Abstract：The netted radar has widely distributed radar nodes, which can deploy radar nodes by tasks and demands. It is a current research hotspot. This paper aims to study different beam pointings and predict the one-dimensional array node combination through neural network to reduce the sidelobe of the beam pointing. The dataset generated by genetic algorithm (GA) is fed to convolutional neural networks (CNN) for training, and the trained CNN can quickly predict. In this paper, a node selection algorithm based on GA-CNN for reducing array sidelobe is proposed, which combines the advantages of GA in dealing with combinatorial explosion problems and deep learning which has good generalization ability and prediction speed, and searches and predicts in the full set. It can be seen from the simulation results of the linear array that the CNN has learned the partial correspondence between the beam pointing and the node selection, and the operation speed is greatly improved. It makes the radar have further research value in the efficient adaptable environment.
陶海红, 闫莹菲. 一种基于GA-CNN的网络化雷达节点遴选算法[J]. 空天防御, 2022, 5(1): 1-5.
TAO Haihong, YAN Yingfei. A Netted Radar Node Selection Algorithm Based on GA-CNN. Air & Space Defense, 2022, 5(1): 1-5.
