ADMM-Net for Communication Interference Removal in Stepped-Frequency Radar
ADMM-Net for Communication Interference Removal in Stepped-Frequency Radar
复制标题
DOI：
10.1109/tsp.2021.3076900
复制
发表时间：
2020-09
影响因子：
5.4
通讯作者：
Jeremy Johnston;Yinchuan Li;M. Lops;Xiaodong Wang
中科院分区：
文献类型：
--
作者：
Jeremy Johnston;Yinchuan Li;M. Lops;Xiaodong Wang
复ADMM-Net是一种基于交替方向乘法器（ADMM）的复值神经网络结构，用于频率步进雷达超分辨角距离多普勒成像中的干扰抑制。我们考虑一个不合作的频谱共享的情况下，雷达的任务是成像稀疏场景中的通信干扰，是频率稀疏，由于频谱利用率不足，激发一个$\ell_1 $-最小化问题，以恢复雷达图像和抑制干扰。问题的ADMM迭代支持神经网络设计，产生一组具有可学习超参数和操作的广义ADMM更新。该网络使用根据雷达和通信信号模型生成的随机数据进行训练。在数值实验中，ADMM-Net表现出明显低于ADMM和CVX的误差和计算成本。
Complex ADMM-Net, a complex-valued neural network architecture inspired by the alternating direction method of multipliers (ADMM), is designed for interference removal in stepped-frequency radar super-resolution angle-range-doppler imaging. We consider an uncooperative spectrum sharing scenario where the radar is tasked with imaging a sparse scene amidst communication interference that is frequency-sparse due to spectrum underutilization, motivating an $\ell _1$-minimization problem to recover the radar image and suppress the interference. The problem's ADMM iteration undergirds the neural network design, yielding a set of generalized ADMM updates with learnable hyperparameters and operations. The network is trained with random data generated according to the radar and communication signal models. In numerical experiments ADMM-Net exhibits markedly lower error and computational cost than ADMM and CVX.