Elastic consistency optimization method and device under agent honeycomb architecture

Abstract

The invention relates to the field of intelligent agent honeycomb architecture technology application, and particularly provides an elastic consistency optimization method and device under an intelligent agent honeycomb architecture, the method is applied to distributed network communication and an intelligent cloud platform honeycomb architecture, and the method comprises the following steps: S1, network attack modeling; injecting error data into a network attack model: xi (k + 1) = fi (xi (k)); xi (k) represents information sent by the agent i to a neighbor agent, and fi (xi (k)) represents any updating mode; a consensus mechanism is introduced into the intelligent agent honeycomb architecture, and a layered topological structure is carried out; s2, assuming that the protected nodes form a dominant connected subset of the network by the agent composition; and S3, carrying out convergence verification on the elastic consistency optimization method. Compared with the prior art, the method can solve the problem that the agents under the honeycomb architecture are consistent in an unreliable communication network.

Description

Technical Field

[0001] The present invention relates to the application field of intelligent agent honeycomb architecture technology, and specifically provides a method and device for optimizing elastic consistency under an intelligent agent honeycomb architecture. Background Art

[0002] The intelligent honeycomb architecture model uses a large number of large-model reasoning agents that work together to solve a problem, with each agent contributing from its own unique perspective. The resulting results demonstrate a collective intelligence that goes beyond what any single artificial intelligence individual can achieve, especially when dealing with complex data sets and scenarios.

[0003] Although the intelligent agent honeycomb architecture pattern has advantages in flexibility, adaptability and compatibility of swarm intelligence, it is challenging to solve the problem of how intelligent agents under the honeycomb architecture can reach a consensus in an unreliable communication network. Summary of the invention

[0004] The present invention aims at the above-mentioned deficiencies of the prior art and provides a highly practical elastic consistency optimization method under an intelligent agent honeycomb architecture.

[0005] A further technical task of the present invention is to provide a rationally designed, safe and applicable elastic consistency optimization device under the intelligent body honeycomb architecture.

[0006] The technical solution adopted by the present invention to solve its technical problem is:

[0007] A method for optimizing elastic consistency under an intelligent agent honeycomb architecture, which is applied to distributed network communication and an intelligent cloud platform honeycomb architecture, has the following steps:

[0008] S1. Network attack modeling;

[0009] Error data injection network attack model:

[0010] x i (k+1)=f i (x i (k));

[0011] Among them, x i (k) represents the information sent by agent i to neighboring agents, f i (x i (k)) indicates any update mode;

[0012] The intelligent honeycomb architecture introduces a consensus mechanism to create a hierarchical topology;

[0013] S2. Agent Composition Assuming that these protected nodes constitute the dominant connected subset of the network, the elastic consistency optimization method is as follows:

[0014] The input is the set of protected node numbers in the network, represented by T = {v i ,|i=1,2,...,n1}, the ordinal set of other nodes is represented as O={v i ,|i=n1+1,...,n0}(n0=n1+n2), the attack node sequence set is represented as A={v i ,|i=n0+1,...,n} and its corresponding status information;

[0015] S3. Verify the convergence of the elastic consistency optimization method.

[0016] Further, in step S1, the consensus mechanism is that the virtual leader layer has one or more leaders, there is no communication between leaders, and only sends information to the nodes in the first layer. By dividing n nodes into m layers, marked as 1 to m from top to bottom, the nodes in the first layer only send information to the second layer, and the nodes in the i-th layer send information to at least one node in the i+1th layer. The information transmission between the nodes in the i-th layer can be bidirectional. Assuming that each layer has r i nodes, then there are r1+r2+K+r m =n.

[0017] Furthermore, the communication network of the intelligent agent honeycomb architecture is represented by an undirected graph G = (V, E, A), and the node i of the intelligent agent has a scalar state at time k, which is expressed as The states of all nodes in the system are represented by the vector x = [x1, x2, K, x n ] T express;

[0018] If graph G = subset G of (V, E, A) d , each not belonging to G d A node has at least one neighbor that belongs to G d , and belongs to G d All nodes of G form a connected graph, G d It is a dominant connected subset of G = (V, E, A).

[0019] Further, it specifically includes:

[0020] S2.1. Each node v i Receive the state of neighbors and form a set

[0021] S2.2, node v i By identifying the status information of the trusted node and combining it with the status information of its own node, the minimum threshold is selected. and the maximum threshold

[0022] S2.3. Judgment set S i The element x in (k) j (k) Whether it meets The set R

[0023] S2.4. Each node v i The update law is:

[0024]

[0025] x j (k)∈R i (k);

[0026] S2.5, repeat steps S2.1-S2.4 until |x i (k)-x j (k)|<ε,v j ∈T i .

[0027] Furthermore, in step S3, each protected node or common node v i , determine the minimum threshold based on the information of the protected node in the neighbor and its own information and the maximum threshold The set R is obtained by threshold screening i (k), node v i In the kth iteration, according to R i The source of the agent in (k) divides the state into three types: The state of the node itself x i (k), and the set from normal nodes or attack nodes

[0028] Further, according to formula (1), it can be rewritten as:

[0029]

[0030] in, The information can be represented by the minimum threshold and the maximum threshold, that is, there exists 0<ρ j <1, satisfying the following equation

[0031] Furthermore, formula (2) can be further rewritten as:

[0032]

[0033] remember but

[0034]

[0035] Among them, M 11 represents the interaction between trust nodes, M 21 Indicates the role of trusted nodes to ordinary nodes, M 22 Indicating the role of the node itself, the consensus mechanism divides the nodes of the communication network into three layers, namely the first layer consisting of trusted nodes, the second layer consisting of ordinary nodes, and the third layer consisting of attack nodes.

[0036] Furthermore, to facilitate the analysis Represents the trusted node status and its own status, then

[0037] It is not difficult to see that M(k) is a row random matrix. The elements in M(k) are non-negative, and the non-zero elements have a lower bound τ. And d M For R i (k) The maximum value of the cardinality, i.e.

[0038] Furthermore, the elastic consistency of the formula mechanism is analyzed:

[0039]

[0040] According to the properties of row random matrices, we have:

[0041] in, is a random vector and is independent of t. In the topology reconstruction-based method, the states of the nodes will reach a consensus and converge to

[0042] An elastic consistency optimization device under an intelligent agent honeycomb architecture includes: at least one memory and at least one processor;

[0043] The at least one memory is used to store a machine-readable program;

[0044] The at least one processor is used to call the machine-readable program to execute a flexible consistency optimization method under an intelligent agent honeycomb architecture.

[0045] Compared with the prior art, the elastic consistency optimization method and device under the intelligent agent honeycomb architecture of the present invention have the following outstanding beneficial effects:

[0046] The intelligent agent honeycomb architecture model of the present invention uses a large number of intelligent agents based on large model reasoning. These intelligent agents work together to solve a problem, and each intelligent agent contributes from its own unique perspective. The resulting comprehensive results show a manifestation of collective intelligence, and its effect exceeds the results that any single artificial intelligence individual can achieve, especially in processing complex data sets and a large number of scenarios. It can solve the problem of intelligent agents under the honeycomb architecture reaching consensus in an unreliable communication network. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0048] Attached Figure 1 It is a schematic diagram of the intelligent agent honeycomb structure in the elastic consistency optimization method under the intelligent agent honeycomb architecture;

[0049] Attached Figure 2 This is a schematic diagram of the consensus mechanism under the honeycomb architecture in the elastic consistency optimization method under the intelligent agent honeycomb architecture. DETAILED DESCRIPTION

[0050] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with specific implementation methods. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0051] A best embodiment is given below:

[0052] like Figure 1-2 As shown, a method for optimizing elastic consistency under an intelligent agent honeycomb architecture in this embodiment has the following steps:

[0053] S1. Network attack modeling;

[0054] False data injection (FDI) attacks are easy to implement, with strong concealment and interference characteristics, posing a potential threat to the consensus decision-making of multiple intelligent agents under the cellular architecture. Under the condition of knowing the current system configuration, the attacker can tamper with the sensor's measurement data or the controller's control input signal through careful design, thereby misleading the system's state estimation and causing the system's state to deviate from stability.

[0055] Consider the following false data injection network attack model:

[0056] x i (k+1)=f i (x i (k))

[0057] Among them, x i (k) represents the information sent by agent i to neighboring agents, f i (x i (k)) indicates any update method.

[0058] The consensus mechanism of the intelligent honeycomb architecture has an important impact. Consider a hierarchical topology. Figure 2 As shown in the figure, the virtual leader layer has one or more leaders. There is no communication between leaders, and they only send information to the nodes in the first layer. By dividing n nodes into m layers, labeled 1 to m from top to bottom. The nodes in the first layer only send information to the second layer, and the nodes in the i-th layer send information to at least one node in the i+1th layer. The information transmission between nodes in the i-th layer can be bidirectional. Assume that each layer has r i nodes, then there are r1+r2+K+r m =n.

[0059] Considering the process of the intelligent honeycomb architecture being injected with false data, by adjusting the original communication topology, a consensus mechanism model with one-way information transmission is constructed. The hierarchical topology reduces the impact of attack nodes on the network and improves the topological robustness of the network. The communication network of the intelligent honeycomb architecture uses an undirected graph G = (V, E, A). The node i of the intelligent agent has a scalar state at time k, which is expressed as The states of all nodes in the system are represented by the vector x = [x1, x2, K, x n ] T express.

[0060] If a graph is represented by G = (V, E, A), its subset G d Each of the d A node has at least one neighbor that belongs to G d , and belongs to G d All nodes of G form a connected graph, G d It is a dominant connected subset of G = (V, E, A).

[0061] S2. Agent Composition Assuming that these protected nodes constitute the dominant connected subset of the network, the elastic consistency optimization method is as follows:

[0062] Under the attack of false data injection network, trust nodes can be set up to prevent being invaded by attacking nodes. Protection measures include: ① Usually improve the firewall security level of some nodes, perform redundant backup of node resources, and ensure that nodes obtain authentic and reliable information; ② Use digital signatures and data encryption technology between nodes to prevent the real information sent by these nodes from being tampered with.

[0063] These partial nodes constitute the assumption that these protected nodes constitute the dominant connected subset of the network, then the consistency optimization method is as follows:

[0064] The input of this method is the set of protected node ordinal numbers in the network, represented as T = {v i ,|i=1,2,...,n1}, the ordinal set of other nodes is represented as O={v i ,|i=n1+1,...,n0}(n0=n1+n2), the attack node sequence set is expressed as and its corresponding status information.

[0065] Specifically include:

[0066] S2.1. Each node v i Receive the state of neighbors and form a set

[0067] S2.2, node v i By identifying the status information of the trusted node and combining it with the status information of its own node, the minimum threshold is selected. and the maximum threshold

[0068] S2.3. Judgment set S i The element x in (k) j (k) Whether it meets The set R

[0069]

[0070] S2.4. Each node v i The update law is:

[0071]

[0072] x j (k)∈R i (k);

[0073] S2.5, repeat steps S2.1-S2.4 until |x i (k)-x j (k)|<ε,v j ∈T i .

[0074] S3. Verify the convergence of the elastic consistency optimization method.

[0075] Each protected node or ordinary node v i , determine the minimum threshold based on the information of the protected node in the neighbor and its own information and the maximum threshold The set R is obtained by threshold screening i (k). Node v i In the kth iteration, according to R i The source of the agent in (k) divides the state into three types: The state of the node itself x i (k), and the set from normal nodes or attack nodes

[0076] According to formula (1), it can be rewritten as:

[0077]

[0078] in, The information can be represented by the minimum threshold and the maximum threshold, that is, there exists 0<ρ j <1, satisfying the following equation

[0079] Formula (2) can be further rewritten as:

[0080]

[0081] remember but

[0082]

[0083] Among them, M 11 represents the interaction between trust nodes, M 21 Indicates the role of trusted nodes to ordinary nodes, M 22 Indicating the role of the node itself, the consensus mechanism divides the nodes of the communication network into three layers, namely the first layer consisting of trusted nodes, the second layer consisting of ordinary nodes, and the third layer consisting of attack nodes.

[0084] For the convenience of analysis Represents the trusted node status and its own status, then

[0085] It is not difficult to see that M(k) is a row random matrix. The elements in M(k) are non-negative, and the non-zero elements have a lower bound τ. And d M For R i (k) The maximum value of the cardinality, i.e.

[0086] Furthermore, the elastic consistency of the formula mechanism is analyzed:

[0087]

[0088] According to the properties of row random matrices, we have:

[0089] in, is a random vector and is independent of t. In the above topology reconstruction-based method, the states of the nodes will reach a consensus and converge to

[0090] Based on the above method, a flexible consistency optimization device under an intelligent agent honeycomb architecture in this embodiment includes: at least one memory and at least one processor;

[0091] at least one memory for storing a machine-readable program;

[0092] At least one processor is used to call the machine-readable program to execute a flexible consistency optimization method under an intelligent agent honeycomb architecture.

[0093] The above-mentioned specific implementations are only specific cases of the present invention. The patent protection scope of the present invention includes but is not limited to the above-mentioned specific implementations. Any technical solutions that conform to the above-mentioned specific implementations of the present invention and any appropriate changes or substitutions made by ordinary technicians in the relevant technical field shall fall within the patent protection scope of the present invention.

[0094] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for optimizing elastic consistency under an intelligent honeycomb architecture, characterized in that: The method is applied to distributed network communication and intelligent cloud platform honeycomb architecture, and has the following steps: S1. Network attack modeling; Error data injection network attack model: x i (k+1)=f i (x i (k)); Among them, x i (k) represents the information sent by agent i to neighboring agents, f i (x i (k)) indicates any update mode; The intelligent honeycomb architecture introduces a consensus mechanism to create a hierarchical topology; S2. Agent Composition Assuming that these protected nodes constitute the dominant connected subset of the network, the elastic consistency optimization method is as follows: The input is the set of protected node numbers in the network, represented by T = {v i ,|i=1,2,...,n1}, the ordinal set of other nodes is represented as O={v i ,|i=n1+1,...,n0}(n0=n1+n2), the attack node sequence set is represented as A={v i ,|i=n0+1,...,n} and its corresponding status information; S3. Verify the convergence of the elastic consistency optimization method.

2. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 1 is characterized in that: In step S1, the consensus mechanism is that the virtual leader layer has one or more leaders, there is no communication between the leaders, and only sends information to the nodes in the first layer. By dividing n nodes into m layers, marked as 1 to m from top to bottom, the nodes in the first layer only send information to the second layer, the nodes in the i-th layer send information to at least one node in the i+1th layer, and the nodes in the i+2nd layer send information to at least one node in the i+1th layer. i The information transmission between nodes in the layer can be bidirectional. Assume that each layer has r i nodes, then there are r1+r2+K+r m =n.

3. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 2 is characterized in that: The communication network of the intelligent agent honeycomb architecture is represented by an undirected graph G = (V, E, A). The node i of the intelligent agent has a scalar state at time k, which is expressed as The states of all nodes in the system are represented by the vector x = [x1, x2, K, x n ] T express; If graph G = subset G of (V, E, A) d , each not belonging to G d A node has at least one neighbor that belongs to G d , and belongs to G d All nodes of G form a connected graph, G d It is a dominant connected subset of G = (V, E, A).

4. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 3 is characterized in that: Specifically include: S2.

1. Each node v i Receive the state of neighbors and form a set S2.2, node v i By identifying the status information of the trusted node and combining it with the status information of its own node, the minimum threshold is selected. and the maximum threshold S2.

3. Judgment set S i The element x in (k) j (k) Whether it is satisfied The set R S2.

4. Each node v i The update law is: x j (k)∈R i (k); S2.5, repeat steps S2.1-S2.4 until |x i (k)-x j (k)|<ε,v j ∈T i .

5. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 4 is characterized in that: In step S3, each protected node or common node v i , determine the minimum threshold based on the information of the protected node in the neighbor and its own information and the maximum threshold The set R is obtained by threshold screening i (k), node v i In the kth iteration, according to R i The source of the agent in (k) divides the state into three types: The state of the node itself x i (k), and the set from normal nodes or attack nodes 6. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 5 is characterized in that: According to formula (1), it can be rewritten as: in, The information can be represented by the minimum threshold and the maximum threshold, that is, there exists 0<ρ j <1, satisfying the following equation 7. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 6 is characterized in that: Formula (2) can be further rewritten as: remember but Among them, M 11 represents the interaction between trust nodes, M 21 Indicates the role of trusted nodes to ordinary nodes, M 22 Indicating the role of the node itself, the consensus mechanism divides the nodes of the communication network into three layers, namely the first layer consisting of trusted nodes, the second layer consisting of ordinary nodes, and the third layer consisting of attack nodes.

8. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 7 is characterized in that: For the convenience of analysis Represents the trusted node status and its own status, then It is not difficult to find that M(k) is a row random matrix. The elements in M(k) are non-negative, and the non-zero elements have a lower bound τ, and τ = 1 / (d M +1), d M For R i (k) The maximum cardinality, i.e. d M =1 / max{|R1(k)|,K,|R n0 (k)|}.

9. The elastic consistency optimization method under the intelligent agent honeycomb architecture according to claim 7 is characterized in that: Analyze the elastic consistency of the formula mechanism: According to the properties of row random matrices, we have: in, is a random vector and is independent of t. In the topology reconstruction-based method, the states of the nodes will reach a consensus and converge to 10. An elastic consistency optimization device under an intelligent agent honeycomb architecture, characterized in that: include: at least one memory and at least one processor; The at least one memory is used to store a machine-readable program; The at least one processor is configured to call the machine-readable program to execute the method according to any one of claims 1 to 9.

Images