Near real time estimation of surveillance gaps
Steven Horn · International Conference on Information Fusion · 2013
A previously developed Bayesian inference algorithm is extended to incorporate multi-target tracking information from one or more sensors in order to generate a near real-time estimation of individual sensor detection performance. The method is also extended to operate in both a historical, and near real-time mode, which provides an up-to-date estimate of the completeness of the surveillance picture. The method is applied to real Automatic Identification System (AIS) maritime vessel traffic data from a network of receivers. Furthermore, the results from this algorithm are compared to a predictive Electro Magnetic (EM) transmission loss model. Applications of this method include surveillance asset optimization, use as a parameter for Multi-Target Tracking (MTT) algorithms, or enhanced Situational Awareness (SA) through the identification of surveillance gaps.