instructions.txt

----------------------------------------

This implementation was tested with iris codes from IITD[1] and LG or QSW feature extraction from USIT[2].
It was executed with Pypy3[3] to achieve a huge speedup in contrast to Python3.

protocolNTRU.py (main):
- example protocol that shows how to use verification and identification

clientNTRU.py
- shift/rotate iris code of the probe 
- encrypt shifted probe

serverNTRU.py
- database server
- init database (encrypt all references)
- handle verification and identification requests
    - calculate sums of ciphertext blocks
    - send encrypted sums to ttp

ttpNTRU.py
- authentication server
- trusted third party (only party that is in possesion of the secret key to decrypt)
- system thresholds are set here (computed for LG algorithm)
- determine Hamming distance of encrypted sum and compare against threshold
- return comparison decision to server

irisNTRU.py
- constant values and helper functions for both client and server usage
- similar to a config file

sortNTRU.py
- wrapper for iris code rearragement, sort most stable bits to the beginning

biometricNTRU.py
- implementation of the cryptographic functionalities of NTRU

timingNTRU.py
- function to measure the execution time


------------------------------------------------------------------
[1] http://www4.comp.polyu.edu.hk/~csajaykr/IITD/Database_Iris.htm
[2] http://wavelab.at/sources/USIT/
[3] http://pypy.org/features.html