Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Joint Power Allocation and User Scheduling in Integrated Satellite-Terrestrial Cell-Free Massive MIMO IoT Systems

Trinh Van Chien, Ha An Le, Ta Hai Tung, Hien Quoc Ngo, Symeon Chatzinotas

TL;DR

This work studies joint power allocation and user scheduling in an integrated satellite-terrestrial CF-MIMO IoT system under practical CSI conditions and pilot contamination.It derives closed-form uplink ergodic throughputs with MRC and proposes two complementary solutions: a model-based alternating-optimization algorithm and a learning-based SHGNN that exploits channel statistics in an unsupervised manner.The AO method delivers a stationary solution with explicit active-set decisions, while SHGNN offers scalable, fast inference with performance close to AO and favorable generalization and runtime properties.Numerical results demonstrate substantial gains from satellite-terrestrial integration over space-only or ground-only baselines and validate the practicality of graph-based learning for large-scale CF-MIMO IoT deployments.

Abstract

Both space and ground communications have been proven effective solutions under different perspectives in Internet of Things (IoT) networks. This paper investigates multiple-access scenarios, where plenty of IoT users are cooperatively served by a satellite in space and access points (APs) on the ground. Available users in each coherence interval are split into scheduled and unscheduled subsets to optimize limited radio resources. We compute the uplink ergodic throughput of each scheduled user under imperfect channel state information (CSI) and non-orthogonal pilot signals. As maximum-radio combining is deployed locally at the ground gateway and the APs, the uplink ergodic throughput is obtained in a closed-form expression. The analytical results explicitly unveil the effects of channel conditions and pilot contamination on each scheduled user. By maximizing the sum throughput, the system can simultaneously determine scheduled users and perform power allocation based on either a model-based approach with alternating optimization or a learning-based approach with the graph neural network. Numerical results manifest that integrated satellite-terrestrial cell-free massive multiple-input multiple-output systems can significantly improve the sum ergodic throughput over coherence intervals. The integrated systems can schedule the vast majority of users; some might be out of service due to the limited power budget.

Joint Power Allocation and User Scheduling in Integrated Satellite-Terrestrial Cell-Free Massive MIMO IoT Systems

TL;DR

This work studies joint power allocation and user scheduling in an integrated satellite-terrestrial CF-MIMO IoT system under practical CSI conditions and pilot contamination.It derives closed-form uplink ergodic throughputs with MRC and proposes two complementary solutions: a model-based alternating-optimization algorithm and a learning-based SHGNN that exploits channel statistics in an unsupervised manner.The AO method delivers a stationary solution with explicit active-set decisions, while SHGNN offers scalable, fast inference with performance close to AO and favorable generalization and runtime properties.Numerical results demonstrate substantial gains from satellite-terrestrial integration over space-only or ground-only baselines and validate the practicality of graph-based learning for large-scale CF-MIMO IoT deployments.

Abstract

Both space and ground communications have been proven effective solutions under different perspectives in Internet of Things (IoT) networks. This paper investigates multiple-access scenarios, where plenty of IoT users are cooperatively served by a satellite in space and access points (APs) on the ground. Available users in each coherence interval are split into scheduled and unscheduled subsets to optimize limited radio resources. We compute the uplink ergodic throughput of each scheduled user under imperfect channel state information (CSI) and non-orthogonal pilot signals. As maximum-radio combining is deployed locally at the ground gateway and the APs, the uplink ergodic throughput is obtained in a closed-form expression. The analytical results explicitly unveil the effects of channel conditions and pilot contamination on each scheduled user. By maximizing the sum throughput, the system can simultaneously determine scheduled users and perform power allocation based on either a model-based approach with alternating optimization or a learning-based approach with the graph neural network. Numerical results manifest that integrated satellite-terrestrial cell-free massive multiple-input multiple-output systems can significantly improve the sum ergodic throughput over coherence intervals. The integrated systems can schedule the vast majority of users; some might be out of service due to the limited power budget.
Paper Structure (23 sections, 6 theorems, 68 equations, 10 figures, 1 table, 1 algorithm)

This paper contains 23 sections, 6 theorems, 68 equations, 10 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model and Uplink Pilot Training
  3. Uplink Pilot Training
  4. Analysis of Ergodic Throughput and Uplink Data Transmission
  5. Uplink Data Transmission
  6. Uplink Ergodic Throughput
  7. Sum Ergodic Throughput Optimization
  8. Problem Formulation
  9. Iterative Algorithm
  10. Deep Learning Framework
  11. Graphical Representation
  12. Heterogeneous Graph Neural Networks (HetGNN) for Integrated Satellite-Terrestrial Cell-Free Massive MIMO
  13. Implementation of SHGNN
  14. Feature Initialization
  15. Data Processing
  16. ...and 8 more sections

Key Result

Lemma 1

By exploiting the MMSE estimation locally at each AP, the estimate of the channel $g_{mk}$ between AP $m$ and user $k$ can be formulated based on eq:ProjChannelypmk as where $c_{mk} = \mathbb{E} \{ y_{pmk}^\ast g_{mk} \} / \mathbb{E} \{ | y_{pmk} |^2 \}$ is computed as From eq:ChanEstgmk, we observe that the channel estimate $\hat{g}_{mk}$ is distributed as $\hat{g}_{mk} \sim \mathcal{CN}(0, \g

Figures (10)

  • Figure 1: The considered satellite-terrestrial cooperative IoT network where $M$ APs and a LEO satellite jointly serve the $K$ users with both active (green color) and inactive (red color) users.
  • Figure 2: Heterogeneous graph representation of the considered satellite-terrestrial cooperative network with $M = 3$ and $K = 2$.
  • Figure 3: CDF of the per user throughput [Mbps] utilizing Monte Carlo simulations versus the analyses with $K=20, \tau_c = 10000,$ and $\tau_p = K/2$.
  • Figure 4: CDF of the sum throughput [Mbps] utilizing Monte Carlo simulations versus the analyses with $K=20, \tau_c = 10000,$ and $\tau_p = K/2$.
  • Figure 5: Per user throughput versus the different number of orthogonal pilot signals with $K = 10$.
  • ...and 5 more figures

Theorems & Definitions (10)

  • Lemma 1
  • proof
  • Theorem 1
  • proof
  • Theorem 2
  • proof
  • Corollary 1
  • Lemma 2
  • Lemma 3
  • proof