A Trust Model for Lightweight Semantic Annotation of Sensor Data in Pervasive Environment

Abstract

Pervasive computing application consists of various types of sensors, actuators and smart devices for monitoring physical, environmental circumstances and happenings by collecting data and act autonomously to serve user. Due to recent advancements of sensors and wireless technologies, pervasive computing is bringing heterogeneous sensors into our everyday life for providing better services. Data collected from heterogeneous sensors and raising number of sensor node manufacturers leads to data heterogeneity problem in pervasive computing applications. The generated data from various sensors depict more conflict in types, formats and representations which arises problem for nodes to process and infer. Because of data heterogeneity, the data cannot be shared with other application which leads to interoperability problem among pervasive environment. To overcome this, Semantic Web Technologies (SWT) are used for semantic annotation of sensor data. Annotating the sensor data with SWT is an important process in making interoperable pervasive applications. Due to resource restriction, harsh and open environments, data generated from sensor network suffers from noisy, faulty data and missing data. Annotating the faulty data with SWT causes unwanted resource consumption, network traffic and affects application performance. To solve these problems, a trust model is proposed to remove noisy, faulty data and reconstruct the missing data. Trust model ensures the annotation process of trustworthy sensor data alone and reduces resource consumption. In order to find the efficiency of proposed approach, we carried out a set of experimentation on medical sensor network prototype of pervasive healthcare application. Results show that the proposed approach is lightweight in semantic data annotation process and suitable for resource restricted nodes in pervasive environment. � 2018 IEEE.