Abstract
Wireless sensor network (WSN) is a dynamic network topology which is formed in a self-organizing way. In the field of WSN, the perception of information is realized by cooperative work between sensors. The wireless sensor network has the characteristics of simple structure, low cost and high reliability. Usually, most of the energy of sensor nodes is consumed in data transmission. So, reducing the data transmission between sensor nodes becomes an effective way to reduce the energy consumption of sensor networks. Data fusion technology can effectively reduce the amount of data transmission in the sensor network, which can reduce the energy consumption and get more accurate information. In this paper, we propose a hybrid data fusion algorithm based on adaptive weighting algorithm and Calman filter algorithm. In the hybrid algorithm, the Pearson similarity algorithm is used to filter the sensor nodes which do not meet the requirements, and then the average values of the two algorithms are obtained. Firstly, this paper introduces the principle of adaptive weighting algorithm and Calman filter algorithm. Secondly, we propose the improved hybrid algorithm by extracting some nodes with low similarity, and then we get the result by weighted calculation. Finally, the experimental results show that the proposed algorithm has higher accuracy and lower energy consumption.
