Internship Topic: Security Analysis of Blockchain Systems using Artificial Intelligence

Context

The world of blockchains is in turmoil and many technologies are emerging on a regular basis. The different consensus protocols (proof-of-work, proof-of-stake, PBFT, etc.) are all alternatives and combinations that must be understood, sorted and mastered in order to direct decision-making and ensure the adequacy between the need and the available technology.

In addition, given the lack of feedback that we have on blockchain technology (due to the youth of the technology) the dependability of blockchain applications has yet to be justified, which is a research subject in itself.

In this internship subject we focus on the Hyperledger Fabric blockchain. Hyperledger Fabric is a permissionned blockchain infrastructure, providing a modular architecture with role-delineation between infrastructure nodes, execution of smart contracts (called chaincode in Fabric), and configurable consensus and membership services.

Objective

Blockchain and smart contract based solutions can facilitate secure exchanges within a critical application. To do this, it is proposed to provide a simulation model to assist in the analysis, development and implementation of safe protocols and smart contracts.

The objective of this internship is to develop an agent-based model to analyze by simulation the operational safety from the angle of the availability of the consensus protocol of the Hyperledger blockchain using MAX (an agent-based simulator developed by LICIA using Java). To this end, various hypotheses will be tested in order to build a solid design for applications that use Hyperledger Fabric.