A neuro‐evolutionary approach for software defined wireless network traffic classification

Abstract Accurate network traffic classification is an essential and challenging issue for wireless management survivability. Existing algorithms, on the other hand, cannot meet required specifications of real networks' in terms user privacy control overhead, latency, above all, speed. For classification, machine learning‐based hybrid optimization techniques have been deployed. This paper takes a software‐defined (SDWN) architecture into account. Because proposed scheme perfectly contained within controller,the SDWN controller's higher processing capability, global visibility, programmability can be used to achieve real‐time, adaptive, precise classification. In this paper, neuro‐evolutionary approach which feed forward neural (FFNN) base classifier particle swarm (PSO) train FFNN accurately classify while minimizing communication overhead between controller switches. Simulation experiments were conducted by acquiring real‐world internet datasets test efficacy scheme. The results state‐of‐the‐art comparisons show that has outperformed accuracy