Generally, least squares (LS) method treats only random errors of observation vector in adjustment function models. However, both observation vector and elements of coefficient matrix of an adjustment function model contain random errors. Therefore, the adjustment result of least square method does not guarantee a global-optimal solution.
Since total least square (TLS) method takes into account both random errors of observation vector and coefficient matrix based on an errors-in-variables (EIV) model, it is possible to improve the accuracy more than the result by LS method. TLS method has been further systematically developed and widely applied to many science and engineering problems, namely some practical problems, such as those in signal processing, statistical calculation and regression analysis.
Kim Jung Hyang, a researcher at the Faculty of Earth Science and Technology, has described a parameter adjustment method based on Weighted Total Least Square (WTLS) method and verified effectiveness of this method through application in simulated network. He has shown its advantage in comparison with classical LS and TLS methods.
The results show that the WTLS method based on an EIV model can further improve the accuracy of adjustment results as it handles simultaneously all kinds of random errors involved in the observation system.
© 2021 KumChaek University of Technology