Parametric algorithms for the estimation of
rapidly-varying Carrier Frequency Offset (CFO) usually
employ pilot symbols multiplexed with the data
transmission. As the CFO variation rate increases so
has to increase the density of pilot symbols transmitted,
thus impairing the bandwidth efficiency. In order
to reduce the number of pilot symbols used in the estimation
of rapidly-varying CFO it was proposed to
use a truncated Taylors series to predict the CFO,
where the derivatives up to order d-1 are recursively
estimated with a d-order Kalman filter (KF).We propose
to compare the performance of a fourth-order
KF predictor in the most popular block transmission
systems: Orthogonal Frequency Division Multiplexing
(OFDM) and Single Carrier Frequency-Domain
Equalization (SC-FDE).
Simulating different transmission scenarios, e.g.,
channel coding and spatial diversity, our results show
that for static multipath fading channels the proposed
receiver for the SC-FDE scheme exhibits better Bit
Error Rate (BER) performance than that of OFDM.