On Channel Estimation Performance in Multi-Cell TDD Systems

In this paper, we analyze the bit error rate (BER) and mean square error (MSE) performance for data uplink of a multi-cell TDD system with imperfect knowledge of channel state information (CSI). We investigate the impact of imperfect CSI, channel tracking methods and pilot contamination on the achievable performance, and derive the closed-form MSE and BER expressions. Performance degradation due to channel aging is analyzed in the context of different channel tracking methods and to what extent we can control it by proper system design. Our results indicate that the channel estimator coherence time should be properly selected given the impact of channel estimation over different blocks on achievable performance. Moreover, decision-feedback channel tracking can be used to optimize transmission efficiency given the target performance.