Title: Bitcoin: A Peer-to-Peer Electronic Cash System
Conference (): Link
Journal (): Link
This paper proposes bitcoin, a peer-to-peer electronic cash system (often referred to blockchain). The system is designed to allow online payments through decentralized consensus (i.e., without a trusted third-party to verify transactions). The key problem is to address the double spending problem in p2p network.
Propose blockchain
Use chained hashing with timestamps and Proof-of-Work to solve double spending problem
Timestamp server (to address double spending): a hash is computed by combining the timestamp of a block and the block itself. The hash is broadcast to peered nodes for verification. It ensures that the block (with transactions) is enforced to be generated after the previous blocks.
Proof-of-Work: use CPU power to scan (enumerate) for a value, such that the hash of the block filled with the scanned value starts with a certain number of 0s. The verification of PoW is simple, but the enumeration process is very difficult (with a exponential complexity of the # of zeros)
Some strategy to reclaim space: pruning on the Merkle tree
The first work of blockchain-based decentralized p2p E-cash system
Known issues: expensive in computation power for the miners