Python-based Simulation of Bit Error Rates

Python-based Simulation of Interconnect Performance and Link Bit Error Rates

The project PhyBER is used for the analysis of link topologies. It will be used for the simulation of statistical eye diagrams based on scattering parameters. The tool will support different line codings and signal shapes. Furthermore, channel coding will be available as well. It will be possible to include jitter and crosstalk in the analysis. The program also contains equalizers and algorithms for their optimization. An interface to metrics like the channel operating margin (COM) [1] will be present as well and could be used for an estimation of signal quality. With the help of the polynomial chaos expansion (PCE) [2] it will be possible to conduct a computational efficient sensitivity analysis. Using external python tools [3] neural networks can be trained and used for the prediction of link behavior.


Picture of eye diagrams

Eye diagram without and with a CTLE.


Picture of BER tool

General program structure of the PhyBER tool. The statistical eye diagram is simulated externally. After post-processing the results can be further analyzed by other tools.

Funding: Free and Hanseatic City of Hamburg
Contact: Katharina Scharff, M.Sc., Ömer Yildiz, M.Sc., Torben Wendt, M.Sc.

Sources:

[1] IEEE Std 802.3bj-2014 – IEEE Standard for Ethernet Amendment 2: Physical Layer Specifications and Management Parameters for 100 Gb/s Operation Over Backplanes and Copper Cables, IEEE Computer Society, September 2014
[2] Feinberg, J. & Langtangen, H. P. Chaospy: An open source tool for designing methods of uncertainty quantification, Journal of Computational Science, Elsevier BV, 2015, 11, 46-57
[3] https://www.tensorflow.org/