Combined sensor information for detection

In this paper we are concerned with the issue of combining sensor information for detection. Two motivating examples are studied: 1) the detection of a static object within a given area and 2) the detection of a faulty tank in a three-tank system. From a very general perspective, both of these problems can be thought of as search problems; the goal in both cases is to determine the presence (or absence) of an “object” within a given area of interest (AOI). It is assumed that there are two classes of sensors: a large sensor capable of searching the entire AOI and a set of small sensors which (collectively) search only a subset of the AOI. Measurements from the small sensors are assumed to follow a Bernoulli distribution (depending on whether they detect the object or not). Measurements collected from the large sensor are allowed to follow any exponential family distribution. In order to combine information we propose a system identification framework based on maximum-likelihood (ML) estimation. This requires collecting several measurements (samples) from each sensor. The ML approach allow us to borrow existing convergence and asymptotic normality results from the literature.