| Title |
Reconstruction Of Human Trajectories Based On Anonymous Sensor Data For Estimating Infection Route |
| ID_Doc |
44682 |
| Authors |
Matsuda K.; Ikeuchi H.; Takahashi Y.; Toyono T. |
| Year |
2022 |
| Published |
Proceedings - IEEE Global Communications Conference, GLOBECOM |
| DOI |
http://dx.doi.org/10.1109/GLOBECOM48099.2022.10001732 |
| Abstract |
The COVID-19 Pandemic has increased the demands of governments for technologies to estimate the route of infection. In this paper, we propose a new smart city framework that collects anonymized passage information from deployed Bluetooth sensors and analyzes them to reconstruct the multiple trajectories of infected people. We formulate recovering multiple trajectories on the basis of anonymized passage information, including passage time and passage position, obtained by sensors in a smart city as a problem of multiple-trajectory reconstruction in general networks. We propose a new method for reconstructing multiple trajectories on the basis of anonymized passage information. Our method assumes that each trajectory follows a Markov process and estimates transit time for each edge in networks and the transition probability of the Markov process. On the basis of its estimation, our method can find multiple trajectories with maximum likelihood by solving a minimum cost flow problem. We evaluate the performance of our method in experiments using simulation data and actual human trajectory data. © 2022 IEEE. |
| Author Keywords |
Contact tracing; Human trajectory reconstruction; Maximum likelihood estimation; Minimum cost flow |
