5G场景下面向位置安全的异常检测算法
Anomaly Detection Algorithms for Location Security in 5G Scenarios
摘要:
基于位置的服务预计将在下一代蜂窝网络中发挥重要作用,从即将到来的5G系统开始。同时,定位技术可能会受到攻击者的严重影响,攻击者能够部署低成本的假基站并使用它们来改变定位信号。在本文中,我们具体集中在两类威胁:噪声干扰,其目标是降低信噪比,欺骗/meaconing攻击,其目标是注入虚假或错误的信息到接收器。然后,我们制定的检测问题的二元假设检验,并解决他们诉诸广义似然比检验设计过程以及潜在变量模型,其中涉及的期望最大化算法来估计未知的数据分布参数。所提出的技术可以应用于一个大类的位置数据,无论所包含的网络架构。性能分析是通过使用文献中的测量模型生成的模拟数据进行的,并强调了所提出的方法在检测上述类别的攻击的有效性。
英文摘要:
Location based services are expected to play a major role in future generation cellular networks, starting from the incoming 5G systems. At the same time, localization technologies may be severely affected by attackers capable to deploy low cost fake base stations and use them to alter localization signals. In this paper, we concretely focus on two classes of threats: noise-like jammers, whose objective is to reduce the signal-to-noise ratio, and spoofing/meaconing attacks, whose objective is to inject false or erroneous information into the receiver. Then, we formulate the detection problems as binary hypothesis tests and solve them resorting to the generalized likelihood ratio test design procedure as well as the Latent Variable Models, which involves the expectation-maximization algorithm to estimate the unknown data distribution parameters. The proposed techniques can be applied to a large class of location data regardless the subsumed network architecture. The performance analysis is conducted over simulated data generated by using measurement models from the literature and highlights the effectiveness of the proposed approaches in detecting the aforementioned classes of attacks.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。