Intelligent fault-tolerant (FT) computing has recently demonstrated significant advantages in predicting and diagnosing faults proactively, thereby ensuring reliable service delivery. However, due to the heterogeneity of fault knowledge, dynamic workloads, and limited data support, existing deep learning-based FT algorithms face challenges in fault detection quality and training efficiency. This is primarily because their homogenization of fault knowledge perception difficuties to fully capture diverse and complex fault patterns. To address these challenges, we propose FT-MoE, a sustainable-learning fault-tolerant computing framework based on a dual-path architecture for high-accuracy fault detection and classification. This model employs a mixture-of-experts (MoE) architecture, enabling different parameters to learn distinct fault knowledge. Additionally, we adopt a two-stage learning scheme that combines comprehensive offline training with continual online tuning, allowing the model to adaptively optimize its parameters in response to evolving real-time workloads. To facilitate realistic evaluation, we construct a new fault detection and classification dataset for edge networks, comprising 10,000 intervals with fine-grained resource features, surpassing existing datasets in both scale and granularity. Finally, we conduct extensive experiments on the FT benchmark to verify the effectiveness of FT-MoE. Results demonstrate that our model outperforms state-of-the-art methods.

With the rapid evolution of 5G and the ongoing development of 6G technologies, the Internet of Vehicles (IoV) is expected to play a critical role in next-generation intelligent transportation systems. Applications such as autonomous driving, augmented reality, and smart mobility not only require ultra-low latency and high computational efficiency, but also demand enhanced trustworthiness and privacy assurance. To address these demands, Vehicular Edge Computing (VEC) has emerged as a foundational paradigm for 6G-IoT, enabling intelligent services by offloading tasks from vehicles to edge nodes. However, task offloading in IoV-VEC systems still faces critical challenges, including the need for responsible AI decision-making under dynamic network conditions and the protection of sensitive vehicular data. This paper proposes FedVTO, a privacy-preserving federated vehicle task offloading framework that integrates Federated Learning (FL) and Deep Reinforcement Learning (DRL) to optimize task offloading decisions and resource allocation strategies in VEC networks. By incorporating information entropy models and dynamically adjusting weighting parameters using an entropy-based method within a three-tier architecture (vehicles, roadside units, and cloud server), FedVTO minimizes latency, energy consumption, and privacy leakage. Experimental results show that FedVTO significantly improves task offloading efficiency and mitigates privacy risks compared to traditional methods in dynamic VEC environments.

In Intent-based Vehicular Edge Computing (VEC) networking, escalating demands for computational offloading and resource management in dynamic urban environments necessitate innovative solutions. This paper proposes a novel RIS-assisted NOMA-VEC framework that empowers vehicle users (VUs) to offload arbitrary task portions to multiple edge servers via any available subcarrier. This approach overcomes limitations posed by heterogeneous local computing capabilities and stringent latency constraints. By leveraging Reconfigurable Intelligent Surfaces (RIS) to enhance channel conditions through both direct and reflected links, our framework significantly improves communication reliability and offloading efficiency. To minimize the average weighted energy consumption of VUs under time-varying channels and traffic dynamics, we formulate a joint optimization problem integrating offloading decisions, power allocation, and transmission time scheduling. Addressing the problem’s inherent complexity, characterized by multi-variable coupling and non-convex constraints, we develop a two-stage decomposition strategy: Offloading decisions are dynamically adapted to environmental fluctuations using a Proximal Policy Optimization (PPO)-based algorithm, while resource allocation is resolved through a hybrid Genetic Algorithm (GA) and Sequential Least Squares Programming(SLSQP) approach, efficiently navigating combinatorial and non-convex landscapes. Extensive simulations demonstrate that our framework reduces VU energy consumption by 11.12% compared to baseline methods, validating its superior efficiency in RIS-enhanced VEC systems.

With the popularity of electric vehicles (EVs), wireless power transmission (WPT) technology has become a hot research topic for next-generation battery charging technology. However, the vulnerability of wireless networks to malicious interference attacks is inherited by WPT. To alleviate the privacy and security issues of WPT, we propose a novel FedDQ, a federated deep reinforcement learning with Q-ensemble, to cope with interference attacks in EV wireless charging network environments. Federated learning protects the security privacy of EVs by training a global model that exploits the property that data and models will not be transmitted. In order to trade-off the training cost and efficiency, we introduce offline-to-online training models by pre-training the offline Q-network with pre-collected data, and the trained model serves as an initialization of the online model. Then, the online Q-network is obtained by weakening or removing the original pessimistic constraints to enhance the training speed. Secondly, we introduce the intelligent reflective surface (IRS) to enhance the security performance of WPT by modifying the IRS phase shift and amplitude to cancel the malicious interference signal. Experimental results show that our proposed FedDQ algorithm has superior performance and outperforms existing baseline methods in terms of anti-jamming metrics.

Intelligent adaptive bitrate (ABR) schemes have been widely recognized for their excellent learning strategies. However, existing intelligent ABR methods have limitations, i.e., the lack of logical reasoning capability for video-aware symbolic representations leads to low sampling efficiency and fails to achieve the optimal performance of Bitrate Adaptation. We introduce NeuroBA, a learning-based approach to realize ABR using neuro-symbolic deep reinforcement learning. NeuroBA trains a neuro-symbolic deep network model without making any assumptions about the edge video scene and without relying on a predefined model. Instead, it enables bitrate decision-making under uncertainty and partial observability by knowledge-driven video quality perception in symbolic first-order logic. To enhance wireless signals, we have introduced Intelligent Reflecting Surface (IRS) technology to address this issue. By dynamically adjusting the phase shift of IRS, the throughput performance of wireless networks is significantly improved. Based on trace-driven and real-world experiments covering a variety of edge video scenarios, and network performance metrics, NeuroBA is compared with state-of-the-art ABR schemes, and NeuroBA exhibits superior performance, with an average QoE improvement of 16.58% (BOLA)-25.34% (Fugu). In particular, it outperforms existing baseline approaches even without pre-programmed models and network scenarios assumed for the edge network.

Video streaming has emerged as a widely used Internet service, in which adaptive bitrate (ABR) algorithms play a critical role in delivering high quality of experience (QoE). However, existing learning-based ABR methods often suffer from limited generalization in unseen and dynamically changing network conditions. Although some meta-reinforcement learning techniques have been proposed to mitigate this issue, they generally depend on additional online training or fine-tuning. To overcome these limitations, this paper introduces EAStream, an environment-aware ABR algorithm based on meta-reinforcement learning for reliable video streaming services. The method employs a variational autoencoder to extract a latent representation of the current network environment from historical interaction data. This latent variable, along with the current system state, is fed into a policy network that perceives network conditions in real time and adapts bitrate decisions accordingly, without requiring further online training. A comprehensive evaluation is conducted using diverse real-world network traces. Experimental results show that EAStream not only achieves leading performance on in-distribution test sets compared to state-of-the-art ABR algorithms, but also demonstrates superior generalization capability on out-of-distribution test scenarios.

Vehicular edge computing (VEC) is an emerging computing paradigm that is rapidly advancing the development of the Internet of Vehicles (IoV). However, edge server has limited data storage capacity and computing resource, making it difficult to handle the massive offloading requests from IoV applications. Moreover, the mobility of vehicles and dynamic data traffic make it highly challenging to design optimal offloading and resource allocation strategies. To address the challenges mentioned above, we design a cloud-edge–vehicle hierarchical architecture for IoV task offloading, introducing a cloud server to assist in computation and alleviate the overload pressure on edge server. Considering the impact of vehicle mobility on task offloading, we propose a mobility detection method to predict which vehicles might leave the communication range of the base station, thereby preventing task offloading failures. Additionally, to achieve efficient task offloading and resource allocation in this complex IoV system, we propose a multiagent-reinforcement-learning-based vehicle proposal offloading algorithm (MVPOA). This algorithm enables vehicles to autonomously decide whether to process tasks locally or propose offloading to edge server. The edge server then decides whether to accept offloading requests based on task priority and sends rejected tasks to cloud server for processing, thereby maximizing the utilization of resources at each layer of the system. Simulation results demonstrate that MVPOA outperforms other baseline approaches in optimizing system delay and energy consumption.

Enhancing the communication rate and quality has become the primary goal for the development of next-generation mobile communication networks, and traditional techniques such as MIMO and increasing the transmit power of the base station (BS) have not achieved a leapfrog effect. The emergence of Intelligent Reflective Surfaces (IRS) provides more reliable technical support for providing high-energy and high-rate communications. However, IRS-aided joint optimization with communication rate and quality of service constraints is a non-convex non-smooth optimization problem, and the optimal global solution cannot be obtained due to its computational complexity. In this paper, we propose an intelligent Advanced Non-convex Non-smooth Scheme (ANNS) in IRS-aided Next Generation Mobile Communication Networks for making the transmission rate of mission communication and communication quality effective. To ensure that the inequality constraints in the joint optimization problem are not violated and the equation constraints are satisfied, a hybrid deep reinforcement learning and data-experience-driven constraint security layer network is proposed, which maps the constraint violations into the safe feasible domain by mapping the constraint variables into the constraint variables mapping method, and the convergence of the algorithm is theoretically demonstrated. Experimental results show that the proposed ANNS performs superior optimization compared to SAC, DDPG, and A2C for solving non-convex non-smooth problems. The proposed ANNS has the potential to be generalized to other mobile computing applications with non-convex non-smooth characteristics.

Human-centric communication metaverse relies on the convergent integration of multiple existing technologies such as 5G and beyond networks, virtual reality, augmented reality, and digital twins, and thus their security vulnerabilities and vulnerability to interference attacks may also be inherited by the metaverse. In particular, existing security policies may be inefficient for communication interference problems encountered in multi-device collaborative computing in a metaverse 5G environment and lack adaptability to metaverse applications. In this paper, we propose a novel intelligent game anti-interference collaborative computing model that accurately describes the interference relationships among source devices, cooperating devices, and interferers in metaverse collaborative computing. We model the offensive and defensive confrontation between multiple metaverse devices as a Stackelberg game, where the source device is the leader, the collaborative computing device acts as the sub-leader, adjusts its antijamming strategy according to the source device’s strategy to improve the source device’s communication anti-jamming performance, and the jammer acts as the follower. We design an intelligent Stackelberg Game-theoretic Policy-based Learning (SGPL) algorithm for jamming resistance in metaverses over 5G and Beyond Networks, where the leaders (co-computing devices) update their training parameters using the total derivatives of the objective function, while the followers (i.e., jammers) update their training parameters using an independent gradient dynamics strategy. Loops are eased and convergence is accelerated by introducing differential dynamics into the leader training network to reflect the interaction structure of the critic and actor network layers. Finally, numerical results demonstrate the effectiveness of the proposed SGPL algorithm in metaverse anti-jamming countermeasures. The proposed SGPL algorithm has the potential to be generalized to other metaverse applications with multi-user attack and defense characteristics.

The multiuser cooperative offloading mobile edge computing (COMEC) system has attracted much attention because it can realize delay-sensitive tasks. However, in the coupling optimization of offloading decision and resource allocation, the existing numerical optimization algorithms are difficult to obtain high-quality optimization solutions. In this paper, we propose a differential dynamic gradient descent (DDGD) optimization algorithm to solve the above optimization problems. DDGD algorithm decomposes the constrained NP-hard optimization problem into two network layers and integrates the constraint function into a larger end-to-end training network. These two-layer networks encode the dependencies and optimization constraints between parameter hidden states, which cannot be captured by a numerical optimization model or a full connection layer neural network. Because the ring learning and self-repeating learning architecture are adopted and the information is stored in the differential dynamics network, the proposed algorithm can achieve better and more intelligent decision-making in searching the solution trajectory without setting accurate parameters in advance and reduce the complexity of the network. We show that compared with the baseline method, the DDGD method has superior optimization performance in the energy consumption optimization of COMEC.