Data security and quality evaluation framework: Implementation empirical study on android devices

Abstract

Tremendous growth of a number of mobile devices and amount of data produced by sensors embedded therein requires new approaches to sensor data management. The main feature of the novel approach proposed in this paper includes an assignment of security and data quality indicators to data entities. These indicators represent the trustworthiness level, which a data consumer may have. Employing them as filters would allow for an optimization of diverse sensors data processing and fusing with a significant reduction in data volumes. The paper describes the developed comprehensive methodology that resulted in the evaluation framework. Framework merges together sensor data collection and security and quality evaluation methods as well as procedures for calculating various data quality metrics such as sensor accuracy, reliability, timeliness, correctness and their integration. The paper desribes main features of this framework and examples of its implementation on Android based smartphone devices. It presents the results of an empirical study of the framework implementation and discusses its application for an anomaly detection in sensor data.