Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Dual Mind World Model Inspired Network Digital Twin for Access Scheduling

Hrishikesh Dutta, Roberto Minerva, Noel Crespi

TL;DR

DMWM addresses adaptive scheduling under deadline and interference constraints in wireless IoT networks by integrating a symbolic planner with a fast reactive module inside a Network Digital Twin. It uses a dual mind architecture that combines a Slow Mind symbolic rollout with a Fast Mind heuristic and an ICN constraint layer to ensure feasibility. The approach is evaluated in a simulated NDT against classical and RL baselines, showing improved throughput, lower delay, and better deadline adherence with interpretable planning. This work advances learning informed, foresight driven network optimization and paves the way for scalable, accountable optimization in networked systems.

Abstract

Emerging networked systems such as industrial IoT and real-time cyber-physical infrastructures demand intelligent scheduling strategies capable of adapting to dynamic traffic, deadlines, and interference constraints. In this work, we present a novel Digital Twin-enabled scheduling framework inspired by Dual Mind World Model (DMWM) architecture, for learning-informed and imagination-driven network control. Unlike conventional rule-based or purely data-driven policies, the proposed DMWM combines short-horizon predictive planning with symbolic model-based rollout, enabling the scheduler to anticipate future network states and adjust transmission decisions accordingly. We implement the framework in a configurable simulation testbed and benchmark its performance against traditional heuristics and reinforcement learning baselines under varied traffic conditions. Our results show that DMWM achieves superior performance in bursty, interference-limited, and deadline-sensitive environments, while maintaining interpretability and sample efficiency. The proposed design bridges the gap between network-level reasoning and low-overhead learning, marking a step toward scalable and adaptive NDT-based network optimization.

Dual Mind World Model Inspired Network Digital Twin for Access Scheduling

TL;DR

DMWM addresses adaptive scheduling under deadline and interference constraints in wireless IoT networks by integrating a symbolic planner with a fast reactive module inside a Network Digital Twin. It uses a dual mind architecture that combines a Slow Mind symbolic rollout with a Fast Mind heuristic and an ICN constraint layer to ensure feasibility. The approach is evaluated in a simulated NDT against classical and RL baselines, showing improved throughput, lower delay, and better deadline adherence with interpretable planning. This work advances learning informed, foresight driven network optimization and paves the way for scalable, accountable optimization in networked systems.

Abstract

Emerging networked systems such as industrial IoT and real-time cyber-physical infrastructures demand intelligent scheduling strategies capable of adapting to dynamic traffic, deadlines, and interference constraints. In this work, we present a novel Digital Twin-enabled scheduling framework inspired by Dual Mind World Model (DMWM) architecture, for learning-informed and imagination-driven network control. Unlike conventional rule-based or purely data-driven policies, the proposed DMWM combines short-horizon predictive planning with symbolic model-based rollout, enabling the scheduler to anticipate future network states and adjust transmission decisions accordingly. We implement the framework in a configurable simulation testbed and benchmark its performance against traditional heuristics and reinforcement learning baselines under varied traffic conditions. Our results show that DMWM achieves superior performance in bursty, interference-limited, and deadline-sensitive environments, while maintaining interpretability and sample efficiency. The proposed design bridges the gap between network-level reasoning and low-overhead learning, marking a step toward scalable and adaptive NDT-based network optimization.
Paper Structure (20 sections, 10 equations, 3 figures, 1 table, 2 algorithms)

This paper contains 20 sections, 10 equations, 3 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Model
  4. Physical Network Model
  5. Network Digital Twin Environment
  6. Scheduling Actions and Rewards
  7. Dual-Mind Decision Architecture
  8. Dual Mind World Model Driven Scheduling
  9. Fast Mind: Reactive Scheduling
  10. Slow Mind: Symbolic Rollout Planning
  11. Combined Decision Rule
  12. Complexity and Feasibility
  13. Integration with the Digital Twin
  14. Experimental Setup
  15. Network Simulation Environment
  16. ...and 5 more sections

Figures (3)

  • Figure 1: System Model
  • Figure 2: DMWM Scheduler Behavior and Model Accuracy: This plot illustrates the node scheduling behavior of the DMWM policy, where each row corresponds to a node, and each column represents a time step. A blue cell (in (a)) indicates that the corresponding node was selected for transmission at that time. Subplot (b) visualizes the absolute difference between the actual queue state transitions and those predicted by the symbolic world model used in the Slow Mind's internal simulations.
  • Figure 3: This figure compares six scheduling policies—DMWM, Random, LQF, Deadline-Priority, Fair Round-Robin, and a Q-learning agent—across four scenarios: Default, Bursty, Deadline-Sensitive, and Interference-Constrained. Results show mean performance over 30 runs with standard deviation bars. DMWM attains the best or near-best throughput while maintaining low delay and queue occupancy, achieving efficient utilization without violating interference or deadline constraints.