14 DAY TRIAL // In a paper released on November 10, researchers revealed a major flaw in the way Bitcoin nodes are rewarded. They also showed how new reward schemes could prevent a phenomenon that can dearly affect the currency that is already on an all-time low.
“On Bitcoin and Red Ballons” is a scientific paper in which researchers from Cornell University and Microsoft mathematically proved their findings which unveil the fact that the current incentives received by nodes for each transaction they verify, could bring out the greed in some of them.
Bitcoin relies on P2P networks to verify and authorize the transactions performed with the currency that at the moment has about 7.5 million units in circulation. So each time someone wants to make a Bitcoin transaction, before the money reaches its destination, the process is taken over by all the nodes in the network to verify the legitimacy of the transfer.
Once a node successfully authorizes a transaction, by inverting the hashes generated by the transfer, it passes the information to its adjacent neighbors. The process continues until all the nodes agree on the fact that the transaction is genuine.
For their effort, each node receives a reward fee, but since the payment becomes limited at one point, some nodes might be tempted not to pass on the information to keep the earnings for themselves.
This is where the main problem steps is. Since only one node will work on each transactions, the authorization process may take a very long time and Sybil attacks, in which a reputation system is subverted by forging identities in peer-to-peer networks, could take place.
“We stress that false identities are a prominent concern in Bitcoin. In fact, the Bitcoin protocol is built around the assumption that nodes can create false identities, and considers a transaction fully approved only once nodes that control a majority of the CPU power in the network have accepted it, rather than just a majority of the nodes. The latter is vulnerable to Sybil attacks,” the paper reads.
The researchers compare their solution to this problem to the 2009 DARPA Network Challenge in which participants had to find 10 red balloons spread all over the US. The winning team won the competition by rewarding individuals not involved in the competition.
That's basically how the solution to the Bitcoin problem works. They propose that each node that solves a transaction, will generate a reward for all the nodes who forward the transaction to others. Sybil attacks in which a single node sends itself the same transaction numerous times can be prevented by canceling rewards when the number of links in the chain reaches a certain limit.