A Deep Reinforcement Learning Approach to IRS-assisted MU-MIMO Communication Systems

<div>The deep reinforcement learning (DRL)-based deep deterministic policy gradient (DDPG) framework is proposed to solve the joint optimization of the IRS phase-shift matrix and the precoding matrix in an IRS-assisted multi-stream multi-user MIMO communication.<br></div><div><br></div><div>The combination of multiple-input multiple-output(MIMO) communications and intelligent reflecting surfaces(IRSs) is foreseen as a key enabler of beyond 5G (B5G) and 6Gsystems. In this work, we develop an innovative deep reinforcement learning (DRL)-based approach to the joint optimization of the MIMO precoders and the IRS phase-shift matrices that is proved to be efficient in high dimensional systems. The proposed approach is termed deep deterministic policy gradient (DDPG)and maximizes the sum rate of an IRS-assisted multi-stream(MS) multi-user MIMO (MU-MIMO) system by learning the best matrix configuration through online trial-and-error interactions. The proposed approach is formulated in terms of continuous state and action spaces, and a sum-rate-based reward function. The computational complexity is reduced by using artificial neural networks (ANNs) for function approximations and it is shown that the proposed solution scales better than other state-of-the-art methods, while reaching a competitive performance.<br></div>

A Deep Reinforcement Learning Approach to IRS-assisted MU-MIMO Communication Systems

<div>The deep reinforcement learning (DRL)-based deep deterministic policy gradient (DDPG) framework is proposed to solve the joint optimization of the IRS phase-shift matrix and the precoding matrix in an IRS-assisted multi-stream multi-user MIMO communication.<br></div><div><br></div><div>The combination of multiple-input multiple-output(MIMO) communications and intelligent reflecting surfaces(IRSs) is foreseen as a key enabler of beyond 5G (B5G) and 6Gsystems. In this work, we develop an innovative deep reinforcement learning (DRL)-based approach to the joint optimization of the MIMO precoders and the IRS phase-shift matrices that is proved to be efficient in high dimensional systems. The proposed approach is termed deep deterministic policy gradient (DDPG)and maximizes the sum rate of an IRS-assisted multi-stream(MS) multi-user MIMO (MU-MIMO) system by learning the best matrix configuration through online trial-and-error interactions. The proposed approach is formulated in terms of continuous state and action spaces, and a sum-rate-based reward function. The computational complexity is reduced by using artificial neural networks (ANNs) for function approximations and it is shown that the proposed solution scales better than other state-of-the-art methods, while reaching a competitive performance.<br></div>

On Channel Estimation and Power Control for Multi-cell Multi-path Massive MIMO TDD System

Some more practical channels that model the networks in a real environment is the multi-path communication channels. In order to investigate these communications channels. This work addressed Channel Estimation (CE) in the Uplink (UL) phase for a multi-cell multi-user massive multipleinput multiple-output (M-MIMO) system that studies multi-path communication between each user and its serving Base Station (BS). We suppose that the network operates under Time-Division Duplex (TDD) protocol. We studied and analyzed the multi-path channels and their benefit over CE since it presents a more realistic channel that displays a real propagation circumstance. on the flip side, we evaluated the CE quality using ideal MinimumMean Square Error (MMSE). This latter relies on an impractical property that can be explicated since the MMSE estimator considers foreknowledge on Large-Scale Fading (LSF) coefficients of interfering users. Thus, the suggested estimator is introduced to overcome this issue, where the suggested estimator tackled this problem and presented result asymptotic approaches to the performance of the MMSE estimator. Besides, we considered a more real communication in which the multi-path channels are either realized using Non-Line-of-Sight (NLoS) only or using both Line-of-Sight (LoS) and NLoS path depending on the distance at which the user is located from his serving BS. Otherwise, in numerous scenarios, users at the cell edge are strongly affected by Pilot Contamination (PC). Hence, we introduced a Power Control (PoC) policy so that the users at the cell edge are less affected by the PC problem. In the simulation results segment, the analytic and simulated results are introduced to assert our theoretical study.

Millimetre wave 3-D channel modelling for next generation 5G networks

<abstract><p>Millimetre wave (mm-wave) spectrum (30-300GHz) is a key enabling technology in the advent of 5G. However, an accurate model for the mm-wave channel is yet to be developed as the existing 4G-LTE channel models (frequency below 6 GHz) exhibit different propagation attributes. In this paper, a spatial statistical channel model (SSCM) is considered that estimates the characteristics of the channel in the 28, 60, and 73 GHz bands. The SSCM is used to mathematically approximate the propagation path loss in different environments, namely, Urban-Macro, Urban-Micro, and Rural-Macro, under Line-of-Sight (LOS) and Non-Line-of-Sight (NLOS) conditions. The New York University (NYU) channel simulator is utilised to evaluate the channel model under various conditions including atmospheric effects, distance, and frequency. Moreover, a MIMO system has been evaluated under mm-wave propagation. The main results show that the 60 GHz band has the highest attenuation compared to the 28 and 73 GHz bands. The results also show that increasing the number of antennas is proportional to the condition number and the rank of the MIMO channel matrix.</p></abstract>

Design and Development of Novel Hybrid Precoder for Millimeter-Wave MIMO System

Power consumption and hardware cost reduction with the use of hybrid beamforming in large-scale millimeter wave MIMO systems. The large dimensional analog precoding integrates with the hybrid beamforming based on the phase shifters including digital precoding with lower dimensionality. The reduction of Euclidean distance between the hybrid precoder and fully digital is the major problem to overcome the minimization of resultant spectral efficiency. The issue formulates as a fully digital precoder’s matrix factorization problem based on the analog RF precoder matrix and the digital baseband precoder matrix. An additional element-wise unit modulus constraint is imposed by the phase shifters on the analog RF precoder matrix. The traditional methods have a problem of performance loss in spectral efficiency. In the processing time and iteration, high complexities result in optimization algorithms. In this paper, a novel low complexity algorithm proposes which maximizes the spectral efficiency and reduces the computational processing time.

Joint Transceiver Optimization of MIMO SWIPT Systems Assisted by Reconfigurable Intelligent Surface

In this work, a reconfigurable intelligent surface (RIS)-assisted multiple-input multiple-output (MIMO) system is studied with wireless energy harvesting (EH). Specifically, we focus on maximizing the harvested power at the receiver by joint optimization of the signal covariance, the phase shifter, and the power splitting factor, subject to the information rate and transmit power constraints. The formulated problem is hard to address due to the nonconcave objective and the nonconvex constraints. To tackle these challenges, an alternating optimization (AO) framework is proposed, where the phase shifter is solved by the penalty-based method. Simulation results validate the performance of the proposed approach.

High-Performance Fingerprint Localization in Massive MIMO System

Location-based service (LBS) for both security and commercial use is becoming more and more important with the rise of 5G. Fingerprint localization (FL) is one of the most efficient positioning methods for both indoor and outdoor localization. However, the positioning time of previous research cannot achieve real-time requirement and the positioning error is meter level. In this paper, we concentrated on high-performance in massive multiple-in-multiple-out (MIMO) systems. Principal Component Analysis (PCA) is applied to reduce the dimension of fingerprint, so that the positioning time is about tens of milliseconds with lower storage. What’s more, a novel fingerprint called Angle Delay Fingerprint (ADF) is proposed. Simulation result of the positioning method based on ADF shows the positioning error is about 0.3 meter and the positioning time is about hundreds of milliseconds, which is much better than other previous known methods. (Foundation items: Social Development Projects of Jiangsu Science and Technology Department (No.BE2018704).)