Analysis and Performance Evaluation of a K-Symbol Pilot Assisted Channel Estimator using Linear Interpolation for OFDM Systems

Abstract

In this work, the performance of a novel linear interpolator scheme that uses K successive orthogonal frequency division multiplexing (OFDM) symbols is examined. The estimator collects K successive OFDM symbols to create a virtual super symbol (VSS) with K-P pilots using P pilots from each OFDM symbol. The (K - 1) middot P pilots are shifted in frequency in such way that in the VSS their distance is divided by K compared to standard symbols. Successive least squares (SLS) channel estimation through linear interpolation is performed by using more than one symbol in the LS estimator introducing the weight channel change (WCC) factor that represents the change of the wireless channel between the successive symbols used. A novel analysis deriving exact expressions for the mean square error (MSE) as a function of pilot spacing, channel statistics and WCC for an OFDM system where data subcarriers also exist before the first and after the last pilot is given. The MSE and the bit error rate (BER) of the estimator are characterized by the number and distance of the pilots, the Doppler spread and the WCC factor. Simulations show the effect of each factor on MSE and BER performance.