.png){kind=icon}
Return
Intelligent Controller Placement in SDN-enabled Mobile Edge Computing Network via Hybrid Optimization Algorithm
.png){kind=icon}
PRE
AI
DOI：10.1016/j.comnet.2026.112720.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Recent advancements in mobile applications such as virtual reality, augmented reality and intelligent multimedia services have significantly increased the demand for low-latency and high-performance computing environments. Software Defined Networking (SDN)-enabled Mobile Edge Computing (MEC) has emerged as an effective framework for integrating mobile devices, edge servers and cloud resources to support efficient resource management and real-time service delivery. However, existing controller placement approaches suffer from limitations such as poor load balancing, high controller-switch communication cost, increased propagation delay and insufficient fault tolerance in large-scale dynamic MEC environments. To address these challenges, this paper proposes an Optimized Controller Placement in Software Defined Mobile Edge Computing Network (OCP-SDMECN) using an Enhanced Mother Assisted Wolverine Optimization Algorithm (EMA-WOA). The proposed framework initially performs optimal controller selection by analyzing controller characteristics such as modularity, graphical user interface support and legacy network compatibility. Subsequently, EMA-WOA is employed for optimal controller placement by jointly considering multiple constraints including load balancing, deployment cost and network reliability. The novelty of the proposed EMA-WOA lies in the integration of mother optimization-assisted position updating with the scavenging and hunting behaviors of Wolverine Optimization Algorithm, which improves global exploration capability and prevents premature convergence toward local optima. Simulation results demonstrate that the proposed EMA-WOA-based OCP-SDMECN framework significantly improves controller placement efficiency compared with existing approaches such as MOA, WaOA, PFO, TSO, CCA-PSO and HAS-PSO. The proposed method achieved a minimum controller-switch cost of 93 after 100 iterations, reduced total placement cost to 98.257 and attained higher reliability of 0.87 under 28 switches. In addition, EMA-WOA achieved superior load balancing performance with a maximum value of 0.947 while reducing execution time to 11.656 s. The overall analysis confirms that the proposed OCP-SDMECN framework provides a scalable, cost-effective and reliable controller placement strategy for SDN-enabled MEC networks.
Keywords:
Controller selection
Controller Placement
EMA-WOA
SDN
MEC
Journal
IF:
4.6
Papers:
1.6K
Citations:
1.6W
Organization
Cited Papers
No cited papers available