Mobile crowdsensing is emerging as a cost-effective solution to conduct extensive monitoring campaigns by exploiting the potential of mobile terminals with unprecedented communication, processing, and sensing capabilities. On the other hand, data provided by a large number of end-users equipped with heterogeneous devices pose trust and accuracy issues that might impair the overall reliability and usability of the system. Trust management techniques developed in the field of online social networks can be effectively used to detect and isolate cheating users, but they cannot avoid the risk of inaccurate data provided by trustworthy agents because of the inherent limitations of their devices or of the adverse conditions in which they operate. This work presents efficient algorithms for compensating the inaccuracy of crowdsensing geospatial data to be reported on a road map. The paper illustrates on a representative case study the main issues of map matching and the effectiveness of the proposed solutions, which belong to the category of incremental online methods targeting dense sampling points to be mapped on road lines without topological annotations.
Efficient Algorithms for Accuracy Improvement in Mobile Crowdsensing Vehicular Applications
DELPRIORI, SAVERIO;FRESCHI, VALERIO;LATTANZI, EMANUELE;BOGLIOLO, ALESSANDRO
2015
Abstract
Mobile crowdsensing is emerging as a cost-effective solution to conduct extensive monitoring campaigns by exploiting the potential of mobile terminals with unprecedented communication, processing, and sensing capabilities. On the other hand, data provided by a large number of end-users equipped with heterogeneous devices pose trust and accuracy issues that might impair the overall reliability and usability of the system. Trust management techniques developed in the field of online social networks can be effectively used to detect and isolate cheating users, but they cannot avoid the risk of inaccurate data provided by trustworthy agents because of the inherent limitations of their devices or of the adverse conditions in which they operate. This work presents efficient algorithms for compensating the inaccuracy of crowdsensing geospatial data to be reported on a road map. The paper illustrates on a representative case study the main issues of map matching and the effectiveness of the proposed solutions, which belong to the category of incremental online methods targeting dense sampling points to be mapped on road lines without topological annotations.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
