| With the development of Internet of Things technology and the improvement of people’s quality of life, traditional home has evolved to the direction of the smart home. Since the popular use of smart phones, the current smart home system is the model of controlling demostic appliance by mobile phone APP. However, this model is over relied on mobile phones, and it has the shortcoming of inadequate intelligence.This thesis focuses on how to improve the intelligence of smart home system. Artificial neural networks was used to design a smart home control system which based on MTK ARM and ZigBee technology by comparing the merits of traditional smart home system in the thesis. Through using artificial neural networks to control appliances equipment, it improved the intelligence of the smart home control systems. To the WiFi network instability problem for traditional smart home control system, it improved the stability of the system by adding GPRS functions in master control module.According to the system design, the corresponding design of hardware module was completed in the thesis, including the circuit and PCB design for control module which has MT6260D core chip, circuit debugging, the design and debugging for ZigBee module, the corresponding underlying sensor. Through the studing for the algorithm which has a learning feature, the algorithm of RBF neural network was applied to the controlling of air conditioning, intelligent lamp and smart socket in the thesis. In the designing of software, a series of work was completed in the thesis, which including the design of the controlling module based on the MTK platform, server software based on VS2008 development platform and APP software based on Android system. Finally, it debugged the entire system of hardware and software modules. Through the user actual usage test, it verified that the system has learning characteristics and other functions, thus demonstrated the usefulness of the system program.
|
PDF Full Text Request