Abstract:
The analysis of physical data in shallow water shows the importance of post processing. The same type of dynamical and measurement model of linear Kalman filter can be used to deal with different state variable, like time delay, bearing and range. These filters can be combined into a wonderful post-processor. Investigation of the noise parameter selection explores that observation noise parameters need not be selected strictly, process noise parameter is sensitive to environment and its adaptation is necessary in the filter. The performance of range filter as second stage filter either polar or cartesian coordinate are analyzed. At last an unexpected phenomenon is revealed from the analysis of physical data which needs further study