Network intrusion detection based on hybrid Fuzzy C-mean clustering

Intrusion of network which couldn´t be analyzed, detected and prevented may make whole network system paralyze while the abnormally detection can prevent it by detecting the known and unknown character of data. A mixed fuzzy clustering algorithm that uses Quantum-behaved Particle Swarm Optimization (QPSO) algorithm and combines with Fuzzy C-means (FCM) is adopted in this paper and used in abnormally detection. The iteration algorithm is replaced by the new hybrid algorithm based on the gradient descent of FCM, which makes the algorithm a strong global searching capacity and avoids the local minimum problems of FCM. At the same time, FCM is no longer in a large degree dependent on the initialization values. The simulation result proves that compared with FCM the new algorithm not only has the favorable convergent capability of the global optimizing but also has been obviously improved the robustness, and has the higher performance in intrusion detection than FCM and K-means algorithm.