Accurate identification of P2P traffic is critical for efficient network management and reasonable utilization of network resources, as P2P applications have been growing dramatically. Fuzzy clustering is more flexible than hard clustering and is practical for P2P traffic identification because of the natural treatment of data using fuzzy clustering. Fuzzy c-means clustering (FCM) is an iteratively optimal algorithm normally based on the least square method to partition data sets, which has high computational overhead. This paper proposes modifications to the objective function and the distance function that greatly reduces the computational complexity of FCM while keeping the clustering accurate. The proposed FCM clustering technology can be incorporated into a Fuzzy Inference System (FIS) to implement real-time network traffic classification by updating the training data set continuously and efficiently.

Additional Metadata
Keywords data transformation, fuzzy c-means clustering, machine learning, network traffic identification, peer-to-peer communications, statistical analysis
Persistent URL dx.doi.org/10.1109/FUZZY.2011.6007613
Conference 2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011
Citation
Liu, D. (Duo), & Lung, C.H. (2011). P2P traffic identification and optimization using fuzzy c-means clustering. Presented at the 2011 IEEE International Conference on Fuzzy Systems, FUZZ 2011. doi:10.1109/FUZZY.2011.6007613