Cost Efficient Data Collection of Sensory Originated Data using Context-Aware Mobile Devices

Sensory originated data collection and processing has always been a big challenge in wireless sensor networks (WSN). WSN represent a distributed producer of large amount of valuable data required by varied number of applications. In this paper we propose the use of context aware data mules (CADAMULE) as a solution for smart data collection within sensor networks. We present an extension to Context Spaces modelling theory by incorporating context discovery at runtime. This facilitates our system to discover new context attributes by looking into previous situations and events when pre-defined context is not sufficient for the reasoning process. We use this model as a base to provide contextual information to the mobile data mule whose spare capacity for communication and processing can be used to collect and process sensor data. The focus of the paper is to propose and evaluate a cost-efficient data collection technique which uses a cost formula computed from the context information obtained by the system. We validate our system by a simulation in which we try to reason out and identify the best and also the most cost efficient data mule. The context aware data mule negotiates with the sensor node collecting and delivering the data to the sink.