Research Of Indoor Localization Algorithm Based On Radio Frequency Identification Technology

As an environmental factor in context-aware system, the target location information plays an important role in governing the effectiveness and accuracy of situation-dependent services. However, typical positioning system GPS is not available for the indoor localization due to its high costs and shielding effect caused by the buildings, though a fully-fledged solution to the outside localization. Hence,the research of the indoor location sensing systems based on WSN, UWB, IF attract lots of attention in the past decade. Among others, RFID provide an excellent approach with its advantages of non line of sight, the non-contacting and low cost.Currently, most indoor location systems based on RFID technology utilize received signal strength indicator to locate objects. LANDMARC algorithm, a common example, reduces system cost and improves the environmental suitability by introducing the active reference tags to replace the reader. After calculating the Euclidian distance in signal strength between a tracking tag and each reference tag, the algorithm chooses k nearest reference tags and uses their coordinates to locate the tracking tag. Compared with other location algorithms, LANDMARC achieves a better overall accuracy of position. However , LANDMARC still faces some drawbacks and disadvantages. On the basis of the analysis of LANDMARC algorithm and building the channel model toward time delay, the major research of this dissertation specifically stated as follows:(1) To investigate the location accuracy of TTBDA (Tracking Tags near the Boundary of Detection Area) is not as good as that of TTCDA (Tracking Tags Covered by Detection Area), an optimized ranging location algorithm was proposed. By introducing the weighted factor, the algorithm rectifies the received signal strength indication values'credibility on each reader. Simulation results showed that the proposed algorithm achieves a higher accuracy and better environmental suitability towards TTBDA compared with LANDMARC.(2) In order to solve the location accuracy of LANDMARC being restricted by the number of reference tags, an enhanced VIRE location algorithm based on Subregion Dynamic Quadratic Regression Interpolation (SDQRI) was proposed. The enhanced algprothm suggests a subregion selcetion mechanism to introduce vitual reference tags dynamically, and puts forward an interpolation algorithm on the basis of quadratic regression to compute the RSSI values of VRTual reference tags. The historical information of localization is utilized to rectify the accidental error as well. Compared with VIRE algprithm, the enhanced algorithms enjoy the less amount of interpolation calculation in the AGVs system and in the case of the tracking tags highly centralized. Simulation results also demonstrated that the location accuracy of optimized algorithm is better.(3) Since the uniform arrangement of the reference tags is not applicabale in rhe case that the probability occurrence of each area differs greatly, or obstacles like wall and partition exist, an event-based occurrence probability based on the multi-gaussian distribution model and Keenan-Motley model was proposed to establish the optimization problem.An improved PSO optimizing mechanism was suggested to solve the optimization problem. Simulation results showed that the proposed algorithm outperforms the arrangement and layout adopted by genetic algorithm(4) Considering the accuracy of LANDMARC system being restricted by the range free mechanism, a location algorithm using 915MHZ UHF signal was proposed. Dual frequency subcarriers were introduced to overcome the ambiguity of whole cycles and reduce the demands of sample rate. On the basis of spectrum analysis of All-Phase FFT, a mechanism was put forward to estimate the phase difference so as to achieve the ranging information with under-sampling.Simulation results showed that the proposed algorithm possesses a higher accuracy and stability.