This paper is about a few implementation aspects of a clustering algorithm for grouping parameters of the channel’s multipath components. Issues as initialization and the computation of distance between multipath components are discussed. Special attention is given to the cluster validation problem dealing with the number of clusters’ estimation. In this context, some validation measures and fusion techniques are described. Synthetic data generated by the Extended Saleh-Valenzuela model was used as input for the clustering algorithm. Results from the clustering algorithm and validation measures are presented.