Context aware data acquisition framework for dynamic data driven applications systems (DDDAS)

Various dynamic data driven applications systems (DDDAS) such as battlefield monitoring, and autonomic control and management of swarms of UAVs often leverage multiple heterogeneous sensors, where the importance of a subset of sensors may increase or decrease due to the change in the execution environment. This may require adaptation of the sampling rate of different sensors accordingly. However, current solutions for optimal rate allocation do not consider the importance (or criticality) metric of sensors, which can cause the algorithms to ignore critical nodes altogether. In this paper, we address this challenge by developing a centralized algorithm that attempts to maximize the overall quality of information for the whole network given the utility functions and the importance rankings of sensor nodes. We also present a threshold based heuristic that may help system administrators to tune the algorithm to prevent omission of highly important nodes at critical times. Extensive evaluation of our algorithm in simulation for various scenarios shows that it can quickly adapt the sampling rate in response to the changed importance of sensor nodes.