Improvement And Simulation Analysis Of BP Neural Network Algorithm In Indoor Environment Monitoring

With the development of information-sensing technology,artificial intelligence technology and network technology,the quality of people's life has been greatly improved.As the indoor environment is the main living environment,indoor environment monitoring system in this context comes into being.Compared with traditional home environment monitoring system,the indoor environment monitoring system presents more demandings,such as the ability to collect indoor environment data,the ability to deal with massive data,and the ability to monitor decision-making,to provide users with intelligent,personalized monitoring services.Based on neural network,embedded technology,ZigBee wireless technology and other related technologies,this thesis presents an improved BP neural network algorithm and apply the algorithm to indoor environment monitoring system.This thesis first introduces the research status of the existing home indoor environment monitoring system at home and abroad,analyzes the limitations of the current system,introduces the key technologies used in the system,including neural network technology,wireless transmission technology and embedded technology.Then,the home environment monitoring system based on improved BP neural network is designed and implemented.The system is divided into four layers including application layer,network layer,sensing layer and control layer.Aiming at the slow convergence rate of standard BP neural network,the thesis presents to improve learning rate adaptive algorithm of BP neural network.Then,the improved BP neural network algorithm is applied to indoor environment monitoring system,and the algorithm is simulated by MATLAB.Comparing the mean square error convergence curve of standard BP neural network and modified BP neural network,the error convergence speed of improved BP neural network algorithm is improved obviously.Finally,the hardware and software of the system are debugged,and the feasibility of applying the improved BP neural network algorithm in the indoor environment monitoring system is demonstrated.