An intelligent reflecting surface (IRS) is a programmable device that can be used to control electromagnetic waves propagation by changing the electric and magnetic properties of its surface. Therefore, IRS is considered a smart technology for the sixth generation (6G) of communication networks. In addition, machine learning (ML) techniques are now widely adopted in wireless communication as the computation power of devices has increased. As it is an emerging topic, we provide a comprehensive overview of the state-of-the-art on ML, especially on deep learning (DL)-based IRS-enhanced communication. We focus on their operating principles, channel estimation (CE), and the applications of machine learning to IRS-enhanced wireless networks. In addition, we systematically survey existing designs for IRS-enhanced wireless networks. Furthermore, we identify major issues and research opportunities associated with the integration of IRS and other emerging technologies for applications to next-generation wireless communication.

This paper proposes an intelligent reflective surface (IRS) design scheme to improve the spectral efficiency (SE) for downlink multi-user (MU) multiple-input-single-output (MISO) system. IRS composed of low-cost reflecting elements is expected to be used in various scenarios in future wireless communication systems. It can be used even when the direct path is blocked by obstacles. Since the reflecting elements can only adjust the phase, the design of IRS matrix causes a non-convex problem. In this paper, the proposed schemes perform optimization by transforming a non-convex problem into a solvable convex function. Typically, base station-IRS (BS-IRS) channel can be assumed as the line-of-sight (LOS) channel environment. In this case, multi-user system suffers from performance degradation due to the LOS channel rank problem. To alleviate this problem, this paper considers deterministic scattering and sufficient spacing between the reflecting elements of the IRS. The simulation results show that the proposed schemes achieve better SE performance than the randomly generated IRS scheme. In addition, proposed minimum mean square error (MMSE)-based scheme can achieve high performance compared to other schemes even in a low-rank channel environment.INDEX TERMS Intelligent reflecting surface, multi-user MISO, phase shift, spectral efficiency.

No abstract

A cell-free massive multiple input multiple output (MIMO) system is an attractive network model that is in the spotlight in 5G and future communication systems. Despite numerous advantages, the cell-free massive MIMO system has a problem in that it is difficult to operate in reality due to its vast amount of calculation. The user-centric cell-free massive MIMO model has a more feasible and scalable benefit than the cell-free massive MIMO model. However, this model has the disadvantage that as the number of users in the area increases, there are users who do not receive the service. In this paper, the proposed scheme creates connections for unserved users under a user-centric scheme without additional access point (AP) installation and disconnection for existing users. A downlink user-centric cell-free massive MIMO system model in which the APs are connected to the central processing unit (CPU) and the APs and users are geographically distributed is considered. First, the downlink spectral efficiency formula is derived and applied to the user-centric cell-free massive MIMO system. Then, the proposed scheme and power control algorithm are applied to the derived formula. The simulation results show that the unserved users within the area disappear by using the proposed scheme, while the bit error rate (BER) performance and sum rate improve compared to the existing scheme. In addition, it is shown that the proposed scheme works well even with a very large number of users in the area, and a significant service performance improvement for the worst 10% of users and the overall improvement of per-user throughput for the bottom 70% of users are ensured.