A novel model of ZigBee in wireless sensor network based on CMOS image sensor and BP neural network

Xiaohong Luo and Mengxing Chan

Abstract

BP neural network is the error back propagation neural network, which consists of an input layer, one or more hidden layers and one output layer, each composed of a certain number of neurons. CMOS image sensor is not only the complete elimination of the fixed pattern noise generated by a circuit, and reduces the random noise, improves the CMOS image sensor sensitivity. ZigBee protocol defined in the IEEE 802.15.4 specification uses the physical layer (PHY) and media access layer dielectric (MAC), and defined on the basis of the network layer (NWK) and application layer (APL) architecture. The paper presents a novel model of ZigBee in wireless sensor network based on CMOS image sensor and BP neural network. Experimental results show the effectiveness of the improved model is better than the traditional model.