Massive computational resources are required by a booming number of artificial intelligence (AI) services in the communication network of the smart grid. To alleviate the computational pressure on data centers, edge computing first network (ECFN) can serve as an effective solution to realize distributed model training based on data parallelism for AI services in smart grid. Due to AI services with diversified types, an edge data center has a changing workload in different time periods. Selfish edge data centers from different edge suppliers are reluctant to share their computing resources without a rule for fair competition. AI services-oriented dynamic computational resource scheduling of edge data centers affects both the economic profit of AI service providers and computational resource utilization. This letter mainly discusses the partition and distribution of AI data based on distributed model training and dynamic computational resource scheduling problems among multiple edge data centers for AI services. To this end, a mixed integer linear programming (MILP) model and a Deep Reinforcement Learning (DRL)-based algorithm are proposed. Simulation results show that the proposed DRL-based algorithm outperforms the benchmark in terms of profit of AI service provider, backlog of distributed model training tasks, running time and multi-objective optimization.

Through offloading computing tasks to visible satellites for execution, the satellite edge computing architecture effectively issues the high-delay problem in remote grids (e.g., mountain and desert) when tasks are offloaded to the urban terrestrial cloud (TC). However, existing works are usually limited to offloading tasks in pure satellite networks and make offloading decisions based on the predefined models. Additionally, runtime consumption for offloading decisions is rather high. Furthermore, privacy information may be maliciously sniffed since computing tasks are transmitted via vulnerable satellite networks. In this paper, we study the task-offloading problem in satellite–terrestrial edge computing networks, where tasks can be executed by satellite or urban TC. A privacy leakage scenario is described, and we consider preserving privacy by sending extra random dummy tasks to confuse adversaries. Then, the offloading cost with privacy protection consideration is modeled, and the offloading decision that minimizes the offloading cost is formulated as a mixed-integer programming (MIP) problem. To speed up solving the MIP problem, we propose a deep reinforcement learning-based task-offloading (DRTO) algorithm. In this case, offloading location and bandwidth allocation only depend on the current channel states. Simulation results show that the offloading overhead is reduced by 17.5% and 23.6% compared with pure TC computing and pure SatEC computing, while the runtime consumption of DRTO is reduced by at least 42.6%. The dummy tasks are exhibited to effectively mitigate privacy leakage during offloading.

This paper investigated a non-orthogonal multiple access (NOMA)-based integrated satellite–terrestrial network (ISTN), where each user can select to access a terrestrial base station (BS) or the satellite according to the capacity of BS and their individual transmission requirements. A two-stage algorithm is proposed to solve the achievable sum rate maximizing resource optimization problem. In the first stage, user associations are determined based on individual preference lists and the backhaul capacities of the access points (APs). In the second stage, the power allocation, and the receiving beamforming vectors are optimized alternately. Within each iteration, the closed-form solution for the transmit power is derived. Simulation results show the effectiveness of the proposed algorithm and the benefits brought by NOMA. When the backhaul link capacity of terrestrial BSs is sufficient, users (UEs) prefer to access these BSs. Otherwise, the satellite can offer QoS guarantees to UEs. Furthermore, the overall system performance reaches its optimum when the number of UEs in the system matches the number of receive antennas at the APs.

A vehicle charging network system has to access large-scale heterogeneous terminals to collect charging pile status information, which may also give malicious terminals an opportunity to access. Though some general access authentication solutions aimed at only allowing trusted terminals have been proposed, they are difficult to work with in a vehicle charging network system. First, among various heterogeneous terminals with significant differences in computing capabilities, there are inevitably terminals that cannot support computations required for cryptography-based access authentication schemes. Second, though access authentication schemes based on device fingerprints are independent of terminal computing capabilities, their authentication performance is weak in robustness and high in overhead. Third, the access authentication delay is huge since the system cannot withstand heavy concurrent access requests from large-scale terminals. To address the above problems, we propose a reliable and lightweight trusted access authentication solution for terminals in the vehicle charging network system. By cloud, edge, and local servers cooperating to execute authentication tasks, our Cloud-Edge-End Collaborative architecture effectively alleviates the authentication delay caused by high concurrent requests. Each server in the architecture deploys our well-designed unified trusted access authentication (UATT) model based on device fingerprints. With ingenious data construction and the powerful swin-transformer network, the UATT model can provide robust and low-overhead authentication services for heterogeneous terminals. To minimize authentication latency, we further design an A2C-based authentication task scheduling scheme to decide which server executes the current task. Comprehensive experiments demonstrate our solution can authenticate terminals with an accuracy higher than 98% while reducing the required data packets by two orders of magnitude, and it can effectively reduce authentication latency.

Channel State Information (CSI)-based Human Activity Recognition (HAR) has emerged as a promising Wi-Fi sensing technology. Particularly in the power grids, ubiquitous sensing is crucial for monitoring personnel safety and terminal states. However, given the heterogeneous nature of communication protocol in real‑world infrastructure (e.g., power grids), the real‑world applicability of CSI‑based HAR is severely hampered by a fundamental limitation: mainstream deep learning methods are rigidly coupled to specific channel bandwidth and its corresponding subcarrier layouts. This lack of downward compatibility prevents models trained on 160 MHz Wi-Fi protocols from being deployed in environments using legacy 20 MHz protocols, despite the latter’s subcarriers being fully contained within the former. In this paper, we address this by learning bandwidth-invariant representations from CSI. Our key insight is that CSI measurements from disparate bandwidths are not disparate signals but rather complementary views of the same underlying activity-induced channel frequency response. We introduce Uni-Fi, a contrastive learning framework that pulls these views together in an embedding space, driving the model to learn shared frequency-domain activity features while being robust to specific bandwidth configuration. Extensive evaluations demonstrate that our framework empowers existing CSI-based HAR models with exceptional cross-bandwidth generalization: a model trained solely on 160 MHz CSI maintains high accuracy in a 20 MHz setting, where standard methods fail completely.