Golden Jackal-Based Intelligent Routing Optimization Framework for QoS- Aware SDN-Enabled IoT Networks

The fast pace of Internet of Things (IoT) development has resulted in ever-increasing complicated network configurations which require very high Quality-of-Service (QoS) requirements with low latency, high throughput, and dependable data delivery requirements. Such dynamic environments usually underperform when using standard routing methods, particularly when dealing with distributed computing on the edge, fog, and cloud layers. In addition, the independent treatment routing solution and the locations of Software-Defined Network (SDN) controllers poses poor performance outcomes, higher latencies and resource wastages. This paper seeks to overcome these problems by introducing a new routing-oriented optimization technique of SDN enabled IoT, wherein intelligent routing direction will be accomplished using Golden Jackal Optimization (GJO). Based on the observations of golden jackals with their interaction in cooperative hunting process, the GJO algorithm is introduced to provide rapid convergence and strong exploration of solution space globally. The framework given has the attributes of parallel optimization of both the routes and controllers’ location, thus being more scalable and with lesser control overheads. The solution based on the GJO has the QoS-based goal namely minimized end to end latencies and packet drops, achieves network load balancing, and anticipates congestion. For finding performance of the network use a novel hybrid model can be created by integrating Capsule Network (CapsNet) with a Deep Belief Network (DBN). Simulated experiments show that our model is much better than the classical and learning-based routing strategies in throughput, delay, and flexibility to adapt to the changing IoT-based network scenarios. The suggested approach meets a low latency of 0.23, high throughput of 0.92, and low packet loss of 0.18 and also meets efficient load balancing with a score of 0.83, showing that it is very efficient and reliable in handling IoT network traffic.