No abstract

COST 2100 channel model (C2CM) wireless propagation multipaths taken from IEEE DataPort are grouped using simultaneous clustering and model selection (SCAMS). SCAMS solves simultaneously the membership and the number of clusters in contrast to just the number of clusters popular clustering approaches provide. The membership and the number of clusters rely on and , the parameters that avoid the trivial solution of the affinity matrix. Jaccard index is used to assess the accuracy of SCAMS by comparing the clustered multipaths with the reference multipath datasets from IEEE DataPort.

Deep divergence-based clustering (DDC) is used to cluster COST 2100 channel model (C2CM) wireless propagation multipaths. The dataset is taken from the IEEE DataPort. DDC solves the membership of the clusters. DDC builds on information theoretic divergence measures and geometric regularization in order to determine the membership of the clusters. The cluster count is then computed through the cluster-wise Jaccard index of the membership of the multipaths to their clusters. The performance of DDC is evaluated using the Jaccard index by comparing the reference multipathdatasets from IEEE DataPort with the calculated multipath clusters obtained by DDC. Results show that DDC can be used as an alternative clustering approach in the field of channel modeling.

Simultaneous Clustering and Model Selection (SCAMS) is introduced to cluster multipaths from COST 2100 channel model (C2CM). SCAMS determines not only the number of clusters but also the membership of the clusters. The study is the first to report clustering of multipaths that consider simultaneously the number of clusters and the membership of the clusters. Cluster identification and cardinality classification are dependent on the values of and , the parameters that weigh the penalty terms to avoid the trivial solution (all 1 matrix) of the affinity matrix. The clustered multipaths are compared with the reference multipaths that can be found in IEEE DataPort. The accuracy of the clustering approach is examined using the Jaccard index ( ). The proposed clustering approach can achieve higher accuracy compared to popular multipath clustering approaches.

Multipaths from COST 2100 channel model (C2CM) semi-urban scenario obtained from IEEE DataPort are clustered using simultaneous clustering and model selection (SCAMS). SCAMS jointly solves the problem on model selection and clustering by determining simultaneously the number of clusters and their membership unlike well-known clustering approaches that give only the number of clusters. The number and membership of clusters depend on and , the parameters that weigh the penalty terms so that the trivial solution of the affinity matrix is avoided. The clustered multipaths from SCAMS are compared with the reference multipath clustering datasets available at IEEE DataPort using Jaccard index which examines the accuracy of the clustering approach.