Research And Implementation Of Indoor Ranging Algorithm For BLE Devices Based On RSSI

With the continuous advancement of communication technology,the Internet of Things and the wide application of mobile devices,location-based services,as a popular network service,have penetrated into all aspects of people’s lives.Devices based on Bluetooth low energy technology have become a hotspot in indoor ranging and positioning research because of their advantages such as low power consumption,low cost,and easy deployment.This paper proposes a BLE device indoor ranging algorithm based on RSSI(Received Signal Strength),a weighted hybrid filtering algorithm based on an improved regression model and an adaptive indoor ranging model based on a RBF(Radial Basis Function)neural network.And use the i OS system to realize the design and implementation of the indoor ranging client.This paper aims at the influence of various problems on the accuracy of RSSI in the propagation process,such as multipath effect,signal attenuation,sudden interference and environmental influence and other adverse factors,which affect the accuracy of ranging.This paper proposes a weighted mixed RSSI filtering algorithm based on regression model to solve the problem.The design,construction,simulation and testing of the algorithm are introduced in detail,and compared with the commonly used multiple linear regression model.In this paper,an adaptive indoor model based on RBF neural network is proposed to solve the deficiencies in the current indoor ranging model,such as the need to manually adjust model parameters and insufficient ranging accuracy in different indoor environments.The design,simulation,construction and testing of the adaptive indoor model are introduced in detail,and a comparative analysis with the commonly used indoor distance measurement model is carried out.This paper discusses the requirements of the indoor ranging system for low-power Bluetooth devices based on RSSI,realizes the overall design architecture of the i OS ranging client,and designs and implements each module.In order to verify the effectiveness of the system,a ranging experiment was carried out on real BLE devices.The test results show that the indoor ranging system equipped with the weighted hybrid filter algorithm based on the improved regression model and the adaptive indoor ranging model based on the RBF neural network can control the error rate within 7% in short and medium distances.In the long distance,the error rate can be controlled within 10%,which has high practicability.