Decoding-Aided Channel Tracking in Time-Varying MIMO Systems: Short or Long Codes?

Multiple-input multiple-output (MIMO) technology significantly improves spectral efficiency and energy efficiency by fully exploiting the spatial resources of the wireless channel. However, the channels can be time-varying due to the mobility of the users in practice, which leads to a non-negligible mismatch of the channel state information (CSI) acquired based on the pilots and the one corresponding to the data transmission stage, and furthermore, the performance deterioration. To handle this, this paper investigates a decoding-aided channel tracking algorithm, where both the detected symbols and the decoded output are utilized to improve the CSI accuracy based on the framework of Kalman filtering. Specifically, instead of the commonly-used long length codes, we adopt short length codes as the coding scheme to enable the fast decoding-aided channel tracking. It is interesting to find via simulations that, the proposed channel tracking algorithm with the use of short length codes can effectively mitigate the effect of channel aging and can even outperform the one using powerful long length codes.