Swift Jamming Attack on Frequency Offset Estimation: The Achilles' Heel of OFDM Systems

Hanif Rahbari, Marwan Krunz, Loukas Lazos

Research output: Contribution to journalArticlepeer-review

25 Scopus citations


Frequency offset (FO) refers to the difference in the operating frequencies of two radio oscillators. Failure to compensate for the FO may lead to decoding errors, particularly in OFDM systems. To correct the FO, wireless standards append a publicly known preamble to every frame before transmission. In this paper, we demonstrate how an adversary can exploit the known preamble structure of OFDM-based wireless systems, particularly IEEE802.11a/g/n/ac, to launch a very stealth (low energy/duty cycle) reactive jamming attack against the FO estimation mechanism. In this attack, the adversary quickly detects a transmitted OFDM frame and subsequently jams a tiny part of the preamble that is used for FO estimation at the legitimate receiver. By optimizing the energy and structure of the jamming signal and accounting for frame detection timing errors and unknown channel parameters, we empirically show that the adversary can induce a bit error rate close to 0.5 , making the transmission practically irrecoverable. Such vulnerability to FO jamming exists even when the frame is shielded by efficient channel coding. We evaluate the FO estimation attack through simulations and USRP experimentation. We also propose three approaches to mitigate such an attack.

Original languageEnglish (US)
Article number7163332
Pages (from-to)1264-1278
Number of pages15
JournalIEEE Transactions on Mobile Computing
Issue number5
StatePublished - May 1 2016


  • IEEE802.11
  • OFDM
  • PHY-layer security
  • USRP implementation
  • frequency offset
  • reactive jamming

ASJC Scopus subject areas

  • Software
  • Computer Networks and Communications
  • Electrical and Electronic Engineering


Dive into the research topics of 'Swift Jamming Attack on Frequency Offset Estimation: The Achilles' Heel of OFDM Systems'. Together they form a unique fingerprint.

Cite this