A global in time existence and uniqueness result for an integrodifferential hyperbolic inverse problem with memory effect (Q711029)

The main focus is on a global existence and uniqueness result for an inverse problem for the following one-dimensional integro-differential hyperbolic system, arising from a simplified model in thermoelasticity: \[ u_{tt}(x,\,t) - \alpha u_{xx}(x,\,t) + \beta \eta_x(x,\,t) = 0,\;\; (x,\,t) \in \Omega \times (0,\,T); \] \[ \eta_t(x,\,t) - \int_0^t k(t-s)\eta_{xx}(x,\,s)ds + \sigma u_{xt}(x,\,t) = 0,\;\; (x,\,t) \in \Omega \times (0,\,T). \] Here, \(\alpha,\,\beta\in {\mathbb R}\), \(\sigma >0\), \(\Omega = (0,\,L)\), \(L>0\), \(u\) is the displacement, \(\eta\) is the temperature and \(k\) is a memory kernel. The above evolution equations are coupled to the following additional information on \(\eta\): \[ \int_{\Omega} \phi(x) \eta(x)dx = g(t), \] where \(g(t)\) is the measurement data representing the average temperature on the domain \(\Omega\) and \(\phi(x)\) is a weight function that is used to measure the temperature. The inverse problem aims to identify \(u\), \(\eta\) and \(k\) simultaneously from the weighted measurement data of temperature. The global existence and uniqueness of the solution to the problem is obtained via a fixed point theorem in suitable Sobolev spaces and employing a splitting process of the convolution term. Moreover, it is shown that this solution depends continuously on the noisy data.