A method for designing and analyzing performance of a cross-chain architecture for a large-scale internet of things based on an IOTA blockchain

Abstract

A cross-chain architecture design and performance analysis method for large-scale Internet of Things based on IOTA blockchain, a multi-field, large-scale distributed Internet of Things, trusted, multi-chain, intelligent and efficient blockchain architecture is designed based on IOTA, which realizes low-cost trusted interconnection of blockchains in different fields and establishes a safe shared ecological model. The delay of one-hop transmission is derived using stochastic calculus theory. From the aspects of consensus time, throughput, security and cross-chain interaction efficiency, the performance of the new architecture is analyzed, providing a theoretical basis for the landing of blockchain. At the same time, in order to speed up information dissemination and reduce network overhead, a directional propagation method is proposed to replace the rumor algorithm of IOTA, reducing network overhead.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of information communication, and particularly relates to a cross-chain architecture design and performance analysis method for large-scale Internet of Things based on IOTA blockchain, which is suitable for large-scale business integration of Internet of Things. BACKGROUND

[0002] With the development of Internet of Things, more and more devices are connected to the Internet, such as smart cars, smart home devices, security monitoring devices, smart medical devices, smart grid monitoring devices, intelligent transportation devices, etc. These devices will generate a large amount of data, and if these data can be effectively combined and applied, it will have great utilization value. For example, by comprehensively utilizing the data of intelligent transportation, the destination of the user can be predicted in advance, so as to schedule and respond to the power supply of the power grid, and further dynamically adjust the power consumption of smart home devices to reduce the cost of power consumption and relieve the pressure of the power grid peak; by utilizing the data of smart home and security monitoring, the intelligent transportation system can assist in overall path planning of vehicle navigation to reduce congestion as much as possible; by comprehensively utilizing all the data of Internet of Things to assist in the construction of smart city and urban operation, management and planning, etc. Therefore, how to integrate different needs of various fields without prior establishment of trust, realize cross-field, secure and reliable sharing of data, and promote autonomous and reliable data interaction and sharing economy system between fields has become a major challenge.

[0003] Blockchain technology is open, transparent, traceable and tamper-proof. Whether it is a person or a machine, it can realize trusted value exchange and data sharing through the blockchain system. By using blockchain technology and its incentive mechanism, an open and shared ecological system of "I for everyone, everyone for me" is established. At present, blockchain has been widely applied in finance, health care, supply chain, asset management and many other fields, but due to the constraints of throughput, network isolation, regulation, scalability and other factors, the current blockchain project is difficult to carry large-scale commercial landing. Among the many problems faced by blockchain, cross-chain interoperability is one of the key challenges that hinder the widespread adoption of blockchain applications.

[0004] Therefore, a new blockchain architecture is urgently needed to break the blockchain island and realize the interoperability of the blockchain. Currently, there are three kinds of cross-chain technologies: notarization mechanism, hash locking and side chain / relay. However, these three kinds of blockchain technologies are all improvements based on the traditional blockchain, and are limited by the inherent defects of the traditional blockchain, so they cannot realize atomic exchange, asset transfer and full sharing of information at the same time, and it is difficult to meet the demand of large-scale Internet of Things multi-service integration. The IOTA blockchain developed for the Internet of Things breaks the chain structure of the traditional blockchain, adopts a directed acyclic graph storage method, allows all nodes to write data into the distributed ledger at any time and anywhere, and the data written into the ledger is no longer verified by the nodes in the blockchain, but by subsequent transactions, so that IOTA breaks the "impossible triangle" of the blockchain, and achieves the optimal combination of security, decentralization and scalability. It has the advantages of 0 transaction fee, low resource occupation, short consensus time, large network throughput and quantum-level encryption. However, the current IOTA is only applicable to a single field and cannot meet the demand of multi-service integration. SUMMARY

[0005] The application designs a multi-field, large-scale distributed Internet of Things, trusted, multi-chain, intelligent and efficient blockchain architecture based on IOTA. The problems of low blockchain throughput and non-universal are effectively solved, the low-cost trusted interconnection of blockchains in different fields is realized, a safe shared ecological model is established, and the system throughput is improved.

[0006] A cross-chain architecture design and performance analysis method based on IOTA blockchain for large-scale Internet of Things, comprising the following steps:

[0007] Step 1: Designing a blockchain architecture based on IOTA, dividing the IOTA blockchain into a distributed identity information DID chain and a plurality of application chains, a node having different DID, and registering the DID information in the DID chain, and selectively uploading part of personal information in different application chains to protect user privacy; for intra-chain transactions, after the device is linked to the blockchain, the DID information is generated by the DID character, the application chain to which the device belongs and the application chain exclusive ID, and the DID file is created; cross-chain interaction includes atomic transaction, asset transfer and cross-chain digital sharing;

[0008] Step 2: Analyzing the performance of the new architecture from the aspects of consensus time, throughput, security and cross-chain interaction efficiency; analyzing the one-hop delay of the node by using random network calculus theory, calculating the consensus time of IOTA in an application chain according to the IOTA consensus process, and deducing the maximum throughput that the network can bear under the condition that the event arrival rate is large, and calculating the event arrival rate required for a successful parasitic chain attack by a malicious node;

[0009] Step 3, the method of directional propagation is used instead of the rumor algorithm of IOTA; the data in an application chain only needs to be transmitted to the nodes in this chain, that is, the nodes in an application chain transmit to his virtual neighbor node each time.

[0010] The beneficial effects achieved by the present application are:

[0011] (1) Based on IOTA, a multi-field fusion, large-scale distributed Internet of Things, trusted, multi-chain, intelligent and efficient blockchain architecture is designed, which realizes low-cost trusted interconnection of blockchains in different fields and establishes a secure shared ecological model.

[0012] (2) The performance of the new architecture is analyzed from the aspects of consensus time, throughput, security and cross-chain interaction efficiency, providing a theoretical basis for the landing of blockchain.

[0013] (3) The rumor algorithm of IOTA is replaced by the method of directional propagation, reducing the network overhead. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 The IOTA-based multi-field fusion, large-scale distributed Internet of Things, trusted, multi-chain, intelligent and efficient blockchain architecture model in the embodiments of the present application.

[0015] Figure 2 The blockchain internal transaction process in the embodiments of the present application.

[0016] Figure 3 The DID distribution diagram in the IOT node A in the embodiments of the present application.

[0017] Figure 4 The cross-chain atomic transaction diagram in the embodiments of the present application.

[0018] Figure 5 The cross-chain asset transfer diagram in the embodiments of the present application.

[0019] Figure 6 The cross-chain data sharing diagram in the embodiments of the present application.

[0020] Figure 7 The blockchain network propagation model in the embodiments of the present application.

[0021] Figure 8 The node service model diagram in the embodiments of the present application.

[0022] Figure 9 The series service diagram in the embodiments of the present application.

[0023] Figure 10 The IOTA consensus process diagram in the embodiments of the present application. DETAILED DESCRIPTION

[0024] The technical solutions of the present application will be further described in detail below in combination with the drawings of the specification.

[0025] The present application designs a multi-field integrated, large-scale distributed Internet of Things oriented, trusted, multi-chain, intelligent and efficient blockchain architecture based on IOTA, and analyzes the performance of the new architecture from the aspects of consensus time, throughput, security, and cross-chain interaction efficiency. At the same time, in order to speed up information propagation and reduce network overhead, a directional propagation method is proposed to replace the rumor algorithm of IOTA by considering the actual network topology, and the one-hop transmission delay is derived using stochastic calculus theory. The network is analyzed from the aspects of network throughput, consensus time, security performance and cross-chain interaction efficiency.

[0026] A multi-field integrated, large-scale distributed Internet of Things oriented, trusted, multi-chain, intelligent and efficient blockchain architecture is designed based on IOTA.

[0027] As shown in Figure 1 The system is mainly divided into two layers, a communication network and a blockchain network mapped by the communication network. The communication network is a large-scale distributed Internet of Things, and the blockchain network is divided into different application chains according to different applications, such as distributed identity information (DID) chain, smart city chain, industrial Internet of Things chain, etc. Each user needs to register personal information in the DID chain to facilitate verification by other users, and at the same time, in order to protect the personal privacy of users, different DID is allowed to be used in different application chains.

[0028] In this system, the operation on the application chain is divided into two types: read operation and write operation. 1) Read operation: each node can freely access all application chains in IOTA to ensure sufficient sharing of data; 2) Write operation: only when it is necessary to write to a certain application chain, it is necessary to join the application chain, store the data of the application chain, and complete the verification work of the transaction. Different application chains can realize atomic exchange, asset transfer and data sharing at low cost, high efficiency and trustworthiness, and fully realize the value sharing of various industries. The scheme mainly includes two parts: intra-chain transaction and cross-chain interaction.

[0029] For intra-chain transaction, as shown in Figure 2 When the device is connected to the blockchain, the DID information is generated by the DID character, the application chain to which it belongs and the application chain exclusive ID (generated by the blockchain, ensuring global uniqueness), and the DID file is created.

[0030] 1) Register the information in the blockchain;

[0031] 2) Apply for a verifiable claim to the identity issuer;

[0032] 3) Identity issuer verifies its identity, issues a certificate to the user, and stores the certificate summary information on the DID blockchain for verification;

[0033] 4) When the device needs a certain service, the corresponding verifiable claim can be selected and sent to the corresponding application chain to apply for a specific service;

[0034] 5) The device that can provide services in the application chain verifies the authenticity of the device in the DID chain; after verification is successful;

[0035] 6) The service provider sends its verifiable claim to the service requester to prove that it can provide the corresponding service;

[0036] 7) The service requester reads the DID blockchain information and verifies the verifiable claim of the service provider, and after verification is successful;

[0037] 8) Payment, and place the transaction record on the corresponding blockchain, and after the transaction is agreed;

[0038] 9) Provide services.

[0039] For cross-chain interaction, the IOTA blockchain is divided into a DID chain and several application chains, as shown in Figure 3 , a node can have different DID, and the DID information is registered in the DID chain, and the node can selectively upload part of the personal information in different application chains to protect the personal privacy of the user. Cross-chain interaction mainly includes three parts, atomic transaction, asset transfer, cross-chain digital sharing, the specific process is as follows:

[0040] Atomic transaction: as shown in Figure 4 , take node A in the industrial Internet of Things chain as an example, which hopes to trade with node B in the smart city, the specific process is as follows:

[0041] 1) A is added to the smart city blockchain;

[0042] 2) Register with the DID chain;

[0043] 3) Node A applies for transaction and sends verifiable claim to node B, proving that the node has the identity verification information required for transaction with node B;

[0044] 4) Node B reads information from the DID chain for verification;

[0045] 5) After verification is completed, node B sends DID and verifiable claim to node A in the smart city chain, and the two nodes can freely trade in the smart city.

[0046] Asset transfer: as shown in Figure 5As shown, taking the example that node A in the industrial internet of things chain hopes to transfer the asset in the industrial internet of things chain to the smart city chain, the specific process is as follows:

[0047] 1) Node A transfers the asset corresponding to the address of the industrial internet of things application chain to the address of the smart city application chain of the node of node A;

[0048] 2) Place the transaction in the industrial internet of things application chain;

[0049] 3) After the transaction is consensus, place the transaction in the smart city chain;

[0050] 4) When the subsequent transaction in the smart city chain verifies this cross-chain transaction, it needs to check the industrial internet of things chain to see if the transaction has reached consensus, and only if the consensus is reached, the transaction in the smart city chain can be verified.

[0051] Cross-chain data sharing: as shown in Figure 6 , any node in the application chain in the IOTA can freely view any application chain in the IOTA, therefore, the data on each application chain in the IOTA can be comprehensively utilized.

[0052] From the consensus time, throughput, security, and cross-chain interaction efficiency, the performance of the new architecture is analyzed, mainly using random network calculus theory to analyze the one-hop delay of the node, calculating the consensus time of IOTA in an application chain according to the IOTA consensus process, and in the case of a large event arrival rate, the maximum throughput that the network can bear is derived, and the event arrival rate required for a successful attack by a malicious node on the parasite chain is calculated, the specific steps are as follows.

[0053] Random network calculus theory: as shown in Figure 8 , in the application chain d, for a node i with m d virtual neighbor nodes, assume that the virtual neighbor nodes of node i are , and the corresponding event arrival rate is The event arrival rate of node i is λ i , which conforms to the Poisson distribution, then the event arrival rate of the through flow from node i to the virtual neighbor node is Then the number of events arriving at the service node within [0, t) is A(t); S(t) represents the cumulative service amount that the service node can provide within the time interval [0, t). When the data flow completes the service and leaves the service node, the cumulative number of packets leaving the service node within [0, t) is Then the transmission delay of the data flow at a service node is: D(t) = inf{d >= 0: A(t) <= A * (t+d)}, wherein t is the event arrival time, and t+d is the event departure time.

[0054] Definition 1 (Random Arrival Curve): For any data stream A(s,t), if there exists a function... in Let be the set of nonnegative generalized increasing (decreasing) functions, such that for all 0 ≤ s ≤ t and any nonnegative number x ≥ 0, the following condition is satisfied:

[0055] P(A(s,t)-α(s,t)>x)≤f(x)

[0056] Then α(s,t) is called the random arrival curve of the data stream A(s,t), and f(x) is the probability upper bound function of α(s,t), which can be denoted as A(t)~<f,α> Here, P(.) represents the probability operator.

[0057] Definition 2 (Random Service Curve) A * (s,t) represents the cumulative amount of data leaving service system S in business flow A(s,t) within [s,t). If for all 0≤s≤t and x≥0, the following condition is met:

[0058]

[0059] The service system is said to provide boundary functions as follows: The service curve β(s,t), β(s,t)∈F, is denoted as S~<g,β> .

[0060] Theorem 1 (Aggregation Flow Theorem): Suppose that an input business flow A(t) consists of A1(t), A2(t), ..., A... Ι (t) is formed by the aggregation of these Ι sub-streams, i.e., A i (t)~ <f i ,α i >, then the aggregated data stream A(t) has A(t)~<f,α> The process of arrival.

[0061]

[0062] Theorem 2 (Serial Service Theorem): Assume a network has K service nodes, and the service curve of each node is S. i ~ <g i ,β i If the services received by the service flow A(t) sequentially through these nodes can be considered as a whole providing services to the service flow A(t), denoted as S(t), and the random service curve satisfies S~<g,β> .

[0063]

[0064] Where β i,-(i-1)θ (t)=β i(t)-(i-1)θt, θ,θ1,θ2,…,θ K These are free parameters that are greater than 0.

[0065] Theorem 3 (Performance Upper Bound Analysis Theorem) Assume that the business flow A(t) ~<f,α> If a user enters service system S to obtain services, then for all t≥0, x≥0, the latency boundary B(t) and backlog boundary D(t) of the business flow are:

[0066]

[0067] Where h(α,β) is the maximum horizontal distance between α(t) and β(t). α(t) represents the maximum vertical distance between α(t) and β(t).

[0068] End-to-end delay T d For the Nodi node, the event arrival rate λ A If the distribution is Poisson, then its moment generating function is, Random arrival curve is Limited by θ i β is any free parameter not less than 0. This embodiment assumes that nodes in the service chain from Node i to its virtual neighbor nodes have the same service process, and their service rates are constant, denoted by C. Therefore, the service process of Node i is β. i (s,t)=C(ts) is constrained by g(x=0).

[0069] like Figure 6 As shown, the event arrival rate of node Nodei and the event arrival rate of its virtual neighbor nodes constitute the flow through Nodei to its virtual neighbor nodes. According to Theorem 1, this is...

[0070]

[0071] The random arrival curves of the throughflow at each node are the same: θ1 = θ2 = ... = θ K =θ, then we have

[0072]

[0073] like Figure 9 As shown, Theorem 2 can be applied to S1→S2→…→S K The three nodes connected in series to provide services for the data flow are equivalent to a single node S providing transmission services for it. Therefore, the service model provided by this node can be derived.

[0074]

[0075] From theorem 3

[0076]

[0077] And

[0078]

[0079] Assuming the service rate of the credential is greater than the average arrival rate, the system does not drop packets, and λ A (e θ -1)-Cθ≤0, at this time, h(α+x,β) is maximum when s=0, and the following can be obtained:

[0080]

[0081] Substituting equation (5) into equation (4) gives

[0082]

[0083] The end-to-end transmission delay T is:

[0084]

[0085] Where, 0<p c <1 is the default probability, which needs to satisfy the condition λ Ai (e θ -1)-Cθ≤0, θ is a parameter greater than 0, if the event arrival rate that the service node can handle is greater, θ should be as small as possible, and because the selection of free parameters should make the corresponding probability boundary function as small as possible, considering these two factors, θ=1 is taken in the embodiment. Equation (11) can be written as:

[0086]

[0087] One-hop delay T in rumor algorithm: in the rumor algorithm, each broadcast is transmitted to the physical neighbor node, therefore, K=1, and equation (12) can be obtained:

[0088]

[0089] (2-2) consensus time:

[0090] Assuming that the number of nodes in the network is N, there are D={1,2,3,…,D} application chains in the network, and the number of nodes of each application chain is N d , the average forwarding hop number of the node to the virtual neighbor node is K, the average delay of the node to the virtual neighbor node is T d , the consensus time of the network is T Cd , and it can be known that the total number of end transactions in the network R0d = 2λ ds h d tends to be stable, wherein, is the total event arrival rate in the application chain d, h d is the average time for writing events in the application chain d into IOTA, and it is known that the consensus time of the application chain d is

[0091]

[0092] wherein a = ω(0.5) is the Lambert ω function, p is the probability of successful reception of a one-hop data packet, L d is the transmission hop number of the data packet transmitted to each node of the application chain d.

[0093] From (12), we have a = ω(0.5) ≈ 0.35, K is the average hop number of the IOTA node to the virtual neighbor node, which is set to 5 in the embodiment, and p ≈ 1 under the condition that the network state is good, C is the data transmission rate of the service node, h d is the time delay for writing a transaction into IOTA, and the transmission speed of the network is much greater than the rate of transaction writing throughput, because a certain amount of proof of work (POW) is required for writing a transaction into IOTA, and therefore Ch d > 1, then, then then wherein is the total number of m d ary nodes, i.e. N d Therefore, η ≈ 1.

[0094] then

[0095]

[0096] wherein for any node A in the application chain d, λ Ai (e-1)-C ≤ 0,

[0097] Similarly, when the application chains are not divided, the consensus time of the network is

[0098]

[0099] wherein, is the total event arrival rate in the network, h is the time delay required for writing into the tangle, L is the hop number required for broadcasting the data written into the tangle to the whole network, and for any node A in the network, λ Ai (e-1)-C ≤ 0,

[0100] From (15) and (16), when the event arrival rate in the network is constant, the consensus time is related to the events of the transaction transmission in the network. Obviously, in a specific network, the more the selected (virtual) neighbor nodes, the shorter the transaction transmission time. Therefore, under the condition of satisfying λ A (e-1)-C≤0, , the larger m d or m, the smaller L d or L, and the shorter the consensus time.

[0101] (2-3) System throughput:

[0102] Since the IOTA transaction consensus is verified by the following transaction, for the application chain d, the total number of Tips in the network is R 0d , the consensus time of the network is T Cd , and the throughput of the application chain d is

[0103]

[0104] Since there are D application chains in the network, the total throughput of the system is

[0105]

[0106] Similarly, in the traditional IOTA blockchain, the throughput of the network is

[0107]

[0108] From the analysis, the larger m d or m, the shorter the consensus time, and R 0d and R0 are independent of m d or m, so the larger m d or m, the larger the system throughput.

[0109] When m d or m is constant, the throughput of the network increases with the total event arrival rate in the network, and for equations (17), (18), and (19), λ A (e-1)-C≤0, but due to the broadcast nature of the IOTA blockchain, any transaction in the network needs to be transmitted to any node in the network, so each node needs to forward all transactions in the network. Therefore, the amount of transactions that each node needs to forward is λ ds , that is

[0110]

[0111] Similarly, in the case of not splitting the application chain,

[0112]

[0113] Therefore, the throughput of the system proposed by the method can reach

[0114]

[0115] The throughput of the traditional IOTA network can reach

[0116]

[0117] (2-4) Security:

[0118] Unlike traditional blockchains, IOTA uses a directed cyclic graph storage method, and transaction verification is no longer verified by nodes in the network, but by subsequent transactions verifying previous transactions. The structure of transactions stored in the database in IOTA is shown in Figure 10 Each transaction is verified by its subsequent transaction, and transactions that have not been verified by subsequent transactions are called end transactions. When a node in the network has a transaction to upload to the database, 1) first select 2 end transactions to verify the legality of the transaction, and after verification, 2) do a certain amount of PoW to prevent malicious nodes from frequently sending information to the database, 3) broadcast the information in the network, and store it as an end transaction on the node in the network, waiting to be verified by subsequent transactions. When all end transactions have verified the transaction, the transaction has completed consensus.

[0119] From the consensus process of the transaction, it can be seen that when the transaction is written to the database, it needs to do PoW, and the complexity of PoW done by each transaction is proportional to the weight of the transaction, as shown in Figure 10 The lower left corner is the weight of the transaction itself, and the middle number is the cumulative weight of the transaction, which is the sum of the weights of the transaction itself and all transactions that directly and indirectly verify the transaction. When selecting end transactions, transactions with large cumulative weights are preferred, i.e. first in the [W, 2W] depth, a transaction is randomly selected using a weighted random walk algorithm, and the next transaction is selected, such as in Figure 10 Assuming that the cumulative weight of the selected transaction is 12, the next transaction is the cumulative weight of 3 and 6, and the cumulative weight of 6 is more likely to be selected, until the end transaction with a weight of 3 is selected as the parent node of the new transaction.

[0120] In summary, IOTA is verified by subsequent transactions, and after the transaction is verified by all end transactions, the transaction completes consensus. However, in the transaction consensus process, how to ensure that there is no double spending transaction is the biggest security problem faced by IOTA.

[0121] Double-spend problem: the most important thing to solve the double-spend problem is to solve the parasitic chain attack, and the probability of success of the parasitic chain attack for application chain d is

[0122]

[0123] wherein is the cumulative weight of the transactions that have reached consensus at time t, is the probability that the next transaction in application chain d is an honest transaction; λ d is the total event arrival rate of honest transactions in application chain d, μ is the event arrival rate of parasitic chain; t 0d is the consensus time. When p≤0.5, if you want to attack successfully, you need the event arrival rate of malicious nodes to be greater than or equal to the event arrival rate of the entire network.

[0124] When p>0.5, μ(t 0d )-φ(t 0d ) is greater, the greater the probability of successful attack, that is, the greater the probability of successful attack, φ(t 0d ), can be known,

[0125] a=w(0.5), then by analysis, that is,

[0126]

[0127] Therefore, the network's ability to resist parasitic chain attacks is mainly proportional to the system's event arrival rate. By bringing equations (20) and (21) into (25), we can get,

[0128]

[0129] wherein, t0 is the consensus time when there is no application chain.

[0130] (2-5) Cross-chain interaction efficiency

[0131] This method realizes atomic exchange, asset transfer, and data sharing for cross-chain interaction. Among them, atomic exchange only needs to add two nodes to be exchanged to the same application chain at the same time, and all application chains allow users to join independently. Therefore, the consensus time of cross-chain atomic exchange and in-chain atomic exchange is the same. For data sharing, since all nodes can freely access all blockchains in the network, there is no additional interaction time. Asset transfer needs to reach consensus in the original application chain first, and then reach consensus again in the new application chain. Therefore, asset transfer requires two consensus times. Assuming that the asset is transferred from application chain i to application chain j, the consensus time for asset transfer is the consensus time delay T CiAnd the application chain consensus delay T transferred Cj The sum, that is, T Cpr = T Ci + T Cj .

[0132] (3) The method of directional propagation replaces the rumor algorithm of IOTA, that is, the data in an application chain only needs to be transmitted to the nodes in this chain, that is, the nodes in an application chain each time transmit to his virtual neighbor node.

[0133] The blockchain network is divided into different application chains, and each node can select different application chains according to the needs to join. For the data in an application chain, it only needs to be transmitted to the nodes in this chain. Since there may be a large part of nodes in the network that do not belong to this application chain and do not need the information of this application chain, therefore, the traditional rumor algorithm of IOTA is no longer suitable for the new system model. As shown in Figure 7 Assuming that there are two application chains 1 and application chain 2 in the network, when transmitting data, the data in application chain 1 only needs to be transmitted to the nodes in application chain 1, for example, in application chain 1, the data of node A only needs to be transmitted to its virtual neighbor nodes B, C and D (the nodes to the virtual neighbor nodes are not physically adjacent, and other nodes may be needed for forwarding), and the nodes B, C and D in turn transmit to their virtual neighbor nodes, thereby transmitting to the entire chain. In this way, the network transmission overhead can be greatly reduced, and the network throughput can be improved; at the same time, the network delay of data transmission to the entire application chain is reduced.

[0134] The above only describes the preferred embodiments of the present application, and the protection scope of the present application is not limited to the above embodiments, but any equivalent modifications or changes made by those skilled in the art according to the disclosed content of the present application shall be included in the protection scope recorded in the claims.

Claims

1. A performance analysis method based on IOTA blockchain for a large-scale Internet of Things cross-chain architecture, characterized in that: The method comprises the following steps: Step 1, based on IOTA design blockchain architecture, IOTA blockchain is divided into distributed identity information DID chain and several application chains, a node has different DID, and the DID information is registered in the DID chain, and the node selectively uploads part of personal information in different application chains to protect the personal privacy of the user; for intra-chain transaction, after the device is linked to the blockchain, first, the DID information composed of DID character, application chain to which the device belongs and application chain exclusive ID is generated, and the DID file is created; cross-chain interaction includes atomic transaction, asset transfer and cross-chain data sharing; Intra-chain transaction comprises the following steps: 1) register information in the blockchain; 2) apply for verifiable claim to the identity issuer; 3) the identity issuer verifies the identity of the user, issues a certificate to the user, and stores the certificate summary information on the DID blockchain for verification; 4) when the device needs a certain service, the corresponding verifiable claim is selected and sent to the corresponding application chain to apply for a specific service; 5) the device that provides the service of the application chain verifies the authenticity of the device in the DID chain; after verification succeeds; 6) the service provider sends the verifiable claim of the service provider to the service requester to prove that the corresponding service is provided; 7) the service requester reads the DID blockchain information, verifies the verifiable claim of the service provider, and after verification succeeds; 8) pay and place the transaction record on the corresponding blockchain, and after the transaction is agreed; 9) provide the service; For atomic transaction, when node A wants to transact with node B, the specific process is as follows: 1) add A to the blockchain of the domain where A is located; 2) register with the DID chain; 3) node A applies for transaction and sends verifiable claim to node B to prove that node A has the identity verification information required for transaction with node B; 4) node B reads the information from the DID chain for verification; 5) after verification is completed, node B sends the DID and verifiable claim to node A, and the two nodes can freely transact in the domain where A is located; For asset transfer, when node A wants to transfer its assets to the application chain of other domains, the specific process is as follows: 1) node A transfers the assets corresponding to the address of the application chain in the domain where node A is located to the address of the application chain in other domains; 2) place the transaction in the application chain of the domain where node A is located, and after the transaction is agreed, place the transaction in the application chain of other domains; 3) when verifying the cross-chain transaction in subsequent transactions, it is necessary to check whether the transaction has been agreed in the application chain of the domain where node A is located, and only when the transaction is agreed can the transaction in the application chain of other domains be verified; For cross-chain data sharing, any node can freely view any application chain in IOTA, and the data on each application chain in IOTA is comprehensively utilized; Step 2, analyze the performance of the new architecture from consensus time, throughput, security, cross-chain interaction efficiency; use random network calculus theory to analyze the one-hop delay of nodes, calculate the consensus time of IOTA in an application chain according to the IOTA consensus process, and derive the maximum throughput of the network in the case of high event arrival rate, and calculate the event arrival rate required for a successful parasitic chain attack by a malicious node; Step 3, use the method of directional propagation to replace the rumor algorithm of IOTA; the data in an application chain only needs to be transmitted to the nodes in this chain, i.e. each node in an application chain transmits to his virtual neighbor node each time.

2. The performance analysis method of the cross-chain architecture for large-scale Internet of Things based on IOTA blockchain according to claim 1, characterized in that: In step 2, for the consensus time, the number of nodes in the network is N, there are D = {1, 2, 3, …, D} application chains in the network, and the number of nodes in each application chain is N d , the average forwarding hop number of the node to the virtual neighbor node is K, the average delay of the node to the virtual neighbor node is T d , the consensus time of the network is T Cd ; the total number of end transactions in the network R 0d = 2λ ds h d tends to be stable, wherein, is the total event arrival rate in the application chain d, h d is the average time for an event in the application chain d to write IOTA, and the consensus time of the application chain d is wherein a = ω(0.5) is the Lambert W function, m d is the number of virtual neighbor nodes, p is the probability of successful reception of a one-hop data packet, d the transmission hop number of each node transmitting to the application chain d; Based on a = ω(0.5)≈0.35, K is the average hop number from IOTA node to virtual neighbor node, p≈1 in the case of good network status, C is the data transmission rate of service node; because POW is needed for transaction writing in IOTA, Ch d >1, then, then then wherein is the total number of full m d tree nodes, that is, N d Therefore, η≈1; then where, for any node A in the application chain d, satisfies λ Ai (e-1)- C≤0, The event arrival rate of node i is λ i The event arrival rate of the through-flow from node i to the virtual neighbor node j is When not divided into application chains, the consensus time of the network is wherein, is the total event arrival rate of the network, h is the latency required to write to the Tangle, L is the number of hops required to broadcast to the full network after a transaction is written to the local Tangle; and for a node A in the network, satisfies Ai (e-1)-C≤0, C is the service rate constant; T is the one-hop latency; When the event arrival rate in the network is constant, the consensus time is related to the events of the transaction transmission in the network. The more the selected neighbor nodes, the shorter the transaction transmission time. Therefore, under the condition of satisfying λ A (e-1)-C≤0, , the greater m d or the smaller L d , the shorter the consensus time. m is the number of neighbor nodes when the partial application chain is used.

3. The performance analysis method of a cross-chain architecture based on IOTA blockchain for a large-scale Internet of Things according to claim 2, characterized in that: In step 2, for the system throughput, since the IOTA transaction consensus is that the later transaction verifies the previous transaction, for the application chain d, the total number of Tips in the network is R 0d , the consensus time of the network is T Cd , then the throughput of the application chain d is Since there are D application chains in the network, the total throughput of the system is In the traditional IOTA blockchain, the throughput of the network is From the analysis, the greater m d or m, the shorter the consensus time, and the greater R 0d and R0 are independent of m d or m, so the greater m d or m, the greater the system throughput; When m d The throughput of the network increases as the total event arrival rate in the network increases; for any node A in the network, satisfies λ A (e-1)-C≤0, since each node needs to forward all transactions in the network, the amount of transactions each node needs to forward is λ ds i.e. Similarly, when not divided into application chains, Therefore, the maximum throughput of the system is The maximum throughput of the traditional IOTA network is 4. The performance analysis method of the cross-chain architecture for large-scale Internet of Things based on IOTA blockchain according to claim 3, characterized in that: In step 2, for security, the purpose is to solve the double-spend problem, and for application chain d, the probability of a successful parasitic chain attack is wherein is the cumulative weight of transactions that have reached consensus at time t, is the probability that the next transaction in application chain d is an honest transaction; λ d is the total event arrival rate of honest transactions in application chain d, μ is the event arrival rate of the parasite chain; t 0d is the consensus time; When p≤0.5, if you want to attack successfully, you need the event arrival rate of malicious nodes to be greater than or equal to the event arrival rate of the entire network. When p > 0.5, μ(t) 0d )-φ(t 0d The larger φ(t) is, the greater the probability of a successful attack. 0d The larger the value, the lower the probability of a successful attack. a = w(0.5) Then from the analysis That is, Therefore, the network's ability to resist parasitic chain attacks is mainly proportional to the system's event arrival rate, and the equation is brought in, Where t0 is the consensus time when not divided into application chains.

5. The performance analysis method of a cross-chain architecture for a large-scale Internet of Things based on an IOTA blockchain according to claim 1, characterized in that: In step 2, for cross-chain interaction efficiency, asset transfer requires two consensus times, assuming that the asset is transferred from application chain i to application chain j, the consensus time of asset transfer is the sum of the consensus time T of the original application chain and the consensus time T of the new application chain, that is, T = T + T Ci Cj Cpr Ci Cj .​​​​ 6. The performance analysis method of a cross-chain architecture for a large-scale Internet of Things based on an IOTA blockchain according to claim 1, characterized in that: In step 3, for the two application chains 1 and 2 in the network, when transmitting data, the data in application chain 1 only needs to be transmitted to the nodes in application chain 1.

Images