IEEE Open Journal of the Communications Society (Jan 2024)

Multibeam Beamforming for Direct Radiating Arrays in Satellite Communications Using Genetic Algorithm

  • Juan Andres Vasquez-Peralvo,
  • Jorge Querol,
  • Flor Ortiz,
  • Jorge Luis Gonzalez Rios,
  • Eva Lagunas,
  • Luis Manuel Garces-Socorras,
  • Juan Carlos Merlano Duncan,
  • Marcele O. K. Mendonca,
  • Symeon Chatzinotas

DOI
https://doi.org/10.1109/OJCOMS.2024.3385021
Journal volume & issue
Vol. 5
pp. 2343 – 2357

Abstract

Read online

Recent advancements in onboard satellite communication have enhanced the capability of dynamically modifying the radiation pattern of a Direct Radiating Array (DRA). This is crucial not only for conventional communication satellites like Geostationary Orbit (GEO) but also for those in lower orbits such as Low Earth Orbit (LEO). Key design factors include the number of beams, beamwidth, Effective Isotropic Radiated Power (EIRP), and Side Lobe Level (SLL) for each beam. However, a challenge arises in multibeam scenarios when trying to simultaneously meet requirements for the aforementioned design factors which are reflected as uneven power distribution. This leads to over-saturation, especially in centrally located antenna elements due to the activation times per beam, commonly referred to as activation instances. In response to this challenge, this paper presents a method to balance the activation instances across antenna elements for each required beam. Our focus is on beams operating at 19 GHz on a CubeSat positioned 500 km above the Earth’s surface. We introduce a Genetic Algorithm (GA)-based algorithm to optimize the beamforming coefficients by modulating the amplitude component of the weight matrix for each antenna element. A key constraint of this algorithm is a limit on activation instances per element, which avoids over-saturation in the Radio Frequency (RF) chain. Additionally, the algorithm accommodates beam requirements such as beamwidth, SLL, pointing direction, and total available power. With the previous key design factors, the algorithm will optimize the required genes to address the desired beam characteristics and constraints. We tested the algorithm’s effectiveness in three scenarios using an $8\times 8$ DRA patch antenna with circular polarization, arranged in a triangular lattice. The results demonstrate that our algorithm not only meets the required beam pattern specifications but also ensures a uniform activation distribution across the antenna array.

Keywords