The WSN Adaptive Positioning Technology For Indoor Environment

With the rapid development of urban construction in China, large public buildings isincreasing, by following the indoor personnel positioning have also been closer attention.Because of this situation, the wireless sensor network technology in the application of theindoor environment is imperative demanding. Among the study on the Wireless sensornetwork technology in the interior, the basic research component is the node’s ownpositioning technology. Because of the particularity of the indoor environment, outdoorenvironment is not suitable for indoor positioning technology environment, so this paperused RSSI-based wireless sensor network indoor positioning technology, designed suitablefor indoor complex environments adaptive node localization model and implement apractical indoor location system. The specific content of this research are summarized asfollows:1) For indoor office environment complexity, made from the source of the errordecreases from the ideological positioning errors. Specific characteristics of the indoorlocation; existing wireless signal propagation model and the impact of the lack ofcommunication between the nodes signal strength value (RSSI) of the main factors.2) Combined with the idea of learning and training to build the communicationbetween nodes RSSI values its distance D adaptive distance model, a three-tier structure ofthe wireless sensor network indoor positioning model. The model is divided into threelayers, the first layer is the use of Gaussian model RSSI value between nodes werescreened to ensure the validity of the initial RSSI value; second layer is based on BP neuralnetwork RSSI-D adaptive distance model, the model inputs are filtered second storeyAdams RSSI values between nodes of the matrix, the distance between the output nodematrix D, the model of the sample data by the elastic gradient descent adaptive learning toachieve the effect; The third layer is the location coordinates of a node refinement usingextended Kalman filter algorithm to coordinate loop refinement. The model of the existingindoor wireless signal propagation model improved on the basis of the measured RSSIbetween nodes based on the accuracy of the distance, and different room environmentsexhibit its adaptability.