chapter  4
Performance Evaluation of Data Fusion Systems, Software, and Tracking
Pages 56

As we discussed in the Joint Directors’ Laboratories (JDL) data model, the evaluation of the performance of a data fusion (DF) system, algorithms, and software is very important for establishing the effectiveness of the DF system and its results. Filter tracking performance is evaluated by computing several metrics [14], some of which have already been described in Chapter 3:

1. Percentage t error (PFE) in position: This is computed as the ratio of the norm of the difference between the true and estimated positions to the norm of the true positions. This will be zero when both true and estimated positions are exactly alike (at the discrete sampling points), and it will increase when the estimated positions deviate from the true positions. When comparing the performance of different tracking algorithms, the algorithm that gives the least PFE is preferable.