Regression CNN Based Fast Fading Channel Tracking Using Decision Feedback Channel Estimation

In a high-speed moving mobile environment, the channel state information (CSI) in last part of packet is different from actual beginning part. Therefore, estimation accuracy degraded, especially when small number pilot symbols are used to ensure transmission efficiency. For above reasons, it necessary compensate for CSIs achieve reliable communication. Decision feedback (DFCE) has been widely considered be one tracking methods. However, presence time and frequency selective fading environments still causes errors due decision-making process. We focused on time-frequency domain response CSIs, which can represented as two-dimensional image. This paper newly proposes regression convolutional neural network (CNN) based scheme using by DFCE training prediction solve these problems. Computer simulation results demonstrate that proposed higher BER performance than conventional schemes.