THE THEORY OF OPTIMAL INFORMATION DIFFUSION ESTIMATION AND ITS APPLICATION

摘要

This paper studies the problem of the least-squares estimation when observations are contaminated by some gross errors. Authors consider that the information-diffusion estimation(IDE) is an intelligent parameter estimation that is beter than the least square estimation. This paper focuses on the method of the optimal information-diffusion estimation(OIDE). The theory and calculations indicate that the new method not only has fine robustness, but also it is very simple and convenient for application. The study is very important to find an intelligent techique instead of the the least square method.