Adjusting measurement error and dynamic parameters by evaluating real-time experiments with electronically steerable radar (ELRA)

The correction of errors in the data processing part of a radar system is discussed, with emphasis on the parameters of the target tracking algorithms. From this point of view, one reason for bad performance of the tracking routine is the theoretical assumption of a stochastic Gauss-Markov process describing the target dynamics and a usually assumed Gaussian measurement noise distribution. A functional relation between measurement error variances and the signal-to-noise ratio as well as the target distance is derived from ELRA real-time experiment evaluation. Improvements of the ELRA system by measurement noise parameter adaptation are discussed. Estimates for dynamic parameters are given.<>