A time-domain-constrained fuzzy clustering method and its application to signal analysis

This paper introduces a new fuzzy clustering method with time-domain-constraints which is used to signal analysis. Proposed method makes it possible to include natural constraints for signal analysis using fuzzy clustering, that is, the neighboring samples of signal belong to the same cluster. This method can be called time-domain-constrained fuzzy clustering. This paper introduces two approaches to include the above kind of constraints. The first approach leads to the time-domain-constrained fuzzy c-regression models method. The second approach leads to the ɛ-insensitive version of the above method, which results in additional robustness for outliers and non-Gaussian noise. Finally, simulations on synthetic as well as real-life signals are realized to evaluate the performance of the time-domain-constrained fuzzy clustering methods. A comparison with the traditional fuzzy c-regression models is also made. Large-scale simulations demonstrate the competitiveness of the proposed methods for signal analysis with respect to the traditional fuzzy clustering methods.