基于优化神经网络的雷达波束分配方法研究  

A Study on the Radar Beam Allocation Method Based on the Optimized Neural Network

作　　者：朱光耀 张贞凯[1] ZHU Guang-yao;ZHANG Zhen-kai(School of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang 212003,China)

机构地区：[1]江苏科技大学电子与信息学院,江苏镇江212003

出　　处：《火力与指挥控制》2021年第1期88-93,共6页Fire Control & Command Control

基　　金：国家自然科学基金资助项目(61871203)。

摘　　要：在多目标的跟踪过程中,自适应地分配雷达波束能够进一步提升雷达的工作效率。为了合理调度雷达波束跟踪目标,根据目标运动状态参数,建立了目标威胁度评估模型。该模型根据目标运动状态,包含类型、速度、加速度、航向角、高度、距离和干扰程度等;然后基于混合遗传粒子群优化算法改进了BP神经网络,分别采用运动状态信息和目标威胁度作为神经网络的输入和输出,建立了目标威胁度评估模型;算法根据威胁度的大小进行波束调度,并采用执行威胁率衡量波束调度的效果。仿真结果表明,该模型具有更高的预测精度,能够准确地实现目标威胁评估,在此基础上具有较好的波束调度效果。During multiple target tracking process,the adaptive allocation of radar beams can further improve working efficiency of the radars.Firstly,a target threat degree evaluation model based on the target motion state parameters is established,which refers to the motion state information of the target,including target type,speed,acceleration,heading angle,altitude,distance,and interference level,etc.Secondly,the Back Propagation(BP)neural network is improved based on the hybrid genetic particle swarm optimization algorithm,the target threat degree evaluation model is built by regarding,the motion state information and target threat degree as the input and output of the neural network respectively.Finally,beam scheduling is realized by the algorithm according to the magnitude of the threat.The execution threat rate is employed as a metric to evaluate the effect of beam scheduling.The simulation results show that the model can accurately realize the target threat assessment with higher prediction accuracy and better beam scheduling effects.

关 键 词：威胁度评估 BP神经网络 粒子群优化 雷达波束调度 

分 类 号：TJ01[兵器科学与技术—兵器发射理论与技术] TN953[电子电信—信号与信息处理]