Efficient data collection through dynamic intensional clustering

Wireless sensor networks consist of a large number of sensor nodes that combine physical sensing (e.g. temperature, light) with networking and computation capabilities. Sensors are small entities where conservation of battery power is a major design factor. In-network processing, such as aggregation, eliminates similar/redundant data and decreases the number of communication messages to be sent to the sink, resulting in an increase of the network lifetime and a decrease of the communication complexity. We propose a dynamic clustering protocol which allows to aggregate efficiently collected data based on intensional destinations. In contrast to classical clustering approaches, the clusters are constructed on the fly by evaluating dynamically the intensional destinations when collected data in messages are traveling. Our simulations over QuestMonitor show that the proposed protocol allows dynamic adaptation to topology changes, persistence of data, as well as a resilience of the system.