Research On Indoor Location Technology Of Bayesian Probability Model Based On Radio Frequency Identification Technology

The growing of ubiquitous computing, wireless sensor network and mobileinternet, location-based services also showed an explosive growth, mobile devicepositioning is the key. The location information of a mobile device can be acquired byGPS, GLONASS and BDS in outdoor environment. In indoor environment，as theexistence of the shade of a complex environment and wall, we can no longer obtaingood positioning accuracy of the location, if we depend on the GPS, GLONASS andBDS. Therefore, the development of new indoor positioning system is very important!Throughout human patterns of life, most human beings live, learn and work timeare concentrated in the interior, far more time outdoors. Through the analysis and datamining of the location information of human indoor activities, not only it is helpful toscientific research, but also be able to promote the commercial prospeirty oflocation-based services. So the research of indoor localization is of important value.Localization algorithm can be divided into two categories: the first category,based on distance (Range-based); and the second category, isn’t based on the Range(Range-free). In general，the effect of the first kind of algorithm to locate is good thanthe second kind, but if you want to get higher position precision，otfen sacrifice theexpense of the hardware cost and algorithm complexity to obtain good results. But theindoor localization technique has the low system cost, the advantages of highreliability and low power consumption, so the indoor positioning system based onRFID technology is the future optimization scheme of indoor positioning system.This paper first introduces the RFID wireless communication, then completefollowing contents:1.Based on the available references and data, this paper introduces the researchbackground and demand. At the same time, we also present relevant data todemonstrate the necessity of the research.2.Analysis of the current commonly used ranging and positioning technologybased on RSSI (Received Signal Strength Indication). And we also compare a variety of systems from the aspects of system overhead, algorithmic complexity and powerconsumption.3.In this part, we introduce our indoor localization system based on RFID，system framework, function of each module and the operation process. Next, wcstudy the wireless signal of RFID system, and choose the single slope propagationmodel as our wireless channel model. Considering the actual signal fluctuation is verybig, it will greatly reduce the system ranging accuracy, this paper proposes a lowcomplexity Gaussian iflter (LCGF-Low Complexity Gaussian Filter), and we analyzethe reason of signal fluctuation. But through using the LCGF, we can obtain reliabledata. This is mainly to solve the issues of data reliability.4.Finally, this paper proposes a lightweight localization algorithm based onBayesian probability model, propagation parameter calibration technology and wheelgraph rough positioning model (WGRPM-Wheel Graph Rough Positioning Model).At the same time, wc utilize Python lo collect data, and Matlab to simulate the data.According to the simulation results, our system can also obtain higher positioningaccuracy under the large scale positioning environment by using propagationparameter calibration technology and wheel graph rough positioning model. This ismainly to solve the issues of the precision of the system on a large scale operation.