A distributed data-centric storage method for hot spot mitigation in wireless sensor networks

In wireless sensor networks, sensor nodes are capable of not only measuring real world phenomena, but also storing, processing, and transferring these measurements. Many techniques have been proposed for storing and retrieving data in these networks. These methods are classified into three main categories. Data-centric storage is the most important one. The name comes from the fact that data's storage location is determined according to it's event type. However, most of the methods proposed in this category use a fixed event-location mapping or don't take nodes' resources into consideration. Therefore, storage nodes may experience unbalanced resource utilization problems which results into hot spot areas. We address this problem by proposing a dynamic mapping scheme which aims to reach two goals: First, to prolong network lifetime by efficient energy consumption and second, increasing quality of service by increasing data availability. The proposed method comes in two phases: First hot spot detection phase where storage nodes' status information is gathered at sink. Applying the proposed hot spot detection mechanism, sink detects a hot spotted node and alarms the node by sending a warning message to it. The hot spotted storage node receives the warning message and triggers the second phase named hot spot mitigation. In this stage it should transfer the responsibility of storing the event to another node in its' h hop neighborhood. Two metrics are considered in selecting a new storage node: 1) minimizing the total energy consumed for storing and retrieving the event 2) distributing network load evenly. The simulation presented here evaluates the proposed method in terms of network lifetime, data availability and energy consumption. The results show that the proposed method will delay hot spot effect with acceptable overhead.