| The advances of communication and computer techniques have enabled the development of low-cost, low-power, multifunctional sensor nodes that are small in size and capable of wireless communicating in short distances. A wireless sensor network is composed of a large number of sensor nodes that are usually densely deployed either inside the phenomenon to be observed or very close to it. They can sample data periodically and communicate with other sensor nodes. Sensor networks open up new opportunities to observe and interact with the physical world around us.Different from traditional networks, wireless sensor networks (WSNs) have limited energy and hard to supply, thus, energy efficiency is always one of the most crucial design goals. Lots of work has been done to save energy to extend the lifetime of WSN. Furthermore, WSNs are data-centric distributed systems, so data characteristic should be considered while design energy-efficient data collection protocol. In this thesis, we give the definition of data consistency and data dynamics in WSN. After analyzing between data consistency and data dynamics, we formally model them respectively in WSN.Based on the definition of data consistency and data dynamics in WSN, we improved the existing adaptive sampling method and adaptive communication method, and propose a data collection protocol called Adaptive Sampling and Communication (ASC). Our protocol can adapt the data sampling rate to the data dynamics and keep communication silence when data consistency is maintained. Besides, to avoid the prediction model has more sensitive to single outlier, we introduce a new metric called quality of prediction model which make the model parameter update decision based on it, more energy can saving from unnecessary parameter update.We evaluate our protocol in TOSSIM. From the results of experiment, we find that the proposed approach indeed reduces the data traffic at the same time guarantee the data reality and data accuracy. |
PDF Full Text Request