Blockchain Consensus Algorithm
This section will finish some definitions that are offered in the previous sections without much clarification, like the nonce field, bits field in Section 2.2, and PoW which is proposed for appending a new block to the chain. Further, the main focus of this section is on how nodes in the verifying network agree with each other about the ledger that they hold. This section is divided into two main subsections: proof-based consensus algorithm and vote-based consensus algorithm.
Proof-Based Consensus Algorithm
This subsection introduces the proof-based consensus algorithm. The original work is PoW, proposed by Nakamoto. To date, many variants of proof-based consensus algorithms have been proposed, which are based on PoW, PoS, their hybrid form, and other variants that are made independently from these two major ones. The basic concept of proof-based consensus algorithm is that among many nodes joining the network, the node that performs sufficient proof will get the right to append a new block to the chain, and receive the reward.
Original proof of work
As mentioned, in the Blockchain network, if every node tries to broadcast their blocks containing the verified transactions, confusion could possibly arise. For example, consider a transaction that is verified by many nodes, who will then put it into their blocks, and broadcast to other nodes. If the broadcasting work is free, this transaction could be duplicated in different blocks, then the ledger is meaningless. In order to get agreement between all nodes about the newly added block, the PoW requires each node to solve a difficult puzzle with adjusted difficulty, to get the right to append a new block to the current chain. The first node who solves the puzzle will have this right.
PoS-based consensus
The proposed PoW is supposed to be unfair: while some miners owning modern and powerful equipment could find the suitable nonce easier, others with poorer condition could find it very difficult to be the first one to find a suitable nonce. PoS is supposed to deal with this inequality. Being firstly introduced and discussed in Bitcoin forum from 2011, PoS has had some variants and research contribution to it. The basic idea of these consensus algorithms is using the stake to decide who will get the chance to mine the next block of the chain. Using stake as a proof has an advantage: anyone who owns much stake would be more trustful. He or she would not want to perform any fraudulent actions to attack the chain that contains much of his or her profits. Furthermore, using PoS would require any attackers to own at least 51% of all stakes in the network to perform a double spending attack, which is very difficult.
There are currently two popular kinds of consensus using PoS: the kind using pure stake to make consensus, and the hybrid kind, which combines PoS and PoW.
Hybrid form of PoW and PoS
Being officially introduced in a paper by King and Nadal, PPcoin could be considered as the first variant of PoS, but it still uses PoW. These authors have proposed a definition called the ‘coin age’ of each miner, which is calculated by his stake multiplied by the time that the miner has owned it. For example, Bob receives 10 coins from Alice and keeps them for 10 days; consequently, he has 100 coinday, and if he sends 2 coins to anyone, the coin age of these 2 coins accumulated will be destroyed. In order to get the right to append a new block to the chain, the miner creates a kind of block called a coin stake, which like before, holds many transactions, but includes a special one from that miner to himself. The amount of money spent on this transaction will provide the miner more chance to mine a new block. Afterwards, he will have to do a puzzle, like PoW. The more money he spends on the transaction, the easier the puzzle he has to solve. If any miner solves the puzzle first, he will get 1% of the amount of coins he has spent in the transaction, but the coin age accumulated by these coins will be reset to 0. Being dissimilar to King and Nadal, Vasin do not employ coin age with their Blackcoin, because they suppose that using coin age could provide the attacker the chance to accumulate enough value to cheat the network. Worse yet, there could be some miners who hold their stake until they have a lot of coin age, while staying offline from the verifying system. Therefore, Vasin propose using the raw stake instead of coin age for providing miners the chance to mine a new block, which could encourage more nodes to be online to get the rewards. Being noticeable with the offline miner, Ren propose using an exponential decay function with the coin age: the more the miner waits for the increase of the coin age, the less speed of increase it has.
The problem of double spending attack is again shown by Duong et al. The problem with miners owning more than 51% mining power raises a high alert for the security of Blockchain. They proposed a method for mitigating the double spending attack by combining PoW and PoS. The aim of this action is to make sure that even if a miner owns more than 51% of the mining power, he still does not have much chance to make a fraudulent action. These authors propose using the PoW first for choosing a winner, who is the first one getting the answer of the puzzle.Subsequently, this winner will not only append a block called PoW block to the chain as usual, but he will also provide a basis to choose another miner who owns stake. This basis is: if the return value of a hash function, whose input parameters are the newly appended PoW block and stake owner’s private key, is below a threshold, the chosen miner will have the chance to append a so-called PoS block to the chain. Duong et al. claim that each PoS block is linked to a single PoW block, and each PoW block is linked to a previous PoS block. Consequently, it is difficult to make a double spending attack. The reason is that assuming that an attacker could own more than 51% of the computing power, then he could get the right to be the miner appending the illegal PoW block to the chain. However, right after that, there is still a PoS block appended by another miner chosen by the mentioned base. This stake miner will see the longest chain at that time, and realize that the latest one is a PoS block, which he could not append his PoS block to. As a result, the attack fails. The double spending attack is supposed to happen only when the attacker owns not only more than 51% of the computing power, but also more than 50% of all the stake holders. This kind of consensus is not based on the amount of stake, but only for miners owning stake. As an improvement of , Chepurnoy et al. have pointed out a problem with their former research. Firstly, the executing environment is static, because at each round, the invested physical hardware for mining the resources and virtual resources known as stake remain the same; which assumption does not equate with reality. Also, the usage of stake has not been employed, which would not be fair for the stakeholders who own more stake than others. The authors suggest that there should be a difficulty adjustment for both the PoW and PoS when the environment changes. Therefore, Chepurnoy et al. propose adjusting the difficulty based on the environment—the rate of block creation. Specifically, their proposed difficulty for proof of work follows the original Bitcoin’s one, while the PoS difficulty is inverse to the stake that the stake owner has. For example, if a miner owning 1 coin has the hash value mentioned from below threshold t, then another miner owning b coin will only have to satisfy the hash value below t * b.
Bentov et al. propose a solution called Proof of Activity. This combines PoW and PoS in order to not only solve the double-spending attack, but also handle some tragedies made by PoW called the tragedies of the common. The first mentioned tragedy is that only the miners solving the puzzle get the reward, while the others who have the role of preserving the ledger, update it and validate the new block; they do not receive anything. The second one is that miners can co-operate with others to raise the transaction fee to be high to charge the users. This action could make the Blockchain useless, because nobody would want to use it. In order to deal with these tragedies, these authors proposed creating an empty block by PoW first: all the miners will try to find a nonce with a block without any transactions. After finding a suitable nonce, the miner will broadcast the block containing it to other miners, who will verify the validation of their received ones. Also, they will check if they are the winner of another lottery, which is designed based on the block they receive. This lucky chance is like followthe- Satoshi procedure in . A hash function is suggested to be used N times, which input parameters are the hash value of the received block, the hash of the latest appended block, and a random number, which is randomized N times. N miners who see that they own the N chosen Satoshi will continue the work: the (N – 1) first miners will sign into this empty block, their signature proving that they own the chosen Satoshi, then broadcast this signature. The last miner will include not only the required signature, but also as many transactions as he wants, then broadcast the block to other miners. The block creater and N chosen miners will get equal rewards. Besides preventing a double spending attack, this kind of consensus also encourages miners to stay online to collect !
