Impact of Sensor Data Patterns on Performance Evaluation of Clustering Schemes in Wireless Sensor Networks

The superiority of clustering schemes proposed so far has been reported by performance evaluation. However, application environments are not effectively considered with respect to the data patterns of sensing attributes. In practice, different categories of applications have different data patterns of sensing attributes. In this paper, we study the data patterns of sensing attributes in various applications and present that one or more appropriate data patterns should be used depending upon the target application for fair evaluation and comparison. As a result of our study, five data patterns of sensing attributes are proposed in this paper. According to our performance analysis, the performance of clustering schemes is significantly affected by the data patterns. Therefore, it is strongly suggested that the performance evaluation of clustering schemes should take the data patterns into consideration on the basis of their application areas. In multi-application sensor networks, according to our evaluation results, this is particularly critical.