Browse Prior Art Database

A System and a Method for Meta-Trust Models and its Application to Meta-Validation of Blockchain

IP.com Disclosure Number: IPCOM000248773D
Publication Date: 2017-Jan-09
Document File: 5 page(s) / 129K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a Meta-Trust model for blockchain. This trust model focuses on validating the validators in order to identify malicious validators.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 38% of the total text.

1

A System and a Method for Meta -Trust Models and its Application to Meta -Validation of Blockchain

Blockchain technology, which originated with bitcoin technology, has created a new way to build trusted distributed systems. A blockchain consists of a shared ledger that is difficult to corrupt. A shared ledger records all relevant transactions. Thus, a blockchain is a distributed database tracks a tamper-proof shared ledger. The shared ledger consists of transactions, which are recorded inside the shared ledger using a consensus algorithm. A transaction is an act of exchanging information or asset among two or more entities, persons, or processes. Since its inception, the blockchain technology has evolved to both a technology for a distributed shared ledger and a trusted model for business transactions.

A shared ledger consists of recorded transactions organized as a linked list of blocks. Each block in the linked list consists of a set of transactions that are "committed'' together. Blockchain uses various forms of consensus algorithms for committing a block to a blockchain or a shared ledger. The consensus algorithm is executed by humans, auto-agents, or processes. Ideally, the processing agents are independent of, and do not influence, each other. A block added to a blockchain is not easily tampered with.

Cryptographic methods are used to guarantee tamper proof of blockchains; therefore, recorded transactions must be correct. Consensus algorithms provide certain trust guarantees for transactions that are recorded in a blockchain. Most consensus algorithms are based on replicating state machines across a network of validators (also termed miners or verifiers). These consensus algorithms provide fault tolerance (i.e., byzantine fault tolerance) due to machine crashes, maliciousness, and other types of faults (i.e., byzantine faults). Most byzantine fault tolerance algorithms use deterministic state machines for replicating across network of validators. Given a set of good validators, with each validator starting from some initial state and then stepping through k steps, the final state is determined. If more than (N-1)/3 validators are good validators, there is agreement on the outcome of the consensus, where N is the number of validators. Bad validators are considered faulty, typically called as the byzantine fault.

Many applications and systems do not have deterministic properties for validation. For instance, when a video or a photo is shown to a set of N good human validators, not all agree on who or what the photo or video is. It is because of inherent non-determinism. Now imagine more than (N-1)/3 are colluding to change the outcome of the consensus. A method is needed to improve the trust model.

The novel contribution to knowledge is a Meta-Trust model for blockchain. This trust model focuses on validating the validators in order to identify malicious validators. The novel trust model comprises:

• Meta-validator agents focused on eva...