Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Service Function Chain Dynamic Scheduling in Space-Air-Ground Integrated Networks

Ziye Jia, Yilu Cao, Lijun He, Qihui Wu, Qiuming Zhu, Dusit Niyato, Zhu Han

TL;DR

This work tackles dynamic service function chain scheduling in space, air, and ground integrated networks by introducing a reconfigurable time extension graph (RTEG) to model time-varying multi-layer resources. It formulates SFC deployment as an ILP to maximize the number of successfully deployed chains under heterogeneous resources, then reformulates the problem as an MDP and solves it with a deep reinforcement learning framework (DRL-MSSNL-SAGIN) based on double DQN. Key contributions include the RTEG modeling, a detailed SFC deployment and scheduling model for SAGIN, and a DRL algorithm with a specialized VNF state transition mechanism, all validated by simulations showing improved convergence, scalability and resource utilization. The approach enables near real time dynamic orchestration of SFCs in highly mobile and heterogeneous SAGIN environments, with potential impact on global coverage and efficient NFV-based service provisioning.

Abstract

As an important component of the sixth generation communication technologies, the space-air-ground integrated network (SAGIN) attracts increasing attentions in recent years. However, due to the mobility and heterogeneity of the components such as satellites and unmanned aerial vehicles in multi-layer SAGIN, the challenges of inefficient resource allocation and management complexity are aggregated. To this end, the network function virtualization technology is introduced and can be implemented via service function chains (SFCs) deployment. However, urgent unexpected tasks may bring conflicts and resource competition during SFC deployment, and how to schedule the SFCs of multiple tasks in SAGIN is a key issue. In this paper, we address the dynamic and complexity of SAGIN by presenting a reconfigurable time extension graph and further propose the dynamic SFC scheduling model. Then, we formulate the SFC scheduling problem to maximize the number of successful deployed SFCs within limited resources and time horizons. Since the problem is in the form of integer linear programming and intractable to solve, we propose the algorithm by incorporating deep reinforcement learning. Finally, simulation results show that the proposed algorithm has better convergence and performance compared to other benchmark algorithms.

Service Function Chain Dynamic Scheduling in Space-Air-Ground Integrated Networks

TL;DR

This work tackles dynamic service function chain scheduling in space, air, and ground integrated networks by introducing a reconfigurable time extension graph (RTEG) to model time-varying multi-layer resources. It formulates SFC deployment as an ILP to maximize the number of successfully deployed chains under heterogeneous resources, then reformulates the problem as an MDP and solves it with a deep reinforcement learning framework (DRL-MSSNL-SAGIN) based on double DQN. Key contributions include the RTEG modeling, a detailed SFC deployment and scheduling model for SAGIN, and a DRL algorithm with a specialized VNF state transition mechanism, all validated by simulations showing improved convergence, scalability and resource utilization. The approach enables near real time dynamic orchestration of SFCs in highly mobile and heterogeneous SAGIN environments, with potential impact on global coverage and efficient NFV-based service provisioning.

Abstract

As an important component of the sixth generation communication technologies, the space-air-ground integrated network (SAGIN) attracts increasing attentions in recent years. However, due to the mobility and heterogeneity of the components such as satellites and unmanned aerial vehicles in multi-layer SAGIN, the challenges of inefficient resource allocation and management complexity are aggregated. To this end, the network function virtualization technology is introduced and can be implemented via service function chains (SFCs) deployment. However, urgent unexpected tasks may bring conflicts and resource competition during SFC deployment, and how to schedule the SFCs of multiple tasks in SAGIN is a key issue. In this paper, we address the dynamic and complexity of SAGIN by presenting a reconfigurable time extension graph and further propose the dynamic SFC scheduling model. Then, we formulate the SFC scheduling problem to maximize the number of successful deployed SFCs within limited resources and time horizons. Since the problem is in the form of integer linear programming and intractable to solve, we propose the algorithm by incorporating deep reinforcement learning. Finally, simulation results show that the proposed algorithm has better convergence and performance compared to other benchmark algorithms.

Paper Structure

This paper contains 34 sections, 37 equations, 8 figures, 2 tables, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Model
  4. Network Model
  5. SFC Dynamic Scheduling Model
  6. Channel Model
  7. Channel Model of G2U and U2G
  8. Channel Model of U2U
  9. Channel Model Related to Satellites
  10. Energy Cost Model
  11. Energy Cost of UAVs
  12. Energy Cost of Satellites
  13. Problem Formulation
  14. Constraints
  15. Deployment Constraints
  16. ...and 19 more sections

Figures (8)

  • Figure 1: Scenario of SAGIN.
  • Figure 2: SFC dynamic scheduling in SAGIN based on RTEG. (a) SFC deployment in SAGIN. (b) SFC offline deployment corresponding to (a). (c) SFC online scheduling.
  • Figure 3: The training process of DRL-MSSNL-SAGIN algorithm.
  • Figure 4: The training effect under different learning rates. (a) Reward versus episode. (b) The number of completed tasks versus episode.
  • Figure 5: Distinctions between different optimizers. (a) Reward versus episode. (b) The number of completed tasks versus episode.
  • ...and 3 more figures