Secure Transmission Scheme Based on Joint Radar and Communication in Mobile Vehicular Networks

IEEE Transactions on Intelligent Transportation Systems, Journal Year: 2023, Volume and Issue: 24(9), P. 10027 - 10037

Published: May 5, 2023

Vehicle-to-vehicle (V2V) communication applications face significant challenges to security and privacy since all types of possible breaches are common in connected autonomous vehicles (CAVs) networks. As an inheritance from conventional wireless services, potential eavesdropping is one the main threats V2V communications. In our work, anti-eavesdropping scheme CAVs networks developed through use cognitive risk control (CRC)-based vehicular joint radar-communication (JRC) system. particular, supplement off-board measurements acquired using links perceptual information has presented enhance traffic target positioning precision. Then, transmission power performed utilizing reinforcement learning, result which determined by a task switcher. Based on threat evaluation, multiple armed bandit problem designed implement secret key switching procedure when it needed. Through constant perception-execution loops (PELs), confidentiality improved for authorized their behavioral interactions with illegal eavesdropper. Numerical experiments have that approach anticipated performance terms some assessment indicators.

Language: Английский

---

URLLC-Awared Resource Allocation for Heterogeneous Vehicular Edge Computing

IEEE Transactions on Vehicular Technology, Journal Year: 2024, Volume and Issue: 73(8), P. 11789 - 11805

Published: Feb. 29, 2024

Vehicular edge computing (VEC) is a promising technology to support real-time vehicular applications, where vehicles offload intensive computation tasks the nearby VEC server for processing. However, traditional that relies on single communication cannot well meet requirement task offloading, thus heterogeneous integrating advantages of dedicated short-range communications (DSRC), millimeter-wave (mmWave) and cellular-based vehicle infrastructure (C-V2I) introduced enhance capacity. The resource allocation may significantly impact ultra-reliable low-latency (URLLC) performance system utility, in this case, how do allocations becoming necessary. In paper, we consider with multiple technologies various types tasks, propose an effective policy minimize utility while satisfying URLLC requirement. We first formulate optimization problem under constraint which modeled by moment generating function (MGF)-based stochastic network calculus (SNC), then present Lyapunov-guided deep reinforcement learning (DRL) method convert solve problem. Extensive simulation experiments illustrate proposed approach effective.

Language: Английский

---

Cooperative Edge Caching Based on Elastic Federated and Multi-Agent Deep Reinforcement Learning in Next-Generation Networks

IEEE Transactions on Network and Service Management, Journal Year: 2024, Volume and Issue: 21(4), P. 4179 - 4196

Published: May 21, 2024

Edge caching is a promising solution for next-generation networks by empowering units in small-cell base stations (SBSs), which allows user equipments (UEs) to fetch users' requested contents that have been pre-cached SBSs. It crucial SBSs predict accurate popular through learning while protecting personal information. Traditional federated (FL) can protect privacy but the data discrepancies among UEs lead degradation model quality. Therefore, it necessary train personalized local models each UE accurately. In addition, cached be shared adjacent networks, thus predicted different may affect cost contents. Hence, critical determine where are cooperatively. To address these issues, we propose cooperative edge scheme based on elastic and multi-agent deep reinforcement (CEFMR) optimize network. We first an FL algorithm UE, adversarial autoencoder (AAE) adopted training improve prediction accuracy, then content proposed SBS trained AAE model. Finally, (MADRL) decide collaboratively Our experimental results demonstrate superiority of our existing baseline schemes.

Language: Английский

---

Delay-Sensitive Task Offloading in Vehicular Fog Computing-Assisted Platoons

IEEE Transactions on Network and Service Management, Journal Year: 2023, Volume and Issue: 21(2), P. 2012 - 2026

Published: Oct. 11, 2023

Vehicles in platoons need to process many tasks support various real-time vehicular applications. When a task arrives at vehicle, the vehicle may not due its limited computation resource. In this case, it usually requests offload other vehicles platoon for processing. However, when resources of all are insufficient, cannot be processed time through offloading platoon. Vehicular fog computing (VFC)-assisted can solve problem VFC which is formed by driving near Offloading delay an important performance metric, impacted both strategy deciding where offloaded and number allocated task. Thus, critical propose minimize delay. VFC-assisted system, adopt IEEE 802.11p distributed coordination function (DCF) mechanism while having resources. Moreover, arrive depart randomly, their also system randomly. paper, we semi-Markov decision (SMDP) based considering these factors obtain maximal long-term reward reflecting Our research provides robust systems, effectiveness demonstrated simulation experiments comparison with benchmark strategies.

Language: Английский

---

Algorithmic Approach to Virtual Machine Migration in Cloud Computing with Updated SESA Algorithm

Sensors, Journal Year: 2023, Volume and Issue: 23(13), P. 6117 - 6117

Published: July 3, 2023

Cloud computing plays an important role in every IT sector. Many tech giants such as Google, Microsoft, and Facebook deploying their data centres around the world to provide computation storage services. The customers either submit job directly or they take help of brokers for submission jobs cloud centres. preliminary aim is reduce overall power consumption which was ignored early days development. This due performance expectations from servers were supposed all services through layers IaaS, PaaS, SaaS. As time passed researchers came up with new terminologies algorithmic architecture reduction sustainability, other anarchies also introduced, statistical oriented learning bioinspired algorithms. In this paper, indepth focus has been done on multiple approaches migration among virtual machines find out various issues existing approaches. proposed work utilizes elastic scheduling inspired by smart algorithm (SESA) develop a more energy-efficient VM allocation algorithm. uses cosine similarity bandwidth utilization additional utilities improve current terms QoS. evaluated service level agreement violation (SLA-V) compared related state art techniques. A presented order solve problems found during survey.

Language: Английский

---

A Power Allocation Scheme for MIMO-NOMA and D2D Vehicular Edge Computing Based on Decentralized DRL

Sensors, Journal Year: 2023, Volume and Issue: 23(7), P. 3449 - 3449

Published: March 25, 2023

In vehicular edge computing (VEC), some tasks can be processed either locally or on the mobile (MEC) server at a base station (BS) nearby vehicle. fact, are offloaded not, based status of vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication. this paper, device-to-device (D2D)-based V2V communication multiple-input multiple-output nonorthogonal multiple access (MIMO-NOMA)-based V2I considered. actual scenarios, channel conditions for MIMO-NOMA-based uncertain, task arrival is random, leading to highly complex environment VEC systems. To solve problem, we propose power allocation scheme decentralized deep reinforcement learning (DRL). Since action space continuous, employ deterministic policy gradient (DDPG) algorithm obtain optimal policy. Extensive experiments demonstrate that our proposed approach with DRL DDPG outperforms existing greedy strategies in terms consumption reward.

Language: Английский

---

A Survey on Video Streaming for Next-Generation Vehicular Networks

Electronics, Journal Year: 2024, Volume and Issue: 13(3), P. 649 - 649

Published: Feb. 4, 2024

As assisted driving technology advances and vehicle entertainment systems rapidly develop, future vehicles will become mobile cinemas, where passengers can use various multimedia applications in the car. In recent years, progress has given rise to immersive video experiences. addition conventional 2D videos, 360° videos are gaining popularity, volumetric which offer users a better experience, have been discussed. However, these place high demands on network capabilities, leading dependence next-generation wireless communication address bottlenecks. Therefore, this study provides an exhaustive overview of latest advancements streaming over vehicular networks. First, we introduce related work background knowledge, provide developments networking types. Next, detail processing technologies, including released standards. Detailed explanations provided for strategies technologies that optimize transmission networks, paying special attention relevant literature regarding current development 6G is applied communication. Finally, proposed research directions challenges. Building upon introduced paper considering diverse applications, suggest suitable architecture transmission.

Language: Английский

---

Distributed Deep Reinforcement Learning Based Gradient Quantization for Federated Learning Enabled Vehicle Edge Computing

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 12(5), P. 4899 - 4913

Published: Aug. 21, 2024

Language: Английский

---

A Critical Analysis of Cooperative Caching in Ad Hoc Wireless Communication Technologies: Current Challenges and Future Directions

Sensors, Journal Year: 2025, Volume and Issue: 25(4), P. 1258 - 1258

Published: Feb. 19, 2025

The exponential growth of wireless traffic has imposed new technical challenges on the Internet and defined approaches to dealing with its intensive use. Caching, especially cooperative caching, become a revolutionary paradigm shift advance environments based technologies enable efficient data distribution support mobility, scalability, manageability networks. Mobile ad hoc networks (MANETs), mesh (WMNs), Wireless Sensor Networks (WSNs), Vehicular (VANETs) have adopted caching practices overcome these hurdles progressively. In this paper, we discuss problems issues in current paradigms as well spotlight versatile potential solution increasing complications We classify multiple schemes distinct communication contexts highlight advantages applicability. Moreover, identify research directions further study enhance mechanisms concerning This extensive review offers useful findings design sound strategies pursuit enhancing next-generation

Language: Английский

---

Asynchronous Federated and Reinforcement Learning for Mobility-Aware Edge Caching in IoV

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(9), P. 15334 - 15347

Published: Jan. 2, 2024

Edge caching is a promising technology to reduce backhaul strain and content access delay in Internet of Vehicles (IoV). It precaches frequently used contents close vehicles through intermediate roadside units. Previous edge works often assume that popularity known advance or obeys simplified models. However, such assumptions are unrealistic, as varies with uncertain spatial-temporal traffic demands IoVs. Federated learning (FL) enables predict popular distributed training. preserves the training data remain local, thereby addressing privacy concerns communication resource shortages. This article investigates mobility-aware strategy by exploiting asynchronous FL deep reinforcement (DRL). We first implement novel framework for local updates global aggregation stacked autoencoder (SAE) Then, utilizing latent features extracted trained SAE model, we adopt hybrid filtering model predicting recommending content. Furthermore, explore intelligent decisions after prediction. Based on formulated Markov decision process (MDP) problem, propose DRL-based solution, neural network-based parameter approximations curse dimensionality RL. Extensive simulations conducted based real-world trajectory. Especially, our proposed method outperforms federated averaging, least recently used, NoDRL, hit rate improved roughly 6%, 21%, 15%, respectively, when cache capacity reaches 350 MB.

Language: Английский

---