Vehicular edge computing (VEC) is a new paradigm that has great potential to enhance the capability of vehicle terminals (VTs) support resource-hungry in-vehicle applications with low latency and high energy efficiency. In this article, we investigate an important computation offloading scheduling problem in typical VEC scenario, where VT traveling along expressway intends schedule its tasks waiting queue minimize long-term cost terms tradeoff between task consumption. Due diverse...

Driven by the tremendous application demands, Internet of Things (IoT) systems are expected to fulfill computation-intensive and latency-sensitive sensing computational tasks, which pose a significant challenge for IoT devices with limited ability battery capacity. To address this problem, edge computing is promising architecture where can offload their tasks servers. Current works on task offloading often overlook unique topologies schedules from devices, leading degraded performance...

Mobile Edge Computing (MEC) is a new computing paradigm with great potential to enhance the performance of user equipment (UE) by offloading resource-hungry computation tasks lightweight and ubiquitously deployed MEC servers. In this paper, we investigate problem decision resource allocation among multiple users served one base station achieve optimal system-wide utility, which defined as trade-off between task latency energy consumption. Mobility in process considered optimization. We prove...

The recent advances in microelectronics and communications have led to the development of large-scale Internet Things (IoT) networks, where tremendous sensory data is generated needs be processed. To support realtime processing for IoT, deploying edge servers with storage computational capability a promising approach. In this paper, we carefully analyze impacting factors key challenges node (EN) deployment. We then propose novel three-phase deployment approach which considers both traffic...

Workflow scheduling plays a critical role in optimizing completion time and throughput distributed cloud environments, leveraging the parallelism of heterogeneous computing resources. However, existing workflow algorithms often fall short due to heuristic limitations challenges adaptability within settings, leading suboptimal solutions. In this paper, we present novel deep reinforcement learning (DRL) framework tailored for continuous environments. First, propose an intelligent scheduler...

In oilfield extraction activities, traditional downhole condition monitoring is typically conducted using dynamometer cards to capture the dynamic changes in load and displacement of sucker rod. However, this method has severe limitations terms real-time performance maintenance costs, making it difficult meet demands modern extraction. To overcome these shortcomings, paper proposes a novel fault detection based on analysis motor power parameters. Through mathematical modeling pumping unit...

Existing production forecasting methods often suffer from limited predictive accuracy due to their reliance on single-source data and the insufficient incorporation of physical principles. To address these challenges, this study proposes a mechanism–data fusion model that integrates mechanistic outputs with data-driven learning techniques. The proposed method first establishes three-phase-separator generate physics-informed simulation data. Then, Global–Local Branch Prediction Model is...

Federated learning (FL) has shown its great potential for achieving distributed intelligence in privacy-sensitive IoT. However, popular FL approaches, such as FedAvg and variants share model parameters among clients during the training process thus cause significant communication overhead Moreover, nonindependent identically (non-IID) data across devices severely affect convergence speed of FL. To address these challenges, we propose a communication-efficient framework based on Two-step...

Inspired by mobile edge computing (MEC), vehicular (VEC) enables vehicle terminals to support resource-hungry on-vehicle applications with significantly lower latency and less energy consumption. In this paper, we investigate the computation offloading problem in a typical VEC scenario, where offloads its tasks servers deployed road side unit (RSU) minimize long-term user cost. The mobility of coupled high dynamics environment makes particularly difficult. To tackle challenge, deep...

Training high-quality machine learning models on distributed systems is a critical issue to achieve edge intelligence in wireless communications. Conventional data-driven approaches are infeasible due non-IID data caused by privacy issues and the limited communication resources networks. Besides, considering complex user identities, training process also faces challenges of Byzantine devices, which can inject poisoning information into models. In this paper, we propose two-step federated...

Transformers are transforming the landscape of computer vision, especially for image-level recognition and instance-level detection tasks. Human-object interaction transformer (HOI-TR) is first transformer-based end-to-end learning system human-object (HOI) detection; vision transformers build a simple multi-stage structure multi-scale representation with single-scale patch patch-based architecture detection. In this paper, we detector (MHOI), novel method to integrate Vision HOI...

Federated learning (FL) is an attractive distributed machine framework due to the property of privacy preservation. The implementation FL encounters challenge Non-Independent and Identically Distributed (Non-IID) data across devices. This work focuses on mitigating impact Non-IID datasets in wireless communications. To achieve this goal, we propose a generative models-based federated augmentation strategy (FedDA) with preservation communication efficiency. In FedDA, Conditional AutoEncoder...

Suspension-based locks are widely used in realtime systems to coordinate simultaneous accesses exclusive shared resources. Although suspension-based have been well studied for sequential real-time tasks, little work has done on this topic parallel tasks. This paper the first time studies problem of how extend existing sequential-task locking protocols and their analysis techniques task model. More specifically, we two OMLP OMIP, which were designed clustered scheduling federated develop...

The quality of experience (QoE) perceived by users is a critical performance measure for Web browsing. "Above-The-Fold" (ATF) time has been recently recognized and widely used as direct user-end QoE number studies. To reduce the ATF time, existing works mainly focus on reducing delay networking. However, we observe that webpage structures content orders can also significantly affect theWeb QoE. In this paper, propose novel optimization framework reorders objects to minimize time. Our core...

Human identification is of great importance for personalized and ubiquitous computing services. As the rapid deployment wireless access, data channel state information (CSI) gathered from networks has become a useful way recognizing human activities. In this paper, we investigate problem based on biometrics using big CSI. The rationale that human's behavioral movements cause unique impacts CSI, which can be used to recognize corresponding further identify different persons in nonintrusive...