Neural Network Based Indoor Fusion Localization And Regional Power Intelligent Allocation Technology

As the information technologies develop rapidly in these years,the demand for localization services in various industries has become increasingly prominent.Fast and accurate location services have become an important part of contemporary society.Therefore,the academia has gradually turned its attention to localization services and their derivative functions in indoor environment.With the rise of the third revolution of information science and technology,the field of indoor positioning is using IOT,achieve its own technological improvement and innovation.Due to the large number of wireless routers usually installed in large public buildings,wireless charging technology has gradually been widely expected by all sectors of society.This technology can realize wireless charging for terminal devices connected to the network by sending radio frequency signals through wireless access point(AP).However,it is accompanied by a serious increase in the system power load.How to rationally allocate the system energy resources while ensuring the wireless charging function becomes the next problem to be solved.Therefore,this paper proposes an indoor fusion location algorithm based on neural network based on ZigBee(Zifeng)technology theory,and designs an intelligent power allocation algorithm for indoor area based on location awareness.According to the actual location information of the terminal,the optimal power allocation of the routing AP group in the area can be achieved,which guarantees the wireless charging performance of the terminal in the network,at the same time minimize the total transmit power of area routing or make the regional terminal receiving power as balanced as possible.Firstly,this paper expounds the research background and current situation of indoor localization technology,and briefly describes some common application scenarios,which indicates the urgent need of indoor localization in all walks of life.The discussion of the principle of the Internet of Things leads to the idea of developing the derivative function of localization services by combining the theory of group intelligence perception with indoor localization technology.Through the analysis of current research hotspots,the wireless energy transmission system is determined as the research scenario of this paper,so the role and significance of this research content are highlighted.Then the research work of this paper is briefly described to facilitate readers to understand the organization and structure of this paper.Then,an indoor fusion localization system based on neural networks is proposed.Twice fingerprinting localization method based on improved Deep Belief Network and the Pedestrian Dead Reckoning localization based on Recurrent Neural Network are implemented in the system,and information fusion is realized by particle filter technology to improve the overall positioning effect.It can be divided into two stages: the training stage trains fingerprint classifier,improved Deep Belief Network and Recurrent Neural Network based on fingerprint samples collected earlier.In the positioning stage,the real-time RSSI of the target to be measured is input into the improved Deep Belief Network to obtain the rough localization results,and the target motion offset angle measured by IMU is input into the Recurrent Neural Network to obtain the predicted position coordinates.Finally,the rough localization results and the predicted position coordinates of the target to be measured are filtered by particle filter to get accurate location information.A large number of experiments and simulations verify the performance of each sub-algorithm and the whole localization fusion system.Finally,this paper studies and designs an intelligent regional power allocation system based on location awareness.In order to solve the problem that the wireless charging function of wireless energy transmission can increase the overall power load of the system,this paper proposes an intelligent regional power allocation algorithm,which uses the location information of the receiving devices to construct the actual wireless channel model and combines the wireless energy transmission model to obtain the system signal on the premise that the energy collection power of all the devices in the network reaches the energy consumption threshold,the minimum total transmission power and transmission power allocation strategy of the AP group are obtained.Then the performance of the system is tested and analyzed through simulation experiments.The proposed intelligent regional power allocation method based on fusion indoor localization improves the accuracy of existing indoor positioning methods by means of neural networks,and reduces the computational complexity and system load of traditional positioning systems.The intelligent allocation strategy of regional power is formulated by using the obtained location information,which effectively reduces the overall power load of the signal transmitting terminal group in wireless energy transmission system and achieves the overall goal of regional energy resource saving.