A novel selective node based aggregation procedure for wireless sensor networks

Traditional data aggregation methods use data originating from all sensor nodes of the wireless sensor network (WSN). Unfortunately, the use of all available sensors is not a resource efficient approach for such environments. Therefore, in this paper, we propose a novel aggregation procedure based on a hypothesis testing, where a subset of reporting nodes is intermittently selected. Neyman-Pearson lemma is used to select reporting nodes and the critical value, which is used in this aggregation procedure. Furthermore, a performance evaluation is discussed to illustrate how the proposed aggregation procedure outperforms the existing detection methods, and the dependency of the aggregation procedure and network attributes.