Biasing emitter location estimates via false location injection

Abstract

We consider the problem of a rogue introducing bias into a network estimating location under the time difference of arrival (TDOA) method. In particular we consider how a rogue by injecting only a single false sensor position can drive the network's location estimate a specified distance away from the true value. The least squares (LS) residuals is minimized to find the false location to inject given the rogue's desired distance offset. In order to illustrate the success of our method, we consider the statistical tools that the locating network might employ to handle our false information injection including least squares and in the presence of outliers robust least median squares (LMS). We show that our method can successfully bias the location estimate of an estimating network when both LS and LMS methods are used.