Joint likelihood aggregation of multiple cluster validity indices for stochastic channel modeling

For cluster-based stochastic channel modeling, selection of clustering method is crucial for the final modeling results. Since a clustering algorithm can generate as many partitions as required, identification of the optimal number of clusters is a vital consideration in clustering, which is called cluster validity. In this contribution, five widely used indices for cluster validity are employed to jointly determine the optimal number of clusters. By putting forward a novel likelihood aggregation approach for combining the decisions of multi-indices, the clustering results are more stable and reasonable. Four kinds of synthetic data are used to illustrate the feasibility of the proposed method in the case where the given data set is either easily clustered or not. Moreover, the performance of the proposed approach is evaluated by using real channel measurement data with convincing results.