TOC-Based Communication-Control Decoupled Optimization for Multi-User Closed-Loop Wireless Networked Control Systems

Abstract

Wireless networked control systems (WNCSs) have emerged as a key enabler for modern industrial automation, interconnecting sensors, control centers, and actuators via wireless communications. However, conventional architectures based on separate optimization of control and communication lead to inefficient resource utilization and limited control stability, making them inadequate for emerging application demands. Especially in multi-user closed-loop WNCS, communication interference leads to inevitable packet loss and latency, which degrade state estimation accuracy and may even compromise overall control stability. In this article, we link the uplink and downlink transmission processes in a closed-loop communication–control co-design framework. Based on this model, we formulate a joint optimization problem that aims to maximize channel capacity while minimizing control cost, subject to constraints including communication transmit power and control convergence rate. To address the resulting highly non-convex problem, we employ the Tornado Optimizer with Coriolis force (TOC) for global exploration, and develop a Hybrid Convex–TOC Optimization (HCTO) framework that integrates convex optimization for efficient local convergence. The proposed framework alternately coordinating convex optimization for efficient local convergence and TOC for enhanced global exploration, which effectively navigates the non-convex solution space while preserving computational tractability. Simulation results demonstrate that the proposed scheme significantly enhances both convergence speed and robustness compared to baseline methods.