Design of optimal time domain filter and iterative channel estimator for LTE-advanced uplink system

Abstract

This thesis deals with channel estimation method for LTE-Advanced uplink system. Due to coexistence of demodulation reference signal (DMRS) from multiple transmit antennas at the same time-frequency resources, channel estimation is performed by using time shift property of DMRS to separate each channel impulse response (CIR) from each transmit antenna. Because the number of resource block (RB) allocated to a user is generally only a small portion of the overall uplink resources, it causes channel energy leakage. Thus we analyze energy leakage phenomenon causing performance degradation of conventional channel estimator. Because performance of the conventional channel estimator mainly depends on time domain filter cutoff value, we also investigate the effect of cutoff value. Furthermore we propose optimal time domain cutoff value and iterative channel estimator with least square fitting algorithm to improve channel estimation accuracy.