Exploiting wireless channel state information structures beyond linear correlations: A deep learning approach

LANGUAGE English

SOURCE  IEEE COMMUNICATIONS MAGAZINE, Vol: 57 No: 3 pp: 28-34, MAR 2019

Published Date: MAR 2019

ABSTRACT

Knowledge of information about the propagation channel in which a wireless system operates enables better, more efficient approaches  for signal transmissions. Therefore, channel state information (CSI) plays a pivotal role in the system performance. The importance of CSI is in  fact growing in the upcoming 5G and beyond systems, for example, for the implementation of massive multiple-input multiple-output (MIMO).  However, the acquisition of timely and accurate CSI has long been  onsidered a major issue, and becomes increasingly challenging due to  the need for obtaining CSI of many antenna elements in massive MIMO systems. To cope with this challenge, existing works mainly focus on  exploiting linear structures of CSI, such as CSI correlations in the spatial domain, to achieve dimensionality reduction. In this article, we first  systematically review the state of the art on CSI structure exploitation. We then extend to seek deeper structures that enable remote CSI inference  wherein a data-driven deep neural network (DNN) approach is necessary due to model inadequacy. We develop specific DNN designs suitable  for CSI data. Case studies are provided to demonstrate great potential in this direction for future performance enhancement.