Dilated Convolution based CSI Feedback Compression for Massive MIMO Systems

TL;DR

The dilated convolutions are used to enhance the receptive field (RecF) of the proposed DCRNet without increasing the convolution size, and advanced encoder and decoder blocks are designed to improve the reconstruction performance and reduce computational complexity.

Abstract

Although the frequency-division duplex (FDD) massive multiple-input multiple-output (MIMO) system can offer high spectral and energy efficiency, it requires to feedback the downlink channel state information (CSI) from users to the base station (BS), in order to fulfill the precoding design at the BS. However, the large dimension of CSI matrices in the massive MIMO system makes the CSI feedback very challenging, and it is urgent to compress the feedback CSI. To this end, this paper proposes a novel dilated convolution based CSI feedback network, namely DCRNet. Specifically, the dilated convolutions are used to enhance the receptive field (RF) of the proposed DCRNet without increasing the convolution size. Moreover, advanced encoder and decoder blocks are designed to improve the reconstruction performance and reduce computational complexity as well. Numerical results are presented to show the superiority of the proposed DCRNet over the conventional networks. In particular, the proposed DCRNet can achieve almost the state-of-the-arts (SOTA) performance with much lower floating point operations (FLOPs). The open source code and checkpoint of this work are available at https://github.com/recusant7/DCRNet.

Figures (4)

• Figure 1: Structure of the CSI feedback in massive MIMO system.
• Figure 2: Architecture of the proposed DCRNet, where (a) and (b) illustrate the structures of the encoder and decoder of DCRNet. In addition, the structures of the encoder block and decoder block in (a) and (b) are illustrated in (c) and (d) with $N_a=N_t=32$, respectively. In general, the encoder block and the decoder block are the main components of DCRNet, where BN layers, activation functions and reshape operations are omitted for brevity.
• Figure 3: Demonstations of the dilated convolution operations with different dilated rates, where solid and shadow areas represent the effective operations and RF, respectively. When $d = 1$, the dilated convolution degenerates into the standard convolution. When $d> 1$, the dilated convolution is able to obtain a larger RF while it still involves the same computational complexity as the standard convolution.
• Figure 4: NSME performance versus FLOPs under indoor scenario when the compression rates are 1/4 and 1/16.