HARVEST: A task-objective efficient data collection scheme in wireless sensor and actor networks

In this paper, we study data fusion in Wireless Sensor and Actor Networks (WSANs). We propose an efficient data collection scheme, called HARVEST to collect data with an application-oriented Mobile Actor (MA). An MA is capable of saving energy of each sensor node and performing advanced computation functions based on the requests of various applications. We consider data collection application in agriculture scenario. An MA determines the next migration based on the uncertainty of sensory data provided by all n-hop neighbor nodes in HARVEST. The uncertainty is important for users to know unexpected events happened in a farmland, which they cannot cope with in advance. HARVEST, not only referring sensory data on immediate neighbors but also involving n-hop neighbors, leads total optimization of an MA's itinerary. The performance of HARVEST is evaluated by simulations of frost prediction which is a real-world application. It is shown that the total execution time of the MA can be reduced significantly while the efficiency of searching interests is maintained.