Related Concept Videos
Short-distance Transport of Resources
Application of Differentiation to Business
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Correction: He et al. An Edge-Computing-Based Emotion-Aware Adaptive Lighting System for Intelligent Cockpits. <i>Sensors</i> 2026, <i>26</i>, 3489.
Related Experiment Video
Updated: Jul 19, 2026

A Gradient-generating Microfluidic Device for Cell Biology
Published on: August 30, 2007
Deep Deterministic Policy Gradient-Based Resource Allocation Considering Network Slicing and Device-to-Device
Hudson Henrique de Souza Lopes1, Lucas Jose Ferreira Lima1, Telma Woerle de Lima Soares2
1Electrical, Mechanical and Computer (EMC) School of Engineering, Federal University of Goias (UFG), Goiânia 74605010, GO, Brazil.
This study introduces DDPG-KRP, a deep reinforcement learning method for dynamic resource allocation in next-generation mobile networks. It efficiently manages network slices and device-to-device communications, outperforming other algorithms.
More Related Videos
Area of Science:
- Telecommunications Engineering
- Computer Science
- Artificial Intelligence
Background:
- Next-generation mobile networks (B5G/6G) require diverse resource management.
- Network slicing (NS) and device-to-device (D2D) communication are key technologies for efficiency.
- Integrating NS and D2D can enhance spectrum utilization and network performance.
Purpose of the Study:
- To address dynamic resource allocation challenges in wireless networks with NS and D2D.
- To develop an efficient deep reinforcement learning (DRL) approach for this complex problem.
- To optimize resource allocation policies for maximizing long-term network rewards.
Main Methods:
- Proposed DDPG-KRP: a DRL approach combining Deep Deterministic Policy Gradient (DDPG), K-Nearest Neighbors (KNNs), and Reward Penalization (RP).
- DDPG-KRP determines resource allocation policies by maximizing cumulative rewards.
- RP is used to eliminate undesirable actions, refining the DRL agent's decision-making process.
Main Results:
- DDPG-KRP demonstrated efficiency in dynamic resource allocation for wireless networks with slicing.
- The proposed method outperformed other DRL algorithms in simulation experiments.
- Effective management of resources for combined NS and D2D communications was achieved.
Conclusions:
- DDPG-KRP offers a robust solution for resource allocation in advanced mobile networks.
- The integration of DRL, NS, and D2D communication presents a promising direction for future network optimization.
- This approach enhances scalability and performance in diverse network environments.

