Deep learning for massive multiple access in 6G

In recent years, the number of massive Internet of Things (mIoT) has grown tremendously, giving rise to the term massive machine-type communications (mMTC). Cellular Internet of Things (IoT) is an economical solution for connecting devices wirelessly because it reuses existing cellular infrastructure. 3rd Generation Partnership Project (3GPP) has recognized mMTC as one of the use cases of 6G. However, providing massive access to the IoT devices within the constraints of limited system resources has been an ongoing challenge in cellular networks. On the other hand, Deep learning (DL) has emerged as a powerful method for various applications, such as image processing and natural language processing. More recently, DL has been successfully applied to a wide range of wireless communication tasks. Given that, this thesis aims to design massive multiple-access protocols using DL algorithms for both cell-based and cell-free networks.