Multi-sensors data fusion for precise measurement based on ZigBee WSN via fuzzy control

This paper presents a storehouse boundary warning system based on multi-sensor information fusion technology. This warning system has fire detection multiple sensor data fusion algorithm based on a fuzzy neural network to compute fire emergency probability. According to the network training and self-learning weight adaptation, the error is least between the output and instruction signal. This allows the network to produce the subjective function and extract the fuzzy rule automatically, thereby improving the warning system precision and intelligence. The methodology, system structure, GSM short message, and software system design are discussed.

[1]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[2]  Y. Chan,et al.  A Kalman Filter Based Tracking Scheme with Input Estimation , 1979, IEEE Transactions on Aerospace and Electronic Systems.

[3]  Wei Sheng,et al.  Neural adaptive Kalman filter's application in MIMU/GPS/MMC integrated navigation systems , 2006, International Symposium on Instrumentation and Control Technology.

[4]  J. Lin.  The adaptation of observation noise covariances and adaptive Kalman filtering , 1973, CDC 1973.

[5]  Peter S. Maybeck,et al.  Stochastic Models, Estimation And Control , 2012 .

[6]  P. J. Escamilla-Ambrosio,et al.  A hybrid Kalman filter-fuzzy logic architecture for multisensor data fusion , 2001, Proceeding of the 2001 IEEE International Symposium on Intelligent Control (ISIC '01) (Cat. No.01CH37206).

[7]  H. P. Blom.  An efficient filter for abruptly changing systems , 1984, The 23rd IEEE Conference on Decision and Control.

[8]  A. H. Mohamed,et al.  Innovation-based Adaptive Kalman Filter for INS/GPS , 1998 .

[9]  Ken-ichi Funahashi,et al.  On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.