Geometrically Optimised Constellations for 12dB, 16dB and 20dB SNR AWGN channel
datasetposted on 26.03.2020 by Eric Sillekens, Domanic Lavery
Datasets usually provide raw data for analysis. This raw data often comes in spreadsheet form, but can be any collection of data, on which analysis can be performed.
Geometrically Shaped Constellations for an additive white Gaussian noise (AWGN) channel. For 12 dB, 16 dB and 20 dB signal-to-noise ratio (SNR) a 64, 256 and 1024 point constellation respectively are numerically optimised to maximise generalised mutual informatio (GMI) for the given SNRs.
These constellations are the result of a numerical optmisation method similar to [https://doi.org/10.1109/ECOC.2018.8535358].
Every text file has the cooridinates for the inphase and quadrature components of each point in the first and the second column respectively. The bit mapping for each point is natural mapping for the line number, i.e., 000 001 010 011 100 etc.
I confirm that I am not uploading any: personal data as defined by data protection legislation, including information that may identify a living individual; information provided in confidence; or information that would contravene a third-party agreement
I have considered whether the data to be published may be licensed commercially before deciding to freely release it to the public. Further information and advice may be sought from UCL Business https://www.uclb.com/about/our-people/