A method for evaluating the anti-attack performance of aircraft swarm networks

Abstract

The present application discloses a method for evaluating the anti-attack performance of an aircraft swarm network, comprising: obtaining a two-dimensional space model, wherein the two-dimensional space model includes a plurality of aircraft; obtaining positioning information of each aircraft in the two-dimensional space model, wherein the positioning information includes speed information of the aircraft in the two-dimensional space model at each moment as position information; generating an aircraft swarm model based on the positioning information of the plurality of aircraft; the aircraft swarm model is configured to determine the probability of successfully establishing a communication connection between any two aircraft in the plurality of aircraft; determining a target aircraft as an attack object based on the aircraft swarm model; performing an attack simulation on the target aircraft using a preset attack method to obtain a swarm network after the attack; performing network recovery on the swarm network after the attack based on a preset recovery method; and performing performance evaluation on the swarm network after network recovery, thereby effectively improving evaluation efficiency and accuracy.

Description

Technical Field

[0001] This specification relates to the technical field of aircraft swarm networks, and in particular to a method for evaluating the anti-attack performance of an aircraft swarm network. Background Art

[0002] An aircraft swarm, also known as a drone swarm, consists of a group of coordinated drones that work together to complete tasks that would be difficult or impossible for a single drone. Drone swarms offer several advantages over individual drones, including enhanced coverage, redundancy, and resilience. The successful deployment and operation of drone swarms depends on a robust and efficient communications network. Due to the high variability of drone swarm networks, drone swarms utilize self-organizing networks, allowing drone nodes to autonomously establish and maintain connections without the need for centralized planning by a network center. However, if some drone nodes in a swarm network fail due to hardware failure, environmental interference, or deliberate attacks, the drone swarm network becomes fragmented.

[0003] To enhance network and system security, methods for protecting communication networks from attacks have been a recent research hotspot. Furthermore, network performance assessments (such as resilience assessment methods) are also a crucial component of strengthening communication network security. However, most of these methods focus on non-UAV domains or propose defense strategies solely against spoofing attacks. Furthermore, many resilience assessment methods often focus on a single dimension, such as system recovery time, performance degradation, or vulnerability, while neglecting comprehensive considerations across multiple dimensions. This can lead to underestimation or overestimation of a system's true resilience.

[0004] Therefore, providing an anti-attack performance evaluation method for an aircraft swarm network can effectively improve the evaluation efficiency and accuracy. Summary of the Invention

[0005] One or more embodiments of the present specification provide a method for evaluating the anti-attack performance of an aircraft cluster network, comprising: obtaining a two-dimensional space model, wherein the two-dimensional space model includes a plurality of aircraft; obtaining positioning information of each aircraft among the plurality of aircraft in the two-dimensional space model, wherein the positioning information includes speed information and position information of the aircraft in the two-dimensional space model at each moment; generating an aircraft cluster model based on the positioning information of the plurality of aircraft; the aircraft cluster model is configured to determine the probability of successfully establishing a communication connection between any two aircraft among the plurality of aircraft; determining a target aircraft as an attack object based on the aircraft cluster model; performing an attack simulation on the target aircraft using a preset attack method to obtain a cluster network after the attack; performing network recovery on the cluster network after the attack based on a preset recovery method; and performing performance evaluation on the cluster network after the network recovery. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, like numbers represent like structures, wherein:

[0007] Figure 1 is an exemplary module diagram of an anti-attack performance evaluation system for an aircraft swarm network according to some embodiments of this specification;

[0008] Figure 2 is an exemplary flow chart of a method for evaluating the anti-attack performance of an aircraft swarm network according to some embodiments of this specification;

[0009] Figure 3a is a schematic diagram of a cluster network before an attack according to some embodiments of this specification;

[0010] Figure 3b is a schematic diagram of a cluster network after an attack according to some embodiments of this specification;

[0011] Figure 4 is an exemplary schematic diagram of a required power determination model according to some embodiments of this specification;

[0012] Figure 5 is a schematic diagram of a curve showing changes in a drone swarm network connectivity indicator after a single attack according to some embodiments of this specification;

[0013] Figure 6a-6c is a schematic diagram showing a connectivity network of a drone swarm after being attacked by a corresponding attack mode according to some embodiments of this specification;

[0014] Figure 7 This is a schematic diagram of energy consumption of drone cluster communication in corresponding attack modes according to some embodiments of this specification;

[0015] Figure 8a-8c This is a schematic diagram comparing the resilience of a cluster network after recovering from attacks of corresponding attack modes according to some embodiments of this specification. DETAILED DESCRIPTION

[0016] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0017] It should be understood that the terms "system," "device," "unit," and / or "module" used herein are a method for distinguishing different components, elements, parts, portions, or assemblies at different levels. However, if other terms can achieve the same purpose, the terms may be replaced by other expressions.

[0018] As used in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not refer to the singular but also include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0019] Flowcharts are used throughout this specification to illustrate the operations performed by systems according to embodiments of this specification. It should be understood that preceding or following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0020] Drones, as a key vehicle for low-altitude flight activities, are a leading sector of the low-altitude economy. Empowered by technologies such as automation, intelligence, and connectivity, drones are poised to play an increasingly important role in various industrial applications, logistics, transportation, urban air mobility, and other fields. However, the effectiveness of individual drones is often limited by their operating range, battery life, and susceptibility to interference or damage. These limitations can severely impact mission accomplishment, especially in complex or large-scale operations. To address these limitations, researchers are increasingly interested in drone swarms. A drone swarm is a group of coordinated drones that collectively complete tasks that would be difficult or impossible for a single drone to accomplish. Compared to individual drones, drone swarms offer several advantages, including enhanced coverage, redundancy, and resilience. For example, when conducting surveillance missions, drone swarms can cover larger areas more quickly and thoroughly than a single drone. If one drone fails, others can continue the mission, increasing the overall success rate. The successful deployment and operation of drone swarms depends on a robust and efficient communication network. Due to the highly variable nature of drone swarm networks, drone swarms employ self-organizing networks, where nodes autonomously establish and maintain connections, eliminating the need for centralized planning by a network center. Once some drone nodes in a drone network fail due to hardware failure, environmental interference or deliberate attacks, the drone swarm network will become "fragmented".

[0021] In order to improve the security of the cluster network, evaluating it is an indispensable part so that targeted performance improvement can be carried out based on the evaluation results. Some embodiments of the present invention provide an anti-attack performance evaluation method, system, and device for an aircraft cluster network, which can realize effective evaluation of the cluster network, effectively improve the speed and efficiency of the evaluation, and at the same time, combine actual needs to conduct multi-dimensional evaluation of the cluster network, so as to further improve the accuracy of the evaluation, and thereby obtain a more accurate and effective security improvement strategy.

[0022] Figure 1 It is an exemplary module diagram of an anti-attack performance evaluation system for an aircraft cluster network (hereinafter referred to as anti-attack performance evaluation system 100 ) according to some embodiments of this specification.

[0023] like Figure 1 As shown, the anti-attack performance evaluation system 100 may include a first acquisition module 110, a second acquisition module 120, a model generation module 130, a target determination module 140, an attack simulation module 150, a network recovery module 160, and a performance evaluation module 170. In some embodiments, all or part of the modules of the anti-attack performance evaluation system 100 may be implemented by a processor.

[0024] In some embodiments, the first acquisition module 110 is configured to: acquire a two-dimensional space model.

[0025] In some embodiments, the second acquisition module 120 is configured to: acquire positioning information of each of the multiple aircraft in the two-dimensional space model.

[0026] In some embodiments, the model generation module 130 is configured to generate an aircraft fleet model based on the positioning information of the plurality of aircraft.

[0027] In some embodiments, the target determination module 140 is configured to determine a target aircraft to be attacked based on the aircraft swarm model. The target aircraft includes at least one of a first target aircraft, a second target aircraft, and a third target aircraft. The first target aircraft is at least one aircraft randomly selected from the plurality of aircraft. The second target aircraft is an aircraft among the plurality of aircraft whose degrees of freedom meet a preset degree of freedom condition. The third target aircraft is an aircraft among the plurality of aircraft whose centrality meets a preset centrality condition, where the centrality is determined based on the aircraft swarm model.

[0028] In some embodiments, the attack simulation module 150 is configured to: use a preset attack method to simulate an attack on the target aircraft to obtain a post-attack fleet network.

[0029] In some embodiments, the attack simulation module 150 is further configured to: in response to the target aircraft being the first target aircraft, interfere with the network of the first target aircraft by means of random hardware failure or simulated environmental interference; in response to the target aircraft being the second target aircraft and / or the third target aircraft, interfere with the network of the second target aircraft and / or the third target aircraft by means of simulated human attack.

[0030] In some embodiments, the network recovery module 160 is configured to perform network recovery on the cluster network after the attack based on a preset recovery method.

[0031] In some embodiments, the network recovery module 160 is further configured to: obtain several connected cluster slices in the cluster network after the attack, wherein the connected cluster slices include at least one aircraft; perform degree of freedom preprocessing on each aircraft in each of the connected cluster slices, and determine the communication capability level of each aircraft; establish communication connections between aircraft whose communication capability level differences are within a preset difference range.

[0032] In some embodiments, the performance evaluation module 170 is configured to perform a performance evaluation on the cluster network after the network is restored.

[0033] In some embodiments, the performance evaluation module 170 is further configured to: perform connectivity level analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain network connectivity before recovery and network connectivity after recovery; and obtain a network connectivity evaluation result based on the network connectivity before recovery and network connectivity after recovery.

[0034] In some embodiments, the performance evaluation module 170 is further configured to: perform communication power demand analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain the power demand before restoration and the power demand after restoration; and obtain a power demand evaluation result based on the power demand before restoration and the power demand after restoration.

[0035] For more information about the first acquisition module 110, the second acquisition module 120, the model generation module 130, the target determination module 140, the attack simulation module 150, the network recovery module 160, and the performance evaluation module 170, please refer to the relevant description below.

[0036] It should be noted that the above description of the anti-attack performance evaluation system 100 and its modules is for convenience only and does not limit this specification to the scope of the embodiments. It is understandable that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the modules or form subsystems connected with other modules without deviating from the principles. In some embodiments, Figure 1 The first acquisition module 110, the second acquisition module 120, the model generation module 130, the target determination module 140, the attack simulation module 150, the network recovery module 160, and the performance evaluation module 170 disclosed in the specification can be different modules in a system, or a module can implement the functions of two or more of the above modules. For example, each module can share a storage module, or each module can have its own storage module. Such variations are all within the scope of protection of this specification.

[0037] The present specification provides a method for evaluating the anti-attack performance of an aircraft swarm network, the method comprising: obtaining a two-dimensional space model; obtaining positioning information of each aircraft among the multiple aircraft in the two-dimensional space model; generating an aircraft swarm model based on the positioning information of the multiple aircraft; determining a target aircraft as an attack object based on the aircraft swarm model; performing an attack simulation on the target aircraft using a preset attack method to obtain a swarm network after the attack; performing network recovery on the swarm network after the attack based on a preset recovery method; and performing performance evaluation on the swarm network after the network recovery.

[0038] Figure 2 This is an exemplary flow chart of a method for evaluating the anti-attack performance of an aircraft fleet network according to some embodiments of this specification. Figure 2 As shown, the process 200 includes the following steps. In some embodiments, the process 200 can be executed by a processor.

[0039] It should be noted that the processor is a processor configured in a control system that integrates one or more modules of the anti-attack performance evaluation system. The control system is a system with computing capabilities, such as a computer, an industrial computer, a computing cloud platform, etc. The processor can execute program instructions to implement the anti-attack performance evaluation method for an aircraft swarm network described in some embodiments of this specification. The processor can include various common general-purpose central processing units (CPUs), graphics processing units (GPUs), microprocessors, application-specific integrated circuits (ASICs), or other types of integrated circuits.

[0040] In some embodiments, the control system may include a storage medium that can store instructions or data. The storage medium may include a mass storage device, a removable memory, a volatile read-write memory, a read-only memory (ROM), or any combination thereof.

[0041] In some embodiments, the control system can establish a data connection with the aircraft via a wired network or a wireless network. The following describes this solution using a drone as an example. In practice, the aircraft can also be an unmanned helicopter, an unmanned airship, an unmanned underwater vehicle, or other intelligent flying device.

[0042] Step 210: Acquire a two-dimensional space model.

[0043] In some embodiments, the processor can obtain a two-dimensional spatial model using various methods. For example, the processor can simplify the drone swarm movement model into an m*m two-dimensional space to obtain a two-dimensional spatial model. The drone swarm movement model can be read from the drone control system or determined based on user input.

[0044] In some embodiments, the two-dimensional space model includes multiple aircraft. For example, n drones are distributed in an m*m two-dimensional space, where each drone uses a unique identifier ID i .

[0045] Step 220: Obtain positioning information of each of the multiple aircraft in the two-dimensional space model.

[0046] In some embodiments, the positioning information includes the speed information of the aircraft in the two-dimensional space model at each time as position information. For example, the speed of the aircraft at time t can be described as (vx i ,vy i ), the position is (x i ,y i ).

[0047] In some embodiments, the processor may obtain positioning information of each aircraft in the two-dimensional space model based on the GPS positioning system of the drone.

[0048] Step 230: Generate an aircraft fleet model based on the positioning information of the multiple aircraft.

[0049] In some embodiments, the processor may construct an aircraft swarm model based on the acquired information and the communication examples between each aircraft.

[0050] In some embodiments, the processor may construct the following aircraft fleet model:

[0051]

[0052] Among them, F(d j→i ) represents the probability of successfully establishing a communication connection between the jth aircraft and the ith aircraft; α represents the item weight; N t is the number of aircraft in the fleet network at time t; ε represents the adjustment coefficient; k i represents the degree of freedom of the i-th aircraft, k l represents the degree of freedom of the lth aircraft; d j→i represents the communication distance between the jth aircraft and the ith aircraft; P j→i It represents the probability of successfully establishing a communication connection between the j-th aircraft and all aircraft in the fleet where the i-th aircraft is located.

[0053] In some embodiments, the degrees of freedom of an aircraft include a communication upper limit of the aircraft, that is, the maximum number of aircraft that the aircraft can connect to communicate with under the current circumstances (such as within the maximum communication range), which can be obtained based on the communication control system of the aircraft.

[0054] In some embodiments, the probability of successfully establishing a communication connection between the j-th aircraft and the i-th aircraft can be represented based on the following formula:

[0055]

[0056] Among them, r c represents the communication range, and η represents the distance impact factor.

[0057] In some embodiments, when When , it means that the probability of the j-th aircraft successfully establishing a communication connection with other aircraft is 0, that is, there is no communicable aircraft within the maximum communication range of the j-th aircraft.

[0058] Step 240: Determine a target aircraft to be attacked based on the aircraft swarm model.

[0059] In some embodiments, the target aircraft includes at least one of a first target aircraft, a second target aircraft, and a third target aircraft.

[0060] Among them, the first target aircraft is at least one aircraft randomly selected from the multiple aircraft; for example, the processor can randomly select m nodes from n nodes (i.e., n aircraft, and subsequent nodes are the alias of aircraft in the swarm network) to make them invalid.

[0061] The second target aircraft is an aircraft among the multiple aircraft whose degrees of freedom meet a preset degree of freedom condition. For example, the second target aircraft is a node with the highest degree of freedom among the n nodes, or a node ranked in the top m in the order of the degrees of freedom.

[0062] The third target aircraft is an aircraft among the plurality of aircraft whose centrality satisfies a preset centrality condition, which may be the highest centrality or the top m in centrality ranking.

[0063] Centrality refers to the centrality of the node in the communication network. Nodes with high centrality play an important bridge role in the network. Attacking these nodes will seriously affect the data transmission efficiency and robustness of the network, and achieve a better attack simulation effect.

[0064] The centrality can be determined based on the aircraft swarm model. For example, the processor can cluster nodes based on the positioning information and degrees of freedom of the aircraft corresponding to each node, as well as the probability mean of each aircraft successfully establishing a communication connection with other aircraft, to obtain a number of clusters, with the cluster center of each cluster serving as the third target aircraft. The probability of each aircraft successfully establishing a communication connection with other aircraft can be determined based on the aircraft swarm model.

[0065] For example only, the clustering algorithm may include but is not limited to K-Means clustering and / or density-based clustering method (DBSCAN), etc.

[0066] Step 250: Use a preset attack method to simulate an attack on the target aircraft to obtain a post-attack fleet network.

[0067] In some embodiments, different attack methods are preset for different types of target aircraft.

[0068] In some embodiments, in response to the target aircraft being the first target aircraft, the processor may interfere with the network of the first target aircraft by simulating random hardware failure or simulating environmental interference.

[0069] For example, random hardware failure simulation may include signal receiving antenna failure, power supply failure, etc. Simulated environmental interference may include simulated stormy weather, simulated high or low temperature weather, etc.

[0070] In some embodiments, in response to the target aircraft being the second target aircraft and / or the third target aircraft, a simulated human attack is used to interfere with the network of the second target aircraft and / or the third target aircraft.

[0071] For example, methods of simulating human attacks may include deception attacks, network intrusions, denial of service attacks, data theft, and the like.

[0072] Step 260: Perform network recovery on the attacked cluster network based on a preset recovery method.

[0073] After an attack occurs, the aircraft in the swarm network will be divided into several connected slices, each of which includes at least one aircraft. The conventional recovery method is to directly perform self-detection and self-recovery through each connected slice, but this method has the problems of poor recovery effect and long recovery time.

[0074] In some embodiments, the preset recovery method includes:

[0075] S262: Acquire several connected cluster slices in the cluster network after the attack.

[0076] For example, interconnected nodes may be considered to be the same connected cluster slice. Each connected cluster slice includes at least one aircraft. The processor may obtain each connected cluster slice separately.

[0077] like Figure 3a and Figure 3b This is a schematic diagram of a cluster network before and after an attack according to some embodiments of this specification; as can be seen from the figure, after the cluster network is attacked, multiple connected cluster slices will be formed.

[0078] S264: Perform degree of freedom preprocessing on each aircraft in each of the connected cluster slices, and determine the communication capability level of each aircraft.

[0079] In some embodiments, the processor may perform degree of freedom pre-processing on each aircraft in the following manner:

[0080] In response to the degree of freedom of the aircraft being less than a minimum threshold, the communication capability level of the aircraft is increased by one level; in response to the degree of freedom of the aircraft being greater than a maximum threshold, the communication capability level of the aircraft is decreased by one level.

[0081] The minimum threshold and maximum threshold can be set based on historical experience, such as the minimum threshold is 2 and the maximum threshold is 50.

[0082] In some embodiments, the communication capability level can be determined based on the aircraft's degrees of freedom. For example, the processor can determine the aircraft's communication capability level by querying a level comparison table based on the aircraft's degrees of freedom. The level comparison table can be a pre-set comparison table based on historical data and historical experience, and includes corresponding communication capability levels for each degree of freedom, e.g., a greater degree of freedom corresponds to a greater communication capability level.

[0083] In some embodiments, each aircraft also includes a signal transmitter with multiple random links, which is communicatively connected to the processor. The signal transmitter can be used to transmit power gears. When the aircraft is in daily use, the gear is small. When the aircraft is attacked and a larger connecting hole appears, the gear gradually increases until the gear threshold is met.

[0084] In some embodiments, the communication capability level of an aircraft is also related to the aircraft's positioning information and its communication distance with other aircraft. For example, a level comparison table also includes a correspondence between different aircraft's positioning information, communication distances with other aircraft, degrees of freedom, and communication capability levels. The processor can further obtain the aircraft's positioning information and its communication distance with other aircraft and determine the aircraft's communication capability level based on the level comparison table.

[0085] S266: Establishing a communication connection between the aircraft whose communication capability level difference is within a preset difference range.

[0086] In some embodiments, the preset difference range can be preset based on demand. For example, the preset difference range can be 0 or 1. For example, when the preset difference range is 0, the processor can control the establishment of a communication connection between two aircraft with equal communication capability levels. In some embodiments, when the number of such communication connections reaches a certain connection threshold, connected swarm slices of the swarm network begin to combine until a large connected component is formed. This large connected component includes most of the peer networks in the attacked swarm network, thereby enabling the gradual recovery of the swarm network.

[0087] Step 270: Perform performance evaluation on the cluster network after the network is restored.

[0088] In some embodiments, the performance evaluation may include a network connectivity performance evaluation, where the network connectivity performance evaluation includes performing a connectivity level analysis on the cluster network after the attack and the cluster network after the network is restored.

[0089] In some embodiments, when performing a network connectivity performance evaluation, the processor may perform a connectivity level analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain the network connectivity before recovery and the network connectivity after recovery, and obtain a network connectivity evaluation result based on the network connectivity before recovery and the network connectivity after recovery.

[0090] In some embodiments, the connectivity level analysis includes:

[0091] S101: Control each of the plurality of aircraft except the target aircraft, and generate a probability μ according to the information at the corresponding time t based on a preset period. t , generate test information and send it to the target aircraft.

[0092] in, μ0 represents the probability of generating initial information, which can be obtained based on the preset. t represents the number of aircraft at time t, and N represents the total number of aircraft in the two-dimensional space model.

[0093] S102: Obtain the total amount of information y(t) acquired by the target aircraft at time t:

[0094] Among them, R i (t) represents the number of messages received by the i-th aircraft, is the shortest path length between the jth aircraft and the ith aircraft obtained by the shortest path algorithm, N t represents the number of aircraft at time t, Δ represents the time sensitivity coefficient, 0<Δ≤1;

[0095] S103: Determine the network connectivity before restoration or the network connectivity after restoration based on the total amount of the information acquired at multiple moments.

[0096] In some embodiments, since the shortest path length of each piece of information may affect the real-time nature of information exchange within the aircraft fleet, the processor may determine the system performance recovery at time t based on the total amount of information acquired by the target aircraft at time t.

[0097] In some embodiments, the data sent by each aircraft are random numbers, and the processor can perform multiple simulations and take the average of the total number of multiple pieces of information obtained from the multiple simulations as the final total number of information.

[0098] In some embodiments, the processor may obtain the total amount of information determined based on data at multiple times before recovery as network connectivity before recovery, and the total amount of information determined based on data at multiple times after recovery as network connectivity after recovery in the above manner.

[0099] In some embodiments, the processor may obtain a network connectivity assessment result based on the network connectivity before and after recovery. For example, the processor may determine the network connectivity assessment result based on the difference in the total number of messages before and after recovery using a first result comparison table. The first result comparison table may be preset based on historical experience, and the first result comparison table may record network connectivity assessment results corresponding to the difference in the total number of messages. For example, if the network connectivity after recovery is greater than the network connectivity before recovery, and the difference between the two is greater, the better the network connectivity assessment result.

[0100] In some embodiments, the performance evaluation may further include a power demand evaluation, where the power demand evaluation includes performing a communication power demand analysis on the cluster network after the attack and the cluster network after the network is restored.

[0101] In some embodiments, when performing a power requirement assessment, the processor can perform communication power requirement analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain the power requirement before recovery and the power requirement after recovery, and obtain a power requirement assessment result based on the power requirement before recovery and the power requirement after recovery.

[0102] In some embodiments, the processor performs communication power demand analysis including: for each aircraft, determining the power demand before recovery and the power demand after recovery by using a power demand determination model based on the communication distance between the aircraft and other aircraft.

[0103] In some embodiments, the required power determination model may be a free space path loss (FSPL) communication model. The free space path loss communication model may be used to determine the increase in energy consumption due to an increase in communication distance.

[0104] Free-space path loss (FSPL) is a metric in wireless communications that describes the attenuation of signal strength as radio waves propagate in free space. The free-space path loss communication model characterizes the decrease in signal power as radio waves propagate over distance, unaffected by obstacles, reflections, or scattering. Because drones typically fly at altitudes between 100 and 1000 meters, with few obstacles blocking communications, a free-space path loss communication model can be constructed to characterize the relationship between communication distance and power.

[0105] In some embodiments, the free space path loss communication model is characterized as follows:

[0106]

[0107] Where d is the communication distance in meters, c is the speed of light in meters per second, and f is the signal frequency in Hz.

[0108] In some embodiments, in order to more intuitively represent the relationship between communication distance and communication power, the processor may further convert the above-mentioned free space path loss communication model and obtain the following converted model:

[0109]

[0110] Among them, G t is the transmitting antenna gain, in decibels (dB); G r is the receiving antenna gain, in decibels (dB); P sens is the receiving sensitivity in decibel milliwatts (dBm).

[0111] In some embodiments, for ease of calculation, the processor may convert the transmit power from decibel milliwatts (dBm) to milliwatts (mW) in the following manner:

[0112]

[0113] From the above content, it can be concluded that as the communication distance increases, the increase in communication power becomes more and more obvious. Therefore, the change in communication distance can be determined based on the communication power, and then the change in its communication capability can be obtained.

[0114] In some embodiments, the required power determination model may be a machine learning model, such as a convolutional neural network model. Figure 4 The corresponding content.

[0115] In some embodiments, the processor may obtain a power demand assessment result based on the power demand before recovery and the power demand after recovery. For example, the processor may determine a difference in communication distance based on the difference in power demand before and after recovery, thereby determining a difference in network performance before and after recovery. For example, if the power demand after recovery is higher than before recovery, it indicates that the communication distance after recovery is higher than before recovery.

[0116] For example only, the processor may determine the power requirement assessment result based on the power requirement difference using a second result comparison table. The second result comparison table may be preset based on historical experience and record power requirement assessment results corresponding to different communication distance differences. For example, if the power requirement after recovery is greater than the power requirement before recovery, and the greater the difference between the two, the better the power requirement assessment result.

[0117] In some embodiments, the processor can jointly perform a performance evaluation on the cluster network after the network is restored based on the network connectivity evaluation result and the power demand evaluation result. For example, the processor can use the weighted value of the network connectivity evaluation result and the power demand evaluation result as the final evaluation result, wherein the weights of the network connectivity evaluation result and the power demand evaluation result can be set based on actual needs. If more emphasis is placed on network connectivity, the weight of the network connectivity evaluation result will be greater. If more emphasis is placed on the recovery or improvement of the communication distance, the weight of the power demand evaluation result will be greater.

[0118] It should be noted that the above description of process 200 is for illustration and purpose only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and variations to process 200 under the guidance of this specification. However, such modifications and variations are still within the scope of this specification.

[0119] Figure 4 FIG. 1 is an exemplary schematic diagram of a required power determination model according to some embodiments of this specification. Figure 4 As shown, in some embodiments, for each aircraft, the processor can determine the power requirement before recovery and the power requirement after recovery 460 based on the communication characteristics 410, weather characteristics 420, communication environment characteristics 440 and equipment characteristics 440 of the aircraft and other aircraft through the power demand determination model 450. For example, the power requirement before recovery can be determined based on the communication characteristics, weather characteristics, communication environment characteristics and equipment characteristics before recovery, and the power requirement after recovery can be determined in the same way.

[0120] In some embodiments, communication characteristics may include the quality of the communication network, weather characteristics may include temperature, humidity, air pressure, etc., communication environment characteristics may include whether there is any obstruction between aircraft, the amount of interference, etc., and equipment characteristics may include aircraft model, age, etc.

[0121] In some embodiments, the inputs of the power requirement determination model 450 include communication characteristics 410 between the aircraft and other aircraft, weather characteristics 420, communication environment characteristics 440, and equipment characteristics 440, and the output includes the power requirement 460 of the aircraft in the corresponding scenario.

[0122] In some embodiments, the processor may train a demand power determination model based on multiple labeled training samples. For example, the processor may input the training samples into an initial demand power determination model, construct a loss function based on the output and labels of the initial demand power determination model, iteratively update the parameters of the initial demand power determination model based on the loss function, and terminate the iteration when an iteration completion condition is met, thereby obtaining a trained demand power determination model. The iterative update method includes, but is not limited to, gradient descent, and the iteration completion condition may be when the loss function converges or the number of iterations reaches a threshold.

[0123] The training samples include sample communication characteristics of the sample aircraft in the sample scenario, sample weather characteristics of the sample scenario, sample communication environment characteristics of the sample scenario, and sample equipment characteristics of the sample aircraft. The output includes the actual power demand of the sample aircraft in the sample scenario.

[0124] The following is an example of a specific evaluation test of the solution of the present invention when a drone is performing forest surveillance. The simulation is based on Matlab 2017a. The details are as follows:

[0125] When a drone swarm is used for tasks such as forest surveillance, a drone may become ineffective due to certain attacks or energy exhaustion. The goal of this embodiment is to prevent the occurrence of isolated drones without changing the planned path of the drone. In this embodiment, it is assumed that there are N drones distributed in a fixed range, the drones' action paths adopt the RandomWaypoint action mode, and the drones' initial communication radius is r c ,like Figure 5 As shown, assuming that the time point of launching the attack is t attack The time point when the drone adopts recovery mode after detecting the attack is t recovery , the simulation stop time is t final , the original value of the connectivity index is y D , after the attack, the connectivity index reaches the lowest value y min , the connectivity index after restoration is y r , the specific parameter configuration is shown in Table 1:

[0126]

[0127] After simulation, the connectivity of the UAV network can be experimentally analyzed. The prediction results of the swarm network under three attack modes can be expressed based on the following formula:

[0128]

[0129] Among them, the number of drone nodes is 200, and the drone network connectivity index is expressed as y(t).

[0130] like Figure 6a-6c is a schematic diagram showing the connectivity of a drone swarm network after being attacked by a corresponding attack mode according to some embodiments of this specification, wherein Figure 6a-6c The connectivity performance of the drone network when 40, 60, and 80 drones are attacked is shown. The period t ranges from 0 to 20 seconds, representing the normal operation of the drone swarm. The fluctuations in y(t) are primarily due to the internal movement of drone nodes. The period t ranges from 21 to 30 seconds, representing the attack phase, in which some drone nodes lose communication capabilities, resulting in a sharp drop in the connectivity index y(t). During the period t ranges from 31 to 50 seconds, the drone swarm uses a pre-set recovery method to restore connectivity. The results show that the pre-set recovery method is effective in restoring the connectivity of the drone network when the number of attacked nodes is small.

[0131] Due to the structural limitations of drones, energy is both important and limited, so it is necessary to evaluate the amount by which the preset recovery method will increase the energy consumption of the drone system. In some embodiments, the processor can convert the communication radius of the drone node into the corresponding required communication energy. The experimental results are shown in Figure 2. Figure 7 As shown, Figure 7 Three colors and three different line types are used to represent the type of node attack and the number of nodes attacked: red represents attack mode I, green represents attack mode II, and blue represents attack mode III. Attack modes I, II, and III can represent different attack methods. For example, attack modes I, II, and III can be selected from random hardware failure simulation, simulated environmental interference, or simulated human attacks. For example, attack mode I can use random hardware failure simulation, attack mode II can use simulated human attacks, and attack mode III can use simulated environmental interference. A solid line indicates 40 nodes attacked, a dashed line indicates 60 nodes attacked, and a denser dashed line indicates 80 nodes attacked.

[0132] The initial smoothed portion represents the total communication energy consumption of the drone swarm before the attack. After the attack, some drone nodes lose communication capabilities, reducing overall communication energy consumption. The observed decrease in the figure reflects that the more drone nodes there are, the more pronounced the decrease. When the same number of nodes are attacked, the green line is consistently higher than the other two colors, indicating that attack mode II requires more energy to recover compared to attack modes I and III. This may be because attack mode II directly disrupts network connectivity but is also easier to recover from, requiring increased energy input. In contrast, it is more difficult for the network to fully recover from attack mode III.

[0133] In this embodiment, the network resilience of drone clusters under different attack types was tested, and the test results are as follows: Figure 8a-8c As shown, Figure 8a-8c The number of drones attacked in the three experimental results was 40, 60 and 80 respectively.

[0134] The blue portion represents the resilience of the drone swarm network when the default recovery method (i.e., the original method) is not used after an attack. The orange portion represents the resilience when the default recovery method (i.e., the present invention) is used. After recovery using the default recovery method, the robustness of the drone swarm network is significantly improved. The most significant improvement in resilience under attack mode III is 5.3612 times, 7.4177 times, and 8.9435 times, respectively. The default recovery method also demonstrates significant effectiveness against attack modes I and II, with resilience increases of 2.5857 times, 3.0313 times, and 3.6188 times, and 2.6322 times, 4.1868 times, and 6.3443 times, respectively.

[0135] While the basic concepts have been described above, it will be apparent to those skilled in the art that the detailed disclosure is merely illustrative and does not limit this specification. Although not explicitly stated herein, various modifications, improvements, and revisions to this specification may be made by those skilled in the art. Such modifications, improvements, and revisions are suggested in this specification and remain within the spirit and scope of the exemplary embodiments of this specification.

[0136] This specification also uses specific terms to describe the embodiments of this specification. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "one embodiment," "an embodiment," or "an alternative embodiment" two or more times in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of this specification may be appropriately combined.

[0137] In addition, unless expressly stated in the claims, the order of the processing elements and sequences, the use of alphanumeric characters, or the use of other names described in this specification are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the spirit and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0138] Similarly, it should be noted that, in order to simplify the presentation of this specification and thus facilitate understanding of one or more embodiments of the invention, the foregoing descriptions of the embodiments of this specification sometimes combine multiple features into a single embodiment, figure, or description thereof. However, this disclosure method does not imply that the subject matter of this specification requires more features than those recited in the claims. In fact, an embodiment may have fewer features than all of the features of a single disclosed embodiment.

[0139] In some embodiments, numbers are used to describe the quantity of components and attributes. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise stated, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the description and claims are approximate values, which may change according to the required characteristics of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the settings of such numerical values ​​are as accurate as possible within the feasible range.

[0140] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, and documents, cited in this specification is hereby incorporated by reference in its entirety. This includes application history documents that are inconsistent with or conflict with the content of this specification, as well as documents (currently or subsequently attached to this specification) that limit the broadest scope of the claims of this specification. It should be noted that if the descriptions, definitions, and / or terminology used in the accompanying materials are inconsistent or conflicting with the content of this specification, the descriptions, definitions, and / or terminology used in this specification will control.

[0141] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A method for evaluating the anti-attack performance of an aircraft swarm network, characterized in that: include: Acquire a two-dimensional space model, wherein the two-dimensional space model includes a plurality of aircraft; Acquire positioning information of each aircraft among the plurality of aircraft in the two-dimensional space model, the positioning information including speed information and position information of the aircraft in the two-dimensional space model at each moment; generating an aircraft swarm model based on the positioning information of the plurality of aircraft; wherein the aircraft swarm model is configured to determine a probability of successfully establishing a communication connection between any two aircraft among the plurality of aircraft; determining a target aircraft to be attacked based on the aircraft swarm model; Performing an attack simulation on the target aircraft using a preset attack method to obtain a post-attack aircraft group network; Performing network recovery on the cluster network after the attack based on a preset recovery method; A performance evaluation is performed on the cluster network after the network is restored.

2. The method according to claim 1, characterized in that The aircraft fleet model is characterized as follows: Among them, F(d j→i ) represents the probability of successfully establishing a communication connection between the jth aircraft and the ith aircraft; α represents the item weight; N t is the number of aircraft in the fleet network at time t; ε represents the adjustment coefficient; k i represents the degree of freedom of the i-th aircraft, k l represents the degree of freedom of the lth aircraft; d j→i represents the communication distance between the jth aircraft and the ith aircraft; P j→i It represents the probability of successfully establishing a communication connection between the j-th aircraft and all aircraft in the fleet where the i-th aircraft is located.

3. The method according to claim 2, characterized in that The degrees of freedom of the aircraft include the communication upper limit of the aircraft. The probability of successfully establishing a communication connection between the j-th aircraft and the i-th aircraft is represented by the following formula: Among them, r c represents the communication range, and η represents the distance impact factor.

4. The method according to claim 2, characterized in that The target aircraft includes at least one of a first target aircraft, a second target aircraft, and a third target aircraft; The first target aircraft is at least one aircraft randomly selected from the plurality of aircraft; The second target aircraft is an aircraft among the multiple aircraft, the degree of freedom of which satisfies a preset degree of freedom condition; The third target aircraft is an aircraft among the multiple aircraft whose centrality satisfies a preset centrality condition, and the centrality is determined based on the aircraft fleet model.

5. The method according to claim 4, characterized in that The attack simulation of the target aircraft using a preset attack method includes: In response to the target aircraft being the first target aircraft, interfering with the network of the first target aircraft by using a random hardware failure or a simulated environmental interference; In response to the target aircraft being the second target aircraft and / or the third target aircraft, a network of the second target aircraft and / or the third target aircraft is interfered with in a manner of simulating a human attack.

6. The method according to claim 1, characterized in that The preset recovery method includes: Acquire a plurality of connected cluster slices in the cluster network after the attack, wherein the connected cluster slices include at least one aircraft; Performing degree of freedom preprocessing on each aircraft in each of the connected cluster slices and determining a communication capability level of each aircraft; the communication capability level is determined based on the degree of freedom of the aircraft, and the degree of freedom preprocessing includes: In response to a degree of freedom of the aircraft being less than a minimum threshold, increasing a communication capability level of the aircraft by one level; In response to a degree of freedom of the aircraft being greater than a maximum threshold, reducing a communication capability level of the aircraft by one level; A communication connection is established between the aircraft whose communication capability level difference is within a preset difference range.

7. The method according to claim 1, characterized in that The performing performance evaluation on the cluster network after the network recovery includes: Performing connectivity level analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain network connectivity before restoration and network connectivity after restoration; A network connectivity evaluation result is obtained based on the network connectivity before restoration and the network connectivity after restoration.

8. The method according to claim 7, characterized in that The connectivity level analysis includes: Control each of the plurality of aircraft except the target aircraft, and generate a probability μ according to the information at the corresponding time t based on a preset period. t , generating test information and sending it to the target aircraft; wherein, μ0 represents the probability of initial information generation, N t represents the number of aircraft at time t, and N represents the total number of aircraft in the two-dimensional space model; Get the total amount of information y(t) acquired by the target aircraft at time t: Among them, R i (t) represents the number of messages received by the i-th aircraft, is the shortest path length between the jth aircraft and the ith aircraft obtained by the shortest path algorithm, N t represents the number of aircraft at time t, Δ represents the time sensitivity coefficient (0<Δ≤1); The network connectivity before restoration or the network connectivity after restoration is determined based on the total amount of the information acquired at multiple moments.

9. The method according to claim 1 or 7, characterized in that The performing performance evaluation on the cluster network after the network recovery includes: Performing communication power demand analysis on the cluster network after the attack and the cluster network after the network is restored, respectively, to obtain power demand before restoration and power demand after restoration; A power demand evaluation result is obtained based on the power demand before recovery and the power demand after recovery.

10. The method according to claim 9, characterized in that The communication power demand analysis includes: For each aircraft, based on the communication distance between the aircraft and other aircraft, a power requirement before restoration and a power requirement after restoration are determined by a power requirement determination model.

Images