利用改进型AlexNet的ADS-B欺骗式干扰检测

ADS-B Spoofing Interference Detection Using Improved AlexNet

广播式自动相关监视(ADS-B)是一种新的空管监视技术,遵循着“空地一体化”和“全球可互用”的指导原则,实现了航迹信息共享。但其开放式的架构特点,使其极易受到各类欺骗式的干扰,严重威胁空中交通安全。本文针对真实ADS-B信号的多普勒频偏变化规律与报告位置的变化规律相符合的特点,结合以深度学习为代表的机器学习方法,提出利用改进型的AlexNet提取特征并检测欺骗干扰。本方法对比传统的信号处理方法,减少了计算复杂度,提高了识别准确率,特别是在航迹长度较短时优势更加明显。仿真实验验证了方法的有效性。

Automatic Dependent Surveillance-Broadcast(ADS-B)technology is a new air traffic surveillance technology that follows the guiding principles of“air-ground integration”and“global interoperability”to achieve track information sharing.However,its open architecture makes it extremely vulnerable to a variety of spoofing attacks,which seriously threatens air traffic safety.In this paper,aiming at the characteristics of Doppler frequency offset variation of ADS-B message and the change rule of reporting position,combined with the machine learning method represented by deep learning,this paper proposes to use the improved AlexNet extraction feature and detect fraud interference.Compared with the traditional signal processing method,the method reduces the complexity of the algorithm and improves the recognition accuracy,especially when the track length is short.Simulation experiments verify the effectiveness of the method.