Numerical algorithms using reproducing kernels for oscillatory boundary value problems (Q6896899)

This interesting paper studies numerical algorithms using reproducing kernels for oscillatory boundary value problems. To set the scene, consider the boundary value ordinary differential equation:\N\[\Nu''(x)+\omega^2u(x)=f(x,u(x)),\, u(0)=\sigma_0, u(1)=\alpha\N\]\Nwhere \(x\in (0,1)\) and the frequency \(\omega>>1\). This equation admits a solution with high frequency oscillations. The equation includes the one-dimensional Helhotz equation.\N\NDespite work by several authors on initial value ordinary differentiation equations that admit solutions with high frequency oscillations, the authors state that not much has been done for equations such as the above which admit a solution with high frequency oscillations. The main purpose of this paper is to tackle this problem using a collection of reproducing kernel methods. Numerous stimulations confirm the results given.\N\NThe paper is nicely written with a nice set of references.