Title :
An Algorithm of Data Fusion Based on Improved BP Natural Network
Author :
Jun-na, Qiu ; Xiao-liang, Xu
Author_Institution :
Sch. of Comput. Sci. & Software Eng., Hangzhou Dianzi Univ., Hangzhou
Abstract :
Data fusion based on feed-forward natural network can get the goal of minimum error, as well as reduce redundant data to transmit and save energy consumption. BP neutral network is considered as one of the most mature algorithm, but the traditional BP algorithm converges slowly and easy traps in a local minimum value, so new improved BP algorithm named algebraic algorithm is put forward. The algebraic algorithm can make a complex nonlinear optimization problems turn to a simple algebraic equation, and the error function will equal to zero. Results show that the algebraic algorithm is higher precision and efficiency than BP algorithm.
Keywords :
algebra; backpropagation; neural nets; optimisation; sensor fusion; wireless sensor networks; BP algorithm; BP natural network; algebraic algorithm; complex nonlinear optimization problems; data fusion; energy consumption; error function; feedforward natural network; local minimum value; redundant data; simple algebraic equation; Computer errors; Computer science; Energy consumption; Neural networks; Sensor fusion; Signal processing algorithms; Software algorithms; Software engineering; Temperature sensors; Wireless sensor networks; BP natural network; algebraic algorithm; data fusion; wireless sensor networks;
Conference_Titel :
Computer Science and Software Engineering, 2008 International Conference on
Conference_Location :
Wuhan, Hubei
Print_ISBN :
978-0-7695-3336-0
DOI :
10.1109/CSSE.2008.895