Optimizing the MIMO Channel Estimation for LTE-Advanced Uplink

Abstract

To achieve good overall performance for wireless communication systems, it is essential to have good channel estimation performance. In this paper, based on a practical MIMO (Multiple-Input Multiple-Output) channel estimation algorithm for LTE-Advanced uplink, we discuss how to optimize its performance when different number of resource blocks (RB) are allocated. It is shown that after optimization, we can achieve better performance at both low and high SNR regions for both small and large number of allocated RBs. When the number of allocated RBs is large, it is shown that, with real channel information obtained by the optimized channel estimation, the frame error rate (FER) is only 1dB away from that with perfect channel information. When the number of allocated RBs is small and when QPSK or 16QAM is used, the performance is only 2 to 3dB away from that with perfect channel information.