Installation

We assume the code is stores in some root folder:

/path/to/code/

Dependencies

• Python 3, tqmd

Build alfk

cd /path/to/code/LatticeHeuristics/alfk
python3 build_alfk.py

Generating cost estimation

Section 3

Naive depth of QPE(W) using the LWE Estimator (page 12).

cd /path/to/code/LWEEstimator 

sage our_kyber_estimates.py

Section 5

cd /path/to/code/Cost_Estimation

python3 main_interface.py [-num-cores X]

This computes all results from the paper, saves intermediate results to costData and final results are written to costResults. Additional informations can be found via

cd /path/to/code/Cost_Estimation

python3 main_interface.py --help

Results

From the paper:

• ... /path/to/code/costResults/Query contains estimations with instantiation of quantum opererator W as in Section 4.1
• ... /path/to/code/costResults/Minimal contains estimations with instantiation of quantum opererator W as in Section 4.2
• The graphs of Figure 4 in Section 5 and Figure 12,13,14,15 of the appendix can be found in the folder with the corresponding blocksize:
  • for Kyber512 in folder 406
  • for Kyber768 in folder 623
  • for Kyber1024 in folder 873
• Table 6 and 9 (Summary of values for Jensen's gap at corssover points) is printed to console, and can also be found in *.critical files

Specific Files:

• *.critical contain the intersection of lines for varying security relevant parameters
• *.pdf are the corresponding plots
• *.costs contain the detailed costs of each point plotted in the corresponding *.pdf

Appendix F

For how to compute the runtime of QPE(W), see /QPE/README.md.

Experiments on the size of enumeration trees

Appendices B, C, G

For computing the experiments on the number of children and descendants in enumeration trees, as well as experiments on the Jensen gap, see /fplll_experiments/README.md.

Code Overview

• Circuit.py: Code for instantiation of quantum operator W as in Section 5
• CostEstimation.py: Code for costing loop over the parameters m, y, z and levels k
• CostFunctions.py: Code for quantum and classical GCost and T-Depth
• plot_interface.py: Matplotlib code to produce figures from estimation results
• Tools.py: Data caching and formatted printing
• TreeHeuristics.py: Acts as an interface to LatticeHeuristics/enumTrees.py and holds functions corresponding to H_k^M, N_kh^M and S_kh^M for the combined enumertion tree