We investigate the iterative methods proposed by Maz'ya and Kozlov (see [3], [4]) for solving ill-posed reconstruction problems modeled by PDE's. We consider linear time dependent problems of elliptic, hyperbolic and parabolic types. Each iteration of the analyzed methods consists on the solution of a well posed boundary (or initial) value problem. The iterations are described as powers of affine operators, as in [4]. We give alternative convergence proofs for the algorithms, using spectral theory and some functional analytical results (see [5], [6]).
翻译:我们调查了Maz'ya和Kozlov(见[3],[4])为解决以PDE为模型的错误重建问题而提出的迭代方法(见[3],[4])。我们考虑了椭圆、双曲和抛物线类型的线性时间依赖问题。所分析方法的每一次迭代都包含着一个井然存在的边界(或初始)价值问题的解决方案。迭代被描述为同系物操作者的力量,如[4]。我们使用光谱理论和一些功能性分析结果为算法提供其他趋同证据(见[5],[6])。