Frame detection and timing acquisition for OFDM transmissions with unknown interference

Luca Sanguinetti, Michele Morelli, H. Vincent Poor

Research output: Contribution to journalArticlepeer-review

25 Scopus citations


Frame detection and timing acquisition are challenging tasks in orthogonal frequency-division multiplexing systems plagued by narrowband interference (NBI). Most existing solutions operate in the time domain by exploiting the repetitive structure of a training symbol and suffer from considerable performance loss in the presence of NBI. In this work, a novel solution in which frame detection is accomplished in the frequency domain on the basis of a suitable likelihood ratio test is presented. In order to increase the resilience to NBI, the interference power is treated as a nuisance parameter that is averaged out from the corresponding likelihood functions. The resulting test statistic depends on the fractional carrier frequency offset (CFO), which is easily estimated. An alternative scheme that dispenses from CFO estimation is also proposed. After frame detection, the test statistic is employed as a timing metric to accurately locate the position of the training symbol within the received data stream. Computer simulations indicate that the proposed solutions are remarkably robust to NBI and outperform existing alternatives in a severe interference scenario. The price for this advantage is a substantial increase in the computational burden.

Original languageEnglish (US)
Article number5427454
Pages (from-to)1226-1236
Number of pages11
JournalIEEE Transactions on Wireless Communications
Issue number3
StatePublished - Mar 2010

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Electrical and Electronic Engineering
  • Applied Mathematics


  • Frame and timing synchronization
  • Generalized likelihood ratio test (GLRT)
  • Inverse-gamma distribution
  • Narrowband interference
  • OFDM


Dive into the research topics of 'Frame detection and timing acquisition for OFDM transmissions with unknown interference'. Together they form a unique fingerprint.

Cite this