Software-defined Networking (SDN) has recently developed and been put forward as a promising and encouraging solution for future internet architecture. Managed, the centralized and controlled network has become more flexible and visible using SDN. On the other hand, these advantages bring us a more vulnerable environment and dangerous threats, causing network breakdowns, systems paralysis, online banking frauds and robberies. These issues have a significantly destructive impact on organizations, companies or even economies. Accuracy, high performance and real-time systems are essential to achieve this goal successfully. Extending intelligent machine learning algorithms in a network intrusion detection system (NIDS) through a software-defined network (SDN) has attracted considerable attention in the last decade. Big data availability, the diversity of data analysis techniques, and the massive improvement in the machine learning algorithms enable the building of an effective, reliable and dependable system for detecting different types of attacks that frequently target networks. This study demonstrates the use of machine learning algorithms for traffic monitoring to detect malicious behavior in the network as part of NIDS in the SDN controller. Different classical and advanced tree-based machine learning techniques, Decision Tree, Random Forest and XGBoost are chosen to demonstrate attack detection. The NSL-KDD dataset is used for training and testing the proposed methods; it is considered a benchmarking dataset for several state-of-the-art approaches in NIDS. Several advanced preprocessing techniques are performed on the dataset in order to extract the best form of the data, which produces outstanding results compared to other systems. Using just five out of 41 features of NSL-KDD, a multi-class classification task is conducted by detecting whether there is an attack and classifying the type of attack (DDoS, PROBE, R2L, and U2R), accomplishing an accuracy of 95.95%.

Graph robustness metrics have been used largely to study the behavior of communication networks in the presence of targeted attacks and random failures. Several researchers have proposed new graph metrics to better predict network resilience and survivability against such attacks. Most of these metrics have been compared to a few established graph metrics for evaluating the effectiveness of measuring network resilience. In this paper, we perform a comprehensive comparison of the most commonly used graph robustness metrics. First, we show how each metric is determined and calculate its values for baseline graphs. Using several types of random graphs, we study the accuracy of each robustness metric in predicting network resilience against centrality-based attacks. The results show three conclusions. First, our path diversity metric has the highest accuracy in predicting network resilience for structured baseline graphs. Second, the variance of node-betweenness centrality has mostly the best accuracy in predicting network resilience for Waxman random graphs. Third, path diversity, network criticality, and effective graph resistance have high accuracy in measuring network resilience for Gabriel graphs.

For many years the research community has attempted to model the Internet in order to better understand its behaviour and improve its performance. Since much of the structural complexity of the Inter- net is due to its multilevel operation, the Internet's multilevel nature is an important and non-trivial feature that researchers must consider when developing appropriate models. In this paper, we compare the normalised Laplacian spectra of physical-and logical-level topologies of four commercial ISPs and two research networks against the US freeway topology, and show analytically that physical level communication networks are structurally similar to the US freeway topology. We also generate synthetic Gabriel graphs of physical topologies and show that while these synthetic topologies capture the grid-like structure of actual topologies, they are more expensive than the actual physical level topologies based on a network cost model. Moreover, we introduce a distinction between geographic graphs that include degree-2 nodes needed to capture the geographic paths along which physical links follow, and structural graphs that eliminate these degree-2 nodes and capture only the interconnection properties of the physical graph and its multilevel relationship to logical graph overlays. Furthermore, we develop a multilevel graph evaluation framework and analyse the resilience of single and multilevel graphs using the flow robustness metric. We then confirm that dynamic routing performed over the lower levels helps to improve the performance of a higher level service, and that adaptive challenges more severely impact the performance of the higher levels than non-adaptive challenges.