Indoor Path Loss Multi-model Identification Based On Wireless Fingerprint

In recent years, with the development of the internet of things, studies of locationinformation are becoming the focus and emphasis of many researchers. GPS technologyis used for people to obtain the location information because of its accuracy and rapidity.However, the GPS signal is difficult to penetrate the obstacles, so it is a great difficultyto locate indoors by GPS, which results in the rapid development of indoor locationtechnology.Indoor radio channel often needs to be modeled in indoor position technology, andfree-space model, the attenuation factor model, Chan model and logarithmic distancemodel is commonly used. However, due to the complexity of the indoor environment,such as reflection, multipath effects and so on, any single model does not exactly matchthe indoor environment, even in fixed indoor environments. When indoor environmentis changed, the result may be bad.To address this problem in this paper, indoor wireless fingerprint model matchingmethod and its application is studied. Based on in-depth analysis of the theoretical andtechnical principles of today’s indoor location, indoor propagation model is discussedand studied. Then a region-partition indoor location algorithm is proposed. First, thisalgorithm uses the offline data to determine the most suitable model for current indoorenvironment. Then the location area is divided into different regions with thereweighted iterative combinational residual minimum algorithm (RICRM). Then anindoor positioning system based on ZigBee is designed and implemented. This systemuses ZigBee sensors to acquire and transmit data, which is processed by the end of users,and the location information is displayed on the screen instantaneously.Verified by simulation and actual environmental test, the indoor location algorithmin this paper can achieve better location accuracy.