Data Fusion: Comparison of Wireless Sensor Network Multi-Sensor Data Fusion Methods

Abstract

Data fusion is one of the main technologies in multi-sensor systems to solve knowledge diversification within sensor networks. The Multi-sensor data fusion is a data processing method that automatically examines then fine tunes the data obtained by various sensors within certain principles to achieve the decisions or actions needed, including fusion of information, fusion of features, fusion of relationships and decision fusion. This increases the service life of wireless sensor networks also enhances the precision of the results. Data fusion is commonly called an automated information processing process. This paper focuses on comparison of Least square error, Random Weighting method, Kalman Filtering, Bayesian Theory, Neural network techniques in multi-sensor data fusion.