A two-step method for out-of-sequence measurements

In multi-sensor environments with latency, reports from the sensors can arrive at the tracking/fusing platform out of sequence. There have been a number of methods for dealing with these reports. The most accurate way of dealing with these reports is to replay them in order. In many applications this is cost prohibitive. Dropping the reports is also not desirable, since, many times, the latent sensor is providing important cross-range information to better specify track location. There have been several approximation methods for processing the out-of-sequence measurements (OOSM). This work presents another approximation method, that in the examples provided, is better than the methods previously appearing in the literature. This method was motivated by the need to produce more accurate information for gating purposes than the previously derived methods produce. In This work we derive the new approximation method, known as the two-step method. We show that when the problem is single lag, i.e., there has only been one report after the OOSM, the method produces the exact result. We analyze the complexity of several approximation methods, and show that the two-step method is the most efficient of those analyzed. Several results are given that shows that the approximations produced are generally better than the existing methods.