An Intelligent Load Balancing Technique for Software Defined Networking based 5G using Macine Learning models

The emergence of two new technologies, namely software defined networking (SDN) and 5G networks, has greatly changed the development network functions topologies. These technologies provide cost benefits for mobile operators, a more flexible scalable network, shorter time to market services applications. Scalability effectiveness are increased when SDN used together. increases reliability by separating control plane from data plane. Incorrect load balancing, lack knowledge traffic, other issues make it difficult Quality Service (QoS) with SDN. This research proposes unique load-balancing method resolve these concerns using Hierarchical Agglomerative Clustering (HAC) Back Propagation Neural Network (BPNN) algorithms. proposed segments into several groups after normalizing requirements. It consists phases: in first phase, there is clustering bandwidth different (e.g., social media, automated homes, cars) inside agglomerative hierarchical (single link technique) implemented clusters work based on minimum distance. After clustering, we allotted bandwidths respective clusters. In second BPNN technique trains choose optimal path check error faults. approach evaluates delay, packet loss, throughput, latency rate, usage evaluate performance Multiple Regression-based Searching (MRBS) Software-defined Sensor Load Balancing (SDSNLB) experimental results model promising as 15%, 23%, 27%, 21%, 30%, respectively, compared existing approaches. addition computational complexity no nodes services, rate varying, solution’s efficiency remains constant.