No label defined (Q4242292)

The article is a continuation of the study of solutions to the following multidimensional evolution equation suggested and preliminarily investigated by \textit{Yu. E. Anikonov}, \textit{M. P. Vishnevskij} [Sib. Mat. Zh. 37, No. 5, 963-976 (1996)]: \[ w_t = Aw +\lambda(x)f(t) \] with the conditions \[ w|{}_{t=a} = w_a(x);\qquad w|{}_{t=b} = w_b(x), \] where \(x=(x^1,\dots, x^n)\in D\subseteq\mathbb R^n\); \(w_a\) and \(w_b\) are given functions; \(w(x,t)\) is a smooth function in \(t\), \(f(t)\) is continuous on \([a,b]\); \(A\) is a linear operator acting in spatial variables and having the real symbol \(A_0(\xi)\), \(Ae^{i\xi x}= A_0(\xi)e^{i\xi x}\), \(\xi\in\mathbb R^n\). The authors call (as in the above-cited article) a formal solution \((\widehat w(\xi,t), \widehat{\lambda}(\xi))\) to the inverse problem correct if both functions \({\mathcal F}^{-1}\widehat w(\xi,t)\) and \({\mathcal F}^{-1}\widehat{\lambda}(\xi)\), where \({\mathcal F}^{-1}\) denotes the inverse Fourier transformation, belong to \(W_2^m(D)\) and \(W_2^{m-2}(D)\) (in the case of Theorem 1) and to the space of \(C^{\infty}\)- rapidly decreasing functions (in the case of Theorem 2). Applying semigroup theory together with Fourier transformations, the authors prove existence of a formal solution to the problem and obtain correctness conditions for the case in which \(A\) is a second-order elliptic-type operator; next, the results obtained are extended to systems of evolution equations. The cases pertaining to nonuniqueness of a solution to the problem are also described.