Verifiable computation for cross-domain information sharing
Techniques for verifiable computation for cross-domain information sharing are disclosed. An untrusted node in a distributed cross-domain solution (CDS) system is configured to: receive a first data item and a first cryptographic proof associated with the first data item; perform a computation on the first data item including one or more of filtering, sanitizing, or validating the first data item, to obtain a second data item; generate, using a proof-carrying data (PCD) computation, a second cryptographic proof that indicates (a) validity of the first cryptographic proof and (b) integrity of the first computation on the first data item; and transmits the second data item and the second cryptographic proof to a recipient node in the distributed CDS system. Alternatively or additionally, the untrusted node may be configured to transmit a cryptographic proof to a trusted aggregator in the CDS system.
1. A method, comprising:
receiving, by a first node in a distributed cross-domain solution (CDS) system, a first data item;
performing, by the first node, a first computation on the first data item comprising one or more of filtering, sanitizing, or validating the first data item, to obtain a second data item;
generating, by the first node, a first cryptographic proof that indicates integrity of the first computation on the first data item;
transmitting, by the first node to a trusted aggregator in the distributed CDS system, the first cryptographic proof; and
transmitting, by the first node to a first recipient node in the distributed CDS system, the second data item.
2. The method of claim 1 , wherein generating the first cryptographic proof is performed in parallel with transmitting the second data item to the first recipient node.
3. The method of claim 1 , further comprising verifying, by the trusted aggregator, integrity of the second data item based at least on the first cryptographic proof.
4. The method of claim 3 , wherein the trusted aggregator is located at a security domain boundary of the distributed CDS system, the operations further comprising:
determining, by the trusted aggregator based on verifying integrity of the second data item, whether to forward the second data item across the security domain boundary.
5. The method of claim 1 , further comprising:
performing, by the first recipient node, a second computation on the second data item comprising one or more of filtering, sanitizing, or validating the second data item, to obtain a third data item;
generating, by the first recipient node, a second cryptographic proof that indicates integrity of the second computation on the second data item;
transmitting, by the first recipient node to the trusted aggregator in the distributed CDS system, the second cryptographic proof; and
transmitting, by the first recipient node to a second recipient node in the distributed CDS system, the third data item.
6. The method of claim 5 , wherein generating the second cryptographic proof is performed in parallel with transmitting the third data item to the second recipient node.
7. The method of claim 5 , further comprising verifying, by the trusted aggregator, integrity of the third data item based at least on the second cryptographic proof.
8. The method of claim 7 , wherein the trusted aggregator is located at a security domain boundary of the distributed CDS system, the method further comprising:
determining, by the trusted aggregator based on verifying integrity of the third data item, whether to forward the third data item across the security domain boundary.
9. The method of claim 1 , wherein the first cryptographic proof is a succinct non-interactive zero-knowledge proof.
10. The method of claim 1 , the first data item comprising an image from a sensor and the first computation comprising redacting the image.
11. The method of claim 1 , the first data item comprising text and the first computation comprising filtering one or more words from the text.
12. A method, comprising:
receiving, by an first node in a distributed cross-domain solution (CDS) system, a first data item and a first cryptographic proof associated with the first data item;
performing, by the first node, a first computation on the first data item comprising one or more of filtering, sanitizing, or validating the first data item, to obtain a second data item;
generating, by the first node using a proof-carrying data (PCD) computation, a second cryptographic proof that indicates (a) validity of the first cryptographic proof and (b) integrity of the first computation on the first data item; and
transmitting, by the first node to a first recipient node in the distributed CDS system, the second data item and the second cryptographic proof.
13. The method of claim 12 , wherein the first recipient node is a trusted node at a security domain boundary of the distributed CDS system, the method further comprising:
determining, by the trusted node based on the second cryptographic proof, whether to forward the second data item across the security domain boundary.
14. The method of claim 12 , further comprising:
receiving, by the first node in the distributed CDS system, a third data item and a third cryptographic proof associated with the third data item;
determining, by the first node, that the third cryptographic proof is invalid; and
based on determining that the third cryptographic proof is invalid, refraining from transmitting the third data item to the first recipient node in the distributed CDS system.
15. The method of claim 12 , further comprising:
performing, by the first recipient node, a second computation on the second data item comprising one or more of filtering, sanitizing, or validating the second data item, to obtain a third data item;
generating, by the first recipient node using the proof-carrying data (PCD) computation, a third cryptographic proof that indicates (a) validity of the second cryptographic proof and (b) integrity of the second computation on the second data item; and
transmitting, by the first recipient node to a second recipient node in the distributed CDS system, the third data item and the third cryptographic proof.
16. The method of claim 15 , wherein the second recipient node is a trusted node at a security domain boundary of the distributed CDS system, the method further comprising:
determining, by the trusted node based on the third cryptographic proof, whether to forward the third data item across the security domain boundary.
17. The method of claim 12 , wherein the first data item comprises an image from a sensor and the first computation comprises redacting the image.
18. The method of claim 12 , wherein the first data item comprises text and the first computation comprising filtering one or more words from the text.
19. A system comprising:
a first digital device serving as a first node in a distributed cross-domain solution (CDS) system;
a second digital device serving as a first recipient node in the distributed CDS system; and
a third digital device serving as a trusted aggregator in the distributed CDS system;
the system being configured to perform operations comprising:
receiving, by the first node, a first data item;
performing, by the first node, a first computation on the first data item comprising one or more of filtering, sanitizing, or validating the first data item, to obtain a second data item;
generating, by the first node, a first cryptographic proof that indicates integrity of the first computation on the first data item;
transmitting, by the first node to the trusted aggregator, the first cryptographic proof; and
transmitting, by the first node to the first recipient node, the second data item.
20. A system comprising:
a first digital device serving as a first node in a distributed cross-domain solution (CDS) system; and
a second digital device serving as a first recipient node in the distributed CDS system;
the system being configured to perform operations comprising:
receiving, by the first node, a first data item and a first cryptographic proof associated with the first data item;
performing, by the first node, a first computation on the first data item comprising one or more of filtering, sanitizing, or validating the first data item, to obtain a second data item;
generating, by the first node using a proof-carrying data (PCD) computation, a second cryptographic proof that indicates (a) validity of the first cryptographic proof and (b) integrity of the first computation on the first data item;
transmitting, by the first node to the first recipient node, the second data item and the second cryptographic proof.