Localizing Missing Entities Using Parked Vehicles: An RFID-Based System

• Published in: IEEE internet of things journal Vol. 5; no. 5; pp. 4018 - 4030
• Main Authors: Griggs, Wynita M., Verago, Rudi, Naoum-Sawaya, Joe, Ordonez-Hurtado, Rodrigo H., Gilmore, Robert, Shorten, Robert N.
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.10.2018; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Dementia; Internet of Things; Internet of Things (IoT); Law enforcement; Monitoring; Passive RFID tags; Pedestrians; Police; Radio frequency identification; Simulation; Urban areas; Vehicles; Wrist
• ISSN: 2327-4662; 2327-4662
• DOI: 10.1109/JIOT.2018.2864590

In this paper, we demonstrate a system for locating missing entities using radio-frequency identification (RFID)-based techniques. A key feature of our system is that we utilize the large, high-density networks of parked vehicles incident to urban areas for the detection and reporting process. RFID readers and antennas are placed within the vehicles, while RFID passive tags are carried on the entity of interest via some means, e.g., a wrist band. If an entity is reported as missing, then the application on board each of the parked vehicles is awoken by an administrative center. The technology on board the vehicles enables those participating in the service to attempt to locate the missing entity, sending useful information back to the administration center, which could be tied to an organization like the police. We demonstrate our system via a use case of a missing Alzheimer's patient in inner-city Dublin, Ireland. One of the key challenges in validating our system is being able to replicate a large-scale, real-world setting. Our technique for obtaining an early evaluation of our system thus employs the use of the microscopic traffic simulation package Simulation of Urban MObility (SUMO). SUMO permits multiple emulations of hundreds or thousands of parked vehicles participating in the service to be carried out, while simulated pedestrians walk random routes. Our results show that a simulated wandering person in need can be detected within a 30-min time frame, in the heart of Dublin city center, during a typical weekday, up to approximately 98% of the time, depending on how various parameters of the system are set.