A Comparative Study of OSPF Metrics in Routing Algorithms for Dynamic Path Selection in Network Security

Open-path first (OSPF) algorithms optimize paths using shortest-path metrics, often overlooking security considerations and adaptability in dynamic network environments. Many OSPF implementations use fixed cost values that do not adapt to network conditions. Research is needed to develop adaptive cost functions that dynamically respond to security threats. This paper compares cost metrics in OSPF routing algorithms to evaluate their effectiveness in dynamic path selection within network security contexts. By incorporating diverse cost functions and assessing their performance across security conditions, this study seeks to identify metrics that evaluate performance. To achieve this, OSPF routing algorithms were analyzed using different cost metrics, including step, linear, and exponential functions. Through simulations, the algorithms were tested under a Barabási-Albert topology, from routine operations to threat-prone scenarios, to evaluate their capabilities in dynamic path selection and resilience against security threats. The numerical results highlight a trade-off among the step, linear, and exponential cost functions, with average delays of 0.553 ms, 0.653 ms, and 0.517 ms, respectively, and average jitters of 0.210 μs, 0.201 μs, and 0.205 μs. The packet delivery success (PDS) rates are also recorded at 87.64%, 86.76%, and 86.18% for the step, linear, and exponential cost functions, respectively. This approach facilitates an analysis aimed at balancing performance with security considerations.