CN114401099A - Block chain PoW selfish mining attack defense method based on network topology - Google Patents

Abstract

The invention belongs to the technical field of blockchain, in particular to a blockchain PoW selfish mining attack defense technology based on network topology. It is characterized by: 1) calculating and updating the central betweenness of nodes in the entire network; 2) in the blockchain consensus mechanism, a branch selection algorithm based on the central betweenness is used to achieve resistance to selfish mining. Under the condition that the network topology structure cannot be changed, the present invention increases the computing power cost that selfish mining nodes need to pay for profit through the negative correlation processing with the betweenness of the network center. There is no impact on the performance of the blockchain.

Description

A blockchain PoW selfish mining attack defense method based on network topology

technical field

The invention belongs to the technical field of blockchain, in particular to a blockchain PoW selfish mining attack defense technology based on network topology.

Background technique

3.1 Partial attribute explanation

Block: Block is the basic unit of blockchain. The blockchain system will generate a block at regular intervals, which is connected with the previous block through the HASH value to form a chain data structure. Each block will include several transactions in the network since the end of the previous block (in Bitcoin, mainly Bitcoin transfer transactions).

Blockchain branch: In an existing block, two or more legal new blocks may appear in the network, and their HASH values can correspond to the existing blocks, but may be due to the release of the block. The content of several new blocks is different due to the different positions of the nodes in the network, different preferences for transaction selection, different content packaged in blocks, and different solutions of block mining. In this case, the network branches. According to the existing Bitcoin protocol, the node will give it the first legal block it receives to continue mining the next blockchain. At this point, multiple branches are formed.

The length of the branch chain can also be any positive integer (including 1, that is, there is only one block).

The longest chain principle: When there is a branch chain with the longest length in the network, when the branch chain is broadcast to the whole network, it will be recognized by all nodes. Even if other nodes are mining based on other branch chains, they will give up at this time. other branches. At the same time, all blocks on the abandoned fork chain will not be rewarded.

3.2 Blockchain Proof of Work (PoW)

Proof of Work (PoW) is currently the most influential consensus mechanism in blockchains, especially public chains. PoW calculates a value (nonce) so that the hash value of the content after the transaction data is put together satisfies a special requirement. In the blockchain, after the node tries and successfully finds a satisfactory Hash value, it will broadcast to the whole network immediately, and obtain the right to package the block and the corresponding incentives. The node of the network receives the broadcast packaged block and will immediately respond to the broadcast. its verified. At present, Bitcoin adopts the PoW mechanism, and Ethereum also started based on the PoW mechanism. Although PoW will generate a lot of power waste in the production process, in the current blockchain public chain field, PoW, especially the Bitcoin system, still has certain benchmarking significance in terms of security and reliability.

3.3 PoW computing power

According to the current PoW algorithm taking Bitcoin as an example, the generation of each block corresponds to a node that successfully finds a solution (or value nonce) that meets the requirements through calculation. In theory, each node in the network has a probability to receive the rewards generated by the blockchain. When we lengthen the running time of the blockchain, we can measure the benefits of our participation in PoW mining by evaluating the probability of each node getting block rewards.

In recent years, with the increase in the number of trading channels between digital currencies such as Bitcoin and Ethereum and legal tender, the scarcity of digital currency itself and the multiple influences of marketing promotion in the field of digital currency, the legal currency purchase price of digital currency Raised to a level that cannot be ignored. Some companies and countries even openly regard digital currency as a tool for asset allocation and value circulation.

Therefore, in the current market environment, it is a commercially profitable practice to obtain PoW computing power and invest it in the blockchain consensus mechanism to obtain incentives. PoW computing power has a very important strategic significance in the field of blockchain public chain digital currency: on the one hand, it marks the expectation of nodes with computing power to obtain incentive benefits in the blockchain, on the other hand, all participants also We are wary of PoW computing power being too concentrated, which will lead to the failure of the distributed properties of the blockchain ledger.

3.4 Selfish mining attack technology

In the case that the entire network nodes are working honestly, the proportion of rewards obtained by each node corresponds to the proportion of computing power provided by the nodes fairly. The attack technology of the blockchain can allow some nodes to obtain excess income, and the income of other nodes who work honestly will resist the proportion of their computing power.

The premise of the establishment of selfish mining is related to the consensus mechanism that the blockchain adopts the longest chain principle, and before people discovered the selfish mining technology [1], the 51% computing power attack was recognized as the only security faced by the PoW blockchain. sexual challenges. Therefore, before introducing selfish mining, briefly introduce the longest chain principle and the "51%" attack.

3.4.1 The longest chain principle

Because in the display network, there is a certain delay in the transmission of information between nodes. Therefore, there is a certain probability that when a node A successfully mines and broadcasts a block in a certain block, when the distant node B in the network has not heard the broadcast node, it also successfully mines a block. At this time, the nodes that are closer to node A in the entire network will recognize the blockchain of node A and continue to mine new blockchains based on the priority of hearing the broadcast content of node A. The same is true for those who hear the blockchain of node B first. At this time, a fork will appear in the network in a short time, and we will divide it into sub-chain A and sub-chain B. At this time, if a block chain is successfully dug out based on sub-chain A, and broadcast to all nodes in the network before sub-chain B fails to dig out a new block, the whole network nodes will give up the B chain and re-enter the chain. The longest A chain starts the calculation of the next block (as shown in Figure 1).

In general, in the blockchain ledger under the PoW longest chain principle, there is no absolutely correct concept of ledger, only all nodes in the network will eventually recognize the result of which ledger. Under this system, a distributed ledger such as Bitcoin can guarantee the security of the ledger with a very high probability.

3.4.2 "51%" Hashrate Attack

as shown in picture 2

When the computing power of the blockchain in the network is too concentrated, when the node that masters the centralized computing power wants to do evil, it is feasible for it to initiate the tampering of the ledger.

According to the longest chain principle, if there is a node in the network that has more than 50% of the computing power of the entire network, it means that if we lengthen the observation time, most of the blockchains in the network will be mined by this node. Taking advantage of this feature, nodes with more than 50% computing power can complete a "double spend" attack.

Assuming that in the main chain, the i-th blockchain records that node A spent a fee and exchanged for the corresponding service or commodity, but from this moment onwards, node A will use all its computing power to start from the i-1 Starting from a block, ignore the main chain (the chain recognized by other nodes) and recalculate the PoW computing power, and generate a new block. The block does not include the cost spent by A. Since the computing power of node A is greater than that of other nodes in the entire network, as long as the time is long enough, node A will be able to obtain a chain that is longer than the "main chain" mined by other nodes through PoW computing power. When this moment comes , node A publishes the longer chain to the network, then according to the longest chain principle, all chains will abandon the original "main chain", and the branch made by node A becomes the new main chain. The block that previously recorded the call fee of A was also discarded along with the original "main chain" and was invalidated together. At this time, the call fee of A in the original i-th blockchain will not be recorded in the ledger. Node A completes a double-spend attack.

3.4.3 Profitability principle of selfish mining

Selfish mining is a principle of using the longest chain of PoW computing power. It is mainly based on hiding the legal blocks that it has dug, and broadcasting the block through a series of delays to optimize its own income and exceed its own computing power. The power accounts for the proportion of the entire network (on the other hand, it also violates the benefits that honest nodes can obtain).

The emergence of selfish mining methods enables nodes with larger computing power (even if it does not exceed 50% of the total network computing power) to obtain a revenue ratio greater than the computing power ratio.

A typical selfish mining principle is as follows:

Assuming that selfish node A (or a "mining pool" node composed of multiple joint execution nodes), the proportion of its computing power to the computing power of the entire network is α. If node A successfully mines a block, it will keep the sequence block information secret, and continue to mine the next block chain based on this block. Then there must be a certain probability that the number of blockchains mined by node A is ahead of the number of remaining nodes in the network. As shown in Figure 3, it is assumed that node A can mine 3 blockchains in a row with a certain probability. At this point, node A hides 3 blocks from publishing, and other nodes continue to work along another branch. Since A can receive the block information announced by other nodes, when the number of blockchains mined by other nodes is only one behind, then node A announces the three previously hidden blocks to the whole network, at this time according to the longest chain In principle, two blockchains processed by other honest nodes will be invalidated.

In this process, the dishonest behavior of node A wastes the computing power of honest nodes, so that the expectation of node A's income is greater than the fair ratio, and a selfish mining attack is completed.

For a detailed introduction to selfish mining, please refer to [1], which points out that as long as the computing power α possessed by node A satisfies:

, selfish mining can obtain additional income. Among them, γ is affected by the network topology. If the proportion of nodes in the network that node A can broadcast more quickly, the greater the value of γ.

3.5 The central betweenness of the network

Central betweenness is a measure that characterizes the importance of a node in the network. Specifically, central betweenness is a measure of graph centrality based on the shortest path. For each pair of nodes in a connected graph, there is at least one shortest path between the two nodes that minimizes the number of edges (for unweighted graphs) or the sum of edge weights (for weighted graphs) that the path passes through. The central betweenness of a node is the number of shortest paths passing through the node or the ratio of the number of shortest paths to the number of all paths.

3.5.1 The relationship between the central betweenness of the network and selfish mining

The greater the betweenness of the network center of the selfish mining node, the greater the profit of selfish mining

The relationship between the betweenness of the network center and selfish mining mainly works in a selfish mining scenario described below (different from the example when the principle is described in 3.4.3)

As shown in Figure 4, the selfish mining node is located in a position with a high central betweenness, and all nodes in the dotted box find the shortest path outside the box through the selfish mining node. This means that selfish mining nodes have the ability to broadcast information to nodes inside the frame before nodes outside the frame.

①Assume that the selfish mining node is only one block ahead of the main chain, and it is hidden and unpublished.

②Other honest nodes (assuming node A) may dig a new legal block before the selfish node digs out a new block, forming a branch chain j, resulting in the selfish mining node not leading the main chain and unable to directly apply the longest chain The principle makes the branch j invalid.

③ At this time, after receiving the branch chain j broadcasted by node A, the selfish mining node immediately broadcasts the existence of branch chain i to the outside world. At this time, due to the network topology position relationship of the selfish mining nodes, the nodes in the dotted box will receive the legal branch i and continue mining on it. At this point, the selfish mining node has achieved the goal of coordinating more computing power to serve its fork chain, and implementing this strategy for a long time will make the selfish mining node's revenue expectation greater than that of the honest node. (Under this strategy, on the one hand, it can increase the probability of continuing to hide and lead the two blockchains in the main chain. On the other hand, even if it is temporarily caught up by the honest nodes, it can coerce some honest nodes to continue mining on the branch i , the probability of hiding the only block to be recognized by yourself will also be greater than the proportion of your own computing power)

To sum up, the profit of selfish mining nodes has a positive relationship with the proportion of nodes it can affect. Therefore, the inventive task of the present invention is to resist selfish mining reasonably and effectively based on the betweenness of the network center.

references

[1] Eyal, Ittay, and Emin Gün Sirer. "Majority is not enough: Bitcoinmining is vulnerable." International conference on financial cryptography and data security. Springer, Berlin, Heidelberg, 2014.

Brandes, Ulrik. "A faster algorithm for betweenness centrality." Journal of mathematical sociology 25.2(2001):163-177.

SUMMARY OF THE INVENTION

The purpose of the present invention is to disclose a method for resisting selfish mining based on the betweenness of the network center. Based on the central betweenness of the network topology, the present invention reduces the benefit of the blockchain PoW selfish mining attack and increases the cost of the blockchain selfish mining attack.

Technical solutions

A blockchain PoW selfish mining attack defense method based on network topology is characterized by including:

1) Calculate and update the central betweenness of nodes in the entire network

In addition to the original blockchain network, a calculation module for calculating the central betweenness of all nodes is added, and the calculation result is passed to all nodes in the blockchain network as the input parameters of the module for resisting selfish attacks;

The calculation module is re-updated at intervals to calculate the central betweenness of the network; during the interval, the value of the last update is used; the calculation result output by the calculation module is the central betweenness of each node in the network, represented by CB(vi) , where vi∈V, V is the set of all nodes in the blockchain network;

2) In the blockchain consensus mechanism, the branch selection algorithm based on center betweenness is used to achieve resistance to selfish mining

The anti-selfish attack module introduces the central betweenness. When the node receives two legal fork chains of the same length, it randomly selects one of them to continue mining. The algorithm process includes the following steps:

2.1) Two nodes in the network broadcast their legal branches to the whole network at close times, and the branch lengths are the same;

2.2) Due to the transmission time difference between the two branches and the transmission time difference caused by the network topology, node 1 receives the two branches in a sequence, assuming that the branch chain information of node 2 and node 3 are broadcast to node 1 first, at this time , Node 1 performs mining based on the branch of Node 2 received first according to the original PoW longest chain principle;

2.3) Node 1 has not mined new block money based on the branch chain of node 2, and received a branch chain from node 3, and the length of the branch chain is the same as the length of the branch chain received from node 2 in step 2.2), At this time, when node 1 stops the current mining behavior, and randomly selects a branch chain again according to the following probability:

That is, the node performs a random operation, the probability of P _v2 will select the node 2 branch, and the probability of P _v3 will select the node 3 branch;

If a node faces more than two branch choices at the same time, the general choice probability expression is as follows:

P _vi : the probability of choosing a branch of node i

C _B (vi ): the network center betweenness of node _i

2.4) Based on the selection result in 2.3), select the corresponding branch for further mining;

2.5) If there is a new legal block in the process of step 2.3) to 2.4), it will still switch to the legal longest chain for mining according to the longest chain principle, so as to ensure that the overall network performance of the blockchain remains unchanged.

beneficial effect

1) Under the condition that the network topology cannot be changed, through the negative correlation processing with the betweenness of the network center, the computing power cost that selfish mining nodes need to pay for profit is increased.

2) Since each node immediately reselects the branch according to the probability, it has no effect on the performance of the blockchain.

Description of drawings

Figure 1 The longest chain principle

Figure 2 "51%" Hashrate Attack

Figure 3 Selfish mining

Figure 4 Selfish mining and network center betweenness

The output of the central betweenness calculation module of the blockchain network shown in Figure 5

Figure 6 Regular update of betweenness calculation in the center of blockchain network

Fig.7 Branch selection algorithm based on center betweenness

Figure 8. Branch selection interruption algorithm based on center betweenness

Detailed ways

1) Calculate and update the central betweenness of nodes in the entire network.

As shown in Figure 5, the present invention adds a calculation module for calculating the central betweenness of all nodes outside the original blockchain network, and transmits the calculation results to all nodes in the blockchain network as a module for resisting selfish attacks input parameters.

Since the topology of the network changes relatively slowly, the calculation module can be re-updated at intervals to calculate the betweenness of the network center (as shown in Figure 6). During the interval, the last updated value is used. The calculation result output by the calculation module is the central betweenness of each node in the network, which is represented by C _B (vi ), where vi _{∈ V} , and V is the set of all nodes in the blockchain network.

2) In the blockchain consensus mechanism, the branch selection algorithm based on center betweenness realizes resistance to selfish mining.

The anti-selfish attack module introduces the center betweenness. When a node receives two legal forks of the same length, it randomly reselects one of them to continue mining; the larger the center betweenness of the branch, the smaller the probability of being randomly reselected. .

As shown in Figure 7, the algorithm of resisting selfish attack module includes the following steps:

2.1) Two nodes in the network broadcast their legal branches to the entire network at relatively close times, and the branch lengths are the same (in most cases, the branch length is 1, that is, a blockchain).

2.2) Due to the transmission time difference between the two branches and the transmission time difference caused by the network topology, the node 1 receives the two branches in a sequence. We assume that the branch chain information of node 2 and node 3 are first broadcast to node 1. At this time, node 1 performs mining based on the branch of node 2 received first according to the original PoW longest chain principle. (It is assumed here that any branch chain of node 2 or node 3 can take precedence, which has no effect on the algorithm)

2.3) Node 1 has not mined new block money based on the branch chain of node 2, and received a branch chain from node 3, and the length of the branch chain is the same as the length of the branch chain received from node 2 in step 2.2), At this time, when node 1 stops the current mining behavior, and randomly selects a branch chain again according to the following probability:

That is, the node performs a random operation. The probability of P _v2 will select the node 2 branch, and the probability of P _v3 will select the node 3 branch.

If a node faces more than two branch choices at the same time, the general choice probability expression is as follows:

P _vi : the probability of choosing a branch of node i

C _B (vi ): the network center betweenness of node _i

2.4) Based on the selection result in 2.3), select the corresponding branch for further mining.

2.5) As shown in Figure 8, it is a special case in the branch selection algorithm, which makes the branch selection interrupted, and the node returns to the longest legal branch chain for mining. Specifically: if a new legal block appears in the process of steps 2.3) to 2.4), it will still switch to the legal longest chain for mining according to the longest chain principle. This ensures that the overall network performance of the blockchain remains unchanged.

Claims (1)

1. A block chain PoW selfish mining attack defense method based on network topology is characterized in that, comprising: 1) Calculate and update the central betweenness of nodes in the entire network In addition to the original blockchain network, a calculation module for calculating the central betweenness of all nodes is added, and the calculation result is passed to all nodes in the blockchain network as the input parameters of the module for resisting selfish attacks; The calculation module is re-updated at intervals to calculate the central betweenness of the network; during the interval, the value of the last update is adopted; the calculation result output by the calculation module is the central betweenness of each node in the network, using C _B (vi _i ), where v _i ∈ V, V is the set of all nodes in the blockchain network; 2) In the blockchain consensus mechanism, the branch selection algorithm based on center betweenness is used to achieve resistance to selfish mining The anti-selfish attack module introduces the central betweenness. When the node receives two legal fork chains of the same length, it randomly selects one of them to continue mining. The algorithm process includes the following steps: 2.1) Two nodes in the network broadcast their legal branches to the whole network at close times, and the branch lengths are the same; 2.2) Due to the transmission time difference between the two branches and the transmission time difference caused by the network topology, node 1 receives the two branches in a sequence, assuming that the branch chain information of node 2 and node 3 are broadcast to node 1 first, at this time , Node 1 performs mining based on the branch of Node 2 received first according to the original PoW longest chain principle; 2.3) Node 1 has not mined new block money based on the branch chain of node 2, and received a branch chain from node 3, and the length of the branch chain is the same as the length of the branch chain received from node 2 in step 2.2), At this time, when node 1 stops the current mining behavior, and randomly selects a branch chain again according to the following probability:

That is, the node performs a random operation, the probability of P _v2 will select the node 2 branch, and the probability of P _v3 will select the node 3 branch; If a node faces more than two branch choices at the same time, the general choice probability expression is as follows:

P _vi : the probability of choosing a branch of node i C _B (vi ): the network center betweenness of node _i 2.4) Based on the selection result in 2.3), select the corresponding branch for further mining; 2.5) If there is a new legal block in the process of step 2.3) to 2.4), it will still switch to the legal longest chain for mining according to the longest chain principle, so as to ensure that the overall network performance of the blockchain remains unchanged.

Images