RSSI Fingerprinting-Based Indoor Localization Algorithm Using an Interval of Estimated Distance

The problem of localization in indoor wireless networks has been actively studied using RSSI(Received Signal Strength Indicator) fingerprinting techniques. In this paper, we propose two localization algorithms, ML(Maximum Likelihood) algorithm and VAML(Valid Area Maximum Likelihood) algorithm, which are based on RSSI fingerprinting, and present key parameters for them. To inspect the effect of the main parameters of the VAML algorithm on the accuracy of the estimated location, the performance of the algorithm was compared by performing simulations and varying the value of key parameters such as the number of iterations the sensor node measures RSSI, the number of reference nodes, and the range of estimated distances calculated by an RSSI value. Lastly, the accuracy of the estimated locations and time complexities of the presented algorithms and wKNN(weighted KNN) algorithm were compared to verify that the VAML algorithm shows better performance compared to the wKNN algorithm and the ML algorithm.