DensE: A MATLAB toolbox for density evolution

Github link:

This toolbox is a collection of MATLAB classes and routines that can be used to analyze iterative decoding schemes via density evolution.

The toolbox is currently limited to deterministic generalized product codes under iterative bounded-distance decoding. It can be used to predict the asymptotic performance of product codes, staircase codes, or other related code classes. For example, the following animation shows the wave-like decoding behavior of a staircase code, which is a type of spatially-coupled code.

staircase window

The toolbox is based on joint work with Henry D. Pfister, Alexandre Graell i Amat, and Fredrik Brännström. If you decide to use the toolbox for your research, please make sure to cite our paper:

C. Häger, Henry D. Pfister, A. Graell i Amat, F. Brännström, “Density Evolution for Deterministic Generalized Product Codes on the Binary Erasure Channel at High Rates, IEEE Trans. Inf. Theory, vol. 63, no. 7, pp. 4357-4378, July 2017
arXiv | CPL | IEEE Xplore