal processor, e.g., the combiner, has to blindly estimate and compensate for these impairments to achieve optimal combining gain. This paper proposes a blind combiner architecture that integrates time-delay synchronization, frequency synchronization, as well as blind diversity combining to work under practical conditions independently of the signal modulation. The architecture is evaluated in a LEO satellite overflight simulation and validated with real measurements from geostationary satellites. We further derive the theoretical requirements for the accuracy under the different impairments and how these impairments degrade the SNR at the combiner output. The proposed architecture shows a near-optimal SNR performance in both the simulation and measurement results.