Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Explainable AI for UAV Mobility Management: A Deep Q-Network Approach for Handover Minimization

Irshad A. Meer, Bruno Hörmann, Mustafa Ozger, Fabien Geyer, Alberto Viseras, Dominic Schupke, Cicek Cavdar

TL;DR

This work tackles UAV mobility management in cellular networks, where frequent handovers threaten service continuity. It proposes a centralized training, distributed execution DQN approach for handover decisions, augmented with SHAP-based feature attributions to explain the policy. An interpretation layer using DeepSHAP and language-model aided natural language explanations is added to bridge AI decisions with human operators. Validation on real-world LTE UAV flight data demonstrates substantial handover reductions and provides insight into which features drive decisions, enhancing trust and practicality of AI-driven mobility management in 3D aerial networks.

Abstract

The integration of unmanned aerial vehicles (UAVs) into cellular networks presents significant mobility management challenges, primarily due to frequent handovers caused by probabilistic line-of-sight conditions with multiple ground base stations (BSs). To tackle these challenges, reinforcement learning (RL)-based methods, particularly deep Q-networks (DQN), have been employed to optimize handover decisions dynamically. However, a major drawback of these learning-based approaches is their black-box nature, which limits interpretability in the decision-making process. This paper introduces an explainable AI (XAI) framework that incorporates Shapley Additive Explanations (SHAP) to provide deeper insights into how various state parameters influence handover decisions in a DQN-based mobility management system. By quantifying the impact of key features such as reference signal received power (RSRP), reference signal received quality (RSRQ), buffer status, and UAV position, our approach enhances the interpretability and reliability of RL-based handover solutions. To validate and compare our framework, we utilize real-world network performance data collected from UAV flight trials. Simulation results show that our method provides intuitive explanations for policy decisions, effectively bridging the gap between AI-driven models and human decision-makers.

Explainable AI for UAV Mobility Management: A Deep Q-Network Approach for Handover Minimization

TL;DR

This work tackles UAV mobility management in cellular networks, where frequent handovers threaten service continuity. It proposes a centralized training, distributed execution DQN approach for handover decisions, augmented with SHAP-based feature attributions to explain the policy. An interpretation layer using DeepSHAP and language-model aided natural language explanations is added to bridge AI decisions with human operators. Validation on real-world LTE UAV flight data demonstrates substantial handover reductions and provides insight into which features drive decisions, enhancing trust and practicality of AI-driven mobility management in 3D aerial networks.

Abstract

The integration of unmanned aerial vehicles (UAVs) into cellular networks presents significant mobility management challenges, primarily due to frequent handovers caused by probabilistic line-of-sight conditions with multiple ground base stations (BSs). To tackle these challenges, reinforcement learning (RL)-based methods, particularly deep Q-networks (DQN), have been employed to optimize handover decisions dynamically. However, a major drawback of these learning-based approaches is their black-box nature, which limits interpretability in the decision-making process. This paper introduces an explainable AI (XAI) framework that incorporates Shapley Additive Explanations (SHAP) to provide deeper insights into how various state parameters influence handover decisions in a DQN-based mobility management system. By quantifying the impact of key features such as reference signal received power (RSRP), reference signal received quality (RSRQ), buffer status, and UAV position, our approach enhances the interpretability and reliability of RL-based handover solutions. To validate and compare our framework, we utilize real-world network performance data collected from UAV flight trials. Simulation results show that our method provides intuitive explanations for policy decisions, effectively bridging the gap between AI-driven models and human decision-makers.

Paper Structure

This paper contains 11 sections, 15 equations, 3 figures, 1 table.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. System Model
  4. Problem Formulation
  5. UAV Handover Management with Reinforcement Learning
  6. Explainable Decision-Making Framework
  7. Shapley Values for Feature Importance
  8. Bridging the Gap with Language Models
  9. Evaluation with Real-World Data
  10. Results
  11. Conclusion

Figures (3)

  • Figure 1: Architecture of the explainable AI (XAI) framework for UAV handover management.
  • Figure 2: Comparison of SHAP value distributions for simulated (left) and real-world (right) data. Each point represents a data instance, colored by feature value magnitude, illustrating similarities and differences in feature contributions to model outputs across simulation and real-world deployments.
  • Figure 3: Waterfall plot depicting SHAP values for simulated data, illustrating the contributions of various features to the prediction of an action.