Advanced Real-Time Indoor Tracking Based on the Viterbi Algorithm and Semantic Data

A real-time indoor tracking system based on the Viterbi algorithm is developed. This Viterbi principle is used in combination with semantic data to improve the accuracy, that is, the environment of the object that is being tracked and a motion model. The starting point is a fingerprinting technique for which an advanced network planner is used to automatically construct the radio map, avoiding a time consuming measurement campaign. The developed algorithm was verified with simulations and with experiments in a building-wide testbed for sensor experiments, where a median accuracy below 2m was obtained. Compared to a reference algorithm without Viterbi or semantic data, the results indicated a significant improvement: the mean accuracy and standard deviation improved by, respectively, 26.1% and 65.3%. Thereafter a sensitivity analysis was conducted to estimate the influence of node density, grid size, memory usage, and semantic data on the performance.