Dynamic Adaptive Deployment of Network Slice for Industrial Internet of Things

With the advancement and maturation of 5G technology, its network slicing capabilities play a crucial role in delivering high-quality network services for industrial Internet of Things (IoT). However, effectively applying 5G network slicing to meet continuously changing network demands remains a challenging endeavor. We propose a heuristic-based dynamic adaptive network slicing deployment algorithm aimed at addressing both resource utilization efficiency and dynamics of network service requests during deployment. We first establish a system model for the network slicing deployment process, framing the deployment issue as a nonlinear integer programming problem. Subsequently, we introduce a two-stage deployment algorithm designed to accommodate the variability of network requests. Finally, we conduct numerical simulations and experiments to assess the algorithm based on criteria such as average deployment cost, number of Virtual Network Functions, and service acceptance rate. Results demonstrate that the proposed algorithm effectively manages resource utilization efficiency and dynamic issues of network service requests.