Opportunistic Sensing and Collaboration for Level Set Estimation
Gagan Raj Gupta and Parmesh Ramanathan
We evaluate a novel low cost sensor network architecture, where sensors are mounted on agents moving without an explicit objective of sensing, for the task of level set estimation. A level set in a planar scalar field is the set of points with field values greater than or equal to a specified value. This problem is of fundamental importance in sensor networking and finds numerous applications in environmental studies, surveillance, habitat monitoring and processing geospatial data.
We model the problem as a classification problem and evaluate a heuristic to reduce the amount of communication, assuming that the base station uses a Support Vector Machine classifier. We then develop a fully distributed, low complexity solution which uses opportunistic information exchange to estimate level set boundaries locally at nodes selected using leader election. We observe that the learning rate of the boundary in a locality is inversely proportional to its complexity. Effectiveness of the proposed scheme is evaluated using simulations with data from both synthetic and measured fields. Random way point mobility model is used for node motion and trade-off of accuracy and of coverage with communication costs is studied.
Keywords: Sensor network, level set, SVM, collaboration, field estimation, opportunistic sensing.