# NEC Europe (2017) ###### tags: `PoS`, `security`, `nothing-at-stake`, `long-range`, `axel` ## Summary This paper addresses the attacks to which proof-of-stake protocols are vulnerable. It also puts forward a state of the art of existing proof-of-stake protocols in relation to these attacks. Finally, it offers two solutions that induce a small loss of performance and some centralization. Nothing-at-stake attack : when nodes can mine conflicting blocks without risking their stake, they have all interest to validate all blocks, which increases the number of forks. => Solution : adding random signatures to block headers in order to detect those who sign multiple blocks at the same height. Long-range attack : the adversary aims to alter the entire history of the blockchain starting from early blocks. He acquires the private keys of older accounts which no longer have any stake at the moment, but that have accrued majority stake at previous block height h; the attacker then constructs a fork starting from block h, leveraging these accounts. => Solution : using digital signatures linked to the registered identities of the nodes. In case blocks in parallel forks are mined by the same node, the private key of that node along with his identity will be immediately revealed. Limitations : more data in blocks, and use of an identity manager ## Comments Interesting state of the art of proof-of-stake protocols. Realistic solutions and transparency regarding their flaws. Worth checking for general knowledge on proof-of-stake blockchains.