In vehicular edge computing (VEC), some tasks can be processed either locally or on the mobile edge computing (MEC) server at a base station (BS) or a nearby vehicle. In fact, tasks are offloaded or not, based on the status of vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication. In this paper, device-to-device (D2D)-based V2V communication and multiple-input multiple-output and nonorthogonal multiple access (MIMO-NOMA)-based V2I communication are considered. In actual communication scenarios, the channel conditions for MIMO-NOMA-based V2I communication are uncertain, and the task arrival is random, leading to a highly complex environment for VEC systems. To solve this problem, we propose a power allocation scheme based on decentralized deep reinforcement learning (DRL). Since the action space is continuous, we employ the deep deterministic policy gradient (DDPG) algorithm to obtain the optimal policy. Extensive experiments demonstrate that our proposed approach with DRL and DDPG outperforms existing greedy strategies in terms of power consumption and reward.
Black-burst based multi-hop broadcast protocols are effective means to disseminate Emergency messages (EMs) in Vehicular ad hoc networks (VANETs). However, Clear to broadcast (CTB) collision will happen and reduce propagation speed. Aiming at this problem, we propose a Black-burst and multi-channel based multihop broadcast (BMMB) protocol. Vehicles with multiple antennas can transmit and sense black-burst at different channels simultaneously based on multi-channel technology. Compared with existing black-burst based methods, BMMB shortens the iterative procedure to find out the optimal segment. Only one relay vehicle can be rapidly selected within the optimal segment without CTB collision. BMMB enables alternative broadcast methods, i.e., unidirectional broadcast and multi-directional broadcast for straight roads and intersections respectively. Theoretical analysis is done for performance evaluation of BMMB, and simulation results demonstrate that BMMB performs better in terms of average one-hop delay and propagation speed.
Sag is an important indicator of the operational health of a transmission line, and its timely measurement is of great significance to maintain the stability and reliability of power systems. However, traditional contact measurements may be affected by the electromagnetic interference of conductors. In contrast, measurement methods without direct electrical contact with the subject provide greater portability and flexibility. This paper presents a study of a transmission line sag measurement and simulation based on non-contact electric field sensing. The finite element method was used to analyze the conductor distribution, establish the coupling relationships among the electric field, transmission line, and measurement point, propose a sag inverse calculation model, and assess the impact of the transmission line parameter on the curved drooping measurement. Simultaneously, sag measurement schemes for single-round and dual-circuit lines were designed for multi-conductive lines, and measurement array studies were conducted. The vertical component of the electric field in space measured by the array was obtained, which could be used to perform conductor sag measurement simply and efficiently. The proposed method will facilitate the monitoring of the overhead transmission line status, which is conducive to the effective operation of the entire system.
We study the temporal and spectral variability properties of the high-redshift blazar B3 1343+451 utilizing Fermi-LAT data from 2008 to 2022 in the energy range of 0.1–300 GeV. We identify six major flares with many substructures and analyze their temporal and spectral properties in detail. The fastest rise and decay timescales are found to be 4.8 ± 0.48 h and 5.28 ± 0.72 h, respectively. The size of the emission region is constrained to be R ∼ 5.18 × –1.56 × cm with the typical Doppler factors of ∼ 10–30. Most of the peaks from the flares exhibit a symmetric temporal profile within the error bars, implying that the rise and decay timescales are dominated by the disturbances caused by dense plasma blobs passing through the standing shock front in the jet region. We also find that four flares are better fitted with a log-parabolic distribution, while two flares are better fitted with a power-law distribution. Our results indicate that the emission regions vary from one flare to another, which is consistent with earlier results.
The deployment of Federated Learning (FL) in edge environments is often impeded by system heterogeneity, non-independent and identically distributed (non-IID) data, and constrained communication resources, which collectively hinder training efficiency and scalability. To address these challenges, this paper presents FLEX-SFL, a flexible and efficient split federated learning framework that jointly optimizes model partitioning, client selection, and communication scheduling. FLEX-SFL incorporates three coordinated mechanisms: a device-aware adaptive segmentation strategy that dynamically adjusts model partition points based on client computational capacity to mitigate straggler effects; an entropy-driven client selection algorithm that promotes data representativeness by leveraging label distribution entropy; and a hierarchical local asynchronous aggregation scheme that enables asynchronous intra-cluster and inter-cluster model updates to improve training throughput and reduce communication latency. We theoretically establish the convergence properties of FLEX-SFL under convex settings and analyze the influence of local update frequency and client participation on convergence bounds. Extensive experiments on benchmark datasets including FMNIST, CIFAR-10, and CIFAR-100 demonstrate that FLEX-SFL consistently outperforms state-of-the-art FL and split FL baselines in terms of model accuracy, convergence speed, and resource efficiency, particularly under high degrees of statistical and system heterogeneity. These results validate the effectiveness and practicality of FLEX-SFL for real-world edge intelligent systems.