Recent Releases of LyceanEM

LyceanEM -

This release marks a major shift in the distribution approach of LyceanEM, moving away from pip to conda in order to simplify the installation experience for users. LyceanEM can now be installed with a single conda command on both windows and linux, with significant improvements in the usability. Gmsh has now been fully integrated not just into the ecosystem but also into the examples and documentation, allowing for a more streamlined workflow when working with the included geometries, including parabolic reflectors.

Now that 0.1.0 is out, the focus will shift to expanding the examples to include examples on modelling measurement setups for validation purposes, and model based systems design approaches. In addition, the use of LyceanEM for wireless power transfer to model scaled scenarios for Space-based Solar Power will also be included.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM 9 months ago

LyceanEM - MeshIO Integration and New Propagation Algorithm

This release has been issued shifting the geometry underpinnings from open3d to meshio, relaxing the dependancy issues mandated by including open3d as a requirement. MeshIO is a powerful IO backend for a wide variety of mesh formats, allowing the LyceanEM project to shift to more flexible formats to transfer field data for scattering, antenna and antenna array patterns. In addition this allows the inclusion of pygmsh to bring support for true constructive geometry and some CAD formats via Gmsh.

Both the Frequency and Time Domain scattering model propagation algorithms have been shifted to a derivation of the Rayliegh-Summerfeld diffraction integral allowing for accurate field modelling in both the near and farfield, as well as power transfer. This employs the propagation constant to allow future releases to support lossy propagation mediums. Currently there have been no significant changes to the model functions, but the way that Calculate Farfield works will be changing in the next release. It will be redefined as a post processing filter based upon Calculate Scattering, allowing a concentration of further development effort on the more general purpose functions.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM over 1 year ago

LyceanEM - Release for Journal of Open Source Software, and Numba Version tolerance

This release has been issued to support submission of LyceanEM to the Journal of Open Source software, including solutions to bugs discovered by the reviewers. This includes applied bugfixes to support Numba 0.57 variable typing support changes. The Frequency Domain Aperture Projection function now supports passthrough of the farfield distance to be considered, so the user can define the range at which the aperture projection operation should be considered.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM over 2 years ago

LyceanEM - Stability Improvements

Updates to the frequency domain scattering function which allow for chunking, relaxing the upper limit to the size of antennas or antenna arrays which can be considered. This is not as time efficient as transferring the arrays to the GPU and back to the host is not optimised, but does enable larger models, even on computers with a low memory GPU. In addition, a bug in the expected behaviour of the mesh_conversion function has now been fixed, with list detection and conversion, so that if there are multiple triangle meshes in the structure object, it will now convert them to the raycaster format appropriately.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM almost 3 years ago

LyceanEM - Stablity and class upgrades

This release includes a number of quality of life upgrades for more advanced modelling, including new pattern plot functions, classes and types to enable consistent manipulation of the antenna structures.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM over 3 years ago

LyceanEM - Initial Release

Initial release of LyceanEM, fixing issues with dependancies in PyPi, and getting the repository ready for conda.

Scientific Software - Peer-reviewed - Python
Published by LyceanEM almost 4 years ago