Power Amplifier Design for Direct-to-Cell Satellite Multi-Beam Active Antenna Arrays

Direct-to-cell satellite communication has been receiving great interest as it plays an important role in future sixth generation (6G) networks. To enable direct-to-cell applications with multibeam support and high-throughput capability, large-scale active antenna arrays with thousands of elements are expected onboard satellites. The unavoidable mutual coupling between antenna elements requires study of the dynamic active reflection coefficients (ARC), corresponding to active loads. Considering that power amplifiers (PAs) matched to 50 Ω are typically used to feed the antenna elements, the active loads impose active load-pulling on the PAs, pulling them away from optimal performance regions. Traditional designs insert an isolator between the PA and the antenna element, forcing the load seen by the PA to be 50 Ω. However, the power delivered to the antenna element is not improved as the mismatching between the isolator and the active loads can also cause power reflections. Meanwhile, the isolator
naturally brings extra insertion loss. Moreover, when considering large-scale antenna arrays, bulky isolators (made from Ferrite materials) occupy considerable space and increase the hardware’s weight, posing significant challenges for satellites with stringent space and weight constraints. We proposed a novel PA design strategy where the PA was matched to the average ARC for the studied multi-beam active antenna array. Simulation results demonstrate that the effective delivered power, and the performance of the PA in terms of efficiency, can be improved. To advance the study experimentally, our current work focuses on the ARC measurement with and without PA using an eight-channel USRP platform. The ARC measurement results will be used as a benchmark for the novel PA matching network design and experimental verification.