The ever-evolving technology might have eradicated the drawbacks of software systems we encountered a decade ago, keeping them secure has become just as challenging. The expansion in artificially intelligent technologies have unlocked the unidentified programming networking terminals that are being flooded by attackers. With the innovation of Software defined networking, programmers acquired the power to manipulate the network by keeping the control plane and data plane isolated from each other. This, however, has become the best bait for DDoS attacks on the data plane owing to its single-point failure. A mechanism is needed that could deal with the dynamic traffic at the point of instigation, thereby preventing the networking attacks.
Many supervised and unsupervised algorithms have been suggested by researchers to protect Software Defined Networking’s drawback of single point failure. Due to the diversity of DDoS attacks, implementing the real solution has been the lost cause. For this, we need an algorithm as diversified as the problem it is targeting. Moreover, it could also potentially stop the need of using multiple algorithms on a single kind of software thereby increasing its optimality and reducing the time overhead that they might cause.
Our proposed Whale Optimization Algorithm Based Clustering for DDoS Detection (WOA-DD) targets the DDoS attack packets by clustering them via a meta-heuristic approach. Though, extensive research has been done in this field, our algorithm improves the measures such as robustness, scalability, accuracy and precision by many folds when compared with other machine learning approaches. The feasibility report of our algorithm has been positive and it was garnered safe to use once used with other prevailing algorithms. However, one might not have to use other algorithms as WOA-DD targets all kinds of DDoS attacks.