• DocumentCode
    3427357
  • Title

    Symbol timing estimation in ultra wideband communications

  • Author

    Tian, Zhi ; Yang, Liuqing ; Giannakis, Georgios B.

  • Author_Institution
    Michigan Technol. Univ., Houghton, MI, USA
  • Volume
    2
  • fYear
    2002
  • fDate
    3-6 Nov. 2002
  • Firstpage
    1924
  • Abstract
    This paper takes on a cyclostationary approach to recover timing of ultra-wideband (UWB) transmissions over rich multipath environments. It is demonstrated that timing dependent cyclic statistics exist without resorting to oversampling, due to the symbol repetition pattern inherent to UWB modulation. Based on the received signal´s second-order cyclic statistics, non-data-aided time offset estimation algorithms are developed. Timing acquisition relies only on frame rate samples, while low-complexity tracking utilizes pulse rate samples. Both acquisition and tracking schemes are tolerant to UWB channel fading, and additive stationary (possibly colored) noise.
  • Keywords
    broadband networks; modulation; multipath channels; noise; parameter estimation; radio networks; signal sampling; statistical analysis; synchronisation; UWB channel fading; UWB modulation; UWB transmissions; additive stationary noise; cyclostationary approach; frame rate samples; low-complexity tracking; multipath environments; nondata-aided time offset estimation algorithms; pulse rate samples; received signal second-order cyclic statistics; short-range wireless communications; symbol repetition pattern; symbol timing estimation; timing acquisition; timing dependent cyclic statistics; timing recovery; ultra wideband communications; Additive noise; Amplitude modulation; Fading; Pulse modulation; Pulse shaping methods; Radar tracking; Sampling methods; Statistics; Timing; Ultra wideband technology;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Signals, Systems and Computers, 2002. Conference Record of the Thirty-Sixth Asilomar Conference on
  • Conference_Location
    Pacific Grove, CA, USA
  • ISSN
    1058-6393
  • Print_ISBN
    0-7803-7576-9
  • Type

    conf

  • DOI
    10.1109/ACSSC.2002.1197115
  • Filename
    1197115