Cell-free massive multi-input-multi-output (CF-mMIMO) provides significant improvement owing to the distributed architecture. However, it suffers from the constraints including information constraints, and computation resources constraints. In this paper, we propose 4 ranks of available information in CF-mMIMO and aim to find a distributed precoding exploiting randomly accessible side information, which is one-step without iterations and robust against the unreliable fronthaul between distributed central processing units. Quadratic team learning (QTL) is devised which is derived from team theory to handle the distributed underdetermined quadratic programming. The extensive 1440 experiments validate the superiority of QTL and we believe QTL is a definitely excellent choice for CF-mMIMO distributed precoding. To the best of our knowledge, this is the first work utilizing team theory to help the design of artificial intelligence architecture for wireless communications. To prompt the development of QTL, we have open-sourced the implementation code on https://github.com/hzy238221seu/QTL4CF-Precoding.git.

Cell-free massive MIMO (CF-mMIMO) represents the pinnacle of distributed antenna systems, offering superior service to all users. However, the distributed nature of access points leads to fragmented information processing, limiting performance in practical deployments. Additionally, existing studies often overlook the impact of limited fronthaul capacity, which introduces noise and degrades the reliability of shared information. In this work, we implement a practical CF-mMIMO prototype under fifth generation new radio standards and propose a diffusion-based uplink scheme that outperforms conventional distributed cell-free systems without cooperation. Our approach adapts to arbitrary fronthaul topologies by leveraging the law of large numbers. We further analyze the linear effects of fronthaul noise and the correlation of uploaded data, demonstrating that the diffusion uplink excels in Rician fading environments while maintaining robust performance in Rayleigh fading. To the best of our knowledge, this is the first work to employ a diffusion model for mitigating fronthaul non-idealities, enabling distributed cooperative uplink in a real-world CF-mMIMO system.

Cell-free is considered a promising technology for the future network, which adopts a large number of distributed antennas to provide a uniformly good service. However, current researches under the long-term evolution standard mostly ignore the problem of asynchronous transmission brought by the different transmission delays due to the geographical distance differences, and assume that the system is perfectly synchronized. On the other, these works often do not consider a distributed method with controllable fronthaul overhead compatible with cell-free. To enable an extensive cell-free in the sixth generation, we derive an asynchronous analysis framework and propose a centralized and a distributed downlink precoding method respectively. What is more important, we have verified that cell-free suffers from inter-carrier-interference and inter-symbol-interference under the 5th generation new radio standard. To the best of our knowledge, this is the first work implementing a distributed asynchronous precoding method in an extensive cell-free, and simulation results demonstrate the effectiveness of the proposed two precoding methods, compared to naive precoding ignoring the asynchronous impact.

Cell-free is seen as one of the most important technology for the future wireless communications. In this paper, we adopt a wideband cell-free to implement low-altitude coverage to serve multiple unmanned aerial vehicles (UAVs) in the city playing the core role of low-altitude economy. For practice, distributed computation, asynchronous effects, beam split and imperfect channel state information are considered. We mainly rely on per-beam synchronization (PBS) and discuss different architecture implementations. A wideband asynchronous architecture that reuses the time delay modules exploited in wideband beam split calibration is proposed. In addition, a semi-synchronized path set (SSP-Set) is derived to eliminate asynchronous interference and a geometric scattering graphic convolutional network is used to acquire the (sub)-optimal SSP-Set. Based on these two technologies, a beam-delay alignment transmission (BDAT) scheme is obtained and we implement it with a distributed paradigm. The numerical results demonstrate the proposed BDAT can benefit from the cooperative downlink beamforming and provide a uniformly good service for UAVs.

Cell-free is seen as an important implementation for future wireless networks, which eliminates the conventional ‘cell’ concept and enables wide deployment. However, previous works mostly ignore the asynchronous effects in such a large distributed antenna system and assume perfect synchronization which is not practical. In this paper, we proposed a user-centric alignment transmission (UCAT) to settle this problem, which has the analytical beamforming vectors in each access point (AP) being computed locally and fits user-centric cell-free well. With cooperative center processing unit power optimization and AP beamforming computation, an asynchronous downlink method is obtained, and finally, numerical results demonstrate the effectiveness of UCAT.

User clustering presents a significant design challenge when implementing non-orthogonal multiple access (NOMA) in multiple-input multiple-output (MIMO) systems. To fully exploit the potential of power domain multiplexing and achieve better performance of NOMA, users must possess significant variations in their channel gains. Consequently, it becomes necessary to cluster users based on their distinct channel gain characteristics. However, using random user clustering often leads to suboptimal outcomes, while the exhaustive search method is burdened with exorbitant complexity. To achieve the balance between system performance and complexity, this paper proposes a user clustering scheme that relies on Jaccard coefficient, which is a measure of differences in user channels, ranging from 0 to 1. A value of 0 indicates complete dissimilarity, while a value of 1 indicates complete similarity. By calculating the Jaccard coefficient, we can identify users who are dissimilar to the centroid as a cluster. Additionally, a new power allocation algorithm is developed to maximize the sum spectral efficiency by employing successive convex approximations (SCA), taking into account both spectral efficiency and user quality of service. The simulation results demonstrate that the algorithm proposed for maximizing the sum spectral efficiency exhibits a higher convergence rate and can substantially enhance the sum spectral efficiency in comparison to the full power control (FPC) scheme.

This paper investigates the distributed uplink for the hierarchically backhaul-linked cell-free network with distributed processors to maximize the advantages of jointly serving under the same time and frequency resources. It is validated in previous works that the performance of fully centralized uplink in a cell-free network overwhelms uplink methods without or with limited coordination. On the other hand, a fully centralized uplink requires extremely high costs on backhaul links and computation capacity on the central processing unit (CPU), which is impractical in a widely deployed large cell-free network. To handle the mentioned problems, the relation between centralized uplink and group sliced distributed uplink is revealed, firstly. With the uniform framework compatible with previous fully centralized and fully distributed minimal mean square error (MMSE) equalization, two theorems are derived as group-joint MMSE complementary and the column space equivalence, which indicate the relation between the local optimal and the global optimal and can include conclusions achieved in previous works. Both computation and backhaul signaling overheads are distributed among the whole network. Simulation results demonstrate the excellent performance of proposed methods based on the derived complementary kernel.

Distributed precoding is a meaningful topic in the cell-free massive multiple input multiple output system. This system faces challenges in performance degradation due to the absence of knowledge from other antennas and several realistic constraints brought by the distributed implementation of the communication system such as the presence of phase noise (PN). In this paper, a robust cascaded team minimum mean square error (RCT-MMSE) precoding based on a hierarchical fronthaul structure is exploited to handle distributed and robust precoding including not only PN but signaling noise, sharing cost constraints and channel aging uncertainty. Such RCT-MMSE precoding, characterized by its avoidance of iterations because we derive the analytic expressions, mitigates the need for high fronthaul level instantaneous information exchange. It also demonstrates scalability with distributed computation burden and flexible signaling overhead, which offers an advantageous performance-cost tradeoff. Simulation results demonstrate the effectiveness of RCT-MMSE to combat several practical constraints and provide a flexible distributed precoding framework compared with previous ones.

In this paper, we study the problem of signal detection for massive multiple-input multiple-output (MIMO) communication systems aided by the intelligent reflecting surface (IRS). In the IRS-assisted massive MIMO systems, the signal between the base station and the users is reflected and hence the signal detection is a challenge problem. This paper proposes a model and data hybrid driven detection method to detect the signal of the receiver through neural network when the channel characteristics change. In the auxiliary transmission of IRS, the signal passes through two hop channels, one hop is the channel between transmitter and the IRS, the other hop is the channel between IRS and receiver. According to the different channel characteristics of two hop channel, a novel neural network detector which is combined the data-driven with model-driven is proposed. Experimental results show that the proposed detection scheme can achieve lower computational complexity and higher detection performance than traditional methods.

Signal detection is one of the fundamental problems in three-dimensional multiple-input multiple-output (3D-MIMO) wireless communication systems. This paper addresses a signal detection problem in 3D-MIMO system, in which matrix approximate inversion techniques is considered. It is well known that the signal detection accuracy of nonlinear signal detection algorithm is improved compared with the linear signal detection algorithm in MIMO systems, but its hardware requirements are higher in practical applications. Therefore, a more easily implemented linear signal detection algorithm is needed in MIMO systems. This paper studies the equivalent matrix inversion method based on linear equation generation and introduces the signal detection scheme based on matrix approximate inversion in large-scale 3D MIMO systems.

3D massive multiple-input multiple-output (MIMO) technology is considered as one of the key technologies of 5G. Compared with traditional 2D massive MIMO systems, the elevation angles of signal propagation are introduced in 3D massive MIMO systems, which make it fully utilize the advantages in horizontal and vertical dimensions at the same time, further increase the freedom of scheduling and resource allocation, and significantly improve the system capacity. To improve the system performance further, beamforming can form high-gain beams to reduce signal interference between user equipments (UEs). Traditionally, beam allocation is regarded as an optimization problem. Since most beam allocation problems are non-convex, it is difficult to obtain an optimal solution in real time. With the rise of artificial intelligence technology, machine learning has become one of the most promising tools. In order to optimize the beam allocation better in a short time, a location aided and machine learning based beam allocation algorithm (LMLBAA) is proposed in 3D massive MIMO systems. The algorithm absorbs the idea of k-NN or SVM to construct the multi-classifier, which uses collected location information of UEs as feature vectors and corresponding precoding codeword indexes as classes. When a new UE joins up, the base station (BS) will select a suitable codeword for the UE to perform precoding and form the corresponding serving beam according to the decision result of the multi-classifier. Simulation results show that the more location information of UEs is collected during the initialization process, the better the performance of the proposed algorithm will be. Meanwhile, as the transmit signal-to-noise ratio (SNR) increases, the average available sum rate will increase and approach the results of the exhaustive search method.

In this paper, we propose a multi-layer precoding scheme for multi-cell three-dimensional multiple-input multiple-output (3D MIMO) system with large-scale uniform planar antenna array at the base station (BS). Based on a lower bound of the average signal-to-leakage-plus-noise ratio (SLNR) and taking the fairness of scheduled users into account, we propose a user grouping algorithm exploiting only statistical channel state information (CSI). Then we apply the user grouping algorithm to the multi-layer precoding framework. The proposed scheme greatly reduces the CSI required by the BS with using only users’ statistical CSI. Simulation results show that the proposed scheme can achieve good fairness and high sum rate.

Massive multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) is one of the promising technology for the fifth generation of wireless communications. As a major radio-frequency impairment, in-phase and quadrature-phase (IQ) imbalance tends to destroy the channel reciprocity and degrade the performance of time division duplex (TDD) massive MIMO-OFDM system. In this paper, a calibration scheme is proposed to restore the channel reciprocity for TDD massive MIMO-OFDM system with IQ imbalance. Two calibration matrices are designed to compensate the uplink and downlink transmission respectively. A low-complexity estimation method of calibration matrices is also proposed in this paper. Numerical results confirm that the proposed scheme can restore the channel reciprocity and provide reliable performance.

Lens-based millimeter-wave massive MIMO technologies could reduce the number of required radio frequency chains at the base station without a loss of performance, but require the beamspace channel information, which is often costly to obtain. In this paper, a low-complexity scheme to estimate the channel for lens-based 3D millimeter wave massive MIMO systems is proposed. Beamspace analysis shows that the main entries of the channel matrix form a dual crossing (DC) shape. Existing DC-based channel estimation algorithms have a very high computational complexity. The proposed algorithm eliminates the process of repeatedly selecting the main row and column that is required in existing algorithms, and provides a solution even when certain entries exceed the matrix boundary. It thus has a much lower computational complexity than existing algorithms. Additionally it achieves a smaller normalized mean-squared error of the channel estimates than existing DC-based algorithms.