Analysis of Channel Estimation in Mobile Communication Systems with Users Moving at Various Speeds

In order to provide a reliable reception and decrease the packet retransmission, the wireless communication system should guaranty the accuracy of channel estimation. This implies that the channel estimation mean-square error (MSE) of the mobile terminals with various speeds should remain lower than a threshold. This usually causes higher pilot overhead in terms of more bandwidth and transmit power. In this paper, a commonly used time multiplexing pilot pattern is reconsidered from a different point of view in contrast with the previous studies. In a frame, the intervals between each data time slot and the nearest pilot time slot are different, and thus the tracking abilities of each data time slot are also different. Therefore, a system is able to support a number of levels of users' moving speed, even though using a fixed pilot pattern. In order to analyze the tracking ability of each time slot, the MSE for two interpolation methods, i.e., linear minimum MSE and linear interpolation, is derived. Based on the derived analytical results, a novel criterion for pilot spacing design is presented, and a novel scheduling algorithm based on user moving speeds is also proposed. Simulation results show that the proposed pilot design and scheduling algorithm can achieve significant performance improvement.