Genetic Neuro-Fuzzy Approach towards Routing in Industrial IoT

The Internet of Things has rapidly grown in the past years as emerging technology. Moreover, 5G networks start to offer communication infrastructure for applications of the Industrial Internet of Things (IIoT). However, due to energy limitations of IIoT devices and heterogeneity of 5G networks, managing IIoT networks is a challenging task. One of the most critical issues in IIoT that requires consideration is traffic routing that has a significant impact on energy consumption, and thus, lifetime of the network. Artificial Intelligence (AI) has been widely employed to solve complex scientific and practical problems. Such AI techniques as neural networks, fuzzy systems, genetic algorithms are commonly employed in wireless networks to promote their optimization, prediction, and management. This study suggests using an Adaptive Neuro-fuzzy Inference System (ANFIS) in 5G networks of IIoT for improving the routing process. A flow-chat of routing protocol was suggested. For input and output values of the ANFIS linguistic variables, terms and membership functions were defined. A rules base was developed. To improve the rule base of the ANFIS, a genetic algorithm was proposed. The operation of ANFIS was simulated in Matlab software.