A Novel Metaheuristic with Optimal Deep Learning-Based Network Slicing in IoT-Enabled Clustered Wireless Sensor Networks in 5G Systems

Abstract

AbstractIn recent times, high rate of 5G networks enables the Internet of Things (IoT) and wireless sensor networks (WSN) to effectively gather the data from the deployed environment. Due to the limitations of energy and network slicing process, the efficiency of the IoT-enabled WSN is considerably affected. Therefore, this study introduces a novel fruit fly optimization-based clustering (FFOC) with optimal gated recurrent unit (OGRU)-based network slicing for IoT-enabled WSN in 5G networks. The proposed FFOC–OGRU technique initially constructs clusters and selects cluster heads using FFOC technique. Besides, the OGRU technique is employed for network slicing process, and the hyperparameters involved in the GRU model are optimally adjusted by the use of Bayesian optimization technique, which results in enhanced performance. For inspecting the improved performance of the FFOC–OGRU technique, a comprehensive experimental analysis is carried out and the outcomes are examined in several aspects. The experimental outcome showcased the betterment of the FFOC–OGRU technique in terms of several measures.KeywordsGated recurrent unitBayesian optimizationClustering5G networksMetaheuristic algorithmsNetwork slicing