Encrypt module provides some encryption methods for data. It contains Paillier, RSA
Paillier encryption is a kind of addition homomorphic encryption which belongs to probabilistic asymmetric algorithm.
Generate key pairs
To generate key pairs for Paillier, two big prime numbers are needed. The generator will randomly pick two prime numbers, (p) and (q), whose bit length has been set. Then, p and q will be used as private keys to generate PrivateKey object.
Then use (n = pq) as public key to generate PublicKey object.
Encrypt by PublicKey
Encryption of Paillier follows the steps below.
Encode: Paillier algorithm is applicable only to integer. Thus the input number will be encoded as a integer.
Encrypt: The principle of encryption can be referred to here or this paper
Apply Obfuscator: Apply an obfuscator so that every encrypted number is different even if the plaintext is the same.
Decrypt by PrivateKey
Decrypt: Same principle introduction with encrypt above.
Decode: decode back as what it is.
Addition and Scalar Multiplication
Please refer to the links above with encrypt step for details.
This encryption method generates three very large positive integers $e$, $d$ and $n$. Let $e$, $n$ as the public-key and $d$ as the privacy-key. While giving data $v$, the encrypt operator will do $$ enc(v) = v^e \pmod{n}, $$ and the decrypt operator will do $$ dec(v) = enc(v) ^ d \pmod{n} $$
Hash factory provides following data encoding methods: "md5", "sha1", "sha224", "sha256", "sha384", "sha512", "sm3". This module is meant to make hashing operation with more convenient. It also supports adding salt and outputing results to base64 format.
Diffie–Hellman key exchange is a method to exchange cryptographic keys over a public channel securely
from federatedml.secureprotol.diffie_hellman import DiffieHellman
p, g = DiffieHellman.key_pair()
import random
r1 = random.randint(1, 10000000)
r2 = random.randint(1, 10000000)
key1 = DiffieHellman.decrypt(DiffieHellman.encrypt(g, r1, p), r2, p)
key2 = DiffieHellman.decrypt(DiffieHellman.encrypt(g, r2, p), r1, p)
assert key1 == key2
SPDZ(Ivan Damg˚ard, Marcel Keller) is a multiparty computation scheme based on somewhat homomorphic encryption (SHE).
init
from fate_arch.session import Session
s = Session()
# on guest side
guest_party_id = 10000
host_party_id = 10001
guest_proxy_ip = "192.168.0.2" # Generally, it is your current machine IP
federation_id = "spdz_demo" # choose a common federation id (this should be same in both site)
session_id = "_".join([federation_id, "guest", str(guest_party_id)])
s.init_computing(session_id)
s.init_federation(federation_id,
runtime_conf={
"local": {"role": "guest", "party_id": guest_party_id},
"role": {"guest": [guest_party_id], "host": [host_party_id]},
},
service_conf={"host": guest_proxy_ip, "port": 9370})
s.as_global()
partys = s.parties.all_parties
# [Party(role=guest, party_id=10000), Party(role=host, party_id=10001)]
# on host side
guest_party_id = 10000
host_party_id = 10001
host_proxy_ip = "192.168.0.1" # Generally, it is your current machine IP
federation_id = "spdz_demo" # choose a common federation id (this should be same in both site)
session_id = "_".join([federation_id, "host", str(host_party_id)])
s.init_computing(session_id)
s.init_federation(federation_id,
runtime_conf={
"local": {"role": "host", "party_id": host_party_id},
"role": {"guest": [guest_party_id], "host": [host_party_id]},
},
service_conf={"host": host_proxy_ip, "port": 9370})
s.as_global()
partys = s.parties.all_parties
# [Party(role=guest, party_id=10000), Party(role=host, party_id=10001)]
from federatedml.secureprotol.spdz import SPDZ
with SPDZ() as spdz:
...
create tenser
We currently provide two implementations of fixed point tensor:
one is based on numpy's array for non-distributed
from federatedml.secureprotol.spdz.tensor.fixedpoint_numpy import FixedPointTensor
# on guest side(assuming local Party is partys[0]):
import numpy as np
data = np.array([[1,2,3], [4,5,6]])
with SPDZ() as spdz:
x = FixedPointTensor.from_source("x", data)
y = FixedPointTensor.from_source("y", partys[1])
# on host side(assuming PartyId is partys[1]):
import numpy as np
data = np.array([[3,2,1], [6,5,4]])
with SPDZ() as spdz:
y = FixedPointTensor.from_source("y", data)
x = FixedPointTensor.from_source("x", partys[0])
from federatedml.secureprotol.spdz.tensor.fixedpoint_table import FixedPointTensor
# on guest side(assuming PartyId is partys[0]):
data = s.computing.parallelize([np.array([1,2,3]), np.array([4,5,6])], include_key=False, partition=2)
with SPDZ() as spdz:
x = FixedPointTensor.from_source("x", data)
y = FixedPointTensor.from_source("y", party[1])
# on host side(assuming PartyId is partys[1]):
data = session.parallelize([np.array([3,2,1]), np.array([6,5,4])], include_key=False, partition=2)
with SPDZ() as spdz:
y = FixedPointTensor.from_source("y", data)
x = FixedPointTensor.from_source("x", party[0])
When tensor is created from a provided data, data is split into n shares and every party gets a different one.
rescontruct
Value can be rescontructed from tensor
x.get() # array([[1, 2, 3],[4, 5, 6]])
y.get() # array([[3, 2, 1],[6, 5, 4]])
You can add or subtract tensors
z = x + y
t = x - y
dot
You can do dot arithmetic:
x.dot(y)
When using numpy's tensor, powerful einsum arithmetic is available:
x.einsum(y, "ij,kj->ik") # dot
FATE implements Oblivious Transfer(OT) protocol based on work by Eduard Hauck and Julian Loss. For more information, please refer here.
Feldman Verifiable secret sharing VSS is an information-theoretic secure method to share secrets between multi-parties.
System parameters
1024-bits prime number $p$ and $g$ , 160-bits prime-order subgroup: $q$
Set share_amount, it is the number of pieces the secret will be split into.
Encrypt
Generate $k-1$ random numbers, which is ${a_0, a_1, a2, ... ,a{k-1}}$, denotes a polynomial of degree $k-1$, which is shown as $f(x)=a_0+a_1x+a2x^2+...+a{k-1}x^{k-1}$. where $a_0$ is the secret number, which requires a number of $k$ points to calculate.
Take $k$ points on the polynomial, generate $k$ sub-keys, such as ${\langle1, f(1)\rangle, \langle 2,f(2)\rangle,\dots, \langle k, f(k)\rangle}$
Generate commitments $c_i$ according to the $k$ coefficients, $c_i=g^{a_i}$
Sub-key holder performs validation: $g^y=c_0c_1c_2c3...c{k-1}$, verifies that the sub-key is valid.
Using Lagrange interpolation to recover secret.
from federatedml.secureprotol.secret_sharing.verifiable_secret_sharing.feldman_verifiable_secret_sharing import FeldmanVerifiableSecretSharing
vss = FeldmanVerifiableSecretSharing()
vss.key_pair()
vss.set_share_amount(3)
s = -5.98
sub_key, commitment = vss.encrypt(s) # generate sub-key and commitment
vss.verify(sub_key[i], commitment) # return True or False
x, y = zip(*sub_key)
secret = vss.decrypt(x,y) # assert s == secret