CN114301918B - Efficient blockchain consensus method based on disqualified node elimination tree - Google Patents
Efficient blockchain consensus method based on disqualified node elimination tree Download PDFInfo
- Publication number
- CN114301918B CN114301918B CN202111635215.7A CN202111635215A CN114301918B CN 114301918 B CN114301918 B CN 114301918B CN 202111635215 A CN202111635215 A CN 202111635215A CN 114301918 B CN114301918 B CN 114301918B
- Authority
- CN
- China
- Prior art keywords
- node
- nodes
- state
- consensus
- stage
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Multi Processors (AREA)
Abstract
The invention discloses a high-efficiency block chain consensus method based on a disqualified node elimination tree, which relates to a single-chain block chain structure, improves the attack cost of an attacker, and ensures the correctness of all newly generated blocks within a reasonable fault tolerance range. The invention provides a logic structure of a disqualified node elimination tree, which uses a mole voting method and a Bayesian fault-tolerant principle. The method of staged election from bottom to top is adopted, the bad and faulty nodes are phased out, the nodes entering the next layer of election are determined, and in the process, all the nodes are divided into different levels through the level and are constructed into a tree structure. After the consensus result is obtained, all sub-nodes corresponding to the root node are updated from top to bottom, and the distributed account book of all the nodes is updated in a recursion mode, so that the speed of generating the block and the safety and reliability of the block chain are effectively improved.
Description
Technical Field
The invention relates to a block chain consensus method. The principles of mole voting and Bayesian fault tolerance are used to ensure the correctness of all new blocks. And (3) adopting a bottom-up staged grouping election mutual trust mechanism to phase out the disuse and fault nodes, and constructing a disuse node elimination tree structure. And a distributed synchronization mechanism updated from top to bottom is adopted to recursively update the distributed account book of all nodes, so that the speed of generating the block and the safety and reliability of the block chain are effectively improved.
Background
Blockchain technology originated in 2009, the overall development path of which can be divided into three stages:
blockchain 1.0 phase: a typical representation is bitcoin. The bitcoin solves the "centralization" problem existing in digital money transactions, so that digital money can be free from third party institutions to conduct point-to-point transactions. It uses blockchain technology as the underlying implementation, where the more important concept is the Proof of Work (POW). This technology is now one of the mainstream consensus mechanisms for cryptocurrency, which guarantees the security of bitcoin-unless half of the total computing power is mastered, transaction histories on the blockchain are almost impossible to maliciously modify.
Blockchain 2.0 phase: a typical representative is Ethernet (ETH), which innovatively implements a smart contract technology, running smart contracts on a blockchain network so that the entire network can be executed by the smart contracts. The Ethernet virtual machine has proved to be complete in design, greatly improves the popularity of the blockchain technology and expands the application scene thereof, and a user can freely design and develop self-centering application on the Ethernet. This stage blockchain technology breaks away from the limitation that it can only be used for point-to-point transactions, and its application range extends to most financial fields.
Blockchain 3.0 phase: this is also the stage that current blockchain technology is experiencing. The method is mainly characterized in that the block chain technology is applied to other fields of the financial industry, and relates to exploration and development of application in aspects of finance, economy, science and technology and even government affairs. But the current stage has more stringent requirements on the ability of the blockchain to protect data privacy, manifested by a need for a more advanced step in its speed and security.
Blockchains belong to distributed systems, and one core problem that exists in distributed systems is that of maintaining data consistency by the system. In a blockchain transaction system, it can be described as: a block identified by most nodes is determined to be added to the distributed ledger. To solve this problem, the present invention classifies the existing methods into four categories: one is the minority of the subject majority methods represented by PBFT, which reduces the complexity of solving the problem from exponential level to polynomial level for the first time, but because of its complexity of O (n 2 ) And thus is difficult to apply to a system having a large number of nodes. The other is an election proxy method represented by a Raft method, wherein the Raft method decomposes a consensus stage into a plurality of subtasks to finish, three roles of a Leader (Leader), a Follower (Follower) and a Candidate (Candidate) are set, and the Leader is electedThe leader performs billing. The Raft method can accommodate 1/2 of the existence of a failed node, but cannot accommodate any offending node, whose temporal complexity can be approximated as O (n). And a method using the principle of cryptography represented by POW and converging in accordance with probability. POW eventually tends to 0 by constructing a fully fair game environment as much as possible for the probability of error occurrence, but POW causes a large consumption of computational resources. The last one is the other methods of the three types, such as RPCA method, and adopts a unique consensus mechanism, thereby improving the consensus efficiency and safety.
While the 99% fault tolerant consensus mechanism recently appears, only at the theoretical level, due to the time complexity of O (n 3 ) And the method has complex flow, so that the method is difficult to be practically applied. The Tetris method published in the YEE project is an efficient method with final consistency, but is essentially a PBFT-like method with a large throughput, but with unsatisfactory time complexity. The Snowflake to Avalanche method has the characteristics of simplicity, high efficiency and safety, but the time complexity is O (n log n) level. These methods are only minor improvements to part of the procedure of the four methods above.
Current blockchain systems commonly employ tree structures that allow forking, and users can trust either chain as the backbone. The subsequent blockchain will automatically eliminate and get a backbone based on the number of its slave chains. This causes uncertainty in the newly generated blocks, and the computational effort to generate the obsolete blocks is equivalent to being consumed. Since the main strategy that users deal with this problem is to wait for six subsequent blocks to be generated by the backbone before validating the transaction, this mechanism also causes a delay in the transaction.
Although the above-mentioned blockchain consensus method guarantees the efficiency and the safety of the system to a certain extent, the blockchain technology is still difficult to be widely applied due to the fact that the chain forming speed is too slow, a series of factors causing transaction delay exist, and the situation that consistency is not strong possibly exists in the system. Meanwhile, the current consensus method has poor popularization. The limitation of consensus methods is the impediment to the development of the blockchain 3.0 phase.
Disclosure of Invention
The invention aims to solve the technical problems that the cost of generating blocks is overlarge, the speed of generating blocks is slower and the nodes are difficult to dislike and fail by the existing consensus method. The invention provides a logic structure of a disqualified node elimination tree, which uses a mole voting method and a Bayesian fault-tolerant principle. The method of staged election from bottom to top is adopted, the bad and faulty nodes are phased out, the nodes entering the next layer of election are determined, and in the process, all the nodes are divided into different levels through the level and are constructed into a tree structure. After the consensus result is obtained, the root node is updated from top to bottom to correspond to all the child nodes, and the distributed ledger of all the nodes is updated in a recursion mode.
The invention relates to a high-efficiency block chain consensus method, which effectively improves the speed and reliability of generating blocks by adopting a disuse node elimination tree method. The method process may be described as a transition of automaton state. The automaton state is a binary truth set of (ready, end). The 00 state of the automaton indicates a preparation phase, 10 indicates an entry into the consensus method phase, and 11 indicates an end phase of the consensus method. The node before entering the consensus method is in the 00 state, then all proposed nodes enter the 10 state through the ready phase (ready phase), and then start the "block chain consensus method phase based on disuse node elimination tree". After the final consensus block is obtained, all nodes go through the synchronization phase (synchronization phase) to enter the 11 state, and then the nodes entering the 11 state are returned to the 00 state through initialization (init phase). Only the node in the 00 state can enter the consensus next, if the node is in the 10 state when the next consensus method starts (the last consensus method must end because the next consensus method has started, and the node must not synchronize its distributed ledger), it is stated that the node needs to participate in the consensus method through two stages of synchronization and init, so as to ensure the correctness of the method.
Drawings
FIG. 1 is a state transition diagram of the present invention
FIG. 2 is a schematic diagram of a disqualified node-based elimination tree according to the invention
FIG. 3 is a graph showing the actual time consumption of the present invention compared with the PBFT and Raft methods
FIG. 4 is a diagram showing the actual time consumption of the present invention
FIG. 5 is a theoretical complexity comparison of the invention with the DumboBFT method
FIG. 6 shows the actual test results of the security values under different probability of aversion to the present invention
Detailed Description
The invention is further described in detail below with reference to the accompanying drawings, and the invention adopts a single-chain block chain structure to improve the attack cost of an attacker and ensure the correctness of all newly generated blocks within a reasonable fault tolerance range. The invention provides a logic structure of a disqualified node elimination tree, which uses a mole voting method and a Bayesian fault-tolerant principle. The method of staged election from bottom to top is adopted, the bad and faulty nodes are phased out, the nodes entering the next layer of election are determined, and in the process, all the nodes are divided into different levels through the level and are constructed into a tree structure. After the consensus result is obtained, the root node is updated from top to bottom to correspond to all the child nodes, and the distributed ledger of all the nodes is updated in a recursion mode. Each node comprises the following structure:
(1) book distributed ledger. Each node maintains a blockchain distributed ledger that records synchronized information.
(2) mail mailboxes. And receiving information sent by a user, wherein the information accords with a certain format.
(3) cur proposes data. The node gathers information from the mailbox and gathers the information to obtain a block structure.
(4) And an encoding value. The binary logic 1 and 0 respectively represent that the synchronous distributed ledger is completed and not completed, and the initial value is 0.
(5) grade value. When a consensus method starts, the grade value will be initially set to 0.
(6) ready value. The binary logic 1 and 0 respectively indicate that the proposal is ready and not ready, and the initial value is 0.
(7) fa upper node pointer. The goal is to mark a node whose grade value is greater than this node as the reference node for the final synchronized ledger. In contrast, this node is referred to as a lower node to this upper node. Initially a null value.
As in FIG. 1, the method process may be described as a transition to an automaton state. The automaton state is a binary truth set of (ready, end).
The 00 state of the automaton indicates a preparation phase, 10 indicates an entry into the consensus method phase, and 11 indicates an end phase of the consensus method. Because of the particularity of the invention and the fact that the system belongs to a distributed system, after the consensus result is obtained, the node state is possibly in any one of three states when the next consensus method starts due to network delay or speed difference among nodes in the synchronization process.
Description of the flow: the node before entering the consensus method is in the 00 state, then all proposed nodes enter the 10 state through the ready phase (ready phase), and then start the "block chain consensus method phase based on disuse node elimination tree". After the final consensus block is obtained, all nodes go through the synchronization phase (synchronization phase) to enter the 11 state, and then the nodes entering the 11 state are returned to the 00 state through initialization (init phase). Only the node in the 00 state can enter the consensus next, if the node is in the 10 state when the next consensus method starts (the last consensus method must end because the next consensus method has started, and the node must not synchronize its distributed ledger), it is stated that the node needs to participate in the consensus method through two stages of synchronization and init, so as to ensure the correctness of the method.
In the ready phase of a round of consensus method, if the state (ready) of the node is in the 00 state, entering the ready phase. In this state, all nodes prepare their own proposals, and after the proposal is ready, the state is changed to 10. In the synchronization stage of a round of method, the node of the highest level finally selected is set as the final upper node. The highest level node first transmits its upper level node (i.e., itself) to its lower level node as its final upper level node, then the nodes respectively transmit their respective upper level nodes to their child nodes as their final upper level nodes of the child nodes of the nodes, and so on until the final upper level nodes of all nodes are determined. All nodes which have determined the final upper node modify and synchronize the respective distributed ledgers according to the information of the upper node, and after the completion, the state of the nodes is modified to 11. In the init phase of a round of method, if the node state is in the 11 state, the node state is modified to be the 00 state. The node qualifies for entry into the next round of consensus method. In the group election mutual trust stage, nodes with the same grade value are limited to be combined into a group, and a common identification process is carried out in the group, wherein the process is as follows: each intra-group node will pass its own proposal to the other three nodes. After verification by the other three nodes, the node that considers the proposal of this node to be correct will return a confirmation signal. When the node receives 2 or more acknowledgement signals, the node obtains the opportunity of competing for participation in the consensus of the next stage, and the node is called a candidate node at this time. Non-candidate nodes in this stage are eliminated. Of all candidate nodes, only 2 of them are eliminated. All the obsolete nodes are deprived of the right to continue to join in groups. While all nodes that are not eliminated are called winning nodes, their grad value is increased by 1. They continue to participate in the process of joining into a union. All nodes eliminated in a consensus process must find one node in the group as its upper node. If there is an insufficient number of final level nodes, such as only 3 remaining nodes, which cannot be combined into a group, the remaining nodes should select a selected node as its upper level node. The node needs to send its own number information to its upper node, so that the next stage finishes the synchronous distributed account book stage from top to bottom, and the node sending the information is called the lower node of the upper node. The operation is performed according to the rule until all nodes can no longer be combined into groups, and the stage is finished. In the final consensus phase, the following two cases will finally occur in this distributed system, 1. Only the node with the highest grade value remains: at this time, all nodes form a tree structure according to the relationship between the upper node and the lower node. 2. Only the nodes with the highest 2 or 3 grad values remain. One node is randomly generated as a root node, and the other two nodes set the root node as an upper node, so that a tree structure is formed.
The blue curve is used for representing the pointing relation of the upper node, the black node and the white node respectively represent honest nodes and disqualified (or fault) nodes, and the stage process diagram of the block chain consensus method based on disqualified nodes eliminating trees is shown in fig. 2.
To verify the performance of an efficient blockchain consensus method of the present invention that is used as a disjunction tree method, a multithreading programming method is used, where the time consumption can be a measure of the communication cost since the multithreading is actually concurrent. The PBFT method and the Raft method in an analog distributed system, which are respectively of complexity O (n 2 ) And O (n) communication complexity, and testing its different time consumption with increasing number of nodes instead of communication cost consumption. Because the PBFT method is difficult to apply because of overlarge complexity when more than 1000 nodes, when the PBFT method is used for 0-800 nodes, the time consumption of the three obtained by test is compared with that of figure 3, for example, when the data size is 0-800 nodes, the time consumption of the method is less than 20ms, the method is represented as a straight line close to a coordinate axis in the figure, the method can be obtained through comparison, and under the condition of the same number of the nodes in the data range, the communication consumption of the method is obviously superior to that of the PBFT method and the Raft method.
To reflect the actual communication complexity of the method, we vary the abscissa node number value by an exponential power of 2 as in fig. 4, and calculate its actual time consumption, the visible line graph grows exponentially. This also illustrates the O (n) level of linearity of the communication complexity of the method. Considering the time complexity of the actual operation of the method, since the intra-group consensus inside each stage is completely parallel, the time complexity is actually the number of rounds of consensus, and the number of rounds is in the order of O (log n). Comparing with the internationally best performing practical bayer fault tolerance method DumboBFT, as shown in fig. 5, it can be seen that in theory the time complexity of this method is the same as it, while the communication complexity is much smaller than this method. Considering network delays, etc., it is anticipated that the actual performance of this approach will be higher than the DumboBFT approach.
According to the maximum-minimum principle, we discuss the reference criteria for fault tolerance as: the outside world must launch a minimum cost for a successful attack on this system.
In the bitcoin blockchain system using the POW method, there is a risk of "51% strength attack". In addition, because the self-private mining attack allows bifurcation, only an attacker is required to control 1/3 of the total number of nodes at minimum. This makes it safe to be hit with a large impact.
In the method, the angle of view is converted into the angle of an attacker, which means that only 75% of nodes are completely controlled, and a successful attack can be launched. Meanwhile, due to the adoption of a single-chain structure, the risk of 'selfish mining attack' does not exist, and the safety of the mining machine is greatly improved. The probability that the node is a disqualified node is given by a fixed probability value p, honest and disqualified situations of all the nodes are randomly simulated by using a Monte Carlo method, data of 100 groups of nodes are generated, and the ratio of the total number of times that the groups of data successfully resist attack to the total simulation number is recorded as a safety value. According to the simulation, the actual test results of calculating the safety value thereof under different probability of aversion are shown in fig. 6.
Claims (1)
1. An efficient blockchain consensus method for use as a disjunction tree elimination method, characterized by: the method adopts a single-chain block chain structure, improves the attack cost of an attacker, ensures the correctness of all newly generated blocks in a more reasonable fault tolerance range, adopts a logic structure of a disuse node elimination tree, adopts a mole voting method and a Bayesian fault tolerance principle, adopts a bottom-up staged election method, gradually eliminates disuse and fault nodes and determines nodes entering the next layer election, divides all nodes into different levels through the levels in the process and constructs the nodes into a tree structure, and after a consensus result is obtained, updates the root node corresponding to all the child nodes from top to bottom, recursively updates the distributed ledger of all the nodes, wherein each node comprises the following structures:
(1) each node maintains a blockchain distributed ledger which is used for recording information after synchronization;
(2) a mail mailbox receives information sent by a user, and the information accords with a certain format;
(3) the cur proposal data is a block structure obtained by collecting information from a mailbox by a node and summarizing;
(4) the encoding value, the binary logic 1 and 0 respectively represent the completed and unfinished synchronous distributed account book, and the initial value is 0;
(5) a grade value, which is initialized to 0 when a first consensus method starts;
(6) a ready value, wherein two-value logic 1 and 0 respectively represent that proposal is ready and not ready, and the initial value is 0;
(7) fa superordinate node pointer, aimed at marking a node whose grade value is greater than that of this node as the reference node of the final synchronous ledger, and opposite this node, called the subordinate node to this superordinate node, initially null,
the method process is described as the transition of automaton state, automaton state is (ready, end) binary truth group;
the 00 state of the automaton represents a preparation stage, 10 represents a stage of entering a consensus method, 11 represents a stage of ending the consensus method, and after a consensus result is obtained, the node state is possibly in any one of three states when the next consensus method starts due to network delay or speed difference among nodes in a synchronization process;
description of the flow: the node before entering the consensus method is in 00 state, then all the nodes put forward enter 10 state through a preparation stage, namely a ready stage, then start to "reject the blockchain consensus method stage of the tree based on the disuse node", after obtaining the final consensus block, all the nodes go through a synchronization stage, namely a synchronization stage, enter 11 state, then return the node entering 11 state to 00 state through initialization, namely an init stage, only the node in 00 state can enter consensus next, if the next consensus method starts, the node is in 10 state, since the next consensus method has already started, the last consensus method must have ended, at this time, the node must not synchronize own distributed account book, it is stated that the node can participate in the consensus method only through the synchronization stage and the init stage, thereby guaranteeing the correctness of the method;
in the ready phase of a round of consensus method, if the state of a node is in 00 state, the ready phase is entered, all nodes are ready for own proposal, after the proposal is ready, the state is modified to 10 state, in the synchronization phase of a round of method, the final highest-level node is set as self, the highest-level node firstly transmits the final upper-level node of the highest-level node, namely, the node itself is used as the final upper-level node of the lower-level node, then the nodes respectively transmit the final upper-level node of the child nodes of the node to the child nodes, and so on, until the final upper-level node of all nodes is determined, all nodes of the final upper-level node are modified according to the information of the upper-level node and the distributed account book of the node is synchronized, after the proposal is completed, the state of the node is modified to 11 state, if the node state is in 11 state, the node state is modified to 00 state, the node obtains the final upper-level node of the node as the final upper-level node of the child node, the node is used as the final upper-level node of the child node, and the node is used as the final upper-level node of the child node of the node, and the node is mutually-shared node, and the node group has the same mutual qualification value, and the mutual consensus process is restricted in the following group 4, and the mutual consensus process is carried out in the following steps: each node in the group transmits own proposal to other three nodes, after the other three nodes are verified, the node which considers that the proposal of the node is correct returns a confirmation signal, when the node receives more than or equal to 2 confirmation signals, the node obtains the opportunity of competing for participation in the consensus of the next stage, the node is called candidate node, non-candidate nodes in the stage are eliminated, only 2 of the candidate nodes are eliminated, all eliminated nodes are deprived of the right of continuing to be combined into groups, all the nodes which are not eliminated are called selected nodes, the grade value of the nodes is increased by 1, and the nodes continue to participate in the combination process, all the nodes eliminated in the primary consensus process must find one node in the group as its upper node, if there is a situation that the number of the last level nodes is insufficient, and the last level nodes cannot be combined into one group, then the remaining nodes should select one selected node as its upper node, the node needs to send own number information to its upper node, so as to complete the synchronous distributed account book stage from top to bottom in the next stage, the node sending the information is called the lower node of the upper node, and operates according to the above rule until all the nodes can not be combined into groups, and this stage ends, in the final consensus stage, the following two cases will occur, 1. Only the node with the highest grade value remains: at this time, all nodes form a tree structure according to the relation between the upper node and the lower node, 2. Only 2 or 3 nodes with highest gradation values are left, one node is randomly generated as a root node, and the other two nodes set the root node as the upper node, so that the tree structure is also formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111635215.7A CN114301918B (en) | 2021-12-29 | 2021-12-29 | Efficient blockchain consensus method based on disqualified node elimination tree |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111635215.7A CN114301918B (en) | 2021-12-29 | 2021-12-29 | Efficient blockchain consensus method based on disqualified node elimination tree |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114301918A CN114301918A (en) | 2022-04-08 |
| CN114301918B true CN114301918B (en) | 2023-09-22 |
Family
ID=80970679
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111635215.7A Active CN114301918B (en) | 2021-12-29 | 2021-12-29 | Efficient blockchain consensus method based on disqualified node elimination tree |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114301918B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106920167A (en) * | 2017-02-24 | 2017-07-04 | 北京航空航天大学 | A kind of block chain mechanism based on P2P finance and common recognition algorithm |
| CN109274692A (en) * | 2018-11-14 | 2019-01-25 | 众安信息技术服务有限公司 | A kind of identification block chain network is done evil the method and device of node |
| CN111106942A (en) * | 2019-12-13 | 2020-05-05 | 南京邮电大学 | Block chain credit mechanism based on AP-PBFT algorithm |
| CN111861464A (en) * | 2020-07-18 | 2020-10-30 | 北京天仪百康科贸有限公司 | Node consensus method and system for block chain |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11483129B2 (en) * | 2018-05-09 | 2022-10-25 | Hefei Dappworks Technology Co., Ltd. | Method and apparatus for reaching blockchain consensus |
| CN109345386B (en) * | 2018-08-31 | 2020-04-14 | 阿里巴巴集团控股有限公司 | Transaction consensus processing method and device based on block chain and electronic equipment |
-
2021
- 2021-12-29 CN CN202111635215.7A patent/CN114301918B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106920167A (en) * | 2017-02-24 | 2017-07-04 | 北京航空航天大学 | A kind of block chain mechanism based on P2P finance and common recognition algorithm |
| CN109274692A (en) * | 2018-11-14 | 2019-01-25 | 众安信息技术服务有限公司 | A kind of identification block chain network is done evil the method and device of node |
| CN111106942A (en) * | 2019-12-13 | 2020-05-05 | 南京邮电大学 | Block chain credit mechanism based on AP-PBFT algorithm |
| CN111861464A (en) * | 2020-07-18 | 2020-10-30 | 北京天仪百康科贸有限公司 | Node consensus method and system for block chain |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114301918A (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Abraham et al. | Solidus: An incentive-compatible cryptocurrency based on permissionless byzantine consensus | |
| JP7184959B2 (en) | Method and apparatus for distributed database in network | |
| US9646029B1 (en) | Methods and apparatus for a distributed database within a network | |
| US20240004869A1 (en) | Methods and apparatus for a distributed database within a network | |
| Ben-Or et al. | Asynchronous secure computations with optimal resilience | |
| CN109819022A (en) | A kind of block chain common recognition method based on credit appraisal | |
| CN111311255B (en) | Intelligent contract formalization verification and error correction method based on prophetic machine | |
| CN109547211B (en) | Grading concurrent byzantine consensus method and system applying digital signature technology | |
| CN109347651B (en) | MSVL (modeling, simulation and verification language) -based block chain system modeling and security verification method and system | |
| CN113676541A (en) | Improved PBFT consensus method | |
| CN112217683B (en) | Cross-heterogeneous chain data reachability processing method, system, medium, equipment and terminal | |
| CN111445247B (en) | Method and apparatus for preventing block chain forking | |
| CN113448694B (en) | Block chain consensus method for improving transaction processing capability | |
| CN116015672B (en) | PBFT consensus mechanism based on reputation model | |
| CN114301918B (en) | Efficient blockchain consensus method based on disqualified node elimination tree | |
| CN106357604B (en) | A kind of consistent data accumulation collaboration assemble method | |
| CN111798234A (en) | Lightweight block chain system and construction method | |
| WO2024066665A1 (en) | Snowman consensus-based consortium blockchain multi-chain system and interaction and deployment method therefor | |
| CN116938523A (en) | Blockchain system, operating method, electronic device and storage medium | |
| CN113743949B (en) | Method for performing transactions in a blockchain, master node and slave node | |
| Zhang et al. | Broadcasting intermediate blocks as a defense mechanism against selfish-mine in bitcoin | |
| Pu et al. | Gorilla: Safe permissionless byzantine consensus | |
| Reynouard et al. | Bar nash equilibrium and application to blockchain design | |
| Hackfeld | A lightweight BFT consensus protocol for blockchains | |
| Shirodkar et al. | Layer 2 Solutions to Improve the Scalability of Blockchain |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |