Efficiency evaluation method and system based on air-ground integrated network analogue simulation confrontation

Abstract

The invention discloses an efficiency evaluation method and system based on air-ground integrated network analogue simulation confrontation, and relates to the technical field of network simulation and efficiency evaluation. The method comprises the following steps: constructing a space-ground integrated network topology simulation model by using a simulation engine; injecting a confrontation event, and simulating a confrontation scene; collecting network performance index data in the confrontation scene; on the basis of the collected index data, a weighted average method is adopted, and project weights included in all levels in a pre-constructed two-level performance evaluation system are utilized, a second-level score is firstly calculated, and then a first-level score is calculated on the basis of the second-level score; based on the score of the first level, utilizing a fuzzy evaluation method to obtain a fuzzy efficiency level of the space-ground integrated network; and taking the score of the first level and the fuzzy efficiency level as an efficiency evaluation result. According to the method, accurate reproduction of a multi-dimensional dynamic scene and closed-loop integration of simulation evaluation can be realized. And meanwhile, a multi-algorithm fusion method is adopted, so that the accuracy and robustness of evaluation are improved.

Description

Technical Field

[0001] This invention relates to the field of network simulation and performance evaluation technology, and in particular to a performance evaluation method and system based on integrated space-ground network simulation and confrontation. Background Technology

[0002] The space-ground integrated network deeply integrates the space segment (various satellite constellations, high-altitude platforms), the air segment (drones), the ground segment (terrestrial cellular networks, the Internet, terminals), and the sea segment through inter-satellite and space-to-ground links, forming a giant composite network that covers the globe, integrates air, space, and sea, supports ubiquitous broadband, provides on-demand access, and is secure and reliable. To measure the ability of the space-ground integrated network to complete its predetermined mission or task in a specific scenario, quantitative or qualitative performance evaluations are generally conducted using a certain indicator system.

[0003] Currently, most performance evaluation models are based on static network environment design, assuming that key network parameters (such as topology, node locations, link status, and resource allocation strategies) remain constant within the evaluation timescale. This is a "snapshot" evaluation of the network at a specific moment or under a specific configuration. However, in reality, the high-speed movement of satellites causes continuous dynamic changes in network topology, link status (latency, bandwidth), and coverage relationships; that is, integrated space-ground networks are dynamic and time-varying. Furthermore, integrated space-ground networks often face complex adversarial environments (such as electromagnetic interference, link failures, and malicious attacks) in scenarios such as military communications and emergency disaster relief. Therefore, static performance evaluation models are difficult to adapt to the highly dynamic topology changes and real-time adversarial requirements (such as inter-satellite link interference and routing spoofing attacks) of integrated space-ground networks, leading to significant deviations between evaluation results and actual network resilience.

[0004] Furthermore, performance evaluation algorithms often rely on a single algorithm (such as the analytic hierarchy process or fuzzy evaluation), which cannot simultaneously address the objectivity of indicator weight allocation, the integration of subsystem performance, and the handling of data uncertainty in adversarial environments. For example, weight calculation may depend on subjective experience or fail to consider indicator volatility, thus affecting the accuracy and robustness of the evaluation results. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the present invention provides the following technical solution.

[0006] This invention provides, in one aspect, a method for evaluating the effectiveness of simulated combat based on a space-ground integrated network, comprising: Construct a simulation model of the integrated space-ground network topology using a simulation engine; Inject adversarial events into the simulation model to simulate adversarial scenarios; Collect network performance metrics data in adversarial scenarios; Based on the collected indicator data, a weighted average method is used. The weights of the items included in each level of the pre-constructed two-level performance evaluation system are used to first calculate the score of the second level, and then the score of the first level is calculated based on the score of the second level. The two-level performance evaluation system includes a first level and a second level, and the second level is a sub-level of the first level. Based on the scores of the second level, the effectiveness level of each item in the first level is obtained using the fuzzy evaluation method; and based on the scores of the first level, the comprehensive effectiveness level of the integrated space-ground network is obtained using the fuzzy evaluation method. It displays the comprehensive effectiveness levels under various adversarial scenarios, the radar chart of the score distribution of each item in the first level, and the change curves of each item before and after the adversarial in the second level.

[0007] Preferably, the performance evaluation method for simulated confrontation based on integrated space-ground network provided by the present invention further includes the step of: standardizing the indicator data.

[0008] Preferably, the first-level items include: basic performance layer, network robustness layer, and assurance performance layer; in the second-level items, the basic performance layer includes throughput, latency, packet loss rate, and end-to-end jitter, the network robustness layer includes link survival probability and fault recovery efficiency, and the assurance performance layer includes critical service completion rate and anti-interference duration.

[0009] Preferably, the weights of the items included at each level are determined based on expert scoring and consistency checks.

[0010] Preferably, the adversarial scenarios include: inter-satellite link interference, satellite-to-ground link deception attacks, satellite node failures, inter-satellite link failures, and / or routing protocol adversarial attacks.

[0011] A second aspect of the present invention provides an effectiveness evaluation system for simulated combat based on a space-ground integrated network, comprising: The simulation model building module is used to build a simulation model of the integrated space-ground network topology using the simulation engine. The adversarial simulation module is used to inject adversarial events into the simulation model to simulate adversarial scenarios; The data acquisition module is used to collect network performance metrics data in adversarial scenarios. Each level of the scoring calculation module is used to calculate the second-level score based on the collected indicator data, using a weighted average method and the item weights included in each level of the pre-constructed two-level performance evaluation system. Then, the first-level score is calculated based on the second-level score. The two-level performance evaluation system includes a first level and a second level, and the second level is a sub-level of the first level. The performance rating module is used to obtain the performance rating of each item in the first level based on the score of the second level using the fuzzy evaluation method; and to obtain the comprehensive performance rating of the integrated space-ground network based on the score of the first level using the fuzzy evaluation method. The performance evaluation result presentation module is used to display the comprehensive performance level under various adversarial scenarios, the score distribution radar chart of each item in the first level, and the change curves of each item before and after the adversarial scenario in the second level. Preferably, the performance evaluation system based on integrated space-ground network simulation adversarial scenario provided by the present invention also includes a data processing module for standardizing the indicator data.

[0012] Preferably, the first-level items include: basic performance layer, network robustness layer, and assurance performance layer; in the second-level items, the basic performance layer includes throughput, latency, packet loss rate, and end-to-end jitter, the network robustness layer includes link survival probability and fault recovery efficiency, and the assurance performance layer includes critical service completion rate and anti-interference duration.

[0013] Preferably, the weights of the items included at each level are determined based on expert scoring and consistency checks.

[0014] Preferably, the adversarial scenarios include: inter-satellite link interference, satellite-to-ground link deception attacks, satellite node failures, inter-satellite link failures, and / or routing protocol adversarial attacks.

[0015] The beneficial effects of this invention are as follows: The performance evaluation method and system for simulated confrontation based on integrated space-ground network provided by this invention have the following advantages compared with the prior art: 1. Significantly enhanced simulation capabilities for dynamic adversarial scenarios. Overcoming the limitations of static assessment: Traditional methods are mostly designed for fixed network environments, making it difficult to simulate the highly dynamic topology changes and real-time adversarial capabilities (such as inter-satellite link interference and route spoofing) of integrated space-ground networks. This invention achieves accurate reproduction of multi-dimensional dynamic scenarios through a configurable adversarial event library (supporting interference scripts and node fault injection), making the assessment results more closely match actual network resilience requirements.

[0016] Closed-loop simulation and evaluation integration: Integrates simulation engines (such as NS3 / Mininet) and performance evaluation modules, avoiding the problem of disconnect between simulation and evaluation tools in traditional platforms, and supports automated analysis of the entire process from scenario construction to performance quantification.

[0017] 2. Multi-algorithm fusion improves evaluation accuracy and robustness Advantages of complementary algorithms: By objectively allocating index weights through the Analytic Hierarchy Process (AHP), combining the weighted average method to integrate the subsystem efficiency, and then using fuzzy comprehensive evaluation to handle data uncertainty in adversarial environments, the algorithm overcomes the shortcomings of single algorithms (such as using only AHP or fuzzy evaluation) in terms of weight subjectivity or fault tolerance.

[0018] The quantitative output is intuitive and reliable: the final output generates standardized performance scores, multi-dimensional radar charts and trend curves, which intuitively present the changes in network performance before and after the confrontation, and provide reliable data support for the optimization of anti-interference strategies.

[0019] It can effectively support the optimization of anti-interference strategies and the verification of network planning, and has the characteristics of standardization, repeatability, and strong intuitiveness. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the effectiveness evaluation method for simulated confrontation based on an integrated space-ground network as described in this invention. Figure 2 This is a functional structure diagram of the performance evaluation system based on integrated space-ground network simulation and confrontation described in this invention. Detailed Implementation

[0021] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0022] The method provided by this invention can be implemented in a terminal environment that may include one or more of the following components: a processor, a memory, and a display screen. The memory stores at least one instruction, which is loaded and executed by the processor to implement the method described in the following embodiments.

[0023] A processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts of the terminal, and performs various functions and processes data by running or executing instructions, programs, code sets or instruction sets stored in memory, and by calling data stored in memory.

[0024] Memory can include random access memory (RAM) or read-only memory (ROM). Memory can be used to store instructions, programs, code, code sets, or instructions.

[0025] The display screen is used to show the user interface of each application.

[0026] In addition, those skilled in the art will understand that the structure of the terminal described above does not constitute a limitation on the terminal. The terminal may include more or fewer components, or combine certain components, or have different component arrangements. For example, the terminal may also include radio frequency circuits, input units, sensors, audio circuits, power supplies, and other components, which will not be described in detail here.

[0027] Example 1 like Figure 1 As shown, this embodiment of the invention provides a method for evaluating the effectiveness of simulated combat based on a space-ground integrated network, which may include the following steps: S101, a simulation model of an integrated space-ground network topology is constructed using a simulation engine. The simulation engine can be built based on Mininet 2.3.0, integrating the ONOS 3.0.0 version SDN controller to enhance the linkage capabilities of dynamic routing and security policies. The integrated space-ground network topology includes a space segment, a ground segment, and link connections. The space segment is used to configure satellites, such as orbital altitude, orbital period, and satellite movement according to preset orbital parameters, forming inter-satellite links. The ground segment is used to configure ground stations, setting their coverage radius and accessible terminal nodes. Link connections are used to configure inter-satellite links and satellite-to-ground links. Furthermore, this invention allows for SDN linkage configuration, such as linkage between the ONOS controller and the simulation engine, and preset route reconstruction strategies to ensure network service continuity.

[0028] S102, Inject adversarial events into the simulation model to simulate adversarial scenarios. Various adversarial events can be preset and stored in an adversarial event library, and custom adversarial events can also be defined as needed. Each adversarial event corresponds to a specific adversarial scenario. Each adversarial event is configured with specific parameters and a preset trigger mode. After injecting the adversarial event, the corresponding adversarial scenario can be simulated according to the configured parameters and trigger mode.

[0029] S103 collects network performance metrics data under adversarial scenarios. The collected performance metrics may include throughput, end-to-end latency, packet loss rate, and end-to-end jitter. Each performance metric can be used to obtain the final data according to a preset collection granularity and calculation method. For example, the throughput collection granularity is 1 second, and the calculation method is: counting the total amount of valid data successfully transmitted end-to-end per second (excluding retransmitted data packets).

[0030] S104, based on the collected indicator data, a weighted average method is used. Utilizing the weights of items at each level in a pre-constructed two-level performance evaluation system, the second-level score is calculated first, and then the first-level score is calculated based on the second-level score. The two-level performance evaluation system includes a first level and a second level, with the second level being a sub-level of the first level. Both the first and second levels include corresponding items. For example, the first level may include items such as: basic performance layer, network robustness layer, and system performance layer. The second level includes sub-items under the first-level items. For example, the basic performance layer may include items such as: throughput, latency, packet loss rate, and end-to-end jitter. In this embodiment, a preset weight is assigned to each item in both the first and second levels. For example, the weight of the basic performance layer in the first level is 0.4. In the second level, the weight of throughput is 0.3, and the weight of latency is 0.25. It should be noted that for any item at any level, the sum of all item weights within that level must be 1. For example, the three items in the first level—basic performance layer, network robustness layer, and system efficiency layer—have weights of 0.4, 0.35, and 0.25 respectively, summing to 1. In this embodiment of the invention, since the second level is a subset of the first level, when calculating the scores for each level, the second-level score is calculated first, and then the first-level score is calculated based on the second-level score. As an example, the score for the basic performance layer is calculated first based on the item data and weights included in the basic performance layer of the second level. Then, the score for the first level is calculated based on the calculated scores and weights of the basic performance layer, network robustness layer, and system efficiency layer. Specifically, for example, the basic performance layer score = throughput standardized value × 0.3 + latency standardized value × 0.25 + packet loss rate standardized value × 0.25 + end-to-end jitter standardized value × 0.2; similarly, the scores for the network robustness layer and system efficiency layer are calculated. The standardized values ​​of each indicator are the data obtained after standardizing the collected indicator data.

[0031] S105, based on the scores of the second level, the efficiency level of each item in the first level is obtained by using the fuzzy evaluation method; and based on the scores of the first level, the comprehensive efficiency level of the integrated space-ground network is obtained by using the fuzzy evaluation method. Since the first-level score includes the scores and weights of the basic performance layer, network robustness layer, and system efficiency layer, it can comprehensively evaluate the efficiency of the integrated space-ground network. Therefore, in this embodiment of the invention, the overall efficiency level of the integrated space-ground network is determined based on the first-level score. Specifically, this embodiment uses a fuzzy evaluation method to determine the corresponding efficiency level. Specifically, based on a set efficiency level, such as excellent, good, average, and poor, a membership function maps quantitative indicators (scores) to the corresponding efficiency level. For example, a basic performance layer score of 0.75 and a membership vector of [0.75, 0.25, 0, 0] correspond to efficiency levels of excellent, good, average, and poor, respectively. In this embodiment of the invention, in addition to determining the overall efficiency level of the integrated space-ground network, the efficiency levels of each item in the first level are also determined. Each item in the first level can be considered a subsystem of the integrated space-ground network; evaluating the efficiency of the subsystems can identify the factors affecting the efficiency of the integrated space-ground network.

[0032] S106 displays the overall performance levels under various adversarial scenarios, a radar chart showing the score distribution of each item in the first level, and a curve showing the changes of each item in the second level before and after the adversarial process. For example, the overall performance levels are: excellent in normal state, good in inter-satellite link interference stage, medium in deception attack and node failure stage, and poor in flooding attack and electromagnetic suppression stage. In this embodiment of the invention, a subsystem performance radar chart can be drawn using the three items included in the first level as three dimensions to show the score distribution of each adversarial stage, so as to clearly present the weak links in the network (such as the network robustness layer score dropping to 0.45 in the deception attack and node failure stage); in addition, in this embodiment of the invention, by drawing the trend change curves of throughput, latency, and overall performance value throughout the adversarial process, the impact trend of the adversarial scenario on network performance can be intuitively reflected.

[0033] In one embodiment of the present invention, the performance evaluation method based on the integrated space-ground network simulation of simulated confrontation may further include the step of: standardizing the indicator data. Specifically, the performance indicators collected in real time at each stage can be stored in the InfluxDB database, and the min-max standardization method can be used to map the indicator values ​​to the [0,1] interval to eliminate the dimensional differences between different indicators (such as the unit difference between throughput "Mbps" and latency "ms"). Positive metrics (higher values ​​indicate better performance, such as throughput and link survivability) are calculated using the following formula: x' = (x - x) min ) / (x max - x min ); Negative metrics (smaller values ​​indicate better performance, such as latency, packet loss rate, and end-to-end jitter) are expressed using the following formula: x' = (xmax - x) / (x max - x min ); In the above formula, the parameters are defined as follows: x: Raw data collected for the metrics (e.g., raw measured throughput of 850 Mbps and raw measured latency of 35 ms in a certain scenario). x': Standardized index value (mapped to the [0,1] interval, used for subsequent weighted average score calculation); x max The historical maximum value of this indicator (or the preset performance upper limit threshold). x min The historical minimum value of this indicator (or the preset performance lower limit threshold).

[0034] In one embodiment of the present invention, the first-level items include: a basic performance layer, a network robustness layer, and a guarantee performance layer; in the second-level items, the basic performance layer includes throughput, latency, packet loss rate, and end-to-end jitter, the network robustness layer includes link survival probability and fault recovery efficiency, and the guarantee performance layer includes critical service completion rate and anti-interference duration.

[0035] The above-mentioned projects are selected based on the core characteristics of the integrated space-ground network: high dynamism, strong adversarial capabilities, and mission orientation. The first-level projects correspond to the network's "basic operational capabilities, damage resistance and recovery capabilities, and mission support capabilities," covering the core dimensions of the assessment and avoiding the one-sidedness of traditional assessments. The second-level projects are all quantifiable and easily collected key indicators, accurately matching the abstract requirements of the first level, ensuring that the assessment is scientific and operable. At the same time, they are adapted to the characteristics of scenarios such as dynamic switching of satellite links and susceptibility to interference, making the assessment results more in line with actual application needs.

[0036] In one embodiment of the present invention, the weights of items at each level can be determined based on expert scoring and consistency checks. Specifically, 5-7 experts in the field of network communication are invited to score the importance of each indicator at the same level pairwise, forming a judgment matrix. The matrix elements are quantitative judgments of indicator importance. The expert scoring forms the judgment matrix, which undergoes a consistency check (calculating a consistency check coefficient CR = 0.07, satisfying CR < 0.1, indicating no logical contradictions in the matrix) to ensure the weight allocation is scientific and objective. Then, the matrix eigenvectors are calculated and normalized to obtain the weights of each indicator (item). As an example, the weight of the basic performance layer is 0.4, and the weight of the network robustness layer is 0.35.

[0037] In one embodiment of the present invention, the adversarial scenario may include: inter-satellite link interference, satellite-to-ground link deception attack, satellite node failure, inter-satellite link failure and / or routing protocol adversarial.

[0038] The following detailed description of the integrated space-ground network simulation and countermeasure effectiveness evaluation system and method of the present invention is provided through specific embodiments 1 and 2, so as to clearly present the specific implementation process, parameter configuration and execution logic of the technical solution. Specific Implementation Example 1 This specific embodiment can be used in low-Earth orbit satellite networks (simulating the Starlink constellation architecture) scenarios.

[0040] I. System Deployment Environment (a) Hardware environment configuration Core server: It adopts Intel Xeon Gold 6330 processor (24 cores and 48 threads), and is equipped with 128GB DDR4 memory, 2TB SSD hard drive and 10Gbps Ethernet adapter to ensure high efficiency of simulation computing and data processing; Auxiliary simulation equipment: signal generator (supports a frequency range of 16 GHz, used to simulate interference signals), spectrum analyzer (to monitor the strength and frequency characteristics of interference signals in real time). Terminal equipment: 3 industrial-grade user terminals (compatible with satellite communication protocols), distributed in different geographical areas, simulating multi-user access scenarios.

[0041] (ii) The software environment configuration is shown in the table below.

[0042] II. System Module Parameter Configuration (I) Network Topology and Link Modeling 1. The topology includes: Space segment: Simulates 36 low-Earth orbit satellite nodes at an altitude of 550km, divided into 6 orbital planes (6 satellites in each orbital plane), with an orbital period of 95 minutes. The satellites move according to preset orbital parameters, dynamically forming inter-satellite links. Ground segment: Four ground stations are deployed (located in Beijing, Shanghai, Guangzhou, and Xi'an respectively), each with a coverage radius of 1000km, connecting 100 user terminal nodes (randomly distributed in the coverage area). Link connection: The inter-satellite link uses laser communication with a bandwidth of 10Gbps, and the propagation delay is dynamically calculated based on the relative position of the satellite (average delay of 20ms); the satellite-to-ground link uses the Ka band with a bandwidth of 5Gbps, and incorporates an atmospheric attenuation model (attenuation coefficient of 0.15dB / km in rainy weather) and Doppler frequency shift effect (maximum frequency deviation ±5kHz).

[0043] 2. SDN linkage configuration: Through the linkage between the ONOS controller and the simulation engine, a route reconstruction strategy is preset. When a link fails or is attacked, the route convergence time threshold is set to 500ms to ensure network service continuity.

[0044] (ii) Adversarial event library: Scene parameters and triggering modes are shown in the table below.

[0045] (III) Data Collection: The rules for collecting indicators are shown in the table below.

[0046] (iv) Configuration of performance evaluation algorithm parameters 1. Analytic Hierarchy Process (AHP) parameters: The performance evaluation system includes a first level and a second level, with the second level being a subset of the first level. The items and their weights in the first level are as follows: basic performance layer (weight 0.4), network robustness layer (weight 0.35), and security performance layer (weight 0.25). The second level includes the following items and weights: the basic performance layer includes throughput (0.3), latency (0.25), packet loss rate (0.25), and end-to-end jitter (0.2); the network robustness layer includes link lifetime (0.5) and fault recovery efficiency (0.5); and the guarantee performance layer includes service completion rate (0.6) and anti-interference duration (0.4).

[0047] Judgment matrix and consistency test: Five experts in the field of network communication were invited to score the importance of each indicator in pairs to form a judgment matrix. The consistency test coefficient CR was calculated to be 0.07 (satisfying CR<0.1), ensuring that the weight allocation is scientific and objective.

[0048] 2. Parameters of the multi-level fuzzy evaluation method: Performance rating scale: {Excellent, Good, Average, Poor}, corresponding to quantitative ranges [0.8, 1.0), [0.6, 0.8), [0.4, 0.6), [0, 0.4); Membership function: Use the triangular membership function, for example: Positive metrics (higher values ​​indicate better performance, such as throughput): When x∈[800,1000] Mbps (“Excellent” level core interval), the membership function is: μExcellent(x) = (x- 800) ÷ (1000 - 800); Additional notes: When x < 800 Mbps, μoptimal(x) = 0; when x > 1000 Mbps, μoptimal(x) = 1, to ensure that the trigonometric function fully covers the range of index values.

[0049] Negative metrics (lower values ​​indicate better performance, such as latency): When x∈[10,50] ms (the core interval of the "Excellent" level), the membership function is: μExcellent(x) = (50 -x) ÷ (50 - 10); Additional explanation: When x < 10ms, μ_excellence(x) = 1, and when x > 50ms, μ_excellence(x) = 0, which aligns with the evaluation logic of negative indicators.

[0050] III. Implementation of the Performance Evaluation Process Step 1: Building the network model and verifying connectivity Launch the NS3 simulation engine, load the above topology configuration parameters, simulate satellite orbital motion and dynamic link connections, and generate an integrated space-ground network model including satellite nodes, ground stations, and user terminals. Verify that the initial network connectivity is ≥98% by issuing connectivity detection commands through the ONOS controller, ensuring the model is effectively built.

[0051] Step 2: Inject the adversarial scenario Injecting combined adversarial scenarios according to time sequence to simulate complex dynamic adversarial environments: 030s: No adversarial response (normal operating state), record network baseline performance; 3060s: Inject random interference into inter-satellite links (interference power 20dBm); 60100s: Injection of a deceptive attack on the satellite-to-ground link + a single point of failure in one satellite node (directed triggering of satellite association with the Beijing ground station); 100150s: Injection of routing protocol flooding attack + custom area electromagnetic suppression interference (covering the Guangzhou ground station area).

[0052] Step 3: Data Acquisition and Standardization Processing The data acquisition module collects performance metrics at each stage in real time and stores them in the InfluxDB database. The min-max normalization method is used to map metric values ​​to the [0,1] interval to eliminate dimensionality differences. Positive metrics (such as throughput) are calculated using the following formula: Positive metrics (higher values ​​indicate better performance, such as throughput and link survivability) are expressed by the following formula: x' = (x - x min ) / (x max - x min ); Negative metrics (smaller values ​​indicate better performance, such as latency, packet loss rate, and end-to-end jitter) are expressed using the following formula: x' = (x max - x) / (x max - x min ); In the above formula, the parameters are defined as follows: x: Raw data collected for the metrics (e.g., raw measured throughput of 850 Mbps and raw measured latency of 35 ms in a certain scenario). x': Standardized index value (mapped to the [0,1] interval, used for subsequent weighted average score calculation); x max The historical maximum value of this indicator (or the preset performance upper limit threshold). x min The historical minimum value of this indicator (or the preset performance lower limit threshold).

[0053] Step 4: Multi-algorithm fusion evaluation 1. Subsystem score calculation: The weighted average method is used to calculate the score of the criteria layer according to the weight of each indicator. For example, the score of the basic performance layer = throughput normalized value × 0.3 + latency normalized value × 0.25 + packet loss rate normalized value × 0.25 + end-to-end jitter normalized value × 0.2; similarly, the scores of the network robustness layer and system performance layer are calculated.

[0054] 2. Fuzzy level mapping: Substitute the scores of each criterion layer into the membership function to obtain the fuzzy level membership vector. For example, if the basic performance layer score is 0.75, the membership vector is [0.75, 0.25, 0, 0] (corresponding to excellent, good, average, and poor).

[0055] 3. Comprehensive performance calculation: Combine the weights of the criterion layer (basic performance layer 0.4, network robustness layer 0.35, and guarantee performance layer 0.25) to perform fuzzy synthesis and calculate the comprehensive performance value. The formula is: comprehensive performance value = Σ (criterion layer score × corresponding weight), and output the performance level according to the quantization interval.

[0056] Step 5: Visualizing the Results Three types of evaluation results are output using visualization tools: Overall performance score: 0.92 (excellent) under normal conditions, 0.71 (good) under inter-satellite link interference, 0.58 (medium) under spoofing attack + node failure, and 0.35 (poor) under flooding attack + electromagnetic suppression. Subsystem Performance Radar Chart: Using the three items of the first level as three dimensions, the score distribution of each stage is displayed, clearly showing the weak links in the network (such as the network robustness layer score dropping to 0.45 during the deception attack + node failure stage). Trend curves: Plot the curves of throughput, latency, and overall performance value throughout the entire adversarial process to intuitively reflect the impact trend of the adversarial scenario on network performance.

[0057] IV. Implementation Results Verification 1. Standardization and repeatability verification: With fixed topology parameters, adversarial scenario parameters, and algorithm parameters, the experiment was repeated 3 times. The error of the comprehensive performance value at each stage was ≤3%, proving that the evaluation process was standardized and the results were repeatable. 2. Accuracy Verification: The performance level changes in adversarial scenarios are consistent with the actual network performance degradation trend. For example, after a node failure, the route reconstruction success rate drops from 99% to 62%, and the corresponding network robustness layer score drops from 0.82 to 0.45. The evaluation results are highly consistent with the actual situation. 3. Practicality Verification: Based on the evaluation results, the anti-interference strategy was optimized (the anti-interference power threshold of the inter-satellite link was increased to 15dBm, and the anti-flooding algorithm of the routing protocol was optimized). During the re-evaluation, the comprehensive performance value of the flooding attack + electromagnetic suppression phase was improved to 0.48 (from good to good), proving the system's supporting role in network optimization.

[0058] This embodiment fully verifies the accuracy, repeatability, and engineering applicability of the present invention in dynamic adversarial environments. Its modular design and standardized process can be flexibly adapted to various integrated space-ground architectures such as low-orbit satellite networks and space-based information networks, providing reliable technical support for anti-interference strategy optimization and network planning verification. Specific Implementation Example 2 This specific embodiment can be used in space-based information network industrial communication scenarios.

[0060] I. Hardware and software environment configuration (a) Hardware environment The core server utilizes an Intel Xeon Platinum 8480C processor with 40 cores and 80 threads, paired with 256GB of DDR5 memory, a 4TB SSD, and a 25Gbps Ethernet adapter, meeting the demands of large-scale simulation and high-density data processing. Auxiliary equipment includes a wideband signal generator (frequency range 0.5-20GHz, supporting the generation of complex electromagnetic interference signals) and a vector network analyzer (used for link parameter calibration and interference monitoring). Five industrial-grade ruggedized satellite communication terminals, compatible with industrial-grade communication protocols, are deployed on land and sea mobile platforms to simulate multi-domain collaborative communication scenarios.

[0061] (ii) Software Environment The simulation engine is built on Mininet 2.3.0 and integrates the ONOS 3.0.0 SDN controller, enhancing the linkage between dynamic routing and security policies. Data acquisition utilizes Mininet's built-in traffic monitoring tool combined with an industrial-grade data acquisition plugin to ensure the security and real-time nature of metric collection. Algorithm calculations rely on the Python 3.10 platform, coupled with NumPy, SciPy, and industrial-grade encrypted computing libraries to ensure that data processing complies with security standards. Visualization tools integrate Matplotlib with an industrial-grade visualization system to generate encrypted evaluation reports, supporting multiple presentation formats such as radar charts and trend curves. Data storage uses an industrial-grade encrypted database with a 30-day retention period, supporting full-process data traceability and auditing.

[0062] II. System Parameter Configuration (I) Simulation Engine: Network Topology and Link Modeling The space segment simulates 18 medium-to-high orbit satellite nodes, including 6 high-orbit geostationary satellites (35,786 km altitude) and 12 medium-orbit satellites (10,000 km altitude). The high-orbit satellites provide wide-area coverage, while the medium-orbit satellites handle regional relay. Inter-satellite links utilize millimeter-wave communication with a bandwidth of 20 Gbps and an average propagation delay of 50 ms. The ground segment deploys 6 industrial-grade ground stations (distributed across different strategic areas), each with a coverage radius of 2,000 km, connecting 200 industrial-grade user terminals (including fixed and mobile terminals). The space-to-ground links utilize industrial-grade UHF bands with a bandwidth of 8 Gbps, incorporating an anti-interference channel model that includes parameters such as electromagnetic shielding attenuation and human interference suppression, with a maximum Doppler shift of ±10 kHz. Through deep integration between the ONOS controller and the simulation engine, a pre-set military priority routing strategy ensures that the convergence time threshold for core operational services is ≤300 ms, guaranteeing the stability of critical communication links.

[0063] (II) Adversarial Event Library: Scene Configuration and Triggering Mode Inter-satellite link jamming employs broadband blocking jamming with a power of -15dBm, covering industrial-grade millimeter-wave communication bands, simulating high-intensity electromagnetic suppression, and using a directional triggering mode (targeting medium-Earth orbit satellite relay links), lasting 30-60 seconds. Satellite-to-ground link deception attacks forge industrial-grade satellite authentication signals and tamper with communication data payloads, achieving a 70% success rate, lasting 20 seconds, and directionally triggering ground station links in specific areas. Satellite node failures simulate energy system attacks and failures, with a failure probability of 0.1 / hour, supporting multi-point cascading failure triggering (2-3 satellites simultaneously failing in the same relay area). Routing protocol countermeasures include industrial-grade routing table tampering (40% tampering rate) and flooding of false routing information (data packet rate 1500pps), both using a directional triggering mode, targeting core command and control links. Customized countermeasure scenarios support a combined scenario of "regional electromagnetic silence + node destruction," with an interference coverage radius of 1000km, an interference intensity of -8dBm, and simultaneous triggering of failures at key ground stations and satellite nodes within the area, manually triggered.

[0064] (III) Data Collection: Indicator Collection Rules Throughput is statistically measured in 500ms increments (Gbps), representing the total amount of successfully transmitted industrial-grade valid data per second (excluding retransmitted packets and invalid data). End-to-end latency is statistically measured in 50ms increments (ms), recording the entire process time of industrial-grade data packets from the sending terminal to the receiving terminal (including encryption processing, routing, and anti-interference verification latency). Packet loss rate is calculated in 3s increments (%), calculated as (number of industrial-grade data packets sent - number of successfully received and verified data packets) / number of industrial-grade data packets sent × 100%. End-to-end jitter is statistically measured in 500ms increments (ms), reflecting latency fluctuations by calculating the standard deviation of the latency of 20 consecutive industrial-grade data packets, with a focus on ensuring the transmission stability of voice and video data for combat command.

[0065] (iv) Performance evaluation: Algorithm parameter configuration 1. Analytic Hierarchy Process (AHP): In the hierarchical structure of indicators, the objective is the integrated combat effectiveness of an industrial-grade space-ground integrated network. The first level includes a basic performance layer (weight 0.35), a network robustness layer (weight 0.4), and a support effectiveness layer (weight 0.25). In the second level, the basic performance layer includes throughput (0.3), latency (0.25), packet loss rate (0.25), and end-to-end jitter (0.2); the network robustness layer includes link survival probability (0.5) and fault recovery efficiency (0.5); and the support effectiveness layer includes critical service completion rate (0.6) and anti-interference duration (0.4). Seven experts and technical backbones in the field of industrial-grade communications were invited to score the indicators, forming a judgment matrix. The consistency test coefficient CR = 0.06 (satisfying CR < 0.1) to ensure that the weight allocation meets the requirements of military applications.

[0066] 2. Multi-level fuzzy evaluation method: The performance evaluation set is set as {Excellent, Good, Average, Poor}, corresponding to the quantification intervals [0.85, 1.0), [0.7, 0.85), [0.5, 0.7), and [0, 0.5). A trapezoidal membership function is used, as shown in the following example: Positive indicators (the larger the value, the better the performance, such as throughput): When x∈[15,20] Gbps (the "excellent" level range), the membership function is μexcel(x) = (x - 15) ÷ (20 - 15); where μexcel(x)=0 when x<15Gbps and μexcel(x)=1 when x≥20Gbps, ensuring that the trapezoidal function fully covers the range of indicator values.

[0067] Negative indicators (smaller values ​​indicate better performance, such as latency): When x∈[30,80] ms (the “excellent” level range), the membership function is μexcel(x) = (80 - x) ÷ (80 - 30); where μexcel(x)=1 when x<30ms and μexcel(x)=0 when x>80ms, which fits the performance evaluation logic of negative indicators.

[0068] This design is adapted to the stringent performance requirements of industrial communication, and can accurately map the actual values ​​of indicators to the hierarchical relationship of "excellent, good, average, and poor" levels, ensuring the scientific and targeted nature of performance level assessment.

[0069] III. Implementation of the Performance Evaluation Process Step 1: Building the network model and verifying connectivity The Mininet simulation engine is launched, and industrial-grade network topology configuration parameters are loaded. The simulation of satellite orbital motion, dynamic connection of industrial-grade links, and encryption authentication processes generates an integrated space-ground network model including medium- and high-orbit satellites, industrial-grade ground stations, and industrial-grade terminals. Industrial-grade connectivity detection commands are issued through the ONOS controller to verify an initial network connectivity rate of ≥99% and a critical link connectivity rate of 100%, ensuring the model meets the requirements of industrial communication scenarios.

[0070] Step 2: Inject the adversarial scenario The following composite adversarial scenarios are injected sequentially: 0-40s is the normal state without adversarial attacks, recording the network baseline performance; 40-90s injects inter-satellite link directional broadband blocking interference; 90-130s injects satellite-to-ground link deception attack + cascading failure of 2 medium-orbit satellites; 130-180s injects routing protocol adversarial attacks + a custom composite scenario of "regional electromagnetic silence + node destruction".

[0071] Step 3: Data Acquisition and Standardization Processing The data acquisition module collects industrial-grade performance indicators at each stage in real time, and stores them in an industrial-grade encrypted database after encryption to ensure data security and traceability. To eliminate dimensional differences between different types of indicators (such as the unit difference between throughput "Gbps" and latency "ms"), a min-max normalization method is used to uniformly map all indicator values ​​to the [0,1] interval. The specific formula is as follows: Positive metrics (higher values ​​indicate better performance, such as throughput and link survivability) are calculated using the following formula: x' = (x - x) min ) / (x max - x min ); Negative metrics (smaller values ​​indicate better performance, such as latency, packet loss rate, and end-to-end jitter) are expressed using the following formula: x' = (x max - x) / (x max - x min ).

[0072] In the above formula, the parameters are defined as follows: x: Raw collected values ​​of industrial-grade performance indicators (e.g., raw measured throughput of 18Gbps and raw measured latency of 45ms in a certain adversarial scenario). x': Standardized index value (range [0,1], used for subsequent weighted average calculation of performance score); x max This indicator corresponds to the upper limit threshold of industrial-grade communication performance (based on preset industrial-grade communication standards). x min The threshold value corresponding to this indicator is the lower limit of industrial-grade communication performance (based on preset values ​​of relevant industrial-grade communication standards).

[0073] Step 4: Multi-algorithm fusion evaluation The weighted average method is used to calculate the criterion layer score. The basic performance layer score = throughput standardized value × 0.3 + latency standardized value × 0.25 + packet loss rate standardized value × 0.25 + end-to-end jitter standardized value × 0.2. Similarly, the network robustness layer and the assurance effectiveness layer scores are calculated. Substituting each criterion layer score into the trapezoidal membership function, a fuzzy level membership vector is obtained. For example, if the network robustness layer score is 0.78, the membership vector is [0.53, 0.47, 0, 0] (corresponding to excellent, good, average, and poor). Fuzzy synthesis is performed by combining the criterion layer weights to calculate the comprehensive combat effectiveness value. The formula is: comprehensive effectiveness value = Σ (criterion layer score × corresponding weight). The effectiveness level is output according to the industrial-grade effectiveness quantification range.

[0074] Step 5: Visualizing the Results In terms of overall performance score, the scores are as follows: 0.95 (Excellent) under normal conditions, 0.76 (Good) during inter-satellite link interference, 0.62 (Medium) during deception attack + node failure, and 0.42 (Poor) during complex adversarial scenarios. The subsystem performance radar chart uses the three items of the first level as three dimensions to clearly present the performance shortcomings at each stage, such as the network robustness layer score dropping to 0.38 during the complex adversarial stage. Trend curves plot the changes in key industrial-grade indicators (throughput, latency, and critical business completion rate) and overall performance values ​​throughout the entire adversarial process, intuitively reflecting the impact of adversarial scenarios on industrial communications.

[0075] IV. Implementation Results Verification 1. Standardization and repeatability verification: Fix industrial-grade topology parameters, adversarial scenario parameters, and algorithm parameters, repeat the experiment 5 times, and the error of the comprehensive performance value at each stage is ≤2.5%, which meets the standardization and repeatability requirements of the industrial evaluation process.

[0076] 2. Accuracy Verification: The changes in effectiveness levels under adversarial scenarios are consistent with the degradation trend of actual combat performance of industrial-grade networks. For example, after multiple satellite failures, the network robustness layer score dropped from 0.91 to 0.38, which is highly consistent with the proportion of critical industrial link interruptions and fault recovery time.

[0077] 3. Practicality Verification: Based on the evaluation results, the industrial-grade anti-interference strategy was optimized, the inter-satellite link anti-interference power threshold was upgraded to -10dBm, and an industrial-grade dynamic encryption and verification mechanism for routing was added. In the re-evaluation, the comprehensive performance value in the complex adversarial scenario stage was improved to 0.57 (from medium to medium-good), effectively enhancing the resilience of the industrial communication network.

[0078] Example 2 like Figure 2 As shown, another aspect of the present invention also includes a functional module architecture that is completely consistent with the aforementioned method flow. That is, the embodiments of the present invention also provide an anti-interference robust spectrum access control system based on fuzzy game theory, including: Simulation model construction module 201 is used to construct a simulation model of the integrated space-ground network topology using the simulation engine; The adversarial simulation module 202 is used to inject adversarial events into the simulation model to simulate adversarial scenarios; Data acquisition module 203 is used to collect network performance index data in adversarial scenarios; The scoring calculation module 204 at each level is used to calculate the second-level score based on the collected indicator data, using a weighted average method and the item weights included in each level of the pre-constructed two-level performance evaluation system. Then, the first-level score is calculated based on the second-level score. The two-level performance evaluation system includes a first level and a second level, and the second level is a sub-level of the first level. The performance rating module 205 is used to obtain the performance rating of each item in the first level based on the score of the second level using the fuzzy evaluation method; and to obtain the comprehensive performance rating of the integrated space-ground network based on the score of the first level using the fuzzy evaluation method. The performance evaluation results presentation module 206 is used to display the comprehensive performance level under various confrontation scenarios, the score distribution radar chart of each item in the first level, and the change curve of each item before and after the confrontation in the second level.

[0079] Furthermore, the performance evaluation system based on space-ground integrated network simulation and confrontation provided by the present invention also includes a data processing module for standardizing the indicator data.

[0080] Furthermore, the first-level projects include: basic performance layer, network robustness layer, and assurance performance layer; in the second-level projects, the basic performance layer includes throughput, latency, packet loss rate, and end-to-end jitter, the network robustness layer includes link survivability probability and fault recovery efficiency, and the assurance performance layer includes critical service completion rate and anti-interference duration.

[0081] Furthermore, the weights of projects at each level are determined based on expert scoring and consistency checks.

[0082] Furthermore, the adversarial scenarios include: inter-satellite link interference, satellite-to-ground link deception attacks, satellite node failures, inter-satellite link failures, and / or routing protocol adversarial attacks.

[0083] The system can be implemented using the anti-interference robust spectrum access control method based on fuzzy game theory provided in Embodiment 1 above. For the specific implementation method, please refer to the description in Embodiment 1, which will not be repeated here.

[0084] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A method for evaluating the effectiveness of simulated combat based on a space-ground integrated network, characterized in that, include: Construct a simulation model of an integrated space-ground network topology using a simulation engine; Inject adversarial events into the simulation model to simulate adversarial scenarios; Collect network performance metrics data in adversarial scenarios; Based on the collected indicator data, a weighted average method is used. The weights of the items included in each level of the pre-constructed two-level performance evaluation system are used to first calculate the score of the second level, and then the score of the first level is calculated based on the score of the second level. The two-level performance evaluation system includes a first level and a second level, and the second level is a sub-level of the first level. Based on the scores of the second level, the effectiveness level of each item in the first level is obtained using the fuzzy evaluation method; and based on the scores of the first level, the comprehensive effectiveness level of the integrated space-ground network is obtained using the fuzzy evaluation method. It displays the comprehensive effectiveness levels under various adversarial scenarios, the radar chart of the score distribution of each item in the first level, and the change curves of each item before and after the adversarial in the second level.

2. The effectiveness evaluation method for simulated confrontation based on a space-ground integrated network as described in claim 1, characterized in that, It also includes the step of standardizing the indicator data.

3. The effectiveness evaluation method for simulated confrontation based on a space-ground integrated network as described in claim 1, characterized in that, The first level of projects includes: basic performance layer, network robustness layer, and assurance performance layer; the second level of projects includes: basic performance layer including throughput, latency, packet loss rate, and end-to-end jitter; network robustness layer including link survivability probability and fault recovery efficiency; and assurance performance layer including critical service completion rate and anti-interference duration.

4. The effectiveness evaluation method for simulated confrontation based on a space-ground integrated network as described in claim 1, characterized in that, The weights of projects at each level are determined based on expert scoring and consistency checks.

5. The effectiveness evaluation method for simulated confrontation based on a space-ground integrated network as described in claim 1, characterized in that, The adversarial scenarios include: inter-satellite link interference, satellite-to-ground link deception attacks, satellite node failures, inter-satellite link failures, and / or routing protocol adversarial attacks.

6. A performance evaluation system based on space-ground integrated network simulation of adversarial warfare, characterized in that, include: The simulation model building module is used to build a simulation model of the integrated space-ground network topology using the simulation engine. The adversarial simulation module is used to inject adversarial events into the simulation model to simulate adversarial scenarios; The data acquisition module is used to collect network performance metrics data in adversarial scenarios. Each level of the scoring calculation module is used to calculate the second-level score based on the collected indicator data, using a weighted average method and the item weights included in each level of the pre-constructed two-level performance evaluation system. Then, the first-level score is calculated based on the second-level score. The two-level performance evaluation system includes a first level and a second level, and the second level is a sub-level of the first level. The performance rating module is used to obtain the performance rating of each item in the first level based on the score of the second level using the fuzzy evaluation method; and to obtain the comprehensive performance rating of the integrated space-ground network based on the score of the first level using the fuzzy evaluation method. The performance evaluation results presentation module is used to display the comprehensive performance level under various confrontation scenarios, the radar chart of the score distribution of each item in the first level, and the change curve of each item before and after the confrontation in the second level.

7. The performance evaluation system for simulated combat based on a space-ground integrated network as described in claim 6, characterized in that, It also includes a data processing module for standardizing the indicator data.

8. The performance evaluation system for simulated combat based on a space-ground integrated network as described in claim 6, characterized in that, The first level of projects includes: basic performance layer, network robustness layer, and assurance performance layer; the second level of projects includes: basic performance layer including throughput, latency, packet loss rate, and end-to-end jitter; network robustness layer including link survivability probability and fault recovery efficiency; and assurance performance layer including critical service completion rate and anti-interference duration.

9. The performance evaluation system for simulated combat based on a space-ground integrated network as described in claim 6, characterized in that, The weights of projects at each level are determined based on expert scoring and consistency checks.

10. The performance evaluation system for simulated combat based on a space-ground integrated network as described in claim 6, characterized in that, The adversarial scenarios include: inter-satellite link interference, satellite-to-ground link deception attacks, satellite node failures, inter-satellite link failures, and / or routing protocol adversarial attacks.

Images