如何通过人工智能方法发现GSMem恶意活动
密码学与安全
2018-01-17 v2
摘要
本文研究如下问题：如何有效地发现GSMem恶意活动。为此，本文提出了一种基于人工智能（AI）的新方法。首先，我们利用大量关于某些电磁波频率和幅值的数据来训练模型。然后，我们将给定的频率和幅值输入所得模型，预测是否发生了GSMem恶意活动。模拟实验表明，该新方法具有检测GSMem活动的潜力，且具有低误报率（FPR）和低漏报率（FNR）。
引用
@article{arxiv.1801.02440,
title = {How to find a GSMem malicious activity via an AI approach},
author = {WeiJun Zhu and ShaoHuan Ban and YongWen Fan},
journal= {arXiv preprint arXiv:1801.02440},
year = {2018}
}
备注
4 pages, 4 figures. We'd like to thank arXiv for giving us a chance to correct some typos in previous version of this paper. For examples, the words "FNR" and "FPR" in Fig.4 are marked incorrectly. And they should be replaced with each other. In addition, the experimental result obtained by a BP neural network is another typo in Fig.4. In this version, we check them carefully and correct them