Stable sequential Lagrange principles in the inverse final observation problem for the system of Maxwell equations in the quasistationary magnetic approximation (Q311184)

In this paper, the authors consider an inverse problem resulting from Maxwell's equations with quasi-static magnetic approximation. The interested model is governed by the equations: \[ \frac{\partial }{\partial t}(\mu{\mathbf H}) + \text{rot}(\sigma^{-1}\text{rot}{\mathbf H}) =\text{rot}{\mathbf E}^{\mathrm{ext}}, \quad\text{in}~\Omega\times (0,T), \] \[ \text{div}(\mu{\mathbf H}) = 0, \quad\text{in}~\Omega\times (0,T), \] subject to the perfectly conducting boundary condition and the initial condition \[ {\mathbf H}({\mathbf x},0)={\mathbf h}({\mathbf x}), \quad {\mathbf x}\in \Omega, \] where \(\Omega\subset \mathbb R^3\) is an open bounded domain with Lipschitz boundary, and \(rot\) denotes the curl operator. The authors want to reconstruct the source \({\mathbf E}^{\mathrm{ext}}\) and initial data \({\mathbf h}({\mathbf x})\) from the observation of the magnetic field \({\mathbf H}\) at the final time \(T\). Due to the instability of the inverse problem, the authors discuss several regularizing algorithms such as a stable sequential Lagrange principle and the method of iterative dual regularization for the inverse final observation problem.