ABSTRACT
In this chapter, a powerful and robust technique is presented, where the only assumption made for the solution of the inverse problem is that the unknown function belongs to Hilbert space of square-integrable functions in the domain of interest. The solution of the inverse problem is obtained here through the minimization of an objective functional, which is defined based on statistical hypotheses for the measurement errors. The solution technique presented in this chapter is not Bayesian, since it does not take into account prior information about the unknown function. The solution of inverse problems of function estimation within the Bayesian framework is addressed in Chapter 6.
