Processing of Noisy Passive Sonar Measurements

In underwater an observer pre-processes the noisy bearing measurements available from passive sonar and then the data is used by Kalman filter to find out target motion parameters. The pre-processing reduces the amplitude of the noise, replaces the missed bearings with estimated bearings, supplies the estimated bearings if the bearing measurement is not available or incorrect and finally it finds out mean and variance of the noisy data. The statistical characteristics of the data are used in Kalman filter which finds out the target motion parameters. On line estimation of bearing measurement is carried out using pseudo linear estimator. Finally, the whole algorithm is evaluated in Monte-Carlo simulation and the results for one typical scenario are presented