Deterministic communication within the wireless domain is essential for industrial applications. However, the stochastic nature of wireless communication introduces substantial challenges for time-sensitive networking (TSN) services, which require strict end-to-end latency bounds. This paper addresses the challenge of minimizing long-term delay jitter in downlink cell-free multi-user multi-input multi-output orthogonal frequency division multiple access (MU-MIMO OFDMA) systems, subject to heterogeneous delay upper bounds and satisfaction rates. The problem involves time-space-frequency precoding constrained by user-specific delay violation probabilities and transmit power limits. To overcome the limitations of model-driven methods in handling implicit system models and the inefficiency of data-driven approaches in large action spaces, we propose a hybrid solution. Specifically, we decompose the problem into two sub-problems: rate scheduling via a constrained Markov decision process (CMDP), and instantaneous precoding through weighted sum-rate (WSR) maximization. We develop a safety reinforcement learning-based algorithm to optimize rate scheduling by allocating user weights, and a weighted minimum mean squared error (WMMSE) algorithm to solve the WSR maximization. Simulation results demonstrate that our approach effectively reduces jitter while meeting stringent delay-related requirements. In diverse TSN scenarios with heavy loading ratio, our proposed co-driven scheme achieves about 45% reduction in delay jitter compared to earliest deadline first (EDF) scheduling, while realizing user-specific delay satisfactory ratios (99.9%-99.999%).

In unmanned aerial vehicle (UAV) communications, beam tracking faces significant challenges due to persistent variations in mobile users’ positions and flight attitudes. Fixed tracking periods result in either unnecessary or insufficient overhead depending on whether the UAV’s relative motion is static or dynamic. To address these challenges, this paper proposes a two-level beam tracking scheme enabled by data-driven beam prediction. In the upper layer, a time-series prediction method is used to adaptively adjust the tracking interval by forecasting confidence intervals of future communication quality. In the lower layer, a threshold-based probing beam selection method is implemented, which selects probing beams whose conditional likelihoods exceeding a predefined threshold. Simulation results demonstrate that the total beam training overhead drop 43.37 %, while the average outage probability decreases from 6.229 % to 0.095 %.

Beam selection for joint transmission in cell-free massive multi-input multi-output systems faces the problem of extremely high training overhead and computational complexity. The traffic-aware quality of service additionally complicates the beam selection problem. To address this issue, we propose a traffic-aware hierarchical beam selection scheme performed in a dual timescale. In the long-timescale, the central processing unit collects wide beam responses from base stations (BSs) to predict the power profile in the narrow beam space with a convolutional neural network, based on which the cascaded multiple-BS beam space is carefully pruned. In the short-timescale, we introduce a centralized reinforcement learning (RL) algorithm to maximize the satisfaction rate of delay w.r.t. beam selection within multiple consecutive time slots. Moreover, we put forward three scalable distributed algorithms including hierarchical distributed Lyapunov optimization, fully distributed RL, and centralized training with decentralized execution of RL to achieve better scalability and better tradeoff between the performance and the execution signal overhead. Numerical results demonstrate that the proposed schemes significantly reduce both model training cost and beam training overhead and are easier to meet the user-specific delay requirement, compared to existing methods.

In imperfect multiple-input multiple-output (MIMO) systems, model-based methods for performance prediction and optimization generally experience degradation in the dynamically changing environment with unknown interference and uncertain channel state information (CSI). To adapt to such challenging settings and better accomplish the network auto-tuning tasks, we propose a generic learnable model-driven framework. We further consider transmit regularized zero-forcing (RZF) precoding as a usage instance to illustrate the proposed framework. The overall process can be divided into three cascaded stages. First, we design a light neural network for refined prediction of sum rate based on coarse model-driven approximations. Then, the CSI uncertainty is estimated on the learned predictor in an iterative manner. In the last step the regularization term in the transmit RZF precoding is optimized. The effectiveness of the generic framework and the derivative method thereof is showcased via simulation results.

The problem of beam alignment and tracking in high mobility scenarios such as high-speed railway(HSR) becomes extremely challenging, since large overhead cost and significant time delay are introduced for fast time-varying channel estimation. To tackle this challenge, we propose a learning-aided beam prediction scheme for HSR networks, which predicts the beam directions and the channel amplitudes within a period of future time with fine time granularity, using a group of observations. Concretely, we transform the problem of high-dimensional beam prediction into a two-stage task, i.e., a low-dimensional parameter estimation and a cascaded hybrid beamforming operation. In the first stage, the location and speed of a certain terminal are estimated by maximum likelihood criterion, and a data-driven data fusion module is designed to improve the final estimation accuracy and robustness. Then, the probable future beam directions and channel amplitudes are predicted, based on the HSR scenario priors including deterministic trajectory, motion model, and channel model. Furthermore, we incorporate a learnable non-linear mapping module into the overall beam prediction to allow non-linear tracks. Both of the proposed learnable modules are model-based and have a good interpretability. Compared to the existing beam management scheme, the proposed beam prediction has (near) zero overhead cost and time delay. Simulation results verify the effectiveness of the proposed scheme.

Beam alignment and tracking for millimeter-wave communication networks in highly mobile scenarios, such as high-speed railway, suffer from large overhead cost and time delay loss. To solve this problem, we propose a learning-aided beam management scheme, which divides the high-dimensional beam prediction procedure into two stages, i.e., parameter estimation and hybrid beamforming. The locations and velocities of the mobile terminals are estimated using the maximum likelihood criterion, and a data fusion module is employed to further improve the estimation accuracy and robustness. Then, the next probable beam directions and the corresponding hybrid precoders are derived based on the estimated parameter set. Numerical simulations show that, the proposed method yields significantly lower overhead cost and time delay compared to the existing beam management scheme.

The model-based power allocation has been investigated for decades, but this approach requires mathematical models to be analytically tractable and it has high computational complexity. Recently, the data-driven model-free approaches have been rapidly developed to achieve near-optimal performance with affordable computational complexity, and deep reinforcement learning (DRL) is regarded as one such approach having great potential for future intelligent networks. In this paper, a dynamic downlink power control problem is considered for maximizing the sum-rate in a multi-user wireless cellular network. Using cross-cell coordinations, the proposed multi-agent DRL framework includes off-line and on-line centralized training and distributed execution, and a mathematical analysis is presented for the top-level design of the near-static problem. Policy-based REINFORCE, value-based deep Q-learning (DQL), actor-critic deep deterministic policy gradient (DDPG) algorithms are proposed for this sum-rate problem. Simulation results show that the data-driven approaches outperform the state-of-art model-based methods on sum-rate performance. Furthermore, the DDPG outperforms the REINFORCE and DQL in terms of both sum-rate performance and robustness.

To meet the spectrum requirement in the future wireless communication system, the m ‐ary phase position shift keying as a type of the spectrally efficient modulation has been considered in this study. However, the inter‐symbol interference (ISI), the waveform distortion and the non‐white noise are introduced by the band‐pass infinite impulse response filters, which is adopted to limit the signal bandwidth and eliminate the out‐band interference. The traditional demodulation methods based on impacting filter or matched filter cannot work well in these scenarios. Therefore, a neural network (NN)‐based demodulator is proposed to solve the problem of waveform distortion. Additionally, a novel NN‐based iterative decision feedback (IDF) demodulator is also proposed to further reduce the ISI iteratively by using the previous estimated symbol information. Simulation results show that both the NN‐based demodulator and the NN‐based IDF demodulator can greatly outperform the traditional demodulation methods in the band‐limited communication system, and the NN‐based IDF demodulator achieves better performance than the normal NN‐based demodulator.