基于超分辨率重建的外源雷达实测信道估计

Super-resolution reconstruction-based channel estimation for measured data of passive radar

针对基于DRM(数字调幅广播)的外辐射源雷达实测参考信道估计面临的精度不足问题,研究了有限实测样本下的超分辨率重建网络估计方法.该方法在深度网络设计上通过将LS(最小二乘)信道估计方法获得的频域信道响应视为低分辨率的图像,经VDSR(非常深的超分辨重建网络)重建为高精度信道响应,最终实现对参考信号的准确估计;在训练数据集构建上,由于DRM外辐射源雷达实测信道估计中难以获取大量对应不同信道特征的实测数据,首先基于实测数据通过LS估计出时域信道响应,粗略确定信道中多径的大致时延和增益,然后在此基础上基于DRM波形特征仿真得到足够的信道数据,训练所提超分辨率重建网络,最后使用预训练模型预测实测数据.实测结果表明:本文方法可以达到比传统信道估计方法更高的精度.

In order to solve the problem of insufficient accuracy in the measured reference channel estimation for DRM(digital radio mondial)based passive radar,this paper studies a super-resolution reconstruction network-based estimation method under limited measured samples.On the design of the deep network,the frequency domain channel response obtained by the LS(least squares)channel estimation method is regarded as a low-resolution image,and is reconstructed into a high-precision channel response through VDSR(very deep super-resolution reconstruction)network,and finally the accurate estimation of the reference signal is obtained.On the construction of the training data set,due to the difficulty of obtaining a large amount of measured data corresponding to different channel environments in DRM-based passive radar,the time domain channel response is first estimated based on the measured data by the LS method to roughly determine the approximate delays and gains of the multipath in the channel.Then,based on the simulation of DRM waveform,enough training data with different channel environments is obtained to train the proposed super-resolution reconstruction network.Finally,the pre-trained model is used to predict the measured data.The experimental results show that the proposed method can achieve higher accuracy than traditional channel estimation methods.