New FCM´s Validity Index for Sorting Radar Signal

Fuzzy C-Means (FCM) is an unsupervised clustering algorithm, which requires a foreknowledge of the number of clusters. In this paper, we have proposed a new validity index with a high speed calculation and good preventive measures against noisy environment, which is necessary for real time sorting signals. The proposed validity index does not make use of the cluster´s centers to measure the compactness and separation indexes, it only uses the memberships´ value. Simulation results show and certify the efficiency of the proposed validity index and compares it with other validity indexes.