The theoretical analysis and algorithm on a class of optimal curve fitting problems (Q864822)

Let us consider the following problem (the so-called optimal curve fitting problem): suppose that \(y_1,y_2,\dots,y_n\), \(y_{n+1}= y_1\) are the average values of measurements on \(x_1< x_2<\cdots< x_n< x_{n+1}\) and seek an optimal curve to fit these data. In this paper the theory and algorithm on a class of optimal curve fitting problems which can be extensively applied to engineering are established. The authors prove the existence of an unique optimal solution and develop its analytical expression. Moreover, they prove that for every integer value of the involved smoothing parareter \(m\) this class of problems is equivalent to solve a quadratic programming with a spherical constraint. They also give as an example the results in the case of \(m =2\) and conduct a numerical test to validate the method.