Dueling DDQN-Based Adaptive Multi-Objective Handover Optimization for LEO Satellite Networks

TL;DR

This paper introduces a DDQN-based adaptive framework for optimizing handovers in LEO satellite networks, balancing multiple objectives like throughput and switching costs under changing conditions.

Contribution

It presents a novel deep reinforcement learning approach that dynamically learns trade-offs among multiple objectives in satellite handover management.

Findings

Achieves up to 10.3% throughput improvement.

Maintains near-zero blocking probability.

Outperforms conventional baseline methods.

Abstract

In this paper, we propose a dueling double deep Q-network (DDQN)-based adaptive multi-objective handover framework for low Earth orbit (LEO) satellite networks. The proposed method enables dynamic trade-off learning among throughput, blocking probability, and switching cost under time-varying network conditions. Simulation results demonstrate that the proposed approach consistently outperforms conventional baselines, achieving up to 10.3% throughput improvement and near-zero blocking under typical operating conditions.