差集与群智能相融合的稀疏天线阵合成方法研究

Thinned antenna arrays have wide applications in communications, radars and remote sensing because of their advantages in costs, weight, consumption, and hardware complexity. Aiming at the drawbacks of the previous methods in convergence, computational costs, and global optimum, a methodology combining difference sets and swarm intelligence (DSSI) for thinned array synthesis is developed in this project. First, by investigating the relationship between difference sets (DSs) sequence and array synthesis objective, a DSs-based thinned antenna array model is constructed. Then the DSSI hybrid optimization strategies are investigated, including population initialization using DSs array model, embedding of DSs model in multi-subgroup structure, and simplifying parameter control using DSs model. Based on these strategies, a DSSI methodology is developed. Finally, the convergence of the DSSI methodology is studied by using Markov-chain model and numerical experiment, and a statistical evaluation system of DSSI convergence is established by analyzing the convergence parameters such as convergence rate, convergence precision and time cost. The DSSI methodology developed in this project will provide reliable and valuable results for antenna engineering, especially for the design of antenna arrays defined over wide apertures, and will be of great importance for the applications of array antenna in the fields such as communications, radars, remote sensing, radio astronomy,and biomedical imaging.

稀疏天线阵由于其成本、重量、功耗、硬件复杂度等方面的优势，在通信、雷达、遥感等领域具有广泛的应用。针对现有稀疏阵列设计方法存在的收敛性、计算代价以及全局寻优精度等方面的问题，本课题拟发展一套差集与群智能相融合（DSSI）的稀疏天线阵合成方法：首先，在探寻差集序列特性与稀疏阵列合成目标之间的关联规律的基础上，建立差集稀疏阵列模型；然后，深入研究差集阵列模型与群智能模型的融合策略，包括种群差集序列初始化、子种群差集序列嵌入、差集简化参数控制方式等，进而发展一套完整的DSSI方法；最后，采用Markov链数学模型和数值实验的手段研究DSSI方法的收敛性，建立一套涵盖收敛速率、收敛精度、时间复杂度等参数的DSSI收敛性统计评估体系。本项目研究成果可为实际的天线工程设计（尤其大规模阵列稀疏）提供可靠的理论依据和有价值的参考结果，从而推动阵列天线在通信、雷达、遥感、射电天文、医学成像等领域的深入应用。

稀疏天线阵由于其成本、重量、功耗、硬件复杂度等方面的优势，在通信、雷达、遥感等领域具有广泛的应用。针对现有稀疏阵列设计方法存在的收敛性、计算代价以及全局寻优精度等方面的问题，本课题发展了一套差集与群智能（DSSI）相融合的稀疏天线阵合成方法通用框架：首先，在探寻差集序列特性与稀疏阵列合成目标之间的关联规律的基础上，建立差集稀疏阵列模型；然后，深入研究差集阵列模型与群智能模型的融合策略，包括种群差集序列初始化、子种群差集序列嵌入、差集简化参数控制方式等，进而发展一套完整的DSSI方法；最后，采用数学分析和数值实验的手段研究DSSI方法的收敛性，建立一套涵盖收敛速率、收敛精度、时间复杂度等参数的DSSI收敛性统计评估体系。将所提DSSI方法应用于最小冗余MIMO雷达虚拟阵列合成和MIMO雷达方向图合成等问题中，理论分析和数值实验结果均表明，DSSI方法在寻优精度、收敛速率和时间代价等方面均优于标准群智能算法、标准差集算法。本项目研究成果可为实际的天线工程设计（尤其大规模阵列稀疏）提供可靠的理论依据和有价值的参考结果，从而推动阵列天线在通信、雷达、遥感、射电天文、医学成像等领域的深入应用。