Motivation

With the increasing need of mobility analytics, Location of Things (LoT) is emerging as a new Internet of Things (IoT) paradigm. By connecting and integrating a variety of location sensing “things”, LoT is able to gather huge amounts of mobility data from physical world. With an annual growing rate of 25% of the devices connected to LoT , we will witness an explosive growth of heterogeneous mobility data in LoT, from outdoor to indoor, from GPS to wireless positioning, and from smartphones to infrastructures. Such data from different sources in LoT contains rich information for mobility analytics, making it possible to realize more powerful location-based services (e.g., traffic planning, logistics, security control, and customer engagement). Nevertheless, location-based services also impose strict requirements on the quality of mobility data in LoT. If the data is of low quality, the extracted knowledge and decisions based on it will be unsound or even wrong.

However, mobility data quality in LoT is compromised by the distributed LoT architecture and heterogeneous mobility data sources. For example, in a LoT for traffic planning, the positioning sensors are geo-distributed in different regions. If the sensors in a hot spot region cannot upload vehicle locations to the backend on time due to network congestion, the overall traffic flow cannot be estimated accurately and the city traffic regulation will be hindered. As another example, in a mobile health application, if there are conflicting values between the data reported by the GPS and the Wi-Fi localization server in a LoT, it is hard to make use of such inconsistent data for identifying a user’s actual mobility. In addition, mobility data is characterized by the inherent spatiotemporal uncertainty. To sum up, advanced data quality management techniques are highly needed for LoT to safeguard the quality of mobility data as well as location-based services.

Abstract

Location of Things (LoT) is an Internet of Things paradigm for mobility analytics. In LoT, massive mobility data is being gathered, processed and transmitted among heterogeneous data nodes in a decentralized architecture. Thus, managing data quality for LoT has become a prominent challenge as traditional techniques cannot cope with the aforementioned characteristics of LoT. In our project MALOT, we aim at designing a set of new techniques to manage data quality for LoT effectively and efficiently. Specifically, MALOT includes 1) a core model for assessing mobility data quality at individual LoT nodes; 2) effective data enhancement algorithms based on the quality model for resolving data heterogeneity and inconsistency; 3) a task scheduling mechanism for improving overall efficiency of data quality management in LoT.

Project Information

This project has received funding from the European Union’s Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement No.882232. More details can be found at https://cordis.europa.eu/project/id/882232.

Huan Li is the Principal Investigator and Prof. Hua Lu acts as his supervisor. This project is hosted by Aalborg University, Denmark.