Deep Decoder CsiNet for FDD Massive MIMO System

In order to achieve a higher gain in frequency division duplex (FDD) massive multiple-input multiple-output (MIMO) system, it is imperative feedback downlink channel state information (CSI) from user equipment (UE) base station (BS). However, excessive overhead makes the task challenging. Hence, recent years, various deep learning (DL)-based models have been introduced effectively compress CSI codeword at UE and then reconstruct back into BS reduce overhead. The authors of this paper new network called Deep Decoder CsiNet (DDCsiNet), which utilizes decoder approach for improving quality reconstructed feedback. proposed DDCsiNet leverages residual on structure facilitate low-frequency flow. Additionally, incorporates feature attention exploit dependencies, resulting significant enhancement reconstruction quality. Numerical results are presented demonstrate that our model can performance outperform other DL methods terms