A Survey of Techniques for the Identification of Mobile Phones Using the Physical Fingerprints of the Built-In Components

被引:84
作者
Baldini, Gianmarco [1 ]
Steri, Gary [1 ]
机构
[1] European Commiss, Joint Res Ctr, I-21027 Ispra, Italy
关键词
Fingerprint recognition; machine learning; radiofrequency identification; counterfeiting; security; telephone equipment; DIGITAL CAMERA IDENTIFICATION; SENSOR FINGERPRINT; FEATURE-EXTRACTION; WIRELESS SECURITY; DEVICE; AUTHENTICATION; MODEL; FORENSICS; NOISE; VERIFICATION;
D O I
10.1109/COMST.2017.2694487
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
In recent years, several research studies have investigated the identification of electronic devices through their physical components and properties, both from a theoretical point of view and through extensive experimental studies. Results have shown that, in many cases, a very high identification accuracy can be obtained by exploiting imperfections and small differences in the electronic components, which are called fingerprints in this context. Part of these studies have focused on a specific category of electronic device, the mobile phone or smartphone, which is usually equipped with components, such as radio frequency front-ends, cameras, micro-electro-mechanical systems, microphones, and speakers that are likely to reveal fingerprints in their digital outputs and then allow the identification of the component and of the mobile phone itself. Keeping the focus on mobile phones, this paper provides a survey of the different techniques for mobile phone identification on the basis of their built-in components. This paper describes the methodology, the classification algorithms, and the types of features that are typically used in literature. Outstanding challenges and research issues are also identified and described, together with an overview of the potential applications of mobile phone fingerprinting. In addition, this paper analyzes the potential privacy risks associated to the tracking of the mobile phone on the basis of its fingerprints and the related mitigation techniques. Finally, it summarizes the main issues and identifies research opportunities and potential future trends for mobile phone fingerprinting.
引用
收藏
页码:1761 / 1789
页数:29
相关论文
共 145 条
[1]   Machine Learning in Wireless Sensor Networks: Algorithms, Strategies, and Applications [J].
Abu Alsheikh, Mohammad ;
Lin, Shaowei ;
Niyato, Dusit ;
Tan, Hwee-Pink .
IEEE COMMUNICATIONS SURVEYS AND TUTORIALS, 2014, 16 (04) :1996-2018
[2]  
Aggarwal Rachit, 2014, 2014 International Conference on Communications and Signal Processing (ICCSP), P1218, DOI 10.1109/ICCSP.2014.6950045
[3]  
Alpaydin E, 2014, ADAPT COMPUT MACH LE, P115
[4]   A Systematic Comparison of Supervised Classifiers [J].
Amancio, Diego Raphael ;
Comin, Cesar Henrique ;
Casanova, Dalcimar ;
Travieso, Gonzalo ;
Bruno, Odemir Martinez ;
Rodrigues, Francisco Aparecido ;
Costa, Luciano da Fontoura .
PLOS ONE, 2014, 9 (04)
[5]  
[Anonymous], SCE0812 CARL U DEP S
[6]  
[Anonymous], 2006, Advances in Neural Information Processing Systems, DOI DOI 10.7551/MITPRESS/7503.001.0001
[7]  
[Anonymous], 2012, Military Communications Conference
[8]  
[Anonymous], 2009, 2009 IEEE INT C COMM, DOI DOI 10.1109/ICC.2009.5199451
[9]  
[Anonymous], P IEEE INT C AC SPEE
[10]   A Survey of Security Challenges in Cognitive Radio Networks: Solutions and Future Research Directions [J].
Attar, Alireza ;
Tang, Helen ;
Vasilakos, Athanasios V. ;
Yu, F. Richard ;
Leung, Victor C. M. .
PROCEEDINGS OF THE IEEE, 2012, 100 (12) :3172-3186