1st International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications

Research Article

Cyclostatilonarilty-Based Test for Detection of Vacant Frequency Bands

  • @INPROCEEDINGS{10.1109/CROWNCOM.2006.363454,
        author={Mohamed  Ghozzi and Francois  Marx and Mischa  Dohler and Jacques  Palicot},
        title={Cyclostatilonarilty-Based Test for Detection of Vacant Frequency Bands},
        proceedings={1st International ICST Conference on Cognitive Radio Oriented Wireless Networks and Communications},
        publisher={IEEE},
        proceedings_a={CROWNCOM},
        year={2007},
        month={5},
        keywords={Cognitive radio Detectors Frequency estimation Matched filters RF signals Radiometry Research and development Signal detection Telecommunications Testing},
        doi={10.1109/CROWNCOM.2006.363454}
    }
    
  • Mohamed Ghozzi
    Francois Marx
    Mischa Dohler
    Jacques Palicot
    Year: 2007
    Cyclostatilonarilty-Based Test for Detection of Vacant Frequency Bands
    CROWNCOM
    IEEE
    DOI: 10.1109/CROWNCOM.2006.363454
Mohamed Ghozzi1,2,*, Francois Marx1,*, Mischa Dohler1,*, Jacques Palicot2,*
  • 1: France Telecom R&D, France.
  • 2: Supelec-campus de Rennes, France
*Contact email: mohamed.ghozzi@francetelecom.com, francois.marx@francetelecom.com, mischa.dohler@francetelecom.com, jacques.palicot@supelec.fr

Abstract

The problem addressed in this paper is the test for presence of any telecommunication signal within a given frequency band, which is of great importance for cognitive radios. Since a telecommunication signal is well modeled as cyclostationary, we transform this problem to the one of testing for presence of the cylostationary property over a range of cyclic frequencies and we propose a constant false alarm rate (CFAR) multicycle detector based on a chi-squared (x2) statistical test of the time varying covariance estimate