An adaptive dual-supervised cross-deep dependency network for pixel-wise classification

IEEE Transactions on Geoscience and Remote Sensing, Journal Year: 2025, Volume and Issue: 63, P. 1 - 13

Published: Jan. 1, 2025

Language: Английский

---

6G virtualized beamforming: a novel framework for optimizing massive MIMO in 6G networks

EURASIP Journal on Wireless Communications and Networking, Journal Year: 2025, Volume and Issue: 2025(1)

Published: April 9, 2025

Language: Английский

---

Survey of Next-generation Optical Wireless Communication Technologies for 6G and Beyond 6G

ICT Express, Journal Year: 2025, Volume and Issue: unknown

Published: April 1, 2025

Language: Английский

---

Network Function Placement in Virtualized Radio Access Network with Reinforcement Learning Based on Graph Neural Network

Electronics, Journal Year: 2025, Volume and Issue: 14(8), P. 1686 - 1686

Published: April 21, 2025

In 5G and beyond networks, function placement is a crucial strategy for enhancing the flexibility efficiency of Radio Access Network (RAN). However, demonstrating optimal splitting to meet diverse user demands remains significant challenge. The problem known be NP-hard, previous studies have attempted address it using Deep Reinforcement Learning (DRL) approaches. Nevertheless, many existing methods fail capture network state in RANs with specific topologies, leading suboptimal decision-making resource allocation. this paper, we propose method referred as GDRL, which deep reinforcement learning approach that utilizes graph neural networks functional problem. To ensure policy stability, design gradient algorithm called Graph Proximal Policy Optimization (GPPO), integrates GNNs into both actor critic networks. By incorporating node edge features, GDRL enhances feature extraction from RAN’s nodes links, providing richer observational data evaluation. This, turn, enables more accurate effective decision outcomes. addition, formulate mixed-integer nonlinear programming model aimed at minimizing number active computational while maximizing centralization level virtualized RAN (vRAN). We evaluate across different scenarios varying configurations. results demonstrate our achieves superior outperforms several overall performance.

Language: Английский

---

Research on Sex Roles & Marital Satisfaction based on Artificial Intelligence Model

Published: June 28, 2024

Language: Английский

---

Deep Reinforcement Learning Explores EH‐RIS for Spectrum‐Efficient Drone Communication in 6G

IET Signal Processing, Journal Year: 2024, Volume and Issue: 2024(1)

Published: Jan. 1, 2024

Reconfigurable intelligent surfaces (RISs) have emerged as a groundbreaking technology, revolutionizing wireless networks with enhanced spectrum and energy efficiency (EE). When integrated drones, the combination offers ubiquitous deployment services in communication‐constrained areas. However, limited battery life of drones hampers their performance. To address this, we introduce an innovative harvesting (EH), that is, EH‐RIS. EH‐RIS strategically divides passive reflection arrays across geometric space, improving EH information transformation (IT). Employing meticulous, exhaustive search algorithm, resources drone‐RIS system are dynamically allocated time space to maximize harvested while ensuring optimal communication quality. Deep reinforcement learning (DRL) is employed investigate performance by intelligently allocating for signal reflection. The results demonstrate effectiveness DRL‐based simultaneous power transfer (SWIPT) system, demonstrating spectrum‐efficient capabilities. Our investigation summarized unleashing potential, which shows how DRL work together optimize next‐generation networks.

Language: Английский

---

Reward-Based Energy-Aware Routing Protocols in Wireless Sensor Networks

Sustainable civil infrastructures, Journal Year: 2024, Volume and Issue: unknown, P. 908 - 914

Published: Jan. 1, 2024

Language: Английский

---