Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Resilience Optimization in 6G and Beyond Integrated Satellite-Terrestrial Networks: A Deep Reinforcement Learning Approach

Dinh-Hieu Tran, Nguyen Van Huynh, Van Nhan Vo, Madyan Alsenwi, Eva Lagunas, Symeon Chatzinotas

TL;DR

This work addresses resilience of 6G ISTNs under base-station outages by integrating LEOS backups with terrestrial gNBs. It formulates a joint, NP-hard optimization to maximize total user throughput, while minimizing LEOS usage under QoS and RSRP constraints, and solves it with a Deep Q-Network that learns policies for antenna downtilt and transmit power and for LEOS engagement. The approach is grounded in realistic channel and 3D antenna models and is framed as an MDP with detailed state/action/reward definitions. Numerical results show that the DRL method outperforms a Q-learning baseline in convergence speed, throughput, and coverage stability, demonstrating enhanced resilience and energy-aware satellite utilization in ISTNs.

Abstract

Ensuring network resilience in 6G and beyond is essential to maintain service continuity during base station (BS) outages due to failures, disasters, attacks, or energy-saving operations. This paper proposes a novel resilience optimization framework for integrated satellite-terrestrial networks (ISTNs), leveraging low Earth orbit (LEO) satellites to assist users when terrestrial BSs are unavailable. Specifically, we develop a realistic multi-cell model incorporating user association, antenna downtilt adaptation, power control, heterogeneous traffic demands, and dynamic user distribution. The objective is to maximize of the total user rate in the considered area by optimizing the BS's antenna tilt, transmission power, user association to neighboring BS or to a LEO satellite with a minimum number of successfully served user satisfaction constraint, defined by rate and Reference Signal Received Power (RSRP) requirements. To solve the non-convex, NP-hard problem, we design a deep Q-network (DQN)-based algorithm to learn network dynamics to maximize throughput while minimizing LEO satellite usage, thereby limiting reliance on links with longer propagation delays and prolonging satellite operational lifetime. Simulation results confirm that our approach significantly outperforms the benchmark one.

Resilience Optimization in 6G and Beyond Integrated Satellite-Terrestrial Networks: A Deep Reinforcement Learning Approach

TL;DR

This work addresses resilience of 6G ISTNs under base-station outages by integrating LEOS backups with terrestrial gNBs. It formulates a joint, NP-hard optimization to maximize total user throughput, while minimizing LEOS usage under QoS and RSRP constraints, and solves it with a Deep Q-Network that learns policies for antenna downtilt and transmit power and for LEOS engagement. The approach is grounded in realistic channel and 3D antenna models and is framed as an MDP with detailed state/action/reward definitions. Numerical results show that the DRL method outperforms a Q-learning baseline in convergence speed, throughput, and coverage stability, demonstrating enhanced resilience and energy-aware satellite utilization in ISTNs.

Abstract

Ensuring network resilience in 6G and beyond is essential to maintain service continuity during base station (BS) outages due to failures, disasters, attacks, or energy-saving operations. This paper proposes a novel resilience optimization framework for integrated satellite-terrestrial networks (ISTNs), leveraging low Earth orbit (LEO) satellites to assist users when terrestrial BSs are unavailable. Specifically, we develop a realistic multi-cell model incorporating user association, antenna downtilt adaptation, power control, heterogeneous traffic demands, and dynamic user distribution. The objective is to maximize of the total user rate in the considered area by optimizing the BS's antenna tilt, transmission power, user association to neighboring BS or to a LEO satellite with a minimum number of successfully served user satisfaction constraint, defined by rate and Reference Signal Received Power (RSRP) requirements. To solve the non-convex, NP-hard problem, we design a deep Q-network (DQN)-based algorithm to learn network dynamics to maximize throughput while minimizing LEO satellite usage, thereby limiting reliance on links with longer propagation delays and prolonging satellite operational lifetime. Simulation results confirm that our approach significantly outperforms the benchmark one.
Paper Structure (12 sections, 15 equations, 2 figures)

This paper contains 12 sections, 15 equations, 2 figures.

Table of Contents

  1. Introduction
  2. System Model
  3. Channel Model
  4. Antenna Model
  5. Integrated Satellite-Terrestrial for Network Resilience Problem
  6. Deep Q-Network-based Solution
  7. State Space
  8. Action Space
  9. Reward Function
  10. Deep Q-Network
  11. Simulation Results
  12. Conclusions

Figures (2)

  • Figure 1: Average rewards vs. number of iterations.
  • Figure 2: Average number of RSRP Users vs. Iterations.