PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
March 29, 2026Journal of Marine Science and EngineeringOpen Access

A Deep Reinforcement Learning Approach for Joint Resource Allocation in Time-Varying Underwater Acoustic Cooperative Networks

View Full Paper
Ask AI
Bookmark
Share

Authors

LZLiangliang ZengTZTongxing ZhengYWYue Wu

Discussion

Loading...

Member takes

Overview

This research investigates joint optimization of relay selection and power allocation in underwater networks, suggesting improved energy efficiency.

Key Points

  • The aim is to optimize relay selection and power allocation to enhance energy efficiency and throughput in underwater acoustic networks.
  • Utilized a deep hybrid reinforcement learning framework with a parameterized deep Q-Network architecture.
  • Integrated deterministic policy and value-based networks for power control and relay evaluation.
  • Incorporated a prioritized experience replay mechanism to enhance sample efficiency.
  • Analyzed the algorithm’s complexity and convergence properties theoretically.
  • The DHRL algorithm surpassed combinatorial bandit algorithms and conventional deep reinforcement learning in energy efficiency.
  • Demonstrated increased robustness against channel estimation errors in simulations.

Cite This Study

Zeng et al. (2026) studied this question.

synapsesocial.com/papers/69c8c371de0f0f753b39e462https://doi.org/10.3390/jmse14070616
View Full Paper
Ask AI
Bookmark
Share