Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
#ifndef _SECP256K1_RANGEPROOF_
|
|
|
|
# define _SECP256K1_RANGEPROOF_
|
|
|
|
|
|
|
|
# include "secp256k1.h"
|
2016-07-07 00:47:41 +02:00
|
|
|
# include "secp256k1_generator.h"
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
# ifdef __cplusplus
|
|
|
|
extern "C" {
|
|
|
|
# endif
|
|
|
|
|
|
|
|
#include <stdint.h>
|
|
|
|
|
2016-07-04 13:04:57 +00:00
|
|
|
/** Opaque data structure that stores a Pedersen commitment
|
|
|
|
*
|
|
|
|
* The exact representation of data inside is implementation defined and not
|
|
|
|
* guaranteed to be portable between different platforms or versions. It is
|
|
|
|
* however guaranteed to be 33 bytes in size, and can be safely copied/moved.
|
|
|
|
* If you need to convert to a format suitable for storage or transmission, use
|
|
|
|
* the secp256k1_pedersen_commitment_serialize_* and
|
|
|
|
* secp256k1_pedersen_commitment_serialize_* functions.
|
|
|
|
*
|
|
|
|
* Furthermore, it is guaranteed to identical signatures will have identical
|
|
|
|
* representation, so they can be memcmp'ed.
|
|
|
|
*/
|
|
|
|
typedef struct {
|
|
|
|
unsigned char data[33];
|
|
|
|
} secp256k1_pedersen_commitment;
|
|
|
|
|
2016-07-06 13:46:23 +02:00
|
|
|
/**
|
|
|
|
* Static constant generator 'h' maintained for historical reasons.
|
|
|
|
*/
|
|
|
|
extern const secp256k1_generator *secp256k1_generator_h;
|
|
|
|
|
2016-07-04 13:04:57 +00:00
|
|
|
/** Parse a 33-byte commitment into a commitment object.
|
|
|
|
*
|
2016-07-06 13:46:23 +02:00
|
|
|
* Returns: 1 if input contains a valid commitment.
|
2016-07-04 13:04:57 +00:00
|
|
|
* Args: ctx: a secp256k1 context object.
|
|
|
|
* Out: commit: pointer to the output commitment object
|
|
|
|
* In: input: pointer to a 33-byte serialized commitment key
|
|
|
|
*/
|
|
|
|
SECP256K1_API SECP256K1_WARN_UNUSED_RESULT int secp256k1_pedersen_commitment_parse(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
secp256k1_pedersen_commitment* commit,
|
|
|
|
const unsigned char *input
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3);
|
|
|
|
|
|
|
|
/** Serialize a commitment object into a serialized byte sequence.
|
|
|
|
*
|
|
|
|
* Returns: 1 always.
|
|
|
|
* Args: ctx: a secp256k1 context object.
|
|
|
|
* Out: output: a pointer to a 33-byte byte array
|
|
|
|
* In: commit: a pointer to a secp256k1_pedersen_commitment containing an
|
|
|
|
* initialized commitment
|
|
|
|
*/
|
|
|
|
SECP256K1_API int secp256k1_pedersen_commitment_serialize(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
unsigned char *output,
|
|
|
|
const secp256k1_pedersen_commitment* commit
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3);
|
|
|
|
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
/** Initialize a context for usage with Pedersen commitments. */
|
|
|
|
void secp256k1_pedersen_context_initialize(secp256k1_context* ctx);
|
|
|
|
|
|
|
|
/** Generate a pedersen commitment.
|
|
|
|
* Returns 1: commitment successfully created.
|
|
|
|
* 0: error
|
|
|
|
* In: ctx: pointer to a context object, initialized for signing and Pedersen commitment (cannot be NULL)
|
|
|
|
* blind: pointer to a 32-byte blinding factor (cannot be NULL)
|
|
|
|
* value: unsigned 64-bit integer value to commit to.
|
2016-07-04 13:04:57 +00:00
|
|
|
* Out: commit: pointer to the commitment (cannot be NULL)
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
*
|
|
|
|
* Blinding factors can be generated and verified in the same way as secp256k1 private keys for ECDSA.
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_pedersen_commit(
|
|
|
|
const secp256k1_context* ctx,
|
2016-07-04 13:04:57 +00:00
|
|
|
secp256k1_pedersen_commitment *commit,
|
|
|
|
const unsigned char *blind,
|
2016-07-06 13:46:23 +02:00
|
|
|
uint64_t value,
|
|
|
|
const secp256k1_generator *gen
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3) SECP256K1_ARG_NONNULL(5);
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
/** Computes the sum of multiple positive and negative blinding factors.
|
|
|
|
* Returns 1: sum successfully computed.
|
|
|
|
* 0: error
|
|
|
|
* In: ctx: pointer to a context object (cannot be NULL)
|
|
|
|
* blinds: pointer to pointers to 32-byte character arrays for blinding factors. (cannot be NULL)
|
|
|
|
* n: number of factors pointed to by blinds.
|
|
|
|
* nneg: how many of the initial factors should be treated with a positive sign.
|
|
|
|
* Out: blind_out: pointer to a 32-byte array for the sum (cannot be NULL)
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_pedersen_blind_sum(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
unsigned char *blind_out,
|
|
|
|
const unsigned char * const *blinds,
|
2016-07-04 13:04:57 +00:00
|
|
|
size_t n,
|
|
|
|
size_t npositive
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3);
|
|
|
|
|
|
|
|
/** Verify a tally of pedersen commitments
|
|
|
|
* Returns 1: commitments successfully sum to zero.
|
|
|
|
* 0: Commitments do not sum to zero or other error.
|
|
|
|
* In: ctx: pointer to a context object, initialized for Pedersen commitment (cannot be NULL)
|
2016-07-04 13:04:57 +00:00
|
|
|
* commits: pointer to array of pointers to the commitments. (cannot be NULL if pcnt is non-zero)
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
* pcnt: number of commitments pointed to by commits.
|
2016-07-04 13:04:57 +00:00
|
|
|
* ncommits: pointer to array of pointers to the negative commitments. (cannot be NULL if ncnt is non-zero)
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
* ncnt: number of commitments pointed to by ncommits.
|
|
|
|
* excess: signed 64bit amount to add to the total to bring it to zero, can be negative.
|
|
|
|
*
|
|
|
|
* This computes sum(commit[0..pcnt)) - sum(ncommit[0..ncnt)) - excess*H == 0.
|
|
|
|
*
|
|
|
|
* A pedersen commitment is xG + vH where G and H are generators for the secp256k1 group and x is a blinding factor,
|
|
|
|
* while v is the committed value. For a collection of commitments to sum to zero both their blinding factors and
|
|
|
|
* values must sum to zero.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_pedersen_verify_tally(
|
|
|
|
const secp256k1_context* ctx,
|
2016-07-04 13:04:57 +00:00
|
|
|
const secp256k1_pedersen_commitment * const* commits,
|
|
|
|
size_t pcnt,
|
|
|
|
const secp256k1_pedersen_commitment * const* ncommits,
|
|
|
|
size_t ncnt,
|
2016-07-06 13:46:23 +02:00
|
|
|
int64_t excess,
|
|
|
|
const secp256k1_generator *gen
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(4) SECP256K1_ARG_NONNULL(7);
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
/** Initialize a context for usage with Pedersen commitments. */
|
|
|
|
void secp256k1_rangeproof_context_initialize(secp256k1_context* ctx);
|
|
|
|
|
|
|
|
/** Verify a proof that a committed value is within a range.
|
|
|
|
* Returns 1: Value is within the range [0..2^64), the specifically proven range is in the min/max value outputs.
|
|
|
|
* 0: Proof failed or other error.
|
|
|
|
* In: ctx: pointer to a context object, initialized for range-proof and commitment (cannot be NULL)
|
2016-07-04 13:04:57 +00:00
|
|
|
* commit: the commitment being proved. (cannot be NULL)
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
* proof: pointer to character array with the proof. (cannot be NULL)
|
|
|
|
* plen: length of proof in bytes.
|
|
|
|
* Out: min_value: pointer to a unsigned int64 which will be updated with the minimum value that commit could have. (cannot be NULL)
|
|
|
|
* max_value: pointer to a unsigned int64 which will be updated with the maximum value that commit could have. (cannot be NULL)
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_rangeproof_verify(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
uint64_t *min_value,
|
|
|
|
uint64_t *max_value,
|
2016-07-04 13:04:57 +00:00
|
|
|
const secp256k1_pedersen_commitment *commit,
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
const unsigned char *proof,
|
2016-07-06 13:46:23 +02:00
|
|
|
size_t plen,
|
|
|
|
const secp256k1_generator* gen
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3) SECP256K1_ARG_NONNULL(4) SECP256K1_ARG_NONNULL(5) SECP256K1_ARG_NONNULL(7);
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
/** Verify a range proof proof and rewind the proof to recover information sent by its author.
|
|
|
|
* Returns 1: Value is within the range [0..2^64), the specifically proven range is in the min/max value outputs, and the value and blinding were recovered.
|
|
|
|
* 0: Proof failed, rewind failed, or other error.
|
|
|
|
* In: ctx: pointer to a context object, initialized for range-proof and Pedersen commitment (cannot be NULL)
|
2016-07-04 13:04:57 +00:00
|
|
|
* commit: the commitment being proved. (cannot be NULL)
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
* proof: pointer to character array with the proof. (cannot be NULL)
|
|
|
|
* plen: length of proof in bytes.
|
|
|
|
* nonce: 32-byte secret nonce used by the prover (cannot be NULL)
|
|
|
|
* In/Out: blind_out: storage for the 32-byte blinding factor used for the commitment
|
|
|
|
* value_out: pointer to an unsigned int64 which has the exact value of the commitment.
|
|
|
|
* message_out: pointer to a 4096 byte character array to receive message data from the proof author.
|
|
|
|
* outlen: length of message data written to message_out.
|
|
|
|
* min_value: pointer to an unsigned int64 which will be updated with the minimum value that commit could have. (cannot be NULL)
|
|
|
|
* max_value: pointer to an unsigned int64 which will be updated with the maximum value that commit could have. (cannot be NULL)
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_rangeproof_rewind(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
unsigned char *blind_out,
|
|
|
|
uint64_t *value_out,
|
|
|
|
unsigned char *message_out,
|
2016-07-04 13:04:57 +00:00
|
|
|
size_t *outlen,
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
const unsigned char *nonce,
|
|
|
|
uint64_t *min_value,
|
|
|
|
uint64_t *max_value,
|
2016-07-04 13:04:57 +00:00
|
|
|
const secp256k1_pedersen_commitment *commit,
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
const unsigned char *proof,
|
2016-07-06 13:46:23 +02:00
|
|
|
size_t plen,
|
|
|
|
const secp256k1_generator *gen
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(6) SECP256K1_ARG_NONNULL(7) SECP256K1_ARG_NONNULL(8) SECP256K1_ARG_NONNULL(9) SECP256K1_ARG_NONNULL(10) SECP256K1_ARG_NONNULL(12);
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
/** Author a proof that a committed value is within a range.
|
|
|
|
* Returns 1: Proof successfully created.
|
|
|
|
* 0: Error
|
|
|
|
* In: ctx: pointer to a context object, initialized for range-proof, signing, and Pedersen commitment (cannot be NULL)
|
|
|
|
* proof: pointer to array to receive the proof, can be up to 5134 bytes. (cannot be NULL)
|
|
|
|
* min_value: constructs a proof where the verifer can tell the minimum value is at least the specified amount.
|
2016-07-04 13:04:57 +00:00
|
|
|
* commit: the commitment being proved.
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
* blind: 32-byte blinding factor used by commit.
|
|
|
|
* nonce: 32-byte secret nonce used to initialize the proof (value can be reverse-engineered out of the proof if this secret is known.)
|
|
|
|
* exp: Base-10 exponent. Digits below above will be made public, but the proof will be made smaller. Allowed range is -1 to 18.
|
|
|
|
* (-1 is a special case that makes the value public. 0 is the most private.)
|
|
|
|
* min_bits: Number of bits of the value to keep private. (0 = auto/minimal, - 64).
|
|
|
|
* value: Actual value of the commitment.
|
|
|
|
* In/out: plen: point to an integer with the size of the proof buffer and the size of the constructed proof.
|
|
|
|
*
|
|
|
|
* If min_value or exp is non-zero then the value must be on the range [0, 2^63) to prevent the proof range from spanning past 2^64.
|
|
|
|
*
|
|
|
|
* If exp is -1 the value is revealed by the proof (e.g. it proves that the proof is a blinding of a specific value, without revealing the blinding key.)
|
|
|
|
*
|
|
|
|
* This can randomly fail with probability around one in 2^100. If this happens, buy a lottery ticket and retry with a different nonce or blinding.
|
|
|
|
*
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_rangeproof_sign(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
unsigned char *proof,
|
2016-07-04 13:04:57 +00:00
|
|
|
size_t *plen,
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
uint64_t min_value,
|
2016-07-04 13:04:57 +00:00
|
|
|
const secp256k1_pedersen_commitment *commit,
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
const unsigned char *blind,
|
|
|
|
const unsigned char *nonce,
|
|
|
|
int exp,
|
|
|
|
int min_bits,
|
2016-07-05 15:46:07 +00:00
|
|
|
uint64_t value,
|
|
|
|
const unsigned char *message,
|
2016-07-06 13:46:23 +02:00
|
|
|
size_t msg_len,
|
|
|
|
const secp256k1_generator *gen
|
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3) SECP256K1_ARG_NONNULL(5) SECP256K1_ARG_NONNULL(6) SECP256K1_ARG_NONNULL(7) SECP256K1_ARG_NONNULL(13);
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
|
|
|
|
/** Extract some basic information from a range-proof.
|
|
|
|
* Returns 1: Information successfully extracted.
|
|
|
|
* 0: Decode failed.
|
|
|
|
* In: ctx: pointer to a context object
|
|
|
|
* proof: pointer to character array with the proof.
|
|
|
|
* plen: length of proof in bytes.
|
|
|
|
* Out: exp: Exponent used in the proof (-1 means the value isn't private).
|
|
|
|
* mantissa: Number of bits covered by the proof.
|
|
|
|
* min_value: pointer to an unsigned int64 which will be updated with the minimum value that commit could have. (cannot be NULL)
|
|
|
|
* max_value: pointer to an unsigned int64 which will be updated with the maximum value that commit could have. (cannot be NULL)
|
|
|
|
*/
|
|
|
|
SECP256K1_WARN_UNUSED_RESULT int secp256k1_rangeproof_info(
|
|
|
|
const secp256k1_context* ctx,
|
|
|
|
int *exp,
|
|
|
|
int *mantissa,
|
|
|
|
uint64_t *min_value,
|
|
|
|
uint64_t *max_value,
|
|
|
|
const unsigned char *proof,
|
2016-07-04 13:04:57 +00:00
|
|
|
size_t plen
|
Pedersen commitments, borromean ring signatures, and ZK range proofs.
This commit adds three new cryptosystems to libsecp256k1:
Pedersen commitments are a system for making blinded commitments
to a value. Functionally they work like:
commit_b,v = H(blind_b || value_v),
except they are additively homorphic, e.g.
C(b1, v1) - C(b2, v2) = C(b1 - b2, v1 - v2) and
C(b1, v1) - C(b1, v1) = 0, etc.
The commitments themselves are EC points, serialized as 33 bytes.
In addition to the commit function this implementation includes
utility functions for verifying that a set of commitments sums
to zero, and for picking blinding factors that sum to zero.
If the blinding factors are uniformly random, pedersen commitments
have information theoretic privacy.
Borromean ring signatures are a novel efficient ring signature
construction for AND/OR admissions policies (the code here implements
an AND of ORs, each of any size). This construction requires
32 bytes of signature per pubkey used plus 32 bytes of constant
overhead. With these you can construct signatures like "Given pubkeys
A B C D E F G, the signer knows the discrete logs
satisifying (A || B) & (C || D || E) & (F || G)".
ZK range proofs allow someone to prove a pedersen commitment is in
a particular range (e.g. [0..2^64)) without revealing the specific
value. The construction here is based on the above borromean
ring signature and uses a radix-4 encoding and other optimizations
to maximize efficiency. It also supports encoding proofs with a
non-private base-10 exponent and minimum-value to allow trading
off secrecy for size and speed (or just avoiding wasting space
keeping data private that was already public due to external
constraints).
A proof for a 32-bit mantissa takes 2564 bytes, but 2048 bytes of
this can be used to communicate a private message to a receiver
who shares a secret random seed with the prover.
Also: get rid of precomputed H tables (Pieter Wuille)
2015-08-05 19:04:14 +02:00
|
|
|
) SECP256K1_ARG_NONNULL(1) SECP256K1_ARG_NONNULL(2) SECP256K1_ARG_NONNULL(3) SECP256K1_ARG_NONNULL(4) SECP256K1_ARG_NONNULL(5);
|
|
|
|
|
|
|
|
# ifdef __cplusplus
|
|
|
|
}
|
|
|
|
# endif
|
|
|
|
|
|
|
|
#endif
|