Memory Based Compensation of Phased Array Beamforming Antenna Gain Loss in LEO Satellite Systems Using Fourier Series

This paper proposes an efficient memory-based prediction method for the antenna gain loss of phased array beamforming antennas in LEO satellite communication systems during overflight situations. The gain loss is caused by the limited set of discrete antenna beam pointing vectors which leads to a nonoptimal antenna steering and thus to a time-varying receiving power. The algorithm uses a pre-computed look-up table (LUT) and the satellite orbit position for prediction. The main purpose of the predictor is to reduce the fixed C/N margin in adaptive coding and modulation systems by eliminating the time delay for signaling and switching of the modulation and code rate. This paper explains how to get the antenna coverage from measured overflights and use a Fourier series based memory allocation algorithm for on the fly predictions.

I. Introduction

Contemporary LEO satellite systems use discrete phased array beamforming antennas in satellites and ground terminals [1] . During the LEO overflight situation the LEO antenna beam needs to be continuously kept accurately pointed towards the ground terminals. Conversely, the terminal beams have to follow the satellite overflight trajectory. The antenna steering is typically realized by means of a discrete set of beam pointing vectors, which is caused by the limited beam resolution of the antenna. This always leads to a nonoptimal beam steering. Through the overflight the system controller selects the best suited pointing vector, which optimizes a given target function. Such a target function can be the average carrier to noise ratio (C/N) of all terminals or the maximum data rate among all terminals in range.