• DocumentCode
    774327
  • Title

    Multipath Fading Models and Adaptive Equalizers in Microwave Digital Radio

  • Author

    Wong, Wai Choong Lawrence ; Greenstein, Larry J.

  • Author_Institution
    AT&T Bell Labs., Holmdel, NJ, USA
  • Volume
    32
  • Issue
    8
  • fYear
    1984
  • fDate
    8/1/1984 12:00:00 AM
  • Firstpage
    928
  • Lastpage
    934
  • Abstract
    In the past few years, a great deal has been learned about the modeling and equalization of multipath fading in microwave digital radio systems. This paper adds a number of new results to this body of knowledge. We begin with a description of two published multipath fading models that were derived from the same collection of data, i.e., the threepath and polynomial models. We then derive outage probability characteristics for a digital radio link using M -level quadrature amplitude modulation, cosine rolloff spectral shaping, and some form of equalization. We obtain results for both models, several kinds of equalization, and various combinations of system parameters. We find that the theoretically optimal equalizer, which is nonlinear, is not measurably better in performance than the best of the linear equalizers, which is simpler. Also, the performance of a properly equalized receiver can be very accurately estimated by knowing just the probability distribution, over many fades, of the gain amplitude at a single frequency. Finally, we show the influences on performance of such parameters as cosine rolloff factor, symbol rate, and unfaded carrier-to-noise ratio.
  • Keywords
    Adaptive equalizers; Digital communications; Microwave radio propagation; Multipath channels; Adaptive equalizers; Degradation; Digital communication; Digital modulation; Dispersion; Fading; Frequency; Performance evaluation; Performance gain; Probability distribution;
  • fLanguage
    English
  • Journal_Title
    Communications, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0090-6778
  • Type

    jour

  • DOI
    10.1109/TCOM.1984.1096158
  • Filename
    1096158