Research On Key Technologies Of Smart Home Based On Wi-Fi Gateway

Nowadays, the Internet technology is profoundly changing the people's life. As the direction of the development of the Internet of things,smart home has attracted much attention in recent years. By the Home Furnishing equipment and various types of sensors to be connected Home Furnishing environment, control platform to realize the control and scheduling of edge equipment, construction of intelligent Home Furnishing system, convenient for people's daily life in a certain extent.The current smart home design, the majority of the integration of the use of the design, the device interconnection. The user is active control of the system, including the sending of operating instructions or query commands. Therefore, the design scheme of intelligent Home Furnishing stayed on the level of operation convenient, intelligent level reflected is not high, intelligent direction of the intelligent Home Furnishing and great research space, by improving intelligence, can achieve the purpose of energy saving.In the smart home, the user as the main body of the home, every day on the home system equipment and sensors to operate. The operating device will record the user's operating data, which contains the user's habits. In this paper, aiming at the problems in the design of smart home,the paper puts forward the improvement scheme of smart home in two aspects. Design and use of BP neural network based on the Wi-Fi neural network to learn user habits. Based on the Wi-Fi gateway design can reduce the wiring of the system, make full use of the Wi-Fi resources in the current home environment, increase the system installation simplicity and equipment mobility. The application of Wi-Fi technology in Wi-Fi probe network, home users, the system can timely perception to the user,based on the detected information of users such as mobile phone MAC and RSSI, combined with the habit of user data, as the user data analysis study. Through the study of artificial neural network, BP neural network applied to intelligent Home Furnishing system, BP neural network with adaptive learning rate improved, the user data collected by gateway learning, characteristics of users daily habits, attributable to the classification of user habits, and build the cloud server platform, to achieve control, the remote storage of user data mining and user habits.The system can achieve a certain degree of understanding of the user's habits,improve the intelligence of the system and can effectively save energy.To verify the feasibility of the program,we build a smart home system, simulation of some simple user habits, in order to validate the data. We mainly select two system indicators for consideration: the refinement degree of user's daily habits; matching accuracy of user's daily habits. Through the experimental results we know, with the continuous learning, the system gradually converge, can effectively distinguish different habits and also reached a very high degree of accuracy in matching, and verified the feasibility of the scheme.