Prediction and Voronoi graph construction with a dense network of simple sensors

It is possible to monitor an environment with a dense network of cheap and simple sensors instead of a few expensive and complex sensors. In this paper, we show how to use the T-pattern algorithm to make sense of a large number of simple sensor readings. We propose two modifications to the basic T-pattern algorithm to use it in this context. In particular, we predict sensor activations given present activity, and we derive a topological layout of the environment automatically.