Practical automotive applications of Cramer-Rao bound analysis

The Cramer-Rao lower bound places a bound, in a mean squared sense, on the performance of all unbiased estimators. In this paper, as a base for discussion, we provide a straight forward derivation of such bounds for estimators of mobile node positions, operating on observations of distances between entities in an asynchronous network. While Cramer-Rao bound analysis is very common in the positioning community, it is mostly used for analytical evaluation of various estimators, for comparison purposes and for sensor information fusion. In this work, we present some more commercial and practical applications of these tools for performance evaluation. We first discuss the deployment of beacons throughout a transportation infrastructure as a first step towards providing global automotive, possibly GPS augmented, positioning services with applications such as collision warning and collision avoidance. We then move on to describe how the cost associated with the deployment of such beacons can be drastically reduced through relying on inter-node range measurements and a dynamic beaconing scheme we call the lighthouse scheme. We also present a method for complexity reduction in the estimation of relative node coordinates and evaluation of Cramer-Rao performance measures.