Greedy recursion parameter selection for the One-Way Navier-Stokes (OWNS) equations

The One-Way Navier-Stokes (OWNS) equations use recursive filtering to construct efficient, well-posed one-way approximations to linear hyperbolic systems. The recursion parameters are critical to the accuracy and stability of the method, and have previously been chosen based on heuristic estimates of key eigenvalues (or their branches), which converges slowly and requires trial-and-error tuning for new systems. We review the projection and recursive OWNS formulations (OWNS-P and OWNS-R) for inhomogeneous equations and propose a greedy algorithm for parameter selection. We show that it converges faster and leads to a net decrease in computational cost.