Hybrid Ant Colony Optimization algorithm for dynamic routing in software defined networking

Abstract

In dynamic network environments where traffic patterns were expected to fluctuate rapidly and unforeseen events like link failures could occur, routing protocols needed to swiftly adapt to maintain optimal performance. To optimize network routing, this research focused on key input parameters such as packet loss ratio, average packet delay, bandwidth utilization, and router processor load, which were essential for understanding network performance. The approach involved developing a mathematical algorithm designed to process these parameters to generate optimization variables, including the required number of routers, necessary link bandwidth capacity, and router processing capacity. This algorithm enabled Internet Service Providers to enhance performance quality through real-time adaptivity, ensuring efficient data transmission, reduced latency, and minimal packet loss. The methodology included conducting interviews with Internet Service Providers in Kenya to gather data and insights on their adoption of technologies like Software Defined Networking and Network Function Virtualization. These interviews provided valuable information on the state of network routing and the challenges faced by Internet Service Providers. By integrating this data with the algorithm, we aimed to create a dynamic routing algorithm that would swiftly adapt to changing network conditions and optimize routing in real-time, ultimately enhancing network reliability and performance. KEYWORDS: Dynamic, Network, Algorithm