From: https://ethresear.ch/t/precompiled-snark-pairing-for-bls-signatures/3196/8
One problem you're may run into with pairing equalty checks on-chain is the ECPAIRING operation doesn't allow you to directly compare arbitrary pairings without some (potentially dangerous) alterations to the verification step, also you can't do scalar multiplication on G2 or GT elements on-chain.
Recap of BLS signatures:
-
$e(P_2,H(m)_1)_T = e(G_2, S_1)_T$ where $_2$ and $_1$ denote points of G1 and G2, and $_T$ for GT.
- Off-chain, you take your secret $x$, and do $xG_2 \to P_2$ (your public key).
- You then provide your public key $P_2$ to the on-chain contract
- You then generate your signature, $xH(m)_1 \to S_1$
- You provide signature to on-chain contract
- It verifies $e(P_2,H(m)_1)_T = e(G_2, S_1)_T$
The ECPAIRING operation works as such: $e(A_2, B_1) * e(C_2, D_1) = 1_T$ - which means you need to modify the pairing equality check in a way which doesn't immediately seem intuitive.
from py_ecc.bn128 import *
p = curve_order
x = randint(1, p-1) # out secret key
H_m = multiply(G1, randint(1, p-1)) # lets pretend it's HashToPoint
P = multiply(G2, x) # our public key in G2
S = multiply(H_m, x) # our signature in G1
a = pairing(P, H_m)
b = pairing(G2, S)
assert a == b # Verify signature
To use equivalent of ECPAIRING, you'd then do:
c = pairing(G2, neg(S))
assert a * c == FQ12.one()
To aggregate them:
y = randint(1, p-1) # second secret key
Q = multiply(G2, y) # second public key
T = multiply(H_m, y) # second signature
d = pairing(add(P, Q), double(H_m))
e = pairing(double(G2, add(S,T))
assert d == e
To verify the aggregates in ECPAIRING style:
d * pairing(double(G2) neg(add(S,T))) == FQ12.one()
From: https://ethresear.ch/t/precompiled-snark-pairing-for-bls-signatures/3196/8
One problem you're may run into with pairing equalty checks on-chain is the
ECPAIRINGoperation doesn't allow you to directly compare arbitrary pairings without some (potentially dangerous) alterations to the verification step, also you can't do scalar multiplication on G2 or GT elements on-chain.Recap of BLS signatures:
The$e(A_2, B_1) * e(C_2, D_1) = 1_T$ - which means you need to modify the pairing equality check in a way which doesn't immediately seem intuitive.
ECPAIRINGoperation works as such:To use equivalent of
ECPAIRING, you'd then do:To aggregate them:
To verify the aggregates in
ECPAIRINGstyle: