Global Energy Efficiency Optimization of a Ka-Band Multi-Beam LEO Satellite Communication System

We investigate the issue of global energy efficiency optimization in a multi-beam low Earth orbit (LEO) satellite communication system. Current terrestrial networks provide high-quality and low cost communication mainly in densely populated areas. However, the cost of extending wideband coverage to remote areas is unaffordable. LEO satellites provide a low cost solution for offering global coverage by supporting terrestrial networks. We consider downlink transmissions in Ka-band, where a LEO satellite transmits to access points or users directly. Impairments due to Ka-band channels, inter-beam interference as well as Doppler effects are taken into account in our mathematical model. We formulate the problem of jointly optimizing beam assignment and power allocation for maximizing global energy efficiency. Given the intractability of this problem, we propose to divide it into two subproblems: first, beam assignment optimization under fixed power per beam, and second, power allocation optimization under fixed beam assignment. We devise two algorithms, beam-wise power optimization and equal power optimization. These two algorithms solve the subproblems, beam assignment and power allocation, in a different way. Numerical results show that our proposed methods can greatly improve the global energy efficiency compared to the baseline method with a fixed power per beam.