Write the integral part as , where and . The kernel is real, so its adjoint transposes the two arguments. Hence
This follows by exchanging the two integrations in the inner product; the exponential factors make the finite-rank operators bounded.
Every eigenvector with eigenvalue different from one lies in , because . The elementary overlaps are , , . Thus
Its characteristic polynomial is , so the finite-multiplicity eigenpairs are
Each eigenspace has dimension one. On , both integrals defining vanish, so . Thus is also an eigenvalue, of infinite multiplicity, and there are no others.
The adjoint also leaves invariant, with and . By adjoint in weighted coordinates, its matrix in the same basis is . Solving the two eigenvector equations gives
The basis is not orthonormal, so one must not simply transpose the previous numerical matrix without accounting for its Gram matrix. Directly transposing the integral kernel avoids that error.
The nonzero kernel of rules out a unique solution for every value of . By the Fredholm alternative, solvability requires orthogonality of the forcing to . Its pairing with is
Thus only is compatible. In that case and a particular solution is . All solutions are
Therefore infinitely many solutions exist for , no solution exists for , and no parameter value gives a unique solution.

Articles by others on the same topic (0)

There are currently no matching articles.