Adaptive channel estimation for LTE uplink

Long Term Evolution (LTE) is designed for high speed data rate, higher spectral efficiency, and lower latency. LTE implements Single Carrier Frequency Division Multiple Access (SC-FDMA) for its uplink. A main challenge for the terminal implementation of LTE uplink is the efficient realization of channel estimation. A Kalman filter-type estimator based on pilot symbols is commonly used to solve the problem. However, this estimator requires statistic prior knowledge of channel and noise that is not easily obtained. In this thesis, novel adaptive and predictive filtering algorithms based on a Kalman recursion are presented for the LTE uplink system. The proposed channel estimator updates coefficients adaptively and it does not need prior knowledge. The novel technique is realized and validated in MATLAB. Simulation results show that the proposed estimators have excellent performance, measured in terms of the mean square error (MSE) in different channel environment, compared to classical channel estimation techniques.