面向可变信道环境的真实射频信号数据集构建  

A Real-world Radio Frequency Signal Dataset Based on LTE System and Variable Channels

作　　者：张树鹏 陈啸锋 陆智怡 陈雪梅 孙金龙 桂冠[1] ZHANG Shupeng;CHEN Xiaofeng;LU Zhiyi;CHEN Xuemei;SUN Jinlong;GUI Guan(College of Telecommunications and Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210023,China;Nanjing Great Information Technology Co.,Ltd.,Nanjing 210046,China;Advanced Technology Research Institute of Beijing Institute of Technology,Beijing Institute of Technology,Beijing 100081 China)

机构地区：[1]南京邮电大学通信与信息工程学院,江苏南京210023 [2]南京桂瑞得信息科技有限公司,江苏南京210046 [3]北京理工大学前沿技术研究院,北京100081

出　　处：《无线电通信技术》2023年第2期248-254,共7页Radio Communications Technology

基　　金：山东省重点研发计划项目(2020CXGC010118);济南市空天信息产业专项,空天地一体化综合智联——北理前沿“双链”创新(QYYZF1112023105)。

摘　　要：基于深度学习的射频指纹(Radio Frequency Fingerprinting, RFF)识别具有增强物理层安全性能的潜力。近年来,为了满足深度学习对大规模数据的需求,提出了几种RFF数据集。然而,这些数据集是从类似的信道环境中收集的,多数仅提供来自接收器的接收数据。针对上述问题,利用软件无线电设备作为无线电信号发生器,通过自定义收发射机系统参数,如频带、调制模式、天线增益等,实现射频信号数据集的个性化定制。由于数据集是通过各种复杂的信道环境生成的,旨在更好地描述现实世界中的射频信号,因此在发射机和接收机处同时收集数据,可以模拟基于长期演进(Long Term Evolution, LTE)的真实RFF数据集。此外,通过一个基于卷积神经网络的射频指纹识别例程,验证了数据集的可用性,所提出的数据集和相关代码均可以在GitHub下载。Radio Frequency Fingerprinting(RFF)identification on account of deep learning has the potential to enhance the security performance of the physical layer.In recent years,several RFF datasets have been proposed to satisfy the demand of large-scale data for deep learning.However,these datasets are collected from similar channel environments and only provide receiving data from the receiver.In order to solve these problems,this paper utilizes a software radio peripheral as a radio signal generator to customize RFF signal data set by customizing system parameters of the receiver and transmitter,such as frequency band,modulation mode,antenna gain,etc.In addition,the proposed dataset is generated through various and complex channel environments,which aims to better characterize radio frequency signals in the real world.We collect the dataset at transmitters and receivers to simulate a real-world RFF dataset based on Long Term Evolution(LTE).Furthermore,we verify the dataset and confirm its reliability by a simple convolutional neural network.Both the dataset and reproducible code of this paper can be downloaded from GitHub.

关 键 词：射频指纹数据集 软件无线电设备 深度学习 设备识别 

分 类 号：TN92[电子电信—通信与信息系统]