MPTCP robustness analysis and optimization method and system

Abstract

The present invention relates to an MPTCP robustness analysis and optimization method and system based on a cascading failure model. By obtaining an MPTCP multi-path transmission system built on a simulation experiment platform, the initial communication load and communication capacity of each node in the MPTCP multi-path transmission system are determined; according to the initial communication load and communication capacity of each node, the corresponding communication state of each node is determined; according to each communication state, the failed node is determined; the load of each failed node is redistributed to obtain the communication load of the adjusted node; according to the communication load of the adjusted node, it is determined whether the node after the load adjustment is failed; if so, the failed node is determined and the load is redistributed again; if not, the cascading failure ends and a cascading failure model is obtained. The present invention can improve the robustness and anti-destruction performance of the MPTCP multi-path transmission system.

Description

Technical Field

[0001] The present invention relates to the field of multipath transmission, and in particular to a MPTCP robustness analysis and optimization method and system. Background Art

[0002] With the booming development of the mobile internet, continuous breakthroughs and innovations in 5G mobile ultra-high-definition video technology, and the increasing popularity of smart mobile devices, demand for content-rich mobile streaming applications is increasing. Personalized streaming services will become a core business in the future mobile internet. The 49th "Statistical Report on China's Internet Development," officially released by the China Internet Network Information Center (CNNIC) on February 25, 2022, shows that as of December 2021, the number of Internet users in my country reached 1.032 billion, an increase of 42.96 million from December 2020, and an Internet penetration rate of 73.0%. Among them, online video and short video users accounted for 94.5% and 90.5% of the total number of Internet users in my country, respectively, with user bases of 975 million and 934 million, respectively. These data indicate a growing demand for personalized streaming applications such as mobile internet short videos and live streaming, which require high data traffic and network transmission performance.

[0003] The TCP and UDP protocols currently used in transmission networks are both single-path transmission protocols, which can no longer meet users' needs for personalized streaming media application data transmission. As more and more terminal devices are equipped with network interfaces of multiple standards (i.e., multi-host) and have the ability to access multiple networks, how to organically integrate multiple wireless access technologies (Bluetooth, Wi-Fi, 4G, 5G, etc.) and rationally utilize heterogeneous wireless network resources to improve the transmission service quality of application data has become a hot topic in current academic research. In recent years, the Internet Engineering Task Force (IETF) has successively proposed a variety of transmission protocols and corresponding configuration standards that support multi-host connections, aiming to enable multi-host terminal devices to access multiple networks at the same time, improve data transmission rates and maximize network resource utilization.

[0004] Multipath TCP (MPTCP) is one of the most representative achievements in multipath transmission technology research and is considered to have broad application prospects in the future development of the mobile internet. MPTCP retains the stability and reliability of the traditional TCP protocol, maintaining backward compatibility with existing Internet devices and TCP ACI, significantly improving the transmission quality of transmission systems. Figure 1 This is a simple example diagram of MPTCP multi-path parallel transmission streaming media application. Figure 1As shown in the figure, a user terminal with multiple network interfaces can simultaneously communicate with a streaming media server using multiple paths (Path 1 and Path 2). Currently, research on multipath transmission theory and algorithms based on MPTCP, both domestically and internationally, focuses on data scheduling optimization, congestion control, energy efficiency optimization, and fairness issues. However, relatively little research has been conducted on MPTCP security issues. In particular, there is a significant lack of discussion on robustness optimization and survivability analysis for MPTCP multipath transmission systems.

[0005] Faced with an increasingly complex and severe network security landscape, MPTCP network security issues are undoubtedly a pressing research challenge. With the continuous expansion of computer scale and application fields, malicious attacks on networks are rapidly developing and employing diverse methods. For example, denial of service (DoS) attacks have evolved from distributed denial of service (DDoS) attacks to low-rate distributed denial of service (LDDoS) attacks, causing immeasurable damage to communication networks. Therefore, rapidly and accurately detecting abnormal network attack traffic within the network traffic of an MPTCP multipath transmission system and responding promptly and appropriately are effective measures to enhance network robustness and resilience, and are a prerequisite for ensuring network security and effective network operation. Summary of the Invention

[0006] The purpose of the present invention is to provide an MPTCP robustness analysis and optimization method and system, thereby realizing network attack abnormal traffic detection and robustness analysis and optimization in the MPTCP multipath transmission system, and improving the robustness and anti-destruction performance of the MPTCP multipath transmission system.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] An MPTCP robustness analysis and optimization method based on a cascading failure model includes:

[0009] Obtain an MPTCP multipath transmission system built on a simulation experiment platform, wherein the MPTCP multipath transmission system includes nodes and edges;

[0010] Determining an initial communication load and a communication capacity of each node in the MPTCP multipath transmission system;

[0011] Determining a corresponding communication state of each node according to the initial communication load and communication capacity of each node, wherein the communication state includes normal, weak communication, and failure;

[0012] Determining a failed node according to each of the communication states;

[0013] Redistributing the load of each failed node to obtain an adjusted communication load of the node;

[0014] Determining whether the node after the load adjustment is invalid according to the communication load of the node after the adjustment;

[0015] If yes, determine the failed node and return to the step of "redistributing the load on each of the failed nodes to obtain the communication load of the adjusted node";

[0016] If not, the cascading failure ends and the cascading failure model is obtained.

[0017] Optionally, the MPTCP multipath transmission system is abstractly represented by an unweighted network graph G containing n nodes and m edges: G = (V, E);

[0018] Where V={v1,v2,v3,…,v n}, V represents the node set, E={e1,e2,e3,…,e m}, E represents the edge set.

[0019] Optionally, determining the initial communication load of each node and the communication capacity of the node in the MPTCP multipath transmission system specifically includes:

[0020] In the MPTCP multipath transmission system, the communication load of each node is defined as the total amount of information passing through the node per unit time in the initial network. k Initial communication load L k (0);

[0021] According to the initial communication load, according to formula C k =L k (0)·(1+α) determines the communication capacity of each node;

[0022] Among them, C k Represents node v k Communication capacity, α represents the tolerance coefficient, α≥0, α is determined according to the actual situation of the network.

[0023] Optionally, determining the corresponding communication state of each node according to the initial communication load and communication capacity of each node specifically includes:

[0024] According to the initial communication load and communication capacity of each node, the formula Determine the corresponding communication status of each node;

[0025] Among them, t represents a certain moment in the transmission process of MPTCP communication network, Q k (t) represents the node v kIdle capacity ratio, 0≤Q k (t)≤1,S k (t) represents the node v k Communication status value, L k (t) represents the node v k Communication load at time t;

[0026] When the communication load of a node is less than the initial communication load, the communication state is normal; when a node in normal state is affected by network attacks or load redistribution factors, the communication load increases but is less than the communication capacity, the communication state is weak communication; when a node in weak communication state is affected by network attacks or load redistribution factors, the communication load is greater than the communication capacity, the communication state is failed.

[0027] Optionally, the redistributing the load of each failed node to obtain the communication load of the adjusted node specifically includes:

[0028] For each of the failed nodes, use the formula L s (T+1)=L s (T)+ΔL s Perform load redistribution to obtain the communication load of the adjusted node;

[0029] Among them, T represents the current number of failure redistribution times, L s Represents node v s Communication load, v s represents the nodes that have not failed, ΔL s Represents node v s The load increment of this round, ΔL s Affected by factors such as its own load capacity and communication link status, it is dynamically calculated and updated in real time. It can be specifically expressed by the following formula:

[0030]

[0031] Among them, Q s Represents node v s Free capacity ratio, C s Represents any node v in the set of non-failed nodes when the number of failure redistribution is T s The communication capacity, Φ T represents the set of failed nodes when the number of failure redistribution is T, Γ T Represents the set of nodes that have not failed when the number of failure redistribution is T, RTT si Represents the node v in the MPTCP multi-path transmission system s and node v i The round trip delay between t Represents any node v in the set of non-failed nodes when the number of failure redistribution is Tt Communication capacity; RTT ti Represents the node v in the MPTCP multi-path transmission system t and node v i The round trip delay between .

[0032] Optionally, the determining, based on the communication load of the adjusted node, whether the node after the load adjustment is invalid specifically includes:

[0033] When L s (T+1)≤C s When node v s When this round does not fail;

[0034] When L s (T+1)>C s When node v s This round is invalid.

[0035] Optionally, the method further includes: performing a performance measurement test on the MPTCP multipath transmission system after the cascade failure ends using an efficiency function E(G); the specific formula of the efficiency function E(G) is as follows:

[0036]

[0037] Among them, E(G) represents the efficiency function, n represents the number of nodes, and RTT ij represents the round-trip delay between nodes i and j, and G represents the unweighted network graph of the MPTCP multipath transmission system.

[0038] Optionally, the method further includes: performing simulation analysis on the cascading failure model in Matlab.

[0039] To achieve the above object, the present invention also provides the following solution:

[0040] An MPTCP robustness analysis and optimization system includes:

[0041] An MPTCP multipath transmission system acquisition module is used to acquire the MPTCP multipath transmission system built on the simulation experiment platform, where the MPTCP multipath transmission system includes nodes and edges;

[0042] An initial communication load and communication capacity determination module, configured to determine an initial communication load and communication capacity of each node in the MPTCP multipath transmission system;

[0043] A node communication state determination module is used to determine the corresponding communication state of each node according to the initial communication load and communication capacity of each node, wherein the communication state includes normal, weak communication and failure;

[0044] A failed node determination module, configured to determine a failed node according to each of the communication states;

[0045] A communication load adjustment module, configured to redistribute the load of each failed node to obtain the communication load of the node after adjustment;

[0046] A failed node judgment module is used to judge whether the node after load adjustment is failed according to the communication load of the node after adjustment;

[0047] A second failed node determination module, configured to determine the failed node when the node after load adjustment fails, and return the information to the communication load adjustment module;

[0048] The cascading failure model determination module is used to determine that the cascading failure ends when the node after load adjustment does not fail, and obtain the cascading failure model.

[0049] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0050] The present invention designs an MPTCP robustness analysis method by constructing a cascading failure model suitable for MPTCP multipath transmission systems. Based on the "load redistribution" distribution mechanism, the MPTCP robustness analysis method is optimized and upgraded. This method can detect abnormal traffic flows during network attacks and analyze and optimize robustness in MPTCP multipath transmission systems, thereby improving the robustness and invulnerability of MPTCP multipath transmission systems. 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 following briefly introduces the drawings required for use in the embodiments. 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 any creative work.

[0052] Figure 1 This is a simple example diagram of MPTCP multi-path parallel transmission streaming media application;

[0053] Figure 2 Schematic diagram of cascading failure of MPTCP multi-path transmission system;

[0054] Figure 3 This is a flow chart of the MPTCP robustness analysis and optimization method based on the cascading failure model of the present invention;

[0055] Figure 4 This is a structural diagram of the MPTCP robustness analysis and optimization system based on the cascading failure model of the present invention. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] The purpose of the present invention is to provide an MPTCP robustness analysis and optimization method and system, thereby realizing network attack abnormal traffic detection and robustness analysis and optimization in the MPTCP multipath transmission system, and improving the robustness and anti-destruction performance of the MPTCP multipath transmission system.

[0058] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0059] To adapt to the future development and application of the mobile internet and meet the urgent need for high-performance transmission of streaming media applications such as mobile ultra-high-definition video, this paper proposes a "MPTCP robustness analysis and optimization method based on a cascading failure model." In a complex, heterogeneous mobile network environment, this paper uses MPTCP, the core transport layer protocol of the future internet, as its foundation and uses LDDoS attacks as an example of network attacks. By integrating signal analysis techniques with the cascading failure model, this paper conducts a detailed design and simulation verification analysis. This method enables detection of abnormal network attack traffic, robustness analysis, and optimization in MPTCP multipath transmission systems, enhancing their robustness and resilience. This research contributes to cutting-edge research in network security management and decision-making for multipath transmission systems.

[0060] Regarding the robustness analysis and optimization of MPTCP multipath transmission systems, scholars at home and abroad have proposed many reasonable and effective research proposals. Tan Qining et al. proposed a robustness evaluation model for multipath transmission protocol networks based on subflow impact factors, addressing the robustness evaluation problem. Luo Yu et al., through research on the TCP / IPv4 address structure and SocketAPI, reconstructed the MPTCP / IPv6 address structure and SocketAPI software. Based on this, they compiled the MPTCP kernel and extended SRS, configured the corresponding routing tables, and ultimately implemented an IPv6+MPTCP-based video live streaming platform, improving network robustness. Li Qian et al., analyzing existing multipath communication methods and incorporating the characteristics of vector networks, designed a multipath communication method for VN switching networks. To address the data transmission reliability issue in transmission networks, Yang Nan et al. proposed a multipath transmission protocol based on a degree-constrained shortest transmission tree. They also proposed intermediate and edge location optimization strategies to address deployment issues in multi-aggregation node networks, analyzing network robustness and node load balancing. Combining the advantages of vector networks (VNs), such as infinite multi-valuedness, security, and lightweight connectivity, Fazal Rabi et al. proposed a VN-supported SCTP protocol approach and designed an SCTP / VN multipath transmission network. They analyzed the advantages of SCTP multipath in a VN-based environment in combating single points of failure, facilitating network congestion control, improving network resource utilization, and effectively increasing network throughput. Li Jinxiang et al. explored a model for multipath transmission of video streams over networks and proposed four decoding methods with varying levels of complexity and robustness to accommodate the diverse needs of network transmission environments. Zhu Ting et al. proposed a graph neural network-based MPTCP performance optimization scheme for heterogeneous networks, elaborating on multipath routing, MPTCP packet scheduling, and congestion control.

[0061] Based on relevant research at home and abroad, it was found that the current research on MPTCP network attack abnormal traffic detection methods mainly focuses on the detection and defense of MPTCP transmission systems against DoS attacks and DDoS attacks, while there is a lack of research on abnormal traffic detection methods for network attacks such as LDDoS attacks. The current research on MPTCP robustness analysis and optimization solutions mainly focuses on leveraging the characteristics of MPTCP multi-path parallel transmission to solve the robustness problem of complex networks, while there is little research on the robustness analysis and optimization of the MPTCP multi-path transmission system itself. In summary, in combination with relevant research trends at home and abroad, adapting to the development and application of the future mobile Internet and meeting the urgent need for high-performance transmission of streaming media applications such as mobile ultra-high-definition video, this paper proposes a "MPTCP robustness analysis and optimization method based on the cascading failure model". Based on MPTCP, the core protocol of the future Internet transport layer, and using LDDoS attacks as an example of network attacks, this paper integrates signal analysis technology and the cascading failure model to enhance the robustness and anti-destruction properties of the MPTCP multi-path transmission system, enriching the research in the field of network security management and decision-making of multi-path transmission systems.

[0062] During the transmission of streaming media application data in MPTCP multipath transmission systems, malicious activities such as network attacks often occur. When a transmission node is attacked, its failure can lead to network load redistribution. This load redistribution causes some nodes to exceed their load capacity and fail. These node failures can then cause cascading failures in other nodes. The symptoms of cascading failures are primarily packet loss and high network latency, affecting not only a single system but the entire network or Internet. Cascading failures in MPTCP transmission systems significantly impact their robustness and resilience, affecting the transmission performance of streaming media applications such as mobile ultra-high-definition video and the user experience quality of service. Therefore, based on the research on abnormal traffic detection methods for network attacks based on signal analysis technology, the cascading failure model is integrated to further explore the attack behavior characteristics of network attacks in MPTCP multi-path transmission systems, and design a robustness analysis and optimization scheme for MPTCP multi-path transmission systems. This includes studying the MPTCP robustness analysis method based on the cascading failure model, establishing a load-capacity cascading failure robustness optimization model based on the node idle capacity ratio, and studying robustness measurement indicators for video real-time requirements. This provides theoretical and model support for future research on robustness analysis and optimization methods of multi-path transmission systems, and enriches related research on network security management decisions of multi-path transmission systems.

[0063] Cascading failure in complex networks occurs when one or a few edges or nodes fail, affecting connected edges or nodes and causing a cascading failure. Sometimes, a small failure in a complex network can trigger a widespread failure of the entire system, and in severe cases, even paralyze the entire network, rendering it inoperable. In MPTCP multipath transmission systems, malicious attacks and other malicious behaviors can have a significant negative impact on node transmission capacity. When a transmission node loses its data transmission capacity due to a network attack, it becomes overloaded. The initial load borne by the overloaded node must be redistributed to other nodes in the network according to specific rules. Specifically, data packets currently or about to be transmitted through the node are redistributed to other paths capable of normal data transmission through a "load redistribution" mechanism. However, these nodes receiving the additional load may also experience a load exceeding their communication capacity, causing them to fail, triggering another round of load redistribution and leading to cascading network failure. Figure 2 Schematic diagram of cascading failure of MPTCP transmission system.

[0064] Therefore, in order to avoid the occurrence of network cascading failure behavior, ensure the normal communication operation of the MPTCP multi-path transmission system, and improve the robustness and stability of the MPTCP multi-path transmission system, it is necessary to study the process of network cascading failure. Based on the research on the MPTCP network attack abnormal traffic detection method based on signal analysis technology, the present invention integrates the load-capacity cascading failure model to study the cascading failure behavior caused by the node being attacked by the network (taking LDDoS as an example) in the MPTCP multi-path transmission system, and further explores the attack behavior characteristics of the network attack in the MPTCP multi-path transmission system. At the same time, the load-capacity model in the network cascading failure is further optimized to reduce the probability of the occurrence of network cascading failure behavior, so that the network is effectively protected, thereby preventing the large-scale spread of cascading failure in the network and improving the network robustness of the MPTCP multi-path transmission system.

[0065] In view of the current situation in which domestic and foreign academic circles have a serious lack of discussion and research on the MPTCP transmission protocol in network security management decision-making, this paper takes the future Internet transport layer core protocol MPTCP as the protocol basis, and conducts research on network attack abnormal traffic detection and robustness analysis and optimization of MPTCP multi-path transmission system in mobile heterogeneous complex network environment. With the goal of improving the robustness and anti-destruction performance of multi-path transmission system, this paper provides an MPTCP robustness analysis and optimization method based on cascading failure model, such as Figure 3 As shown, the method includes:

[0066] Step 101: Obtain an MPTCP multipath transmission system built on a simulation experiment platform, where the MPTCP multipath transmission system includes nodes and edges.

[0067] Since the communication between the sender, receiver, and each communication node in the MPTCP multipath transmission system is bidirectional, the MPTCP multipath transmission system built on the NS2 simulation platform can be abstractly represented by an unweighted network graph G containing n nodes and m edges: G = (V, E);

[0068] Where V={v1,v2,v3,…,v n}, V represents the node set, E={e1,e2,e3,…,e m}, E represents the edge set.

[0069] Step 102: Determining the initial communication load and communication capacity of each node in the MPTCP multipath transmission system, specifically including:

[0070] In the MPTCP multipath transmission system, the communication load of each node is defined as the total amount of information passing through the node per unit time in the initial network (network without attack), and its value is set to the total amount of data packets forwarded by the node. k For example, get the kth node v k Initial communication load L k (0);

[0071] According to the initial communication load, according to formula C k =L k (0)·(1+α) determines the communication capacity of each node;

[0072] Among them, C k Represents node v k Communication capacity, α represents the tolerance coefficient, α≥0, α is determined according to the actual situation of the network.

[0073] Step 103: Determine the corresponding communication state of each node based on the initial communication load and communication capacity of each node, specifically including:

[0074] According to the initial communication load and communication capacity of each node, the formula Determine the corresponding communication status of each node;

[0075] Among them, t represents a certain moment in the transmission process of MPTCP communication network, Q k (t) represents the node v k Idle capacity ratio, 0≤Q k (t)≤1,S k (t) represents the node vk Communication status value, L k (t) represents the node v k Communication load at time t;

[0076] To better provide real-time feedback on the transmission capacity of each communication node, it is planned to assign a corresponding communication status to each communication node based on its communication load, with three communication states: "normal," "weak communication," and "failed." In the initial state, the initial communication load of each node in the network is less than its corresponding communication capacity. When a node's communication load is less than its initial communication load, its communication status is "normal." When a node in a normal state experiences an increase in communication load but is less than its communication capacity due to a network attack or load redistribution, its communication status is "weak communication." When a node in a weak communication state experiences a load greater than its communication capacity due to a network attack or load redistribution, its communication status is "failed."

[0077] Step 104: Determine a failed node according to each of the communication states;

[0078] The most important thing in building a load-capacity cascading failure model is to determine how the load on a node is redistributed when the node fails. A reasonable and efficient load redistribution scheme is an effective measure to prevent cascading failures in MPTCP multipath transmission systems. Assume that S k (t) = n, that is, node v k The state has changed from "weak communication" to "failed", v s For a valid node ("normal" or "weak communication" status)

[0079] Step 105: performing load redistribution on each of the failed nodes to obtain the communication load of the adjusted nodes, specifically including:

[0080] For each of the failed nodes, use the formula L s (T+1)=L s (T)+ΔL s Perform load redistribution to obtain the communication load of the adjusted node;

[0081] Among them, T represents the current number of failure redistribution times, L s Represents node v s Communication load, v s represents the nodes that have not failed, ΔL s Represents node v s The load increment of this round, ΔL s It is affected by factors such as its own load capacity (node ​​idle capacity ratio) and communication link status, and is dynamically calculated and updated in real time. It can be expressed by the following formula:

[0082]

[0083] Among them, Q s Represents node v s Free capacity ratio, C s Represents any node v in the set of non-failed nodes when the number of failure redistribution is T s The communication capacity, Φ T represents the set of failed nodes when the number of failure redistribution is T, Γ T Represents the set of nodes that have not failed when the number of failure redistribution is T, RTT si Represents the node v in the MPTCP multi-path transmission system s and node v i The round trip delay between t Represents any node v in the set of non-failed nodes when the number of failure redistribution is T t Communication capacity; RTT ti Represents the node v in the MPTCP multi-path transmission system t and node v i The round-trip time between them.

[0084] Step 106: Determine whether the node after the load adjustment is invalid based on the communication load of the node after the adjustment.

[0085] After each round of load redistribution, determine the node v that adjusts the load s Is it invalid? s (T+1)≤C s , then node v s This round does not fail; if L s (T+1)>C s , then node v s This round of failure triggers the next round of load redistribution. When the communication load of all nodes does not exceed their communication capacity (when no new nodes are turned into "failed" state), the cascading failure ends.

[0086] Step 107: If yes, determine the failed node and return to the step of "redistributing the load on each of the failed nodes to obtain the communication load of the adjusted node";

[0087] Step 108: If not, the cascading failure ends and a cascading failure model is obtained. The cascading failure model is a load-capacity cascading failure model based on the node idle capacity ratio.

[0088] In addition to the above steps 101-108, the present invention further includes step 109: performing a performance measurement test on the MPTCP multipath transmission system after the cascade failure using an efficiency function E(G); the specific formula of the efficiency function E(G) is as follows:

[0089]

[0090] Among them, E(G) represents the efficiency function, n represents the number of nodes, and RTT ij represents the round-trip delay between nodes i and j, and G represents the unweighted network graph of the MPTCP multipath transmission system.

[0091] And step 110: simulate and analyze the cascading failure model in Matlab. The load-capacity cascading failure model based on the node idle capacity ratio is simulated and analyzed in Matlab, and the constructed cascading failure model is simulated and compared with the traditional cascading failure model to compare the robustness and stability of the communication network under the two cascading failure models.

[0092] Figure 4 This is the structure diagram of the MPTCP robustness analysis and optimization system based on the cascading failure model of the present invention. Figure 4 As shown, an MPTCP robustness analysis and optimization system includes:

[0093] An MPTCP multipath transmission system acquisition module 201 is configured to acquire an MPTCP multipath transmission system constructed on a simulation experiment platform, wherein the MPTCP multipath transmission system includes nodes and edges;

[0094] An initial communication load and communication capacity determination module 202 is configured to determine an initial communication load and communication capacity of each node in the MPTCP multipath transmission system;

[0095] The node communication state determination module 203 is used to determine the corresponding communication state of each node according to the initial communication load and communication capacity of each node, wherein the communication state includes normal, weak communication and failure;

[0096] A failed node determination module 204 is configured to determine a failed node according to each of the communication states;

[0097] The communication load adjustment module 205 is configured to redistribute the load of each failed node to obtain the communication load of the node after adjustment;

[0098] A failed node determination module 206 is configured to determine whether the node after load adjustment is failed based on the communication load of the node after adjustment;

[0099] A second failed node determination module 207 is configured to determine the failed node when the node after load adjustment fails, and return the information to the communication load adjustment module;

[0100] The cascading failure model determination module 208 is configured to determine that the cascading failure ends when the node after load adjustment does not fail, and obtain a cascading failure model.

[0101] The present invention has certain scientific significance: In the past, research on multi-path transmission theory and algorithms based on MPTCP protocol, domestic and foreign researchers mainly focused on aspects such as data scheduling optimization, congestion control, energy consumption optimization, and fairness issues, while research on MPTCP security-related issues was relatively small. In particular, the current academic community is seriously lacking in discussions on issues related to robustness optimization and anti-destruction analysis of MPTCP multi-path transmission systems. Existing research on MPTCP network security management decisions focuses more on using the characteristics of MPTCP multi-path parallel transmission to solve security management decision-making problems in complex networks, while there is little research on network security management decisions of MPTCP multi-path transmission systems themselves. The present invention integrates signal analysis technology and cascading failure models, proposes research on "complex network security management decisions for MPTCP multi-path parallel transmission", studies MPTCP network attack abnormal traffic detection methods, and designs MPTCP robustness analysis and optimization solutions, enriching the research on network security management decisions of multi-path transmission systems and providing theoretical guidance for future research.

[0102] This invention has certain practical significance: With the vigorous development of the mobile internet, the continuous breakthroughs and innovations in 5G mobile ultra-high-definition video technology, and the increasing popularity of smart mobile terminal devices, demand for personalized streaming media applications such as mobile internet short videos and live streaming, which have high requirements for traffic and network transmission performance, is constantly increasing. Personalized streaming media application services will become the core services of the future mobile internet. Multipath transmission protocols represented by MPTCP can implement multi-path parallel data transmission, greatly improving the transmission quality of the transmission system and the user's service experience quality, and are considered to have broad application prospects in the future development of the mobile internet. However, faced with the increasingly complex and severe network security situation, MPTCP also faces serious network security issues. Therefore, implementing network attack abnormal traffic detection in MPTCP multipath transmission systems, designing robustness analysis and optimization schemes suitable for personalized streaming media applications, and further exploring the attack behavior characteristics of network attacks in MPTCP multipath transmission systems are prerequisites for ensuring network security and effective network operation. These studies are of practical significance for adapting to the future development and application of the mobile internet and meeting the urgent need for high-quality transmission of streaming media applications such as mobile ultra-high-definition video, and provide theoretical guidance for security management decisions in complex networks.

[0103] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0104] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The above examples are only intended to help understand the method and core concept of the present invention. At the same time, those skilled in the art will find that the specific implementation methods and application scopes may vary based on the concept of the present invention. In summary, the contents of this specification should not be construed as limiting the present invention.

Claims

1. A method for analyzing and optimizing MPTCP robustness based on a cascading failure model, characterized in that: include: Obtain an MPTCP multipath transmission system built on a simulation experiment platform, wherein the MPTCP multipath transmission system includes nodes and edges; Determining an initial communication load and a communication capacity of each node in the MPTCP multipath transmission system; Determining a corresponding communication state of each node according to the initial communication load and communication capacity of each node, wherein the communication state includes normal, weak communication, and failure; Determining a failed node according to each of the communication states; Redistributing the load of each failed node to obtain an adjusted communication load of the node; Determining whether the node after the load adjustment is invalid according to the communication load of the node after the adjustment; If so, determine the failed node and return to the step of "redistributing the load on each of the failed nodes to obtain the communication load of the adjusted node"; If not, the cascading failure ends and the cascading failure model is obtained; The redistributing the load of each failed node to obtain the communication load of the adjusted node specifically includes: For each of the failed nodes, use the formula L s (T+1)=L s (T)+ΔL s Perform load redistribution to obtain the communication load of the adjusted node; Among them, T represents the current number of failure redistribution times, L s Represents node v s Communication load, v s represents the nodes that have not failed, ΔL s Represents node v s The load increment of this round, ΔL s Affected by its own load capacity and communication link status factors, and dynamically calculated and updated in real time, it can be specifically expressed by the following formula: Among them, Q s Represents node v s Free capacity ratio, C s Represents any node v in the set of non-failed nodes when the number of failure redistribution is T s The communication capacity, Φ T represents the set of failed nodes when the number of failure redistribution is T, Γ T Represents the set of nodes that have not failed when the number of failure redistribution is T, RTT si Represents the node v in the MPTCP multi-path transmission system s and node v i The round trip delay between t Represents any node v in the set of non-failed nodes when the number of failure redistribution is T t Communication capacity; RTT ti Represents the node v in the MPTCP multi-path transmission system t and node v i The round trip delay between .

2. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 1 is characterized in that: The MPTCP multipath transmission system is abstractly represented by an unweighted network graph G containing n nodes and m edges: G = (V, E); Where V={v1,v2,v3,…,v n }, V represents the node set, E={e1,e2,e3,…,e m }, E represents the edge set.

3. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 1 is characterized in that: The determining of the initial communication load of each node and the communication capacity of the node in the MPTCP multipath transmission system specifically includes: In the MPTCP multipath transmission system, the communication load of each node is defined as the total amount of information passing through the node per unit time in the initial network. k Initial communication load L k (0); According to the initial communication load, according to formula C k =L k (0)·(1+α) determines the communication capacity of each node; Among them, C k Represents node v k Communication capacity, α represents the tolerance coefficient, α≥0, α is determined according to the actual situation of the network.

4. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 3 is characterized in that: Determining the corresponding communication state of each node according to the initial communication load and communication capacity of each node specifically includes: According to the initial communication load and communication capacity of each node, the formula Determine the corresponding communication status of each node; Among them, t represents a certain moment in the transmission process of MPTCP communication network, Q k (t) represents the node v k Idle capacity ratio, 0≤Q k (t)≤1,S k (t) represents the node v k Communication status value, L k (t) represents the node v k Communication load at time t; When the communication load of a node is less than the initial communication load, the communication state is normal; when a node in normal state is affected by network attacks or load redistribution factors, the communication load increases but is less than the communication capacity, the communication state is weak communication; when a node in weak communication state is affected by network attacks or load redistribution factors, the communication load is greater than the communication capacity, the communication state is failed.

5. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 1 is characterized in that: The determining, based on the communication load of the adjusted node, whether the node after the load adjustment is invalid specifically includes: When L s (T+1)≤C s When node v s When this round does not fail; When L s (T+1)>C s When node v s This round is invalid.

6. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 1 is characterized in that: Also includes: The efficiency function E(G) is used to measure the performance of the MPTCP multipath transmission system after the cascade failure. The specific formula of the efficiency function E(G) is as follows: Among them, E(G) represents the efficiency function, n represents the number of nodes, and RTT ij represents the round-trip delay between nodes i and j, and G represents the unweighted network graph of the MPTCP multipath transmission system.

7. The MPTCP robustness analysis and optimization method based on the cascading failure model according to claim 1 is characterized in that: Also includes: The cascading failure model is simulated and analyzed in Matlab.

8. A system based on the method according to any one of claims 1 to 7, characterized in that: include: An MPTCP multipath transmission system acquisition module is used to acquire the MPTCP multipath transmission system built on the simulation experiment platform, where the MPTCP multipath transmission system includes nodes and edges; An initial communication load and communication capacity determination module, configured to determine an initial communication load and communication capacity of each node in the MPTCP multipath transmission system; A node communication state determination module is used to determine the corresponding communication state of each node according to the initial communication load and communication capacity of each node, wherein the communication state includes normal, weak communication and failure; A first failed node determination module, configured to determine a failed node according to each of the communication states; A communication load adjustment module, configured to redistribute the load of each failed node to obtain the communication load of the node after adjustment; A failed node judgment module is used to judge whether the node after load adjustment is failed according to the communication load of the node after adjustment; A second failed node determination module, configured to determine the failed node when the node after load adjustment fails, and return the information to the communication load adjustment module; The cascading failure model determination module is used to determine that the cascading failure ends when the node after load adjustment does not fail, and obtain the cascading failure model.

Images