CE-Optimized Cross-Beam Combining for GEO Satellite in Space-Air-Ground Integrated Network

Geostationary Earth orbit (GEO) satellites constitute critical components in space-air-ground integrated network (SAGIN). This paper presents a cross-beam signal combining framework that leverages beam-overlapping regions using direct sequence spread spectrum (DSSS) in multi-beam GEO satellite systems. By synchronously transmitting identical information across multiple beams and performing coherent combining at the receiver, the proposed scheme effectively enhances the system's low-probability-of-detection (LPD) and anti-jamming capabilities. To mitigate phase offsets and channel gain mismatches caused by transponder non-idealities and beam-edge effects, a cross-entropy (CE)-optimized joint estimation algorithm is proposed for simultaneous estimation of multiple parameters across signal branches. Simulation results demonstrate that the proposed algorithm achieves notable improvements in signal-to-noise ratio (SNR), estimation accuracy, and computational efficiency, confirming its robustness and scalability.