Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Accelerating Graph Neural Networks via Edge Pruning for Power Allocation in Wireless Networks

Lili Chen, Jingge Zhu, Jamie Evans

TL;DR

The paper tackles scalable power allocation in D2D wireless networks by replacing full-graph GNNs with edge-pruned, threshold-based variants. It introduces a novel neighbour-based threshold, supported by a stochastic-geometry analysis that guides threshold selection and demonstrates substantial time complexity reductions from $O(|\mathcal{V}|^2)$ to $O(|\mathcal{V}|)$ while preserving strong performance. The approach yields a N-GNN that generalises across varied spatial dimensions and network densities, approaching the WMMSE benchmark with significantly lower inference time. This work offers a practical pathway to real-time, scalable GNN-based power control in dense wireless networks.

Abstract

Graph Neural Networks (GNNs) have recently emerged as a promising approach to tackling power allocation problems in wireless networks. Since unpaired transmitters and receivers are often spatially distant, the distance-based threshold is proposed to reduce the computation time by excluding or including the channel state information in GNNs. In this paper, we are the first to introduce a neighbour-based threshold approach to GNNs to reduce the time complexity. Furthermore, we conduct a comprehensive analysis of both distance-based and neighbour-based thresholds and provide recommendations for selecting the appropriate value in different communication channel scenarios. We design the corresponding neighbour-based Graph Neural Networks (N-GNN) with the aim of allocating transmit powers to maximise the network throughput. Our results show that our proposed N-GNN offer significant advantages in terms of reducing time complexity while preserving strong performance and generalisation capacity. Besides, we show that by choosing a suitable threshold, the time complexity is reduced from O(|V|^2) to O(|V|), where |V| is the total number of transceiver pairs.

Accelerating Graph Neural Networks via Edge Pruning for Power Allocation in Wireless Networks

TL;DR

The paper tackles scalable power allocation in D2D wireless networks by replacing full-graph GNNs with edge-pruned, threshold-based variants. It introduces a novel neighbour-based threshold, supported by a stochastic-geometry analysis that guides threshold selection and demonstrates substantial time complexity reductions from to while preserving strong performance. The approach yields a N-GNN that generalises across varied spatial dimensions and network densities, approaching the WMMSE benchmark with significantly lower inference time. This work offers a practical pathway to real-time, scalable GNN-based power control in dense wireless networks.

Abstract

Graph Neural Networks (GNNs) have recently emerged as a promising approach to tackling power allocation problems in wireless networks. Since unpaired transmitters and receivers are often spatially distant, the distance-based threshold is proposed to reduce the computation time by excluding or including the channel state information in GNNs. In this paper, we are the first to introduce a neighbour-based threshold approach to GNNs to reduce the time complexity. Furthermore, we conduct a comprehensive analysis of both distance-based and neighbour-based thresholds and provide recommendations for selecting the appropriate value in different communication channel scenarios. We design the corresponding neighbour-based Graph Neural Networks (N-GNN) with the aim of allocating transmit powers to maximise the network throughput. Our results show that our proposed N-GNN offer significant advantages in terms of reducing time complexity while preserving strong performance and generalisation capacity. Besides, we show that by choosing a suitable threshold, the time complexity is reduced from O(|V|^2) to O(|V|), where |V| is the total number of transceiver pairs.
Paper Structure (21 sections, 16 equations, 6 figures, 7 tables)

This paper contains 21 sections, 16 equations, 6 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Preliminaries
  3. System Model
  4. Optimisation Problem
  5. Threshold-based Graph Neural networks for power allocation
  6. Traditional Graph Representation
  7. The Proposed Guideline for Choosing Appropriate Threshold
  8. Distance-based Threshold
  9. Neighbour-based Threshold
  10. The Proposed Guideline
  11. Threshold-based Graph Neural Networks
  12. Graph Representation
  13. The Structure of Graph Neural Network
  14. Simulations and results
  15. Simulation Setup
  16. ...and 6 more sections

Figures (6)

  • Figure 1: D2D communication network.
  • Figure 2: D2D Communication Network with distance-based threshold.
  • Figure 3: The structure of threshold-based GNN.
  • Figure 4: Normalized average weighted sum rate with thresholds that achieve $95\%$ interference (Tables \ref{['Tab:DistanceThreshold']} and \ref{['Tab:expected95']}) under $\lambda=0.01$.
  • Figure 5: Average inference time with thresholds that achieve $95\%$ interference (Tables \ref{['Tab:DistanceThreshold']} and \ref{['Tab:expected95']}) under $\alpha =5.5$ and $\lambda=0.01$.
  • ...and 1 more figures