Spatial Ultra-Sparse Array Formation on LEO Distributed Satellite Cluster: An Enhanced Hybrid Particle Swarm Method

Abstract

The rapid development of Direct-to-device (D2D) services has put forward higher requirements for the performance of satellite antenna systems. The Spatial Ultra-Sparse Distributed Array (SUSDA) constructed by Distributed Satellite Cluster (DSC) has the characteristics of strong directivity, high flexibility and strong anti-jamming ability, which can better meet the communication requirements in future D2D scenarios. However, the non-uniform arrangement of SUSDA leads to the increase of the side lobe level (SLL) and the decrease of the overall antenna performance. To solve this problem, this paper proposes for the first time a configuration design method for a Low Earth Orbit (LEO) SUSDA capable of supporting D2D services in future 6G scenarios. It constructs a mathematical model related to the configuration design of the LEO SUSDA and provides a rapid prediction of the performance of the SUSDA radiation pattern function based on a probabilistic model. Then, an Enhanced Hybrid Particle Swarm Optimization (EHPSO) algorithm is proposed to solve the configuration design problem, which overcomes the slow convergence problem of traditional HPSO algorithm particularly when the array scale is large. The EHPSO algorithm adapts to the search requirements of different stages by adjusting parameters adaptively. It introduces a single suboptimal particle solution to enhance competition and cooperation among particles and employs a local search strategy to precisely narrow the search domain. Simulation results show that the algorithm can significantly reduce the number of iterations and running time of the algorithm while ensuring computational accuracy, which provides a new solution to the configuration design problem of large-scale LEO SUSDA in the future.