Logo der Uni Stuttgart
Pilot Based Channel Estimation

Communication in wireless channels may get corrupted by fading. It is required that these fading characteristics of the channel to be known at the receiver to implement effective receiver techniques.

In this demo pilot-based channel estimation techniques are implemented. In pilot based estimation the transmitter sends known symbols called pilots at regular intervals.

The relation between the transmitted signal $X_k$ and received signal $Y_k$ at k-th instance is given by:

$$Y_k = H_k*X_k + Z_k$$

where
$Z_k$ is the noise at the k-th instance, and
$H_k$ is the channel transfer function at the k-th instance

At the pilot locations the channel can be estimated by: $$\hat{H_k} = Y_k/X_k + Z_k/X_k = H_k + Z_k'$$ where $Z_k'$ is a scaled noise at k-th instance.

Different methods can then be applied to estimate the channel at the non-pilot locations. The channel estimation techniques presented in this demo are:

Linear interpolation method

Least-squares (LS) interpolation method

Wiener/LMMSE interpolation method

The performance of all these schemes is investigated with BPSK modulation on Rayleigh fading channel based on Clarks scattering model.