Diagnosing the root causes of Quality of Experience (QoE) degradations in operational mobile networks is challenging due to complex cross-layer interactions among kernel performance indicators (KPIs) and the scarcity of reliable expert annotations. Although rule-based heuristics can generate labels at scale, they are noisy and coarse-grained, limiting the accuracy of purely data-driven approaches. To address this, we propose DK-Root, a joint data-and-knowledge-driven framework that unifies scalable weak supervision with precise expert guidance for robust root-cause analysis. DK-Root first pretrains an encoder via contrastive representation learning using abundant rule-based labels while explicitly denoising their noise through a supervised contrastive objective. To supply task-faithful data augmentation, we introduce a class-conditional diffusion model that generates KPIs sequences preserving root-cause semantics, and by controlling reverse diffusion steps, it produces weak and strong augmentations that improve intra-class compactness and inter-class separability. Finally, the encoder and the lightweight classifier are jointly fine-tuned with scarce expert-verified labels to sharpen decision boundaries. Extensive experiments on a real-world, operator-grade dataset demonstrate state-of-the-art accuracy, with DK-Root surpassing traditional ML and recent semi-supervised time-series methods. Ablations confirm the necessity of the conditional diffusion augmentation and the pretrain-finetune design, validating both representation quality and classification gains.

In cellular networks, cell handover refers to the process where a device switches from one base station to another, and this mechanism is crucial for balancing the load among different cells. Traditionally, engineers would manually adjust parameters based on experience. However, the explosive growth in the number of cells has rendered manual tuning impractical. Existing research tends to overlook critical engineering details in order to simplify handover problems. In this paper, we classify cell handover into three types, and jointly model their mutual influence. To achieve load balancing, we propose a multi-agent-reinforcement-learning (MARL)-based scheme to automatically optimize the parameters. Experimental results show that our proposed scheme outperforms existing benchmarks in balancing load and improving network performance.

In this paper, we introduce a novel user association (UA) and resource block group (RBG) allocation (RA) method utilizing multi-agent reinforcement learning (MARL) for a multi-user multiple-input multiple-output (MU-MIMO) downlink system. Unlike traditional MARL radio resource management (RRM) approaches, which utilize user equipment (UE) as learning agents, base stations (BS) are deployed as agents for practical consideration. However, this will significantly enlarge the action space and bring about action constraint violation problems. We employ dual-actor neural networks to separate UA and RBG allocation actions, thereby effectively reducing the joint action space and accelerating exploration. In addition, a Q-value-rank-based action projection algorithm is proposed to address the cross-agent coupling constraints. The simulation results demonstrate that the proposed MARL framework with action projection outperforms other baselines in terms of RRM performance.

Optimizing antenna parameters like azimuth, down-tilt, and power is crucial for coverage and capacity optimization (CCO) in next-generation wireless networks. However, traditional expert knowledge-based methods struggle to maintain optimal results when faced with changing environments. To address this, we propose a guided deep reinforcement learning (DRL) algorithm that learns a policy to dynamically adjust antenna parameters based on the evolving environment. Our approach employs proximal policy optimization-based DRL and integrates a problem-specific pretraining process using zero-order gradient descent. The pretrain policy serves as a guiding policy, enabling the agent to explore and discover high-reward regions, thus accel-erating the learning process. The performance of our solution is validated by numerical experiments conducted on a 5G simulation platform with real-world topological properties. The results show that our approach achieves significantly faster convergence and outperforms baseline methods in terms of CCO performance.

In this paper, we propose a novel joint trajectory and communication scheduling scheme for multiple unmanned aerial vehicles (UAVs) enabled wireless caching networks. To exploit the favorable propagation of air-to-ground channels and spatial multiplexing gains, we consider an ultra dense UAVs enabled content-centric wireless transmission network, where massive UAVs are deployed to transmit cached contents to a group of random distributed ground users. We formulate this problem as a infinite horizon ergodic stochastic differential game (SDG) for optimizing the users’ quality-of-experience (QoE) on the concept of request queues for cached contents. By means of mean-field game (MFG) analysis, we derive a reduced-complexity control solution. In addition, we further propose a novel model-specific deep neural network (DNN) to learn the PDEs numerically by exploiting a homotopy perturbation method (HPM).

In this paper, we propose a novel joint trajectory and communication scheduling scheme for multiple unmanned aerial vehicles (UAVs) enabled wireless caching networks. To exploit the favorable propagation of air-to-ground channels, we consider an ultra dense UAVs enabled content-centric wireless transmission network, where massive UAVs are deployed to transmit cached contents to a group of random distributed ground users. We formulate the problem as an infinite horizon ergodic stochastic differential game (SDG) for optimizing the users' quality-of-experience (QoE). In particular, stochastic dynamics of channel states, UAVs' mobility, energy queues and content request queues are modeled in this game. To deal with the state coupling between the UAVs, we consider a limiting problem for large number of UAV based on mean field analysis. A reduced-complexity decentralized solution can be obtained through mean-field equilibrium analysis. To further reduce the solution complexity on each UAV, we propose a model-specific deep neural network (DNN) to learn the optimal control solution in an online manner. The DNN is not arbitrarily generated but tailored to the structural properties of the value function and stationary distribution based on the homotopy perturbation method analysis. Finally, simulation results are provided to show that the proposed solution can achieve significant gain over the existing baselines.