Communication method and device of unmanned aerial vehicle, storage medium and electronic equipment

Abstract

The invention discloses a communication method and device of an unmanned aerial vehicle, a storage medium and electronic equipment. The method relates to the technical field of communication security, and comprises the following steps: classifying target communication signals received by a target unmanned aerial vehicle to obtain a first type of communication signals and a second type of communication signals; performing risk assessment on the second type of communication signals based on the first type of communication signals and a target neural network model to obtain a target risk value; according to the target risk value, judging whether to adjust a communication channel of the target unmanned aerial vehicle for receiving the communication signal; and under the condition that the communication channel for receiving the communication signal of the target unmanned aerial vehicle is adjusted, the communication channel for receiving the communication signal of the target unmanned aerial vehicle is adjusted to be a target channel, and the communication signal is received through the target channel. Through application of the method and the device, the problem that the communication security of the unmanned aerial vehicle is relatively poor due to the fact that the communication security of the unmanned aerial vehicle is ensured by depending on a static security protocol in related technologies is solved.

Description

Technical Field

[0001] The present application relates to the field of communication security technology, and more specifically, to a communication method and device for an unmanned aerial vehicle, a storage medium, and an electronic device. Background Art

[0002] With the rapid development and widespread application of drone technology, drones have played an important role in reconnaissance, logistics and transportation, environmental monitoring, disaster relief and other fields. Especially in complex and dangerous environments, drones are able to perform tasks that are difficult for humans to complete. However, the widespread use of drones has also raised new security issues, especially in terms of communication security. Drones usually rely on wireless remote control signals for navigation and control, which makes them vulnerable to the threat of electronic interference and signal intrusion.

[0003] With the advancement of hacker technology and electronic tactics, the security of drone remote control signals faces severe challenges. Malicious attackers can control drones through signal interference, forgery or hijacking, thereby threatening the safety and efficiency of drone missions. For example, through signal hijacking, attackers can force drones to deviate from their intended routes or even perform unauthorized operations. Therefore, how to effectively detect and prevent the intrusion and hijacking of drone remote control signals has become a technical problem that needs to be solved urgently.

[0004] Traditional drone communication security solutions mainly rely on encryption technology and static security protocols. Although they can improve the security of communications to a certain extent, these methods often lack sufficient flexibility and adaptability when facing a complex and changing threat environment.

[0005] Currently, no effective solution has been proposed to address the problem that the relevant technology relies on static security protocols to ensure the security of drone communications, resulting in poor security of drone communications. Summary of the invention

[0006] The main purpose of this application is to provide a communication method and device, a storage medium and an electronic device for a drone, so as to solve the problem that the related technology relies on static security protocols to ensure the security of drone communications, resulting in poor security of drone communications.

[0007] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a communication method for a drone is provided. The method includes: classifying the target communication signal received by the target drone to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; judging whether to adjust the communication channel for the target drone to receive the communication signal based on the target risk value; in the case of adjusting the communication channel for the target drone to receive the communication signal, adjusting the communication channel for the target drone to receive the communication signal to the target channel, and receiving the communication signal through the target channel.

[0008] Furthermore, performing risk assessment on the second category of communication signals based on the first category of communication signals and the target neural network model to obtain a target risk value includes: performing risk assessment on the second category of communication signals through the target neural network model to obtain a first risk value; performing risk assessment on the second category of communication signals based on the first category of communication signals to obtain a second risk value; and calculating based on the first risk value and the second risk value to obtain the target risk value.

[0009] Furthermore, performing risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value includes: processing the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate adjustment of the communication parameters of the target UAV; after adjusting the communication parameters of the target UAV according to the target execution action, obtaining an updated second type of communication signal received by the target UAV; evaluating the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of influence of the target execution action; performing risk assessment on the second type of communication signal according to the target evaluation value to obtain the first risk value.

[0010] Furthermore, performing risk assessment on the second type of communication signal based on the first type of communication signal to obtain a second risk value includes: calculating the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal to obtain a first similarity; calculating the difference between the signal strength value corresponding to the first type of communication signal and the signal strength corresponding to the second type of communication signal to obtain a target difference; calculating the similarity between the signal attribute corresponding to the second type of communication signal and the signal attribute corresponding to multiple historical communication signals to obtain a second similarity; and obtaining the second risk value based on the first similarity, the target difference and the second similarity.

[0011] Further, calculating based on the first risk value and the second risk value to obtain the target risk value includes: obtaining a first weight value corresponding to the first risk value; obtaining a second weight value corresponding to the second risk value; and calculating based on the first weight value, the second weight value and the first risk value and the second risk value to obtain the target risk value.

[0012] Furthermore, based on the target risk value, determining whether to adjust the communication channel for the target UAV to receive communication signals includes: determining whether the target risk value is greater than a preset threshold; if the target risk value is greater than the preset threshold, determining to adjust the communication channel for the target UAV to receive communication signals; if the target risk value is less than or equal to the preset threshold, determining not to adjust the communication channel for the target UAV to receive communication signals.

[0013] Furthermore, the target neural network model is obtained by adopting the following steps: outputting a sample execution action based on an initial sample communication signal through a first network sub-model in an initial neural network; determining a true evaluation value based on an updated sample communication signal and the sample execution action, wherein the updated sample communication signal is an updated sample communication signal received by the UAV after adjusting the communication parameters of the UAV based on the sample execution action; processing the updated sample communication signal and the sample execution action through a second network sub-model in the initial neural network to obtain a predicted evaluation value; training the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and training the first network sub-model in the initial neural network based on the predicted evaluation value, so as to obtain the target neural network model.

[0014] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a communication device for a drone is provided. The device includes: a classification unit, which is used to classify the target communication signal received by the target drone to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; an evaluation unit, which is used to perform risk evaluation on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; a judgment unit, which is used to judge whether to adjust the communication channel for the target drone to receive the communication signal according to the target risk value; an adjustment unit, which is used to adjust the communication channel for the target drone to receive the communication signal to the target channel when adjusting the communication channel for the target drone to receive the communication signal, and receive the communication signal through the target channel.

[0015] Furthermore, the evaluation unit includes: a first evaluation subunit, used to perform risk evaluation on the second type of communication signal through the target neural network model to obtain a first risk value; a second evaluation subunit, used to perform risk evaluation on the second type of communication signal based on the first type of communication signal to obtain a second risk value; and a calculation subunit, used to perform calculation based on the first risk value and the second risk value to obtain the target risk value.

[0016] Furthermore, the second evaluation subunit includes: a processing module, used to process the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate the adjustment of the communication parameters of the target UAV; a first acquisition module, used to obtain the updated second type of communication signal received by the target UAV after adjusting the communication parameters of the target UAV according to the target execution action; an evaluation module, used to evaluate the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of influence of the target execution action; an evaluation module, used to perform a risk assessment on the second type of communication signal based on the target evaluation value to obtain the first risk value.

[0017] Furthermore, the first evaluation subunit includes: a first calculation module, used to calculate the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal, to obtain a first similarity; a second calculation module, used to calculate the difference between the signal strength value corresponding to the first type of communication signal and the signal strength corresponding to the second type of communication signal, to obtain a target difference; a third calculation module, used to calculate the similarity between the signal attribute corresponding to the second type of communication signal and the signal attribute corresponding to multiple historical communication signals, to obtain a second similarity; a determination module, used to obtain the second risk value based on the first similarity, the target difference and the second similarity.

[0018] Furthermore, the calculation subunit includes: a second acquisition module, used to obtain a first weight value corresponding to the first risk value; a third acquisition module, used to obtain a second weight value corresponding to the second risk value; and a fourth calculation module, used to calculate based on the first weight value, the second weight value, the first risk value and the second risk value to obtain the target risk value.

[0019] Furthermore, the judgment unit includes: a judgment subunit, used to judge whether the target risk value is greater than a preset threshold; a first determination subunit, used to determine whether to adjust the communication channel for the target UAV to receive communication signals if the target risk value is greater than the preset threshold; and a second determination subunit, used to determine not to adjust the communication channel for the target UAV to receive communication signals if the target risk value is less than or equal to the preset threshold.

[0020] Furthermore, the target neural network model is obtained by using the following devices: an output unit, used to output a sample execution action based on an initial sample communication signal through a first network sub-model in an initial neural network; a determination unit, used to determine a true evaluation value based on an updated sample communication signal and the sample execution action, wherein the updated sample communication signal is an updated sample communication signal received by the drone after adjusting the communication parameters of the drone based on the sample execution action; a processing unit, used to process the updated sample communication signal and the sample execution action through a second network sub-model in the initial neural network to obtain a predicted evaluation value; a training unit, used to train the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and to train the first network sub-model in the initial neural network based on the predicted evaluation value, so as to obtain the target neural network model.

[0021] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a computer-readable storage medium is provided, which stores a program, wherein when the program is running, the device where the storage medium is located is controlled to execute any one of the above-mentioned drone communication methods.

[0022] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is also provided, the electronic device includes one or more processors and a memory, and the memory is used to store the communication method of the drone implemented by one or more processors as described above.

[0023] Through the present application, the following steps are adopted: classifying the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; judging whether to adjust the communication channel for the target UAV to receive the communication signal based on the target risk value; in the case of adjusting the communication channel for the target UAV to receive the communication signal, adjusting the communication channel for the target UAV to receive the communication signal to the target channel, and receiving the communication signal through the target channel, thereby solving the problem in the related art that the security of the UAV communication is ensured by relying on a static security protocol, resulting in poor security of the UAV communication. In this scheme, the target communication signals received by the target UAV are classified into first-class communication signals and second-class communication signals. The source of the second-class communication signals is not the target source and is considered to be a communication signal that may be risky. The risk of the second-class communication signals is assessed through both the first-class communication signals and the target neural network model. Then, based on the obtained target risk value, it is determined whether to adjust the communication channel for the target UAV to receive the communication signal. The security of the communication signal can be accurately judged through the first-class communication signals and the target neural network model, thereby achieving the effect of improving the security of the UAV's communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0025] Figure 1 The following is a flow chart of the communication method of the drone provided in the embodiment of the present application: Figure 1 ;

[0026] Figure 2 The communication method of the drone provided in the embodiment of the present application is as follows: Figure 2 ;

[0027] Figure 3The following is a flow chart of the communication method of the drone provided in the embodiment of the present application: Figure 3 ;

[0028] Figure 4 is a schematic diagram of a communication device of a drone provided according to an embodiment of the present application;

[0029] Figure 5 is a schematic diagram of a drone safety detection and warning system provided according to an embodiment of the present application;

[0030] Figure 6 is a schematic diagram of an electronic device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0031] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0033] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0034] It should be noted that the relevant information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for display, data for analysis, etc.) involved in this disclosure are all information and data authorized by the user or fully authorized by all parties. For example, an interface is set between this system and the relevant user or organization. Before obtaining relevant information, it is necessary to send an acquisition request to the aforementioned user or organization through the interface, and obtain relevant information after receiving the consent information fed back by the aforementioned user or organization.

[0035] The present invention is described below in conjunction with preferred implementation steps. Figure 1 The following is a flow chart of the communication method of the drone provided in the embodiment of the present application: Figure 1 ,like Figure 1 As shown, the method comprises the following steps:

[0036] Step S101, classifying the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source.

[0037] Optionally, the target communication signals received by the target UAV are collected, and the target communication signals are classified and processed to obtain primary communications (i.e., the first type of communication signals mentioned above) with known sources (i.e., the target sources mentioned above) and secondary communications (second type of communication signals) with unknown sources and unknown intentions.

[0038] Step S102, performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value.

[0039] Optionally, since the source of the second type of communication signal is unknown and may be risky, risk assessment of the second type of communication signal includes independent analysis and comparative analysis, that is, risk assessment of the second type of communication signal is performed through the first type of communication signal and the target neural network model to obtain a target risk value.

[0040] In an optional embodiment, in the communication method of the drone provided in the embodiment of the present application, a risk assessment is performed on a second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value, including: performing a risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value; performing a risk assessment on the second type of communication signal based on the first type of communication signal to obtain a second risk value; and calculating based on the first risk value and the second risk value to obtain a target risk value.

[0041] In an optional embodiment, the risk assessment of the second type of communication signal based on the first type of communication signal and the target neural network model includes the following steps: the risk assessment of the second type of communication signal is performed through the target neural network model to obtain a first risk value, and then the risk assessment of the second type of communication signal is performed through the first type of communication signal to obtain a second risk value. Finally, the target risk value is obtained by calculating the first risk value and the second risk value.

[0042] In an optional embodiment, the comparative analysis may be performed by calculating the similarity between the first type of communication signal and the second type of communication signal, and then obtaining the comparative analysis risk (i.e., the second risk value mentioned above) according to the similarity. The independent analysis is to perform risk assessment on the second type of communication signal through the target neural network model to obtain the independent analysis risk (i.e., the first risk value mentioned above), and finally determine the risk value of the second type of communication signal according to the comparative analysis risk and the independent analysis risk.

[0043] In an optional embodiment, a weighted calculation method or a maximum value method may be used to calculate the final risk value.

[0044] Step S103, judging whether to adjust the communication channel for the target UAV to receive the communication signal according to the target risk value.

[0045] In an optional embodiment, the calculated target risk value is used to determine whether to adjust the communication channel through which the target UAV receives communication signals. For example, if the target risk value exceeds a threshold, the communication channel through which the target UAV receives communication signals is adjusted.

[0046] Step S104, when adjusting the communication channel through which the target UAV receives the communication signal, the communication channel through which the target UAV receives the communication signal is adjusted to the target channel, and the communication signal is received through the target channel.

[0047] In an optional embodiment, when adjusting the communication channel through which the target UAV receives communication signals, the communication channel through which the target UAV receives communication signals is adjusted to a safe channel (i.e., the target channel mentioned above), and the communication signals are received through the target channel.

[0048] In an optional embodiment, the target channel and the communication channel before the target drone changes the communication channel use different parts of the electromagnetic spectrum; the different parts of the electromagnetic spectrum include: L band, C band, Ku band, Ka band, Wi-Fi channel, LTE band, Bluetooth band, cellular band and UWB band.

[0049] To summarize, the target communication signals received by the target UAV are classified into first-class communication signals and second-class communication signals. The source of the second-class communication signals is not the target source and is considered to be a communication signal that may be risky. The risk of the second-class communication signals is assessed through the first-class communication signals and the target neural network model. Then, based on the obtained target risk value, it is determined whether to adjust the communication channel for the target UAV to receive the communication signal. The security of the communication signal can be accurately judged through the first-class communication signals and the target neural network model, thereby achieving the effect of improving the security of the UAV's communication.

[0050] Optionally, in the communication method of the unmanned aerial vehicle provided in the embodiment of the present application, the risk assessment of the second type of communication signal is performed through the target neural network model to obtain the first risk value, which includes: processing the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate the adjustment of the communication parameters of the target unmanned aerial vehicle; after adjusting the communication parameters of the target unmanned aerial vehicle according to the target execution action, obtaining an updated second type of communication signal received by the target unmanned aerial vehicle; evaluating the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of influence of the target execution action; and performing a risk assessment on the second type of communication signal according to the target evaluation value to obtain a first risk value.

[0051] In an optional embodiment, the target neural network model can be a network model of the actor-critic framework. Actor network: The actor network is responsible for mapping directly from state to action, that is, determining the optimal action that the agent should take in a given state; based on the current strategy, this network attempts to explore and exploit the environment to find actions that can obtain higher rewards. Critic network: The critic network evaluates the quality of the action selected by the actor, that is, it gives an estimate of the action value (Q value), which reflects the expected return that can be obtained by taking an action and following the current strategy based on the current state; the actor network adjusts its own strategy based on the feedback given by the critic network (that is, the value evaluation of the action) to produce better actions; in this process, the actor network optimizes the policy function through the policy gradient method; the critic network updates its value function estimate by observing the feedback of the environment (that is, the actual reward obtained) and the Bellman equation, which helps to more accurately evaluate the actor's actions.

[0052] Therefore, the risk assessment of the second type of communication signal through the target neural network model includes: processing the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model, that is, outputting the current execution action (that is, the above-mentioned target execution action) according to the signal attributes corresponding to the second type of communication signal, such as signal strength, frequency, etc. through the actor network (that is, the above-mentioned first network sub-model). It should be noted that the range space corresponding to the execution action can be [frequency adjustment, modulation mode change, encryption state switching, channel hopping]. That is, the communication parameter space of the target drone can be adjusted to P = [signal strength, frequency, modulation mode, encryption state, signal consistency], where the meaning of each parameter is: Signal strength: The power level of the signal, which can be used to evaluate the quality of the signal and possible interference or blocking. Frequency: The frequency used by the signal. Different frequencies may correspond to different communication protocols and uses. Abnormal frequency changes may indicate signal intrusion. Modulation mode: the modulation method of the signal. Different modulation technologies have different anti-interference capabilities and transmission characteristics. Abnormal changes in modulation mode may be a sign of intrusion; Encryption status: whether the signal is encrypted and the type of encryption. Encrypted communication is generally more secure. Unencrypted signals or signals with sudden changes in encryption methods may pose security risks; Signal consistency: compared with the consistency of previous communication modes. Inconsistencies such as frequency hopping and intermittent signals may indicate that the signal has been interfered with or tampered with.

[0053] In an optional embodiment, the random noise function in the target neural network model can be shown as formula (1):

[0054] N = 10 + 0.05 rand (n, 1) (1)

[0055] Among them, 10 represents the noise mean, 0.05 represents the noise standard deviation, rand(n,1) represents an n*1 random noise matrix, and n is a positive integer; the above formula generates an n*1 random noise matrix with a Gaussian distribution of 10 and a standard deviation of 0.05.

[0056] In an optional embodiment, the actor network can output the above-mentioned current execution action through formula (2):

[0057] Action=π θ (Φ(S)+N)(2)

[0058] Among them, Action represents the currently executed action. For example, the above-mentioned Action includes specific action values ​​of frequency adjustment, modulation mode change, encryption state switching, and channel hopping, which can be considered as a four-dimensional vector; π θ() is the strategy function of the actor network based on the actor network parameter θ; Φ(P) represents the characteristic vector of the current state P (ie, signal attribute) of the second type of communication signal in the state space, and N represents the added random noise function.

[0059] Then, the communication parameters of the target UAV are adjusted according to the target execution action, and the updated second type of communication signal is obtained again, and the reward can also be calculated according to the new state P′ of the updated second type of communication signal. For example, the reward value can be determined according to the change value of the communication signal strength, the change of the communication error rate, and the change rate of the communication frequency.

[0060] Finally, the target execution action, the updated second-category communication signal, and the above-mentioned reward are evaluated by the critic network (i.e., the second network sub-model mentioned above) to obtain the target evaluation value, and then the risk of the second-category communication signal is evaluated by the target evaluation value to obtain the first risk value. It should be noted that the target evaluation value is used to indicate the degree of influence of the target execution action, that is, whether the target execution action is effective. If the target execution action is effective, it indicates that the risk level of the second-category communication signal is relatively low. If the target execution action is invalid or the effectiveness is poor, it indicates that the risk level of the second-category communication signal is relatively high.

[0061] The second type of communication signal is analyzed and evaluated by the above-mentioned target neural network model, thereby improving the accuracy of determining the risk of the second type of communication signal.

[0062] Optionally, in the communication method of the drone provided in the embodiment of the present application, a risk assessment is performed on the second type of communication signal based on the first type of communication signal to obtain a second risk value, including: calculating the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal to obtain a first similarity; calculating the difference between the signal strength value corresponding to the first type of communication signal and the signal strength corresponding to the second type of communication signal to obtain a target difference; calculating the similarity between the signal attribute corresponding to the second type of communication signal and the signal attribute corresponding to multiple historical communication signals to obtain a second similarity; and obtaining a second risk value based on the first similarity, the target difference and the second similarity.

[0063] In an optional embodiment, performing risk assessment on the second type of communication signal through the first type of communication signal includes the following steps:

[0064] First, the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal is calculated. For example, a text comparison algorithm (such as Levenshtein distance) can be used to quantify the degree of difference between the two groups of instructions. The difference in urgency or priority markings in the first instruction and the second instruction can also be evaluated; the execution parameters involved in the first instruction and the second instruction (such as speed, direction, height, etc.) can also be compared to calculate the degree of difference in parameter values, thereby obtaining the above-mentioned first similarity.

[0065] Then, the first type of communication signal is a signal required for transmitting normal operations to the drone, and its radio frequency power level (i.e., the above-mentioned signal strength) usually operates stably within a predetermined range; while the second type of communication signal, due to its unknown source, its radio frequency power level may fluctuate abnormally or deviate significantly from the normal value, suggesting potential interference or intrusion. Therefore, the difference between the signal strength value corresponding to the first type of communication signal and the signal strength corresponding to the second type of communication signal is calculated to obtain the above-mentioned target difference.

[0066] Secondly, the second similarity is obtained by calculating the similarity between the signal attributes corresponding to the second type of communication signal and the signal attributes corresponding to multiple historical communication signals. The historical signal data covers a wide range of communication characteristics collected from various drones, such as signal mode, frequency distribution, modulation technology, RF power level and encryption protocol, etc. Therefore, the normality of the second type of communication signal can be determined based on past actual cases and scenarios.

[0067] Finally, the second risk value is calculated by the first similarity, the target difference and the second similarity. For example, the final comparative analysis risk index (i.e., the second risk value mentioned above) can be calculated by weighted summation. The distribution of weights can be adjusted according to actual conditions and security policies; for example, if it is believed that the difference in instruction content has the greatest impact on security, a higher weight can be given.

[0068] The accuracy of the second type of communication signal can be accurately evaluated by calculating the degree of difference between the second type of communication signal and the first type of communication signal.

[0069] Optionally, in the communication method of the drone provided in the embodiment of the present application, calculation is performed based on the first risk value and the second risk value to obtain the target risk value, including: obtaining a first weight value corresponding to the first risk value; obtaining a second weight value corresponding to the second risk value; and calculating based on the first weight value, the second weight value and the first risk value and the second risk value to obtain the target risk value.

[0070] In an optional embodiment, after obtaining the values ​​of both the independent analysis risk index and the comparative analysis risk index (i.e., the first risk value and the second risk value mentioned above), a weighted calculation method can be used to calculate the risk value, that is, setting a first weight value corresponding to the first risk value and a second weight value corresponding to the second risk value, and calculating through the first weight value, the second weight value and the first risk value and the second risk value to obtain the target risk value.

[0071] Optionally, in the communication method for a drone provided in an embodiment of the present application, judging whether to adjust the communication channel for the target drone to receive communication signals based on the target risk value includes: judging whether the target risk value is greater than a preset threshold; if the target risk value is greater than the preset threshold, determining to adjust the communication channel for the target drone to receive communication signals; if the target risk value is less than or equal to the preset threshold, determining not to adjust the communication channel for the target drone to receive communication signals.

[0072] In an optional embodiment, the following steps are used to determine whether to adjust the communication channel for the target drone to receive the communication signal: determine whether the target risk value is greater than a preset threshold, if the target risk value is greater than the preset threshold, determine to adjust the communication channel for the target drone to receive the communication signal, if the target risk value is less than or equal to the preset threshold, determine not to adjust the communication channel for the target drone to receive the communication signal. By judging the target risk value and the preset threshold, it is possible to evaluate whether the second type of communication signal is a malicious signal, and then accurately determine whether it is necessary to switch the communication channel.

[0073] Optionally, in the communication method of the drone provided in the embodiment of the present application, the following steps are adopted to obtain the target neural network model: output a sample execution action based on the initial sample communication signal through the first network sub-model in the initial neural network; determine the true evaluation value based on the updated sample communication signal and the sample execution action, wherein the updated sample communication signal is the updated sample communication signal received by the drone after the communication parameters of the drone are adjusted based on the sample execution action; process the updated sample communication signal and the sample execution action through the second network sub-model in the initial neural network to obtain a predicted evaluation value; train the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and train the first network sub-model in the initial neural network based on the predicted evaluation value to obtain the target neural network model.

[0074] In an optional embodiment, the target neural network model is obtained by the following steps: randomly initialize the actor network parameters (i.e., the first network sub-model in the above-mentioned initial neural network) θ and the critic network parameters w (i.e., the second network sub-model in the above-mentioned initial neural network). Initialize the state space of the initial sample communication signal; the state space P is designed as: P = [signal strength, frequency, modulation mode, encryption state, signal consistency].

[0075] The first network sub-model in the initial neural network outputs a sample to perform an action Action based on the initial sample communication signal. It should be noted that the initial sample communication signal may be a historical second-category communication signal.

[0076] Then, the reward can be calculated based on the updated sample communication signal after the drone performs the sample execution action Action, and the true evaluation value can be determined based on the updated sample communication signal and the sample execution action. For example, the above true evaluation Target is calculated using the Bellman equation j .

[0077] The updated sample communication signal and sample execution action are then processed by the second network sub-model in the initial neural network, namely the critic network, to obtain a predicted evaluation value. For example, Q(Φ(P), Action, w) represents the predicted evaluation value, and Φ(P) represents the feature vector of the current state of the updated sample communication signal.

[0078] Secondly, the second network sub-model in the initial neural network is trained by predicting the evaluation value and the true evaluation value, that is, the mean square error loss function is used to calculate the Target j The loss Loss between and Q(Φ(P), Action, w) is shown in formula (3):

[0079]

[0080] Among them, m represents the number of sampled samples, and j represents the jth sample collected.

[0081] Through the approximate policy gradient method, the predicted evaluation value of the current state, the sample execution action, and the reward are combined to calculate the gradient of the current execution action, and the gradient of the sample execution action is back-propagated through the gradient of the neural network to update the actor network parameters. For example, the gradient of the sample execution action is calculated by formula (4):

[0082]

[0083] in, represents the gradient, represents the gradient vector of the policy gradient, m represents the number of sampled samples, and π(a|p;θ) represents the probability that the UAV communication system chooses action a in state p when the actor network parameters θ.

[0084] It should be noted that the training termination signal can be set according to at least one of the following conditions: the iteration performance is saturated or a preset number of iterations is reached, etc.

[0085] In an optional embodiment, the Figure 2 The schematic diagram shown realizes the communication control of the UAV, including: receiving and distinguishing the primary communication from the known source on the primary channel (i.e., the first type of communication signal mentioned above), and the secondary communication from the unknown source and unknown intention (i.e., the second type of communication signal mentioned above). Performing threat analysis on the received secondary communication. Threat analysis includes independent analysis and comparative analysis; in the independent analysis, the potential threat of the secondary communication is determined by a neural network based on a deep reinforcement learning algorithm to obtain an independent analysis risk index; in the comparative analysis, the communication risk level is evaluated by calculating at least one of the following parameters: comparison of the instructions of the primary communication and the secondary communication, comparison of the radio frequency power levels of the primary communication and the secondary communication, comparison of the communication characteristics in the secondary communication and the historical communication database, etc., to obtain a comparative analysis risk index. The total risk index is obtained based on the independent analysis risk index and the comparative analysis risk index, and it is determined whether to switch the communication channel of the UAV according to the total risk index.

[0086] In an optional embodiment, the Figure 3 The flowchart shown implements the neural network model training in independent analysis. Specifically, it includes the following steps:

[0087] S100: Get the current secondary communication status of the drone;

[0088] S200: Initialize the neural network parameters of the threat analysis module, clear the experience pool, and initialize the state space of the current communication state of the drone;

[0089] S300: Calculate the current execution action according to the current communication state of the drone, and calculate the reward according to the communication state of the new secondary communication after executing the current execution action;

[0090] S400: putting the communication status, current execution action, and reward representing the secondary communication of the drone into the experience pool as experience samples;

[0091] S500: Selecting a number of experience samples from the experience pool, combining the selected experience samples with the target value function, and updating the neural network parameters of the threat analysis module using the back propagation of the neural network;

[0092] S600: Repeat steps S300 to S500 to perform neural network iteration until the training is terminated after receiving a termination signal.

[0093] The communication method for a drone provided in an embodiment of the present application classifies a target communication signal received by a target drone to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; a risk assessment is performed on the second type of communication signal based on the first type of communication signal and a target neural network model to obtain a target risk value; based on the target risk value, it is determined whether to adjust the communication channel through which the target drone receives the communication signal; in the case of adjusting the communication channel through which the target drone receives the communication signal, the communication channel through which the target drone receives the communication signal is adjusted to the target channel, and the communication signal is received through the target channel, thereby solving the problem in the related art of relying on static security protocols to ensure the security of drone communications, resulting in poor security of drone communications. In this scheme, the target communication signals received by the target UAV are classified into first-class communication signals and second-class communication signals. The source of the second-class communication signals is not the target source and is considered to be a communication signal that may be risky. The risk of the second-class communication signals is assessed through both the first-class communication signals and the target neural network model. Then, based on the obtained target risk value, it is determined whether to adjust the communication channel for the target UAV to receive the communication signal. The security of the communication signal can be accurately judged through the first-class communication signals and the target neural network model, thereby achieving the effect of improving the security of the UAV's communication.

[0094] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0095] The embodiment of the present application also provides a communication device for a drone. It should be noted that the communication device for a drone in the embodiment of the present application can be used to execute the communication method for a drone provided in the embodiment of the present application. The communication device for a drone provided in the embodiment of the present application is introduced below.

[0096] Figure 4 Schematic diagram of a communication device of a drone according to an embodiment of the present application. Figure 4 As shown, the device includes: a classification unit 401, an evaluation unit 402, a judgment unit 403 and an adjustment unit 404.

[0097] The classification unit 401 is used to classify the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source;

[0098] An evaluation unit 402, configured to perform risk evaluation on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value;

[0099] A determination unit 403 is used to determine whether to adjust a communication channel for receiving communication signals of the target UAV according to the target risk value;

[0100] The adjustment unit 404 is used to adjust the communication channel through which the target UAV receives the communication signal to the target channel and receive the communication signal through the target channel.

[0101] The communication device of the drone provided in the embodiment of the present application classifies the target communication signal received by the target drone through the classification unit 401 to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; the evaluation unit 402 performs risk evaluation on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; the judgment unit 403 judges whether to adjust the communication channel for the target drone to receive the communication signal according to the target risk value; the adjustment unit 404 adjusts the communication channel for the target drone to receive the communication signal to the target channel when adjusting the communication channel for the target drone to receive the communication signal, and receives the communication signal through the target channel, thereby solving the problem in the related art that the security of the drone's communication is ensured by relying on static security protocols, resulting in poor security of the drone's communication. In this scheme, the target communication signals received by the target UAV are classified into first-class communication signals and second-class communication signals. The source of the second-class communication signals is not the target source and is considered to be a communication signal that may be risky. The risk of the second-class communication signals is assessed through both the first-class communication signals and the target neural network model. Then, based on the obtained target risk value, it is determined whether to adjust the communication channel for the target UAV to receive the communication signal. The security of the communication signal can be accurately judged through the first-class communication signals and the target neural network model, thereby achieving the effect of improving the security of the UAV's communication.

[0102] Optionally, in the communication device of the drone provided in the embodiment of the present application, the evaluation unit includes: a first evaluation subunit, used to perform risk evaluation on the second type of communication signal through a target neural network model to obtain a first risk value; a second evaluation subunit, used to perform risk evaluation on the second type of communication signal based on the first type of communication signal to obtain a second risk value; and a calculation subunit, used to perform calculation based on the first risk value and the second risk value to obtain a target risk value.

[0103] Optionally, in the communication device of the drone provided in the embodiment of the present application, the second evaluation subunit includes: a processing module, which is used to process the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate the adjustment of the communication parameters of the target drone; a first acquisition module, which is used to obtain the updated second type of communication signal received by the target drone after adjusting the communication parameters of the target drone according to the target execution action; an evaluation module, which is used to evaluate the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of influence of the target execution action; an evaluation module, which is used to perform a risk assessment on the second type of communication signal based on the target evaluation value to obtain a first risk value.

[0104] Optionally, in the communication device of the drone provided in the embodiment of the present application, the first evaluation subunit includes: a first calculation module, used to calculate the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal, to obtain a first similarity; a second calculation module, used to calculate the difference between the signal strength value corresponding to the first type of communication signal and the signal strength corresponding to the second type of communication signal, to obtain a target difference; a third calculation module, used to calculate the similarity between the signal attribute corresponding to the second type of communication signal and the signal attribute corresponding to multiple historical communication signals, to obtain a second similarity; a determination module, used to obtain a second risk value based on the first similarity, the target difference and the second similarity.

[0105] Optionally, in the communication device of the drone provided in the embodiment of the present application, the computing subunit includes: a second acquisition module, used to acquire a first weight value corresponding to the first risk value; a third acquisition module, used to acquire a second weight value corresponding to the second risk value; and a fourth computing module, used to perform calculations based on the first weight value, the second weight value, the first risk value, and the second risk value to obtain a target risk value.

[0106] Optionally, in the communication device of the drone provided in the embodiment of the present application, the judgment unit includes: a judgment subunit, used to judge whether the target risk value is greater than a preset threshold; a first determination subunit, used to determine whether to adjust the communication channel for the target drone to receive communication signals if the target risk value is greater than the preset threshold; a second determination subunit, used to determine not to adjust the communication channel for the target drone to receive communication signals if the target risk value is less than or equal to the preset threshold.

[0107] Optionally, in the communication device of the drone provided in the embodiment of the present application, the following devices are used to obtain the target neural network model: an output unit, used to output a sample execution action based on the initial sample communication signal through the first network sub-model in the initial neural network; a determination unit, used to determine the true evaluation value based on the updated sample communication signal and the sample execution action, wherein the updated sample communication signal is the updated sample communication signal received by the drone after the communication parameters of the drone are adjusted based on the sample execution action; a processing unit, used to process the updated sample communication signal and the sample execution action through the second network sub-model in the initial neural network to obtain a predicted evaluation value; a training unit, used to train the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and to train the first network sub-model in the initial neural network based on the predicted evaluation value, so as to obtain the target neural network model.

[0108] The communication device of the drone includes a processor and a memory. The above-mentioned classification unit 401, evaluation unit 402, judgment unit 403 and adjustment unit 404 are all stored in the memory as program units, and the processor executes the above-mentioned program units stored in the memory to realize corresponding functions.

[0109] The processor contains a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the communication control of the drone can be achieved by adjusting the kernel parameters.

[0110] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0111] In an optional embodiment, a drone safety detection and early warning system is also provided. Figure 5 As shown, the drone safety detection and warning system includes a signal processing module 10, a threat analysis module 20, a warning decision module 30 and a communication management module 40. The signal processing module 10 is configured to receive and analyze signals from the drone remote control channel, including receiving and distinguishing primary communications from known sources on the primary channel, and secondary communications from unknown sources and unknown intentions; the threat analysis module 20 is configured to perform threat analysis on the received secondary communications; the warning decision module 30 is configured to make warning decisions; the communication management module 40 is configured to automatically adjust the communication parameters of the drone according to the warning decision, and when it is confirmed that the communication on the secondary channel has malicious intentions, guide the drone communication system to switch from the current primary channel to a preset safety channel to avoid potential intrusion threats.

[0112] An embodiment of the present invention provides a computer-readable storage medium having a program stored thereon, and the program implements a communication method for a drone when executed by a processor.

[0113] An embodiment of the present invention provides a processor, which is used to run a program, wherein the communication method of a drone is executed when the program is run.

[0114] like Figure 6 As shown, an embodiment of the present invention provides an electronic device, the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: classifying the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; judging whether to adjust the communication channel for the target UAV to receive the communication signal based on the target risk value; in the case of adjusting the communication channel for the target UAV to receive the communication signal, adjusting the communication channel for the target UAV to receive the communication signal to the target channel, and receiving the communication signal through the target channel.

[0115] Optionally, performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value includes: performing risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value; performing risk assessment on the second type of communication signal based on the first type of communication signal to obtain a second risk value; and calculating based on the first risk value and the second risk value to obtain a target risk value.

[0116] Optionally, performing risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value includes: processing the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate adjustment of the communication parameters of the target UAV; after adjusting the communication parameters of the target UAV according to the target execution action, obtaining an updated second type of communication signal received by the target UAV; evaluating the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of impact of the target execution action; performing risk assessment on the second type of communication signal according to the target evaluation value to obtain a first risk value.

[0117] Optionally, performing a risk assessment on a second type of communication signal based on a first type of communication signal to obtain a second risk value includes: calculating the similarity between a first instruction content corresponding to the first type of communication signal and a second instruction corresponding to the second type of communication signal to obtain a first similarity; calculating the difference between a signal strength value corresponding to the first type of communication signal and a signal strength corresponding to the second type of communication signal to obtain a target difference; calculating the similarity between a signal attribute corresponding to the second type of communication signal and a signal attribute corresponding to multiple historical communication signals to obtain a second similarity; and obtaining a second risk value based on the first similarity, the target difference, and the second similarity.

[0118] Optionally, calculating based on the first risk value and the second risk value to obtain the target risk value includes: obtaining a first weight value corresponding to the first risk value; obtaining a second weight value corresponding to the second risk value; and calculating based on the first weight value, the second weight value and the first risk value and the second risk value to obtain the target risk value.

[0119] Optionally, based on the target risk value, determining whether to adjust the communication channel for the target UAV to receive communication signals includes: determining whether the target risk value is greater than a preset threshold; if the target risk value is greater than the preset threshold, determining to adjust the communication channel for the target UAV to receive communication signals; if the target risk value is less than or equal to the preset threshold, determining not to adjust the communication channel for the target UAV to receive communication signals.

[0120] Optionally, the following steps are adopted to obtain the target neural network model: outputting a sample execution action based on an initial sample communication signal through a first network sub-model in the initial neural network; determining a true evaluation value based on the updated sample communication signal and the sample execution action, wherein the updated sample communication signal is an updated sample communication signal received by the UAV after adjusting the communication parameters of the UAV based on the sample execution action; processing the updated sample communication signal and the sample execution action through a second network sub-model in the initial neural network to obtain a predicted evaluation value; training the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and training the first network sub-model in the initial neural network based on the predicted evaluation value, so as to obtain the target neural network model.

[0121] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0122] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program that is initialized with the following method steps: classifying the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; performing a risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; judging whether to adjust the communication channel through which the target UAV receives the communication signal based on the target risk value; in the case of adjusting the communication channel through which the target UAV receives the communication signal, adjusting the communication channel through which the target UAV receives the communication signal to the target channel, and receiving the communication signal through the target channel.

[0123] Optionally, performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value includes: performing risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value; performing risk assessment on the second type of communication signal based on the first type of communication signal to obtain a second risk value; and calculating based on the first risk value and the second risk value to obtain a target risk value.

[0124] Optionally, performing risk assessment on the second type of communication signal through the target neural network model to obtain a first risk value includes: processing the signal attributes corresponding to the second type of communication signal through the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to indicate adjustment of the communication parameters of the target UAV; after adjusting the communication parameters of the target UAV according to the target execution action, obtaining an updated second type of communication signal received by the target UAV; evaluating the target execution action and the updated second type of communication signal through the second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the degree of impact of the target execution action; performing risk assessment on the second type of communication signal according to the target evaluation value to obtain a first risk value.

[0125] Optionally, performing a risk assessment on a second type of communication signal based on a first type of communication signal to obtain a second risk value includes: calculating the similarity between a first instruction content corresponding to the first type of communication signal and a second instruction corresponding to the second type of communication signal to obtain a first similarity; calculating the difference between a signal strength value corresponding to the first type of communication signal and a signal strength corresponding to the second type of communication signal to obtain a target difference; calculating the similarity between a signal attribute corresponding to the second type of communication signal and a signal attribute corresponding to multiple historical communication signals to obtain a second similarity; and obtaining a second risk value based on the first similarity, the target difference, and the second similarity.

[0126] Optionally, calculating based on the first risk value and the second risk value to obtain the target risk value includes: obtaining a first weight value corresponding to the first risk value; obtaining a second weight value corresponding to the second risk value; and calculating based on the first weight value, the second weight value and the first risk value and the second risk value to obtain the target risk value.

[0127] Optionally, based on the target risk value, determining whether to adjust the communication channel for the target UAV to receive communication signals includes: determining whether the target risk value is greater than a preset threshold; if the target risk value is greater than the preset threshold, determining to adjust the communication channel for the target UAV to receive communication signals; if the target risk value is less than or equal to the preset threshold, determining not to adjust the communication channel for the target UAV to receive communication signals.

[0128] Optionally, the following steps are adopted to obtain the target neural network model: outputting a sample execution action based on an initial sample communication signal through a first network sub-model in the initial neural network; determining a true evaluation value based on the updated sample communication signal and the sample execution action, wherein the updated sample communication signal is an updated sample communication signal received by the UAV after adjusting the communication parameters of the UAV based on the sample execution action; processing the updated sample communication signal and the sample execution action through a second network sub-model in the initial neural network to obtain a predicted evaluation value; training the second network sub-model in the initial neural network based on the predicted evaluation value and the true evaluation value, and training the first network sub-model in the initial neural network based on the predicted evaluation value, so as to obtain the target neural network model.

[0129] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0130] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0131] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0132] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0134] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0135] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0136] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0137] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0138] The above are only embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included within the scope of the claims of the present application.

Claims

1. A communication method for an unmanned aerial vehicle, characterized in that: include: Classifying the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; Performing risk assessment on the second type of communication signal based on the first type of communication signal and the target neural network model to obtain a target risk value; According to the target risk value, determining whether to adjust the communication channel through which the target UAV receives the communication signal; When adjusting the communication channel through which the target UAV receives the communication signal, the communication channel through which the target UAV receives the communication signal is adjusted to the target channel, and the communication signal is received through the target channel.

2. The method according to claim 1, characterized in that The risk of the second type of communication signal is evaluated based on the first type of communication signal and the target neural network model, and the target risk value is obtained, including: Performing risk assessment on the second type of communication signal by using the target neural network model to obtain a first risk value; Performing a risk assessment on the second type of communication signal based on the first type of communication signal to obtain a second risk value; The target risk value is obtained by performing calculation according to the first risk value and the second risk value.

3. The method according to claim 2, characterized in that Performing risk assessment on the second type of communication signal by using the target neural network model to obtain a first risk value includes: Processing the signal attribute corresponding to the second type of communication signal by the first network sub-model in the target neural network model to obtain a target execution action, wherein the target execution action is used to instruct to adjust the communication parameters of the target UAV; After adjusting the communication parameters of the target UAV according to the target execution action, obtaining an updated second type of communication signal received by the target UAV; The target execution action and the updated second type of communication signal are evaluated by a second network sub-model in the target neural network model to obtain a target evaluation value, wherein the target evaluation value is used to indicate the influence degree of the target execution action; Perform risk assessment on the second type of communication signal according to the target evaluation value to obtain the first risk value.

4. The method according to claim 2, characterized in that: Performing risk assessment on the second type of communication signal according to the first type of communication signal to obtain a second risk value includes: Calculating the similarity between the first instruction content corresponding to the first type of communication signal and the second instruction corresponding to the second type of communication signal to obtain a first similarity; Calculating a difference between a signal strength value corresponding to the first type of communication signal and a signal strength value corresponding to the second type of communication signal to obtain a target difference; Calculating the similarity between the signal attribute corresponding to the second type of communication signal and the signal attributes corresponding to multiple historical communication signals to obtain a second similarity; The second risk value is obtained according to the first similarity, the target difference and the second similarity.

5. The method according to claim 2, characterized in that: Calculating according to the first risk value and the second risk value to obtain the target risk value includes: Obtaining a first weight value corresponding to the first risk value; Obtaining a second weight value corresponding to the second risk value; The target risk value is obtained by performing calculation according to the first weight value, the second weight value, the first risk value and the second risk value.

6. The method according to claim 1, characterized in that Judging whether to adjust the communication channel for receiving the communication signal of the target UAV according to the target risk value includes: Determine whether the target risk value is greater than a preset threshold; If the target risk value is greater than the preset threshold, determining to adjust the communication channel through which the target UAV receives the communication signal; If the target risk value is less than or equal to the preset threshold, it is determined not to adjust the communication channel for the target UAV to receive the communication signal.

7. The method according to claim 3, characterized in that The target neural network model is obtained by following the steps below: Outputting a sample to perform an action based on an initial sample communication signal through a first network sub-model in the initial neural network; Determining a true evaluation value according to the updated sample communication signal and the sample execution action, wherein the updated sample communication signal is an updated sample communication signal received by the drone after the communication parameters of the drone are adjusted based on the sample execution action; Processing the updated sample communication signal and the sample execution action through the second network sub-model in the initial neural network to obtain a predicted evaluation value; The second network sub-model in the initial neural network is trained according to the predicted evaluation value and the true evaluation value, and the first network sub-model in the initial neural network is trained according to the predicted evaluation value to obtain the target neural network model.

8. A communication device for an unmanned aerial vehicle, characterized in that: include: A classification unit, configured to classify the target communication signal received by the target UAV to obtain a first type of communication signal and a second type of communication signal, wherein the source of the first type of communication signal is the target source, and the source of the second type of communication signal is not the target source; an evaluation unit, configured to perform risk evaluation on the second type of communication signals based on the first type of communication signals and a target neural network model to obtain a target risk value; A judgment unit, configured to judge whether to adjust a communication channel for receiving communication signals of the target UAV according to the target risk value; The adjustment unit is used to adjust the communication channel through which the target UAV receives communication signals to a target channel and receive communication signals through the target channel when adjusting the communication channel through which the target UAV receives communication signals.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the storage medium is controlled to execute the communication method of the drone according to any one of claims 1 to 7 on the device.

10. An electronic device, characterized in that: It includes one or more processors and a memory, wherein the memory is used to store one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the communication method of the drone described in any one of claims 1 to 7.

Images