An Adaptive Cooperative Caching Strategy for Vehicular Networks

IEEE Transactions on Mobile Computing, Journal Year: 2024, Volume and Issue: 23(10), P. 9502 - 9517

Published: Feb. 20, 2024

Edge caching has emerged as an effective solution to the challenges posed by massive content delivery in vehicular network. In networks, vehicles and roadside units (RSUs) can serve intermediate relays with capabilities. However, due mobility of vehicles, topology edge network changes frequently, which leads frequent link interruptions increases transmission delay. This paper proposes adaptive cooperative (ACC) strategy adapt describes optimization problem minimize average Then, is transformed into two sub-optimization problems: multiplechoice knapsack (MCK) multiple minimum-weight dominating set (MMWDS) problem. Finally, greedy algorithms low complexity are designed solve above problems obtain approximate solutions optimal decision. Simulation results show that ACC effectively improve cache hit rate reduce delay communication overhead compared other strategies.

Language: Английский

---

Joint Communication and Sensing Design in Coal Mine Safety Monitoring: 3-D Phase Beamforming for RIS-Assisted Wireless Networks

IEEE Internet of Things Journal, Journal Year: 2023, Volume and Issue: 10(13), P. 11306 - 11315

Published: Feb. 9, 2023

This article investigates the resource allocation of a reconfigurable intelligent surface (RIS)-aided joint communication and sensing (JCAS) system in coal mine scenario. In JCAS system, an RIS is implemented at corner zigzag tunnels to improve complicated wireless environment, where ground obstacles frequently block direct links. addition, backhaul base station with limited energy budget deployed depth sense target area provide Internet Things (IoT) services for users. Furthermore, data center placed on analyze obtained route data. Under this deployment, optimization problem phase-shift matrix, element switches, time proposed. We aim maximize successful sensed bits under total completion time, maximum transmit power constraints. order solve problem, iterative algorithm The successive convex approximation (SCA)-based used matrix subproblem. For subproblem, quadratic method proposed optimize number perceptions. coordinate descent utilized switches. Simulation results show that efficiency improved by up 38%, 7% increases specific size compared benchmark solutions.

Language: Английский

---

ISFL: Federated Learning for Non-i.i.d. Data With Local Importance Sampling

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(16), P. 27448 - 27462

Published: May 8, 2024

As a promising learning paradigm integrating computation and communication, federated (FL) proceeds the local training periodic sharing from distributed clients. Due to non-i.i.d. data distribution on clients, FL model suffers gradient diversity, poor performance, bad convergence, etc. In this work, we aim tackle key issue by adopting importance sampling (IS) for training. We propose (ISFL), an explicit framework with theoretical guarantees. Firstly, derive convergence theorem of ISFL involve effects sampling. Then, formulate problem selecting optimal IS weights obtain solutions. also employ water-filling method calculate develop algorithms. The experimental results CIFAR-10 fit proposed theorems well verify that reaps better efficiency, as explainability data. To best our knowledge, is first solution aspect which exhibits compatibility neural network models. Furthermore, approach, can be easily migrated into other emerging frameworks.

Language: Английский

---

Semantic-Aware Spectrum Sharing in Internet of Vehicles Based on Deep Reinforcement Learning

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(23), P. 38521 - 38536

Published: Aug. 23, 2024

Language: Английский

---

AFL-DMAAC: Integrated Resource Management and Cooperative Caching for URLLC-IoV Networks

IEEE Transactions on Intelligent Vehicles, Journal Year: 2023, Volume and Issue: 9(6), P. 5101 - 5117

Published: Aug. 10, 2023

In this paper, we propose a novel approach for optimal resource management and caching in ultra-reliable low-latency communication (URLLC)-enabled Internet of Vehicles (IoV) networks. The proposed framework includes mobile edge computing (MEC) servers integrated into roadside units (RSUs), unmanned aerial vehicles (UAVs), base stations (BSs) hybrid vehicle-to-vehicle (V2V) vehicle-to-infrastructure (V2I) communication. To enhance the accuracy global model while considering mobility characteristics vehicles, leverage an asynchronous federated learning (AFL) algorithm. problem allocation is formulated to achieve best frequency, computation, resources complying with delay restrictions. solve non-convex problem, multi-agent actor-critic type deep reinforcement algorithm called DMAAC introduced. Additionally, cooperative scheme based on AFL Co-Ca proposed, utilizing Dueling Deep-Q-Network (DDQN) predict frequently accessed contents cache them efficiently. Extensive simulation results show effectiveness algorithms compared existing schemes.

Language: Английский

---

Efficient Vehicular Edge Computing: A Novel Approach With Asynchronous Federated and Deep Reinforcement Learning for Content Caching in VEC

IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 13196 - 13212

Published: Jan. 1, 2024

Vehicular Edge Computing (VEC) technology holds great promise, but also poses significant challenges to the limited computing power of in-vehicle devices and capacity Roadside Units (RSUs). At same time, highly mobile nature vehicles frequent changes in content requests from make it critical offload applications edge servers effectively predict cache most popular content, so that can be cached advance RSU. And considering protecting privacy vehicle user vehicular users (VUs), traditional data-sharing methods may not suitable for this work, we use an asynchronous Federated learning (FL) approach update global model time at protect personal VUs. Unlike synchronous FL, FL no longer needs wait all finish training uploading local models before updating model, which avoids problem long time. In paper, propose caching scheme based on federated deep reinforcement learning(AFLR), prefetches possible contents caches them nodes or according vehicle's location moving direction while reducing latency requests. After extensive experimental comparisons, AFLR outperforms other benchmark schemes.

Language: Английский

---

Multi-Agent Reinforcement Learning for Efficient Resource Allocation in Internet of Vehicles

Electronics, Journal Year: 2025, Volume and Issue: 14(1), P. 192 - 192

Published: Jan. 5, 2025

The Internet of Vehicles (IoV), a burgeoning technology, merges advancements in the internet, vehicle electronics, and wireless communications to foster intelligent interactions, thereby enhancing efficiency safety transportation systems. Nonetheless, continual high-frequency among vehicles, coupled with regional limitations system capacity, precipitate significant challenges allocating resources for vehicular networks. In addressing these challenges, this study formulates resource allocation issue as multi-agent deep reinforcement learning scenario introduces novel actor-critic framework. This framework incorporates prioritized experience replay mechanism focused on distributed execution, which facilitates decentralized computing by structuring training processes defining specific reward functions, thus optimizing allocation. Furthermore, prioritizes empirical data during phase based temporal difference error (TD error), selectively updating network high-priority at each sampling point. strategy not only accelerates model convergence but also enhances efficacy. validations confirm that our algorithm augments total capacity vehicle-to-infrastructure (V2I) links 9.36% success rate vehicle-to-vehicle (V2V) transmissions 6.74% compared benchmark algorithm.

Language: Английский

---

Federated deep reinforcement learning-based cost-efficient proactive video caching in energy-constrained mobile edge networks

Computer Networks, Journal Year: 2025, Volume and Issue: unknown, P. 111062 - 111062

Published: Jan. 1, 2025

Language: Английский

---

A survey on resource scheduling approaches in multi-access edge computing environment: a deep reinforcement learning study

Cluster Computing, Journal Year: 2025, Volume and Issue: 28(3)

Published: Jan. 21, 2025

Language: Английский

---

Intelligent Inspection of Electronic Devices in Specific Environments via a Novel Cascade Network of Combining Mixed Sampling and Nonstrided Convolution

IEEE Transactions on Systems Man and Cybernetics Systems, Journal Year: 2025, Volume and Issue: 55(5), P. 3287 - 3299

Published: Feb. 20, 2025

Language: Английский

---