基于MIMO阵列的混合波束赋形算法

低轨(low earth orbit，LEO)卫星通信可以克服恶劣地形影响实现远距离通信，面对突发灾害能快速部署解决通信问题，但是，卫星在保持高速运动的同时采用高频段的信号，容易导致通信时产生严重的信号衰落。针对此问题，基于传统遗传算法提出一种适用于混合波束赋形架构的算法。固定模拟域对数字域进行迭代优化，然后固定数字域对模拟域进行迭代优化，通过共同的待优化矩阵进行联合求解，在固定功率的前提下不断迭代优化矩阵，采用混合波束赋形方法有效减少硬件复杂度，更加接近数字波束赋形的性能。仿真表明，在混合波束赋形架构应用上，相对于传统的正交匹配追踪(orthogonal matching pursuit,OMP)算法，所提算法性能提升了6.61%。

Hybrid beamforming algorithm based on MIMO array

Low earth orbit (LEO) satellite communications is capable of providing the long-distance communication even within harsh terrain, which can be deployed instantly to solve the communication interruption problem incurred by the unpredictable natural disaster. However, the high-speed movement of satellites and the adopted high-frequency signals lead to serious signal fading. Beamforming technology, based on large-scale multiple input multiple output (MIMO) antenna array, can mitigate signal fading effectively through adjusting the weighted coefficients of antenna array. This paper modifies the traditional genetic algorithm and further proposes an algorithm suitable for hybrid beamforming architecture. The fixed analog domain iteratively optimizes the digital domain, and then the fixed digital domain iteratively optimizes the analog domain. The joint solution is carried out through the common matrix to be optimized, and the matrix is optimized through continuous iteration under the premise of fixed power. The hybrid beamforming architecture is utilized to reduce hardware complexity, and the proposed algorithm has the performance closer to that of digital beamforming. Compared with the traditional o rthogonal matching pursuit (OMP) algorithm, the performance of the proposed algorithm applied on hybrid beamforming architecture is improved by 6.61%.