Statistical-Based Data Quality Model for Mobile Crowd Sensing Systems

TL;DR: This paper proposes a statistical MCS data quality model which can be used to collect the sensory data based on the data requester requirements to improve the quality of the data and selects the best users to participate in the sensing task for collecting the requested sensory data.

Abstract: Quality of information is an emerging issue in the mobile crowd sensing (MCS). MCS is an essential computing paradigm that tasks everyday mobile devices to form crowd sensing networks. Nowadays, there is an increasing demand to provide real-time environmental information such as air quality, noise level, traffic condition. However, the openness of crowd sensing exposes the system to malicious and erroneous participation, inevitably resulting in poor data quality. This brings forth an important issue of false data detection and correction in crowd sensing. Furthermore, data collected by participants normally include considerable missing values, which poses challenges for accurate false data detection. To improve the quality of the collected sensory data, the system server needs to consider the factors which are influencing the quality of the collected data. So, to acquire a high-quality sensory data, the MCS needs some efficient platforms to enhance the collected data, select the best participants from a group of users, and determine the perfect coverage type of sensing location and the exact sensing time which will achieve high-quality sensory data with low cost. In this paper, we will study the factors which are affected the MCS data quality, and propose a statistical MCS data quality model which can be used to collect the sensory data based on the data requester requirements to improve the quality of the data and selects the best users to participate in the sensing task for collecting the requested sensory data.