Optimized routing strategy applied to scale-free network

Abstract

The invention discloses an optimized routing strategy applied to a scale-free network, relates to the technical field of network communication, and solves the technical problem that the survivability of the whole network is difficult to improve through the routing strategy when the network is attacked or has a fault. The method comprises the following steps: introducing an NCI network constraint index and an adjustable parameter of a routing function through a node-based degree, constructing the routing function of a scale-free network, and defining the weight of an edge between nodes through the constructed routing function of the scale-free network; the characteristic of the scale-free network enables a small number of'hub 'nodes which are highly connected to bear most flow, and the structure shows high robustness when facing random faults. The key nodes are better utilized by adjusting a routing strategy, so that the survivability of the whole network under attack or fault is improved, and the importance of each node can be more reasonably evaluated by taking the NCI as a constraint condition, so that overload is avoided, and the overall performance is improved.

Description

Technical Field

[0001] The invention belongs to the field of network communication, and in particular relates to an optimized routing strategy applied to a scale-free network. Background Art

[0002] With the rapid advancement of science and technology, the load in the network continues to rise, such as the increasing traffic volume in the traffic network, countless messages, photos, and video sharing in social networks. The network has the ability to process data volume. When it exceeds its capacity, congestion will occur. This phenomenon will gradually spread from the local part of the network to the entire network, causing network paralysis. With the progress of human society, the load flow is also increasing, and the network is becoming more and more complex. As a result, the research on the robustness of the network has also received attention. Because after one or several nodes in the network are attacked, it may cause more nodes to crash or even the entire network to paralyze.

[0003] At present, the research on network robustness mainly focuses on the network's resistance to random attacks and intentional attacks. However, there is little research on the robustness of routing strategies, and it is difficult to improve the survivability of the entire network when it is attacked or faulty through routing strategies. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art; to this end, the present invention proposes an optimized routing strategy applied to scale-free networks, which is used to solve the technical problem that it is difficult to improve the survivability of the entire network when it is attacked or fails through a routing strategy.

[0005] To solve the above problems, a first aspect of the present invention provides an optimized routing strategy applied to a scale-free network, comprising the following steps:

[0006] Construct a BA scale-free network model and define the capacity setting rules for a single node;

[0007] Based on the degree of nodes, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network. The weights of the edges between nodes are defined by the constructed routing function of the scale-free network.

[0008] Conduct random and deliberate attacks on the nodes in the constructed scale-free network, count the number of network nodes before and after the cascading failure of the scale-free network, and calculate the ratio of the number of network nodes before and after the cascading failure;

[0009] The robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascading failure, and the capacity setting rules of a single node are adjusted according to the robustness of the routing strategy.

[0010] As a further solution of the present invention: defining a capacity setting rule for a single node includes the following steps:

[0011] Set the node capacity to:

[0012] c i =βB i (0),β≥1

[0013] Among them, c i is the capacity of node i, B i (0) is the initial load of node i, that is, the betweenness of the node, and β is an adjustable parameter of node capacity.

[0014] As a further solution of the present invention: based on the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network, including the following steps:

[0015] According to the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function, the probability of a data packet waiting on the routing path is obtained, and the routing function of the scale-free network is constructed by the following formula:

[0016]

[0017] Among them, P is the probability of a data packet waiting on the routing path, k i is the degree of node i, α is an adjustable parameter, and NCI is the C i Value, C i =∑C ij ;

[0018] Among them, node j is the neighbor node of node i, C ij is the constraint index between node i directly or indirectly and node j, and the formula is as follows:

[0019]

[0020] Among them, node f is the third node besides node i and node j, C ij is the constraint index of node i on node j, U ij The number of paths from node i to node j, U if The number of paths that pass through node i to reach node j, U fi The number of paths that go through node i to reach node j.

[0021] As a further solution of the present invention: the variable parameter α is set by the following steps:

[0022] Set an initial value for the variable parameter α, and the initial value of the variable parameter α is set to 1;

[0023] The network transmission capacity is defined as:

[0024] F=∑ (i,j) R ij P ij

[0025]

[0026] Where F is the network traffic between node i and node j, R ij is the routing probability between node i and node j, ε is a constant;

[0027] With the constraint that the node load does not exceed its capacity, the optimal variable parameter α is solved by the gradient descent method;

[0028] As a further solution of the present invention: solving the optimal variable parameter α by the gradient descent method includes the following steps:

[0029] The gradient of the optimal variable parameter α is calculated as:

[0030]

[0031] Set the learning rate η to 0.05, and α n+1 =α n +ηg(α n );

[0032] Repeat the above steps until the update amplitude of the parameter α is less than the preset threshold value, and the α value at this time is the optimal variable parameter α.

[0033] As a further solution of the present invention: defining the weights of the edges between nodes by the constructed routing function of the scale-free network comprises the following steps:

[0034] According to the routing function of the constructed scale-free network, the weight of the edge between nodes is defined by the following formula:

[0035]

[0036] Among them, q ij is the edge between node i and node j, P i is the probability that a data packet waits on all routing paths of node i, P j is the probability that a data packet waits on all routing paths of node j.

[0037] As a further solution of the present invention: performing random attacks and deliberate attacks on nodes in the constructed scale-free network, counting the number of network nodes before and after the cascading failure of the scale-free network, and calculating the ratio of the number of network nodes before and after the cascading failure, including the following steps:

[0038] Construct a scale-free network model;

[0039] The local effective routing strategy method is applied on the constructed scale-free network, and the transmission capacity of the network under this routing strategy is obtained;

[0040] Conduct random and deliberate attacks on nodes in the network;

[0041] Simulate the transmission capacity of the network after the attack under the constructed routing strategy;

[0042] The number of network nodes before and after the cascading failure is defined as S 0 and S 1 , S 1 With S 0 The ratio is defined as the ratio H of the number of network nodes before and after the cascading failure.

[0043] As a further solution of the present invention: analyzing the robustness of the network under the routing strategy by the ratio of the number of network nodes before and after the cascade failure, and adjusting the capacity setting rule of a single node according to the robustness of the routing strategy, including the following steps:

[0044] The ratio of the number of network nodes before and after the cascading failure is counted. When the ratio of the number of network nodes before and after the cascading failure is greater than a preset threshold range, the robustness of the routing strategy is good and there is no need to adjust the capacity of the node. Otherwise, the rule is set according to adjusting the capacity of a single node.

[0045] The capacity setting rules for a single node are as follows:

[0046] When the ratio of the number of network nodes before and after the cascading failure is less than the preset threshold range, the node capacity adjustable parameter β increases by 2%β;

[0047] When the ratio of the number of network nodes before and after the cascading failure is within a preset threshold range, the node capacity adjustable parameter β is increased by 1% β.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention introduces NCI network constraint indicators and adjustable parameters of routing functions based on node degrees to construct a routing function of a scale-free network, and defines the weights of edges between nodes through the constructed routing function of the scale-free network; the characteristics of the scale-free network enable a small number of highly connected "hub" nodes to carry most of the traffic, and this structure exhibits strong robustness in the face of random failures. By adjusting the routing strategy, these key nodes can be better utilized, thereby improving the survivability of the entire network when it is attacked or fails.

[0050] The present invention can more reasonably evaluate the importance of each node by taking NCI as a constraint condition, which means that traffic can be dynamically allocated according to the importance of the node, thereby avoiding overload and improving overall performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] 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 only 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.

[0052] Figure 1 It is a schematic diagram of the process of the present invention. DETAILED DESCRIPTION

[0053] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. 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.

[0054] See also Figure 1 The first aspect of the present invention provides an optimized routing strategy for a scale-free network, comprising the following steps:

[0055] Construct a BA scale-free network model and define the capacity setting rules for a single node;

[0056] Based on the degree of nodes, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network. The weights of the edges between nodes are defined by the constructed routing function of the scale-free network.

[0057] Conduct random and deliberate attacks on the nodes in the constructed scale-free network, count the number of network nodes before and after the cascading failure of the scale-free network, and calculate the ratio of the number of network nodes before and after the cascading failure;

[0058] The robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascading failure, and the capacity setting rules of a single node are adjusted according to the robustness of the routing strategy.

[0059] Specifically, the BA scale-free network model: In many networks such as the WWW network and the protein interaction network, there is a phenomenon to varying degrees: most nodes have only a few edges with other nodes, while some nodes have hundreds or tens of millions of edges. Networks with this phenomenon are called scale-free networks. Unlike the two networks introduced above, scale-free networks are non-homogeneous. In layman's terms, a small number of nodes have a large number of edges while many nodes have only a few edges. Scale-free networks have better recovery capabilities when attacked, that is, the network elasticity is higher, but the problem of reduced network efficiency is inevitable.

[0060] In this embodiment, a scale-free network model is established. Then, a routing strategy algorithm is established, and a new indicator is introduced at the cost of node degree and NCI. The new indicator is used to define the weight of the edge between nodes in the network, thereby determining the routing table for data packet transmission. At the same time, a parameter α is also added, so that the parameter α with the largest network transmission capacity is the optimal parameter.

[0061] Based on the degree of nodes, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network. The weight of the edge between nodes is defined by the constructed routing function of the scale-free network. The characteristics of the scale-free network enable a small number of highly connected "hub" nodes to carry most of the traffic. This structure shows strong robustness in the face of random failures. By adjusting the routing strategy, these key nodes can be better utilized, thereby improving the survivability of the entire network when attacked or failed.

[0062] By using NCI as a constraint, the importance of each node can be evaluated more reasonably, which means that traffic can be dynamically allocated according to the importance of the node, thereby avoiding overload and improving overall performance.

[0063] Through random attacks and deliberate attacks on nodes in the constructed scale-free network, the number of network nodes before and after the cascading failure of the scale-free network is counted, and the ratio of the number of network nodes before and after the cascading failure is calculated; the robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascading failure, and the capacity setting rules of a single node are adjusted according to the robustness of the routing strategy. By observing the node survival rate under different attack modes, the network's resistance to various types of attacks can be quantified. A high ratio indicates that the network can effectively maintain connections, while a low ratio indicates potential vulnerability. Analyzing the ratio of the number of nodes before and after the cascading failure is convenient for helping to analyze the performance of the routing strategy in the face of a specific type of attack, and is convenient for providing a basis for subsequent adjustment of the routing strategy. According to the analysis results, the capacity setting rules of a single node are adjusted in real time. If it is found that the survival rate of some key nodes after the attack is low, it can be considered to increase the processing power or redundant configuration of these nodes to enhance their carrying capacity and stability.

[0064] In one embodiment of the present invention, defining a capacity setting rule for a single node includes the following steps:

[0065] Set the node capacity to:

[0066] c i =βB i (0),β≥1

[0067] Among them, c i is the capacity of node i, B i (0) is the initial load of node i, that is, the betweenness of the node, and β is an adjustable parameter of node capacity.

[0068] Specifically, degree is certainly an indicator that can measure the importance of a node, but not entirely. For example, in a transportation network, there are two regions with well-developed internal transportation, but there is a place with only two roads leading to the outside, but this place is located on the only road connecting the two regions. The degree value of the node representing this location is obviously very small, but if this node is removed, the two regions will not be able to communicate with each other, so this node plays a very important role in the network. At this time, another parameter from a global perspective needs to be defined, that is, betweenness. After a certain amount of expansion, when the routing strategy is selected, it is assumed that the number of all valid paths in the network is δ sd , where the number of valid paths passing through node i is δ sd (i), then the effective betweenness of node i is defined as:

[0069]

[0070] In one embodiment of the present invention, based on the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network, including the following steps:

[0071] According to the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function, the probability of a data packet waiting on the routing path is obtained, and the routing function of the scale-free network is constructed by the following formula:

[0072]

[0073] Among them, P is the probability of a data packet waiting on the routing path, k i is the degree of node i, α is an adjustable parameter, and NCI is the C i Value, C i =∑C ij ;

[0074] Among them, node j is the neighbor node of node i, C ij is the constraint index between node i directly or indirectly and node j, and the formula is as follows:

[0075]

[0076] Among them, node f is the third node besides node i and node j, C ij is the constraint index of node i on node j, U ij The number of paths from node i to node j, U if The number of paths from node i to node j, U fi The number of paths that go through node i to reach node j.

[0077] Specifically, degree and degree distribution: As a node information that cannot be ignored, degree is an extremely important parameter, which indicates the number of first-order neighbors a node has, usually represented by the letter k. The number of first-order neighbors a node has determines the importance of this node. The adjacency matrix of the network is A, and the degree of node i can be obtained by the sum of the elements in the i-th row or i-th column of A, that is:

[0078] Among them, a ij is the element in row i and column j of A.

[0079] In one embodiment of the present invention, the variable parameter α is set by the following steps:

[0080] Set an initial value for the variable parameter α, and the initial value of the variable parameter α is set to 1;

[0081] The network transmission capacity is defined as:

[0082] F=∑ (i,j) R ij P ij

[0083]

[0084] Where F is the network traffic between node i and node j, R ij is the routing probability between node i and node j, ε is a constant;

[0085] With the constraint that the node load does not exceed its capacity, the optimal variable parameter α is solved by the gradient descent method;

[0086] In one embodiment of the present invention, solving the optimal variable parameter α by the gradient descent method includes the following steps:

[0087] The gradient of the optimal variable parameter α is calculated as:

[0088]

[0089] Set the learning rate η to 0.05, and α n+1 =α n +ηg(α n );

[0090] Repeat the above steps until the update amplitude of the parameter α is less than the preset threshold value, and the α value at this time is the optimal variable parameter α.

[0091] In one embodiment of the present invention, defining the weights of the edges between nodes by constructing a routing function of a scale-free network includes the following steps:

[0092] According to the routing function of the constructed scale-free network, the weight of the edge between nodes is defined by the following formula:

[0093]

[0094] Among them, q ij is the edge between node i and node j, P i is the probability that a data packet waits on all routing paths of node i, P j is the probability that a data packet waits on all routing paths of node j.

[0095] In one embodiment of the present invention, random attacks and deliberate attacks are performed on nodes in a constructed scale-free network, the number of network nodes before and after a cascading failure occurs in the scale-free network is counted, and the ratio of the number of network nodes before and after the cascading failure is calculated, including the following steps:

[0096] Construct a scale-free network model;

[0097] The local effective routing strategy method is applied on the constructed scale-free network, and the transmission capacity of the network under this routing strategy is obtained;

[0098] Conduct random and deliberate attacks on nodes in the network;

[0099] Simulate the transmission capacity of the network after the attack under the constructed routing strategy;

[0100] The number of network nodes before and after the cascading failure is defined as S 0 and S 1 , S 1 With S 0 The ratio is defined as the ratio H of the number of network nodes before and after the cascading failure.

[0101] In one embodiment of the present invention, the robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascade failure, and the capacity setting rule of a single node is adjusted according to the robustness of the routing strategy, including the following steps:

[0102] The ratio of the number of network nodes before and after the cascading failure is counted. When the ratio of the number of network nodes before and after the cascading failure is greater than a preset threshold range, the robustness of the routing strategy is good and there is no need to adjust the capacity of the node. Otherwise, the rule is set according to adjusting the capacity of a single node.

[0103] The capacity setting rules for a single node are as follows:

[0104] When the ratio of the number of network nodes before and after the cascading failure is less than the preset threshold range, the node capacity adjustable parameter β increases by 2%β;

[0105] When the ratio of the number of network nodes before and after the cascading failure is within a preset threshold range, the node capacity adjustable parameter β is increased by 1% β.

[0106] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An optimized routing strategy applied to scale-free networks, characterized in that: The following steps are involved: Construct a BA scale-free network model and define the capacity setting rules for a single node; Based on the degree of nodes, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network. The weights of the edges between nodes are defined by the constructed routing function of the scale-free network. Conduct random and deliberate attacks on the nodes in the constructed scale-free network, count the number of network nodes before and after the cascading failure of the scale-free network, and calculate the ratio of the number of network nodes before and after the cascading failure; The robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascading failure, and the capacity setting rules of a single node are adjusted according to the robustness of the routing strategy.

2. The optimized routing strategy for scale-free networks according to claim 1, characterized in that: Defining the capacity setting rules for a single node includes the following steps: Set the node capacity to: c i =βB i (0),β≥1 Among them, c i is the capacity of node i, B i (0) is the initial load of node i, that is, the betweenness of the node, and β is an adjustable parameter of node capacity.

3. The optimized routing strategy for scale-free networks according to claim 1, characterized in that: Based on the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function are introduced to construct the routing function of the scale-free network, which includes the following steps: According to the degree of the node, the NCI network constraint index and the adjustable parameters of the routing function, the probability of a data packet waiting on the routing path is obtained, and the routing function of the scale-free network is constructed by the following formula: Among them, P is the probability of a data packet waiting on the routing path, k i is the degree of node i, α is an adjustable parameter, and NCI is the C i Value, C i =∑C ij ; Among them, node j is the neighbor node of node i, C ij is the constraint index between node i directly or indirectly and node j, and the formula is as follows: Among them, node f is the third node besides node i and node j, C ij is the constraint index of node i on node j, U ij The number of paths that pass through node i to reach node j, U if The number of paths from node i to node j, U fi The number of paths that go through node i to reach node j.

4. The optimized routing strategy for scale-free networks according to claim 3, characterized in that: The variable parameter α is set by the following steps: Set an initial value for the variable parameter α, and the initial value of the variable parameter α is set to 1; The network transmission capacity is defined as: F=∑ (i,j) R ij P ij Where F is the network traffic between node i and node j, R ij is the routing probability between node i and node j, ε is a constant; With the constraint that the node load does not exceed its capacity, the optimal variable parameter α is solved by the gradient descent method.

5. The optimized routing strategy for scale-free networks according to claim 4, characterized in that: Solving the optimal variable parameter α by the gradient descent method includes the following steps: The gradient of the optimal variable parameter α is calculated as: Set the learning rate η to 0.05, and α n+1 =α n +ηg(α n ); Repeat the above steps until the update amplitude of the parameter α is less than the preset threshold value, and the α value at this time is the optimal variable parameter α.

6. The optimized routing strategy for scale-free networks according to claim 5, characterized in that: The weights of the edges between nodes are defined by the routing function of the constructed scale-free network, which includes the following steps: According to the routing function of the constructed scale-free network, the weight of the edge between nodes is defined by the following formula: Among them, q ij is the edge between node i and node j, P i is the probability that a data packet waits on all routing paths of node i, P j is the probability that a data packet waits on all routing paths of node j.

7. The optimized routing strategy for scale-free networks according to claim 1, characterized in that: Performing random attacks and deliberate attacks on nodes in the constructed scale-free network, counting the number of network nodes before and after the cascading failure of the scale-free network, and calculating the ratio of the number of network nodes before and after the cascading failure, includes the following steps: Construct a scale-free network model; The local effective routing strategy method is applied on the constructed scale-free network, and the transmission capacity of the network under this routing strategy is obtained; Conduct random and deliberate attacks on nodes in the network; Simulate the transmission capacity of the network after the attack under the constructed routing strategy; The number of network nodes before and after the cascading failure is defined as S0 and S1, and the ratio of S1 to S0 is defined as the ratio H of the number of network nodes before and after the cascading failure.

8. The optimized routing strategy for scale-free networks according to claim 2, characterized in that: The robustness of the network under the routing strategy is analyzed by the ratio of the number of network nodes before and after the cascading failure, and the capacity setting rule of a single node is adjusted according to the robustness of the routing strategy, including the following steps: The ratio of the number of network nodes before and after the cascading failure is counted. When the ratio of the number of network nodes before and after the cascading failure is greater than a preset threshold range, the robustness of the routing strategy is good and there is no need to adjust the capacity of the node. Otherwise, the rule is set according to adjusting the capacity of a single node. The capacity setting rules for a single node are as follows: When the ratio of the number of network nodes before and after the cascading failure is less than the preset threshold range, the node capacity adjustable parameter β increases by 2%β; When the ratio of the number of network nodes before and after the cascading failure is within a preset threshold range, the node capacity adjustable parameter β is increased by 1% β.

Images