Comb-type pilot-aided OFDM channel estimation for underground WLAN communications

by Yuan, Xin

Most analytical studies of Orthogonal Frequency Division Multiplexing (OFDM) channel estimation for fading channels assume a constant number of multiple paths. However, in an underground environment, the number of significant multipath components can be variable and follow a spécifie distribution. This thesis addresses the problem of channel estimation based on a wideband statistical channel model in a mining environment, where the number of arrivai paths follows a modifïed Poisson distribution and their amplitudes undergo Rayleigh or Rician fading. Channel estimations for OFDM Systems are based on a comb-type pilot arrangement, which is divided into pilot signal estimation and channel interpolation. The pilot signal estimation based on Least Mean Square (LMS) or Least Square (LS) criteria, together with a variety of interpolation schemes, is studied. Performances for ail conditions are presented and compared by measuring the Bit Error rate (BER) with différent modulation schemes derived from the IEEE802.il WLAN standard.