Title :
Update to the Hybrid Conditional Averaging Performance Prediction of the IMM Algorithm
Author :
Osborne, R.W. ; Blair, W.D.
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Connecticut, Fairfield, CT, USA
fDate :
10/1/2011 12:00:00 AM
Abstract :
Traditionally the performance evaluation of a target tracking algorithm is accomplished via Monte Carlo simulations for each specific scenario of interest. For some applications, the time and computational resource requirements of performing the necessary simulations for algorithm design is excessive; so the need for performance prediction becomes paramount. One method of performance prediction developed during the early 1990s is the hybrid conditional averaging (HYCA) technique, which can be used to predict the performance of the interacting multiple model (IMM) algorithm. Applying the HYCA technique to the IMM algorithm as originally developed leads to poor performance prediction in certain situations. A new extension used in these circumstances is shown to lead to superior performance prediction without an increase in computational complexity compared with the originally developed algorithm for such situations.
Keywords :
performance evaluation; signal processing; target tracking; IMM algorithm; computational complexity; computational resources; hybrid conditional averaging performance prediction; interacting multiple model algorithm; target tracking algorithm; Filtering algorithms; Mathematical model; Monte Carlo methods; Prediction algorithms; Trajectory; Uncertainty;
Journal_Title :
Aerospace and Electronic Systems, IEEE Transactions on
DOI :
10.1109/TAES.2011.6034677