CN109088741B

CN109088741B - Formalized modeling and verification method for block chain system

Info

Publication number: CN109088741B
Application number: CN201710450435.XA
Authority: CN
Inventors: 胡凯; 段张博; 白晓敏; 陈志东
Original assignee: Beihang University
Current assignee: Beihang University
Priority date: 2017-06-14
Filing date: 2017-06-14
Publication date: 2021-01-01
Anticipated expiration: 2037-06-14
Also published as: CN109088741A

Abstract

The invention provides a block chain system formalized modeling and verification method, which comprises the following steps: step 1: establishing an SDL (Specification and Description Language) model of a single node by utilizing a layered modeling mode according to a finite state machine; step 2: establishing an SDL model of a block chain system by utilizing a modular modeling mode according to the SDL model of a single node; and step 3: carrying out simulation analysis on the SDL model of the whole block chain system; and 4, step 4: formally verifying the nature of the SDL model of the blockchain system.

Description

Formalized modeling and verification method for block chain system

Technical Field

The invention relates to the field of block chain verification, in particular to a modeling and verifying method of a block chain link point and a network.

Background

Since the 90 s of the 20 th century, the birth and rapid growth of the internet have addressed the fundamental problem of information manufacture and transmission. The internet connects countries and regions around the world, allowing information to be quickly disseminated and shared. Through the information internet, the information of the whole internet grows exponentially when people enter an age of 'information explosion'. In recent years, as the information and digitalization progress of various industries in society continues to advance, and the comprehensive application and development of big data technology in various industries, data frequently flows among organizations and entities in various industries, wherein various confidential information and information containing privacy are not lacked, and if the information is leaked or tampered, a very serious problem is caused. Moreover, each of the interested parties of information sharing has a distrust problem, however, the introduction of a third party and a centralized data center leads to new problems of low information flow efficiency, easy information tampering and attack, and the like. Therefore, the demand of the social informatization process for ensuring the safety and credibility of information is more and more urgent. In addition, the information internet only solves the problem of information sharing, but cannot support value transfer, in which a part of value is transferred from one place to another in a network in a manner that a person can control and confirm, rather than copy, and the current internet does not support value transfer, which is a problem that needs to be solved urgently in many industries (particularly in the financial field). Therefore, a safe and reliable protocol for information sharing and value transfer is constructed, and the development of the digital society and the comprehensive application of big data can be greatly promoted.

In this context, Blockchain (Blockchain) technology is gradually entering the human vision and is more and more popular and is considered to be one of the most innovative applications in the 21 st century. The blockchain was widely applied in Bitcoin (Bitcoin) system, which was a P2P (Peer-to-Peer) electronic cash system invented by the minwis in 2008. The purpose of bitcoin protocol is to solve the Double-spending (Double-spending) problem in electronic money without the need of a third party. However, since the bitcoin is an autonomous system, is not owned by any organization or individual, and is difficult to supervise, most governments adopt a pressing attitude for the bitcoin, and the bitcoin is always used as a value bearing object similar to stocks and is not widely applied. However, as the progress of social digitization has advanced in recent two years, blockchain technology has moved away from the limitations of bitcoin applications, and, in addition to the introduction of concepts such as Smart contracts (Smart track), DAO/DACs (Decentralized Autonomous Organization/Corporation), blockchain technology has bloomed throughout various industries. Not only in the fields of finance, medical treatment, energy and the like, but also in the industries of insurance, supply chain, law, evidence storage, copyright, social assistance and the like, people are actively exploring the application scene, business mode and implementation mode of a block chain, and the block chain technology can bring huge innovation and even subversive change to the industries. Therefore, the blockchain technology is gradually becoming the cornerstone of the future digital society as a perfect solution for secure and credible information sharing and value transfer.

Superficially, a blockchain is a linked list of data linked by data blocks (blocks), whereas in essence, a blockchain is a decentralized, public, autonomous, tamper-proof, distributed ledger. The method comprises the steps of adding a timestamp to transaction or data, hashing through a Hash Function (Hash Function), guaranteeing data safety and privacy through encryption and signature technologies, packaging the data into individual data blocks, linking the data blocks together in a time sequence, transmitting the data blocks in a Peer-to-Peer (P2P) network, performing redundancy backup on each Peer node, achieving global consistency through a consensus algorithm and guaranteeing that the data cannot be tampered. Through the mechanism and the huge innovation of the intelligent contract, a plurality of application scenes of various industries are supported, and a safe and autonomous database system with public trust is formed.

Therefore, applications in such areas with high security requirements make the research on the security credibility of the blockchain system very important. Etherns lost approximately 4500 million dollar worth of Etherns in a previous severe DAO attack. Therefore, formal verification methods are introduced to improve the security and reliability of the blockchain system.

The formalization method is a special technology based on mathematics and is suitable for describing, developing and verifying software and hardware. Using formal methods for software and hardware design is desirable to be able to use appropriate mathematical analysis to improve the reliability and robustness of the design, as well as other engineering disciplines. An important research item of the formalization method is a formal specification (also called formal specification or formalization description), which is a mathematical description of what is done to a program, a functional description of the program written in a formal language with precise semantics, which is a starting point for designing and programming, and a basis for verifying whether the program is correct. The formal conventions are usually discussed in terms of their consistency (no contradictions in themselves) and completeness (whether the objects to be described are completely and exhaustively depicted). The methods of formal specification can be mainly divided into two categories: one type is a model-oriented method, also known as system modeling, that characterizes different behaviors of a system by constructing a computational model of the system; another type is a property-oriented approach, also called property description, which describes a system by defining some properties that the system must satisfy.

Another important research content of formal verification formalization methods is formal verification. The formal verification is to verify whether the existing program (system) P meets the specification of the existing program (system) P or not

Is (i.e. P)

) It is also the core problem that formalization approaches are to solve. The traditional verification method includes simulation and test, which are both to check the system by experiment. Simulation and testing are performed on the system abstract model and the actual system, respectively, and the general method is to give input at one point of the system and observe output at another point, which is very costly and difficult to find out all potential errors due to the limited system behavior that can be covered by the experiment. Based on this, formal verification mainly studies how to strictly prove the correctness of a program using a mathematical method (i.e., program verification).

In a sense, a blockchain network can be considered a highly trusted distributed communication network. Therefore, the formalization method can effectively improve the trust degree of the user on the safety of the blockchain system. We introduce a hierarchical and modular modeling approach based on SDL (Specification and Description Language) to formally model the blockchain system. The individual nodes are modeled by hierarchical modeling, while the entire blockchain network architecture is modeled by modular modeling. On the basis, the whole block chain system can be simulated and formally verified to find out the logic error of the system. The mode is beneficial to improving the development automation level, shortening the development time and reducing the possibility of errors in the coding process. Conventional blockchain experiments typically require deployment of a large number of computers to meet the experimental conditions. By the block chain form modeling method, multi-node block chain simulation can be realized on a single computer. Furthermore, we can study some properties of different consensus algorithms on a given blockchain network by building blockchain models. Meanwhile, the method can also be beneficial to the formulation of the future block chain standard.

Disclosure of Invention

Accordingly, the present invention introduces formal verification methods to improve the security and reliability of the blockchain system.

A block chain system formalized modeling and verification method comprises the following steps:

step 1: establishing an SDL (Specification and Description Language) model of a single node by utilizing a layered modeling mode according to a finite state machine;

step 2: establishing an SDL model of a block chain system by utilizing a modular modeling mode according to the SDL model of a single node;

and step 3: carrying out simulation analysis on the SDL model of the whole block chain system;

and 4, step 4: formally verifying the nature of the SDL model of the blockchain system.

Preferably, before step 1, the method further comprises:

analyzing the environment and the function of the block chain system;

a finite state machine of the blockchain system is established.

Preferably, in step 3, a simulation analysis is performed on the SDL model of the whole block chain system using a Simulator in Telelogic Tau.

Preferably, step 3 further comprises:

(1) compiling the SDL model of the block chain system, executing an option of Microsoft RealTimeSimuling, and opening a compiled executable file in a simulator;

(2) generating a first MSC graph by using an MSC (Message Sequence Chart) function in the simulator;

(3) aiming at the block chain system requirement function, inputting a command script of an excitation signal to a simulator;

(4) executing the command script to generate a second MSC graph;

(5) and analyzing the second MSC graph, judging whether the SDL model of the block chain system realizes the block chain system requirement function, if not, debugging is carried out, and simulating again until the block chain system requirement function is simulated to pass.

Preferably, step 4 further comprises:

verifying the SDL model of the blockchain system by using a Validator (verifier) in Telelogic Tau, wherein the verified property comprises at least one of the following items: reachability, deadlock free, livelock free, weak livelock, time dependent livelock, bounded, recoverability or self-synchronization, stateless ambiguity, mutual exclusion, termination or progression, no redundant description, fairness, integrity, consistency.

Preferably, the verification is performed using one of the following traversal algorithms in the Validator: bit state traversal, Random walk traversal, Tree search traversal, Tree walk traversal, and explicit traversal.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings described below are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to the drawings without creative efforts.

FIG. 1 is a block chain system formalized modeling and verification method flow of the present invention;

FIG. 2 is a block chain hierarchy;

FIG. 3 is a model page of the consensus layer SDL model;

fig. 4 is a simple blockchain network topology model.

Detailed Description

Referring to fig. 1, a block chain system formalized modeling and verification method mainly includes the following steps:

● performing environmental and functional analysis on the existing blockchain system;

● establishing a finite state machine of the blockchain system;

●, establishing an SDL model of a single node by using a layered modeling mode according to a finite state machine;

●, building an SDL model of the block chain network by using a modular modeling mode according to the single node model;

● performing simulation analysis on the whole block chain system;

● formally verify the properties of the blockchain system model.

Block chain system modeling

And establishing a model of the block chain system through a formal specification document of the block chain system. A modeling tool and corresponding modeling language need to be selected. The SDL language is used here to describe the blockchain system according to the above generated formal specification document, i.e. complete the modeling process.

The modeling is divided into two processes:

1. aiming at the characteristics of an SDL language, a block chain hierarchical modeling method based on the SDL is provided by combining the attributes of a block chain single node system, and the block chain single node system is formally modeled. For a single-node blockchain system, the data layer, the network layer, the consensus layer, the intelligent contract layer and the application layer can be hierarchically modeled by a hierarchical architecture as shown in fig. 2.

The data layer is responsible for block generation and block chain construction and storage. The data layer hashes public or encrypted transactions or data into a Merkle tree through a hash function, then encapsulates fields such as a tree root, a timestamp and a workload certification result (if a PoW consensus mechanism is adopted) into a block, and links a newly generated block to a block chain by storing a hash value of a previous block. Since the data blocks are linked according to the hash of the previous block, the method is more secure and reliable compared with the traditional pointer linking.

The network layer is typically a P2P network, and may be a conventional network for private links. The network layer needs to have the functions of unicast, broadcast, multicast and the like, can filter the data packet, and simultaneously completes the functions of authorized or open communication node addition, deletion and the like.

The common recognition layer is one of the most core layers of the blockchain model, defines rules of the blockchain model, and specifies how to achieve agreement among nodes, how to backup, how to Fault-tolerant, and how to balance data consistency, partition Fault-Tolerance and availability, common public chains include PoW (Proof of Work), PoS (Proof of rights and benefits), and DPoS (delete Proof of rights and certifications), etc., common alliance chains and private chains generally adopt a Byzantine Fault-tolerant algorithm and a conventional consistency algorithm as their common recognition, and common recognition algorithms include PBFT (active Byzantine Fault-tolerant algorithm), Paxos, Raft, etc.

The intelligent contract layer realizes the mechanisms of compiling, running and the like of the intelligent contract. The intelligent contract management system comprises rules of an intelligent contract language, how an intelligent contract script is compiled into instruction codes, and how the intelligent contract instruction codes are safely isolated from running in an intelligent contract virtual machine.

The application layer is various applications of the upper layer, such as applications of various industries including a first generation block chain-an electronic money system, a second generation block chain-a programmable finance, medical treatment, supply chain and the like, and a future third generation block chain-a programmable society.

One SDL model page of the consensus layer model is shown in fig. 3.

2. On the basis of a single node formalized model, a whole block chain network is subjected to modular modeling, various block chain network topological structures can be verified and the characteristics of the block chain network topological structures can be analyzed, and a simple SDL model of the block chain network topological structure is shown in FIG. 4.

Real-time simulation of block chain system

The Simulator in Telelogic Tau (an SDL tool) can simulate the established block chain model. The simulation includes two cases: real-time simulation and non-real-time simulation, and real-time simulation is needed when the block chain has timeliness. In real-time simulation, the time lapse in the simulation system is kept consistent with the actual environment. When real-time simulation is adopted and a block chain model needs to be compiled, the parameter is set to be Microsoft RealTimeSimulination.

In the simulation process, various signals can be input through a simulator and then run; the MSC can display the state transition, signal input and output of the block chain model, the operation condition of the block chain model under the condition of inputting an excitation signal can be known through observation and analysis of the MSC, and whether the block chain model realizes the corresponding function or not can be judged.

Therefore, the real-time simulation of the blockchain model comprises the following steps:

(1) the established block chain model is compiled, the option is Microsoft RealTimeSimulination, and the compiled executable file is opened in the simulator.

(2) The MSC function in the emulator is turned on to generate a first MSC map at the time of emulation.

(3) For a certain protocol function, a command script simulating an excitation signal to be input is edited.

(4) And starting simulation, executing the command script and generating a second MSC graph.

(5) And analyzing the second MSC graph, judging whether the blockchain model realizes the function required by the blockchain system, if not, debugging, and simulating again until the function simulation passes.

Model verification for blockchain systems

We will use Validator (verifier) in the SDL tool Telelogic Tau to do verification work on the blockchain model, mainly verifying the following properties:

(1) general properties of the blockchain system:

● accessibility (accessibility). The reachable relationships between the various possible states of the blockchain system are verified. If a state of the blockchain system is not reachable from the initial state, an error is indicated in the blockchain system. If a transition from state A to state B is not possible (directly or indirectly), then state A to state B is unreachable.

● has no Deadlock (Deadlock free). The most typical deadlock is one in which each entity in the blockchain system is in a wait state in which further action can only be taken after an "event" occurs, but in which state this "event" is unlikely to occur. When a deadlock occurs, the state in which the blockchain system is located is called a deadlock state.

● has no Livelock (Livelock free). Livelock refers to a blockchain system in an infinite dead loop, and no other event can cause the blockchain system to disengage from this loop. For example, the blockchain system performs a timeout retransmission operation without limitation, but always does not receive acknowledgement information of the counterpart. The state is still changing, but cannot be left out of this dead-cycle state.

● Weak livelock (week livelock). It means that the blockchain system is in the dead loop, and the blockchain system exits the dead loop only when the relative speed of the blockchain system exchange commands reaches a certain state.

● Time-dependent livelock (Time-dependent livelock). Also known as temporary blocking (Tempo-blocking). It means that the blockchain system is in the dead loop, but when the relative speed of the two communicating parties exchanging messages reaches a certain state, the blockchain system can come out of the dead loop.

● bounded nature (Boundedness). It is checked whether the capacity (e.g. channel capacity, window size) of certain components or parameters of the blockchain system is bounded. The boundedness is in terms of the properties of the elements of the blockchain system and the properties of the channels.

● recoverability or self-synchronization (Recovery from failures). This is whether the block chain system can return to normal state (including initial state) execution within a limited number of steps after an error occurs.

● stateless ambiguity (State ambiguities free). A process is only allowed to have one steady state at a time. The steady state is a process state when the channels of both communication parties are empty. If a process can have multiple stable states at a time, the state of the process is said to be ambiguous.

● Mutual exclusion (Mutual exclusion). Mutual exclusivity refers to the fact that some blockchain system actions cannot be performed simultaneously by multiple users. For example, multiple users cannot request the same resource at the same time.

● terminate or Progress (Termination or Progress). Meaning that the services provided by the blockchain system must be completed in a limited time. Termination is for terminating block chain systems (terminating protocols), meaning that a block chain system can always reach the desired end state. While progress is made with respect to cyclic blockchain systems (cyclic protocols), this means that the blockchain system always reaches its initial state.

● description of redundancy-free (Absence of Overspecification). There is no useless, redundant description in the blockchain system specification, e.g., no unpractical message reception (absence of unused message receptions).

● Fairness (Fairness). It means that each blockchain system entity should get an equal chance of running, no matter what other blockchain system entities want to do.

(2) The special properties of the blockchain system:

● integrity (completensiss). It is meant that the blockchain system design takes into account all events, options, and services that may occur. It is checked whether the blockchain system can handle all possible inputs, i.e. lack of processing by the application, or unexpected receipt or input (i.e. miscarriage).

● consistency (consistency). It means that the service behavior (or property) of the blockchain system is consistent with the service behavior (or property) of the blockchain system, that is, the blockchain system should provide the service required by the user without providing the service which is not required by the user.

Meanwhile, several verification algorithms in validators are used, Bit state traversal, Random walk traversal, Tree search traversal, Tree walk traversal, and explicit traversal. To perform the verification work. The algorithms have different priorities, so that a proper algorithm needs to be selected. For the verification of the blockchain system, we generally choose Bit state implementation.

The foregoing is merely an example of the present invention and is not intended to limit the invention in any manner. Those skilled in the art can make various other improvements or modifications equivalent to the above-described embodiments without departing from the scope of the present invention, and any simple modification, equivalent change or modification made to the above embodiments according to the technical essence of the present invention will still fall within the scope of the present invention.

Claims

1. A block chain system formalized modeling and verification method is characterized in that:

step 1: establishing a standard description language model of a single node by utilizing a layered modeling mode according to a finite state machine;

and step 3: performing simulation analysis on the SDL model of the whole block chain system, wherein a simulator in Telelogic Tau is used for performing simulation analysis on the SDL model of the whole block chain system; wherein the step 3 further comprises:

(1) compiling the SDL model of the block chain system, executing a Microsoft real-time simulation option, and opening a compiled executable file in a simulator;

(2) generating a first MSC graph by using a message sequence diagram function in the simulator;

(4) executing the command script to generate a second MSC graph;

(5) analyzing the second MSC graph, judging whether the SDL model of the block chain system realizes the block chain system requirement function, debugging if the block chain system requirement function is not realized, and simulating again until the block chain system requirement function simulation passes the judgment;

and 4, step 4: formally verifying the nature of the SDL model of the blockchain system, said step 4 further comprising:

verifying the SDL model of the blockchain system by using a verifier in Telelogic Tau, wherein the verified properties comprise at least one of the following: reachability, deadlock free, livelock free, weak livelock, time dependent livelock, bounded nature, recoverability or self-synchronization, stateless ambiguity, mutual exclusion, termination or progression, no redundant description, fairness, integrity, consistency; wherein

Performing a verification using one of the following traversal algorithms in the verifier: bit state traversal, random walk traversal, tree walk traversal, or exhaustive traversal.

2. The block chain system formal modeling and verification method of claim 1, wherein: before step 1, the method further comprises the following steps: analyzing the environment and the function of the block chain system; a finite state machine of the blockchain system is established.