Purdue University Graduate School
Browse
Thesis-MS-Shaan.pdf (8.69 MB)

Side-Channel Attacks on Encrypted 5G/4G Voice Calls

Download (8.69 MB)
thesis
posted on 2024-05-01, 13:52 authored by Shaan ShekharShaan Shekhar

5G/4G voice calls are encrypted for the purpose of confidentiality, secrecy and privacy. Although protected by well-examined security measures we unveil several vulnerabilities previously unreported in the 5G/4G voice calls that unintentionally leak 5G/4G call state information despite encryption protection and device proof of concept attacks in this thesis. Unlike existing attacks, these new attacks are significantly more threatening because they are completely contactless without requiring any malware, access or compromise on the victim's phones, the 5G/4G network and the other call party. Instead, the attacker only needs to deploy a radio sniffer to eavesdrop on 5G/4G communication and infer confidential call information.

Interestingly, such confidentiality breaches are technically feasible due to recent 5G/4G call enhancement technologies standardized in the 3GPP specifications and adopted by mobile network operators. While effective in enhancing 5G/4G call quality and efficiency, they, unfortunately, expose extra call information, which can be exploited to infer call states and launch side-channel attacks precisely. Another major contributor to this attack is the IVR technology, which uses a computer-operated telephone system to help companies answer customer calls. In this thesis, we focus on snooping Pay-over-the-Phone transactions done over IVR calls and optionally inferring the company involved in the transaction. The attacks exploit technologies designed to enhance the call quality and efficiency and develop several attack modules to (1) detect voice calls over encrypted 5G/4G traffic, (2) infer the use of IVR over limited call information leaked in the air, and (3) spy on sensitive payment transactions in real-time. We have implemented this proof-of-concept attack using an SDR-based sniffer only. We have validated its effectiveness and assessed damages in various experiments with 5G operators in the US. Lastly, we have discussed the lessons learned from the attacks and the future work that can be done to improve the efficiency of the attacks and make them more threatening.

History

Degree Type

  • Master of Science

Department

  • Computer Science

Campus location

  • West Lafayette

Advisor/Supervisor/Committee Chair

Dr. Chunyi Peng

Additional Committee Member 2

Dr. Vasileios Zikas

Additional Committee Member 3

Dr. Zeynel Berkey Celik

Usage metrics

    Licence

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC