Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Learning Based Dynamic Cluster Reconfiguration for UAV Mobility Management with 3D Beamforming

Irshad A. Meer, Karl-Ludwig Besser, Mustafa Ozger, Dominic Schupke, H. Vincent Poor, Cicek Cavdar

TL;DR

This work addresses dynamic cluster reconfiguration and power allocation for UAV communications in interference-laden downlink with time-varying reliability requirements. It introduces a Masked-SAC reinforcement learning framework that uses action masking to cope with evolving active user sets and leverages 3D beamforming to manage interference. The approach is formulated as a multi-objective optimization balanced by a learnable reward, and empirical results show MSAC outperforms Opportunistic and Closest baselines by meeting stringent reliability in high-demand zones, reducing total transmit power, and forming compact clusters. The proposed framework offers a scalable, model-free solution for mobility management in cell-less UAV networks with dynamic service requirements.

Abstract

In modern cell-less wireless networks, mobility management is undergoing a significant transformation, transitioning from single-link handover management to a more adaptable multi-connectivity cluster reconfiguration approach, including often conflicting objectives like energy-efficient power allocation and satisfying varying reliability requirements. In this work, we address the challenge of dynamic clustering and power allocation for unmanned aerial vehicle (UAV) communication in wireless interference networks. Our objective encompasses meeting varying reliability demands, minimizing power consumption, and reducing the frequency of cluster reconfiguration. To achieve these objectives, we introduce a novel approach based on reinforcement learning using a masked soft actor-critic algorithm, specifically tailored for dynamic clustering and power allocation.

Learning Based Dynamic Cluster Reconfiguration for UAV Mobility Management with 3D Beamforming

TL;DR

This work addresses dynamic cluster reconfiguration and power allocation for UAV communications in interference-laden downlink with time-varying reliability requirements. It introduces a Masked-SAC reinforcement learning framework that uses action masking to cope with evolving active user sets and leverages 3D beamforming to manage interference. The approach is formulated as a multi-objective optimization balanced by a learnable reward, and empirical results show MSAC outperforms Opportunistic and Closest baselines by meeting stringent reliability in high-demand zones, reducing total transmit power, and forming compact clusters. The proposed framework offers a scalable, model-free solution for mobility management in cell-less UAV networks with dynamic service requirements.

Abstract

In modern cell-less wireless networks, mobility management is undergoing a significant transformation, transitioning from single-link handover management to a more adaptable multi-connectivity cluster reconfiguration approach, including often conflicting objectives like energy-efficient power allocation and satisfying varying reliability requirements. In this work, we address the challenge of dynamic clustering and power allocation for unmanned aerial vehicle (UAV) communication in wireless interference networks. Our objective encompasses meeting varying reliability demands, minimizing power consumption, and reducing the frequency of cluster reconfiguration. To achieve these objectives, we introduce a novel approach based on reinforcement learning using a masked soft actor-critic algorithm, specifically tailored for dynamic clustering and power allocation.
Paper Structure (11 sections, 12 equations, 4 figures, 1 table, 1 algorithm)

This paper contains 11 sections, 12 equations, 4 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model and Problem Formulation
  3. SINR Outage
  4. Mobility Model
  5. User Handling
  6. Users Entering the Coverage Area
  7. User Leaves the Coverage Area
  8. Problem Formulation
  9. Proposed Solution With Reinforcement Learning
  10. Numerical Evaluation
  11. Conclusion

Figures (4)

  • Figure 1: Considered communication scenario with fixed and moving at an altitude above the ground. Within the highlighted zone in the center, the reliability constraint is $\varepsilon_{\text{max},2}$, otherwise it is $\varepsilon_{\text{max},1}$.
  • Figure 2: of the outage probability $\varepsilon$ experienced by the .
  • Figure 3: Numerical results of the distribution of the total transmit power of the system normalized by the maximum available power.
  • Figure 4: Average cluster size for serving mobile