Linear Unbiased Channel Estimation and Data Detection in Superimposed OFDM Systems

Abstract

Reliable channel estimation is necessary for orthogonal frequency-division multiplexing (OFDM) systems employing coherent detection to achieve high data rates. Pilot-aided detection algorithms benefit from rapid convergence, but suffer from non-efficient bandwidth usage. The idea of superimposed data transmission can be enabled for channel estimation without sacrificing the data rate. In this paper, we first derive the best linear unbiased estimator (BLUE) for channel impulse response and then, we propose an iterative joint channel estimator and data detector to improve the performance of both the estimator and the detector. Furthermore, we derive the variance of the proposed estimator and show that the equispace and equipower conditions hold for the superimposed pilots to attain the minimum variance of the estimator. Simulations show that the performance with perfect knowledge of the channel impulse response can be achieved closely by the proposed iterative data detector.