Use policyThe full-text may be used and/or reproduced, and given to third parties in any format or medium, without prior permission or charge, for personal research or study, educational, or not-for-prot purposes provided that:• a full bibliographic reference is made to the original source • a link is made to the metadata record in DRO • the full-text is not changed in any way The full-text must not be sold in any format or medium without the formal permission of the copyright holders.Please consult the full DRO policy for further details.
SUMMARYThe boundary element method (BEM) is a popular technique for wave scattering problems given its inherent ability to deal with infinite domains. Recently, the partition of unity BEM, in which the approximation space is enriched with a linear combination of plane waves, has been developed; this significantly reduces the number of degrees of freedom required per wavelength. It has been shown that the element ends are more susceptible to errors in the approximation than the mid-element regions. In this paper the authors propose that this is due to the reduced order of continuity in the Lagrangian shape function component of the basis functions. It is demonstrated, using numerical examples, that choosing trigonometric shapes functions, rather than classical quadratic shape functions, provides accuracy benefits. It is also demonstrated that the somewhat arbitrary choice of collocating at equally spaced points about the surface of a scatterer is, in fact, the optimum choice of collocation scheme.