metadata-query-using-abe

this is a proof of concept, it is not designed to be used in production or enterprise context

this project requires charm-crypto (a python wrapper for stanford pbc elliptic curve cryptography)
- see http://pages.cs.wisc.edu/~ace/install-charm.html

idea:

The idea of this project was to create a cryptographically secure and privacy preserving matching of stakeholders in a medial context.

Lets say that we have a number of patients that have a number of metrics assigned to them:

their ethnic origin
their age
a condition/disease

Lets also say that there are researchers that would like to get participants for medical studies. The participants should be:

from asia or europe
between 18 and 35
should have autism or cancer

It is obvious why having a central and easily decryptable database of patients, with their respective information, is a bad idea.

With the encrypthion-scheme provided in this example, matching of patiens and researchers happens completely encrypted. The matching of patients and researchers dose not reveal any more information than needed to fulfill the query. Researchers know that the patients given back are within their parameters, but they know no more.

This is possible with the usage of Attribute-Based-Encryption.

Have a look at [~repo]/prototype/actors/patient.py to get a feel for how patients are created

Patient(asian, 22, autism)

Have a look at [~repo]/prototype/actors/query.py to get a feel for how queries work

Queue(
	"(asian or european or north_american) and (22 to 100) and (cancer or autism)"
)

Originally this was designed with Blockchain in mind, so that matching happens in a decentralized way. The class Smartcontract ([~repo]/prototype/actors/smartcontract.py) is a dummy implementation of a smartcontract that is able to do the matching completely encrypted. This could be implemented in Solidity (for the Ethereum Blockchain).

[~repo]/prototype/ contains a prototype to query bit wise encoded metadata from nodes (or people) in a privacy preserving matter
[~repo]/tests/ contains tests for encoding- and decoding-speed of different ABE-Schemes
[~repo]/string_encryption/ contains a prototype for encoding strings with ECC (Elliptic Curve Cryptography). This could be very handy once the prototype is used in a more complex manner

requirements:

python3
charm-crypto

installation:

install python3
- https://www.python.org/downloads/
install charm-crypto
- http://pages.cs.wisc.edu/~ace/install-charm.html

run the prototype:

cd [~repo]/prototype/
python3 pyrototype.py

the prototype is implemented with CPabe09. Because FAME has better performance the prototype should be updated to use FAME

run string encoding with EEC

cd [~repo]/string_encoding/
python3 concept.py

run the tests:

for testing CPabe09 (slightly modified version to fit needs) (https://jhuisi.github.io/charm/charm/schemes/abenc/abenc_waters09.html#module-abenc_waters09)

cd [~repo]/speed_tests/cp_abe/
python3 test.py

for testing FAME (https://jhuisi.github.io/charm/charm/schemes/abenc/ac17.html#ac17.AC17CPABE)

cd [~repo]/speed_tests/fame_abe/
python3 test.py

notes:

If you have any questions, you are welcome to write me: schuler.simon@gmx.net

SchulerSimon/metadata-query-using-abe