Air-ground cooperative online federated learning method for unreliable communication

Abstract

The invention provides an unreliable communication-oriented air-ground collaborative online federated learning method. The method comprises the following steps: an unmanned aerial vehicle collects real-time image data according to a track preset in advance and distributes and screens samples according to dynamic data to update a local model; a model parameter error occurs due to the influence of unstable channels such as path loss and shadow fading in a local model uploading process, and the central server screens unmanned aerial vehicles to participate in global aggregation according to the model parameter error; through joint optimization of client scheduling and sample screening, the technical purpose of minimizing training loss under the condition that time delay conditions, energy consumption constraints and decision variable basic limitations are met is achieved. The method provided by the invention has the characteristics of high efficiency and intellectualization, and the precision of unmanned aerial vehicle cooperative target detection is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of air-ground collaborative online federated learning, and specifically to an air-ground collaborative online federated learning method for unreliable communication. Background Art

[0002] Recent advances in drone technology have shown great potential in multiple fields due to its high flexibility, three-dimensional mobility, and ease of deployment. In addition to collecting data, drones can train machine learning models onboard, which is essential for tasks such as target recognition. Equipped with sensors and intelligent computing modules, drones can cover larger areas in less time, perform real-time intelligent data processing, and promote the expansion of intelligent network services from the ground to low-altitude airspace.

[0003] Despite the huge potential of drones, there are inherent challenges with limited onboard resources. IoT applications, such as autonomous driving and target tracking, involve highly distributed datasets, making it challenging to centrally process the data. This poses significant challenges for drones, and a distributed collaborative learning paradigm is urgently needed to address these issues. Federated learning excels in collaboration, efficiency, and edge processing, making it an ideal choice for collaborative target detection in drones. Its principle is to collaboratively train the same machine learning model among a group of drones using local datasets.

[0004] The drone acts as a client to train a local model and uploads its local model to the base station for global aggregation to form air-ground collaborative federated learning. However, current work mainly focuses on offline training of existing datasets, which cannot be applied to real-time target detection tasks enabled by drones. Therefore, online federated learning takes precedence over adapting to dynamic environments, including changes in participating clients and changes in data distribution. In practice, incorporating all real-time new sample data into training will impose huge computational costs on resource-limited drones. In addition, the unreliability of air-ground wireless channel transmission, including path loss and small-scale fading, increases the model parameter error rate. That is, during the uplink transmission of the local model, each layer of the neural network faces the problem of parameter error, which ultimately leads to a decrease in the overall accuracy of the model. Summary of the invention

[0005] The present invention provides an online federated learning method for air-ground collaboration in unreliable communication, which achieves the technical purpose of minimizing training loss under the conditions of delay, energy consumption and basic restrictions on decision variables by jointly optimizing client scheduling and sample screening. The method proposed in the present invention is efficient and intelligent, and greatly improves the accuracy of cooperative target detection of UAVs.

[0006] In order to achieve the above technical objectives, the technical solution adopted by the present invention is:

[0007] In a first aspect, the present invention discloses an air-ground collaborative online federated learning method for unreliable communication, comprising the following steps:

[0008] Step 1: The drone screens sample images collected in real time in a dynamic environment, uses meta-gradient update training to generate a local personalized model, introduces the Bernoulli distribution to describe the probability that errors will affect the accuracy of the global model when the drone uploads model parameters in an unstable channel, and analyzes the convergence of air-ground collaborative online federated learning with parameter errors. The expected convergence bound of air-ground collaborative online federated learning with parameter errors is used to characterize the global training loss.

[0009] Step 2: Calculate the uplink communication delay and downlink communication delay of the UAV in one round of global training according to the uplink / downlink link rate of the UAV and the model parameter quantity, and calculate the calculation delay of the UAV in one round of global training according to the calculation frequency of the UAV's onboard embedded device and the size of the training data set;

[0010] Step 3: Calculate the communication energy consumption of the user in one round of global training based on the transmission power and communication delay of the drone, and calculate the computing energy consumption of the drone in one round of global training based on the computing frequency of the drone's onboard embedded device and the size of the training data set;

[0011] Step 4: Based on the idea of ​​alternating optimization, the problem of minimizing the global training loss is decomposed into the client scheduling problem and the sample screening problem. According to the energy consumption constraints, delay constraints and basic restrictions on optimization variables, the optimal solution for drone scheduling and sample screening is obtained through iterative optimization.

[0012] As a preferred example, in step 1, the expected convergence bound G of air-ground collaborative online federated learning with parameter errors is loss The calculation formula is:

[0013]

[0014] in,

[0015]

[0016] In the formula, η1 is the learning rate, L is the Lipschitz constant, and D i is the size of the local dataset of drone i, D is the size of the training set after filtering samples, N is the number of drones, α i ∈(0,1) is the sample screening variable, β i ∈{0,1} is the client screening variable, e i is the parameter error rate during the transmission process of UAV i, ε is a positive parameter, η2 is the meta-learning rate, t is the training round, θ is the waterfall threshold, E|h i | 2is the channel gain, B U is the uplink bandwidth, p i is the transmission power of UAV i, N0 is the power spectrum density of Gaussian noise, φ is the intermediate variable expressed in the simplified formula, t is the global training round, φ t is the result of recursively summing the intermediate variable φ in the global round.

[0017] As a preferred example, in step 2, the uplink communication delay T i U , Downlink communication delay T i D and calculation delay T i C The calculation formula is:

[0018]

[0019] in,

[0020]

[0021] In the formula, S i (ω) is the number of model parameter bits uploaded by UAV i, S(Ω) is the number of global model parameter bits broadcasted, B U , B D are uplink / downlink bandwidth, respectively, i 、p B are the transmission powers of drone i and base station respectively, N0 is the power spectrum density of Gaussian noise, c i The number of CPU cycles required to process a single data sample for drone i, g i is the CPU frequency of drone i, s is the speed of light, f c is the carrier frequency, v i is the flight speed of drone i, d i is the distance from drone i to the base station, are the line-of-sight link / non-line-of-sight link losses, ζ i is the probability of line-of-sight link between UAV i and the base station, α i ∈(0,1) is the sample screening variable.

[0022] As a preferred example, in step 3, the communication energy consumption and calculate energy consumption The calculation formula is:

[0023]

[0024] In the formula, Calculate the chip effective capacitance coefficient, S, for drone i i (ω) is the number of bits of model parameters uploaded by UAV i, pi is the transmission power of UAV i, B U is the uplink bandwidth, E|h i | 2 is the channel gain, α i ∈(0,1) is the sample screening variable, g i is the CPU frequency of drone i, β i ∈{0,1} is the client screening variable, and N0 is the power spectral density of Gaussian noise.

[0025] As a preferred example, step 4 specifically includes the following sub-steps:

[0026] Step 4.1: According to the first delay constraint, the first energy consumption constraint, the basic restrictions of the client screening variables and the initial feasible solution of the sample screening variables, the client scheduling is optimized by the branch and bound method to obtain the optimal solution of the current client screening variables;

[0027] Step 4.2: Based on the optimal solution of the client screening variable obtained in step 4.1, the sample screening variable is optimized by the Dinkelbach algorithm in combination with the second delay constraint, the second energy consumption constraint, the gradient error constraint and the basic restrictions of the sample screening variable to obtain the optimal solution of the current sample screening variable;

[0028] Step 4.3: Repeat the alternating iterations of steps 4.1 to 4.2 until the relative error of the optimization target of two iterations is less than the predefined threshold, the global training loss is minimized, and the optimal solution for client scheduling and sample screening is obtained.

[0029] As a preferred example, in step 4.1, the first delay constraint is calculated as follows:

[0030]

[0031] Where, T th represents the upper limit of the delay of a round of training, Screen the initial feasible solutions of variables for the sample;

[0032] The first energy consumption constraint is calculated as follows:

[0033]

[0034] In the formula, E th Indicates the upper limit of energy consumption for one round of training;

[0035] The basic restrictions on client-side filter variables are as follows:

[0036] β i ∈{0,1}.

[0037] As a preferred example, in step 4.2, the second delay constraint is calculated as follows:

[0038]

[0039] In the formula, Filter the optimal solution of variables for the client;

[0040] The second energy consumption constraint is calculated as follows:

[0041]

[0042] The gradient error is calculated as follows:

[0043]

[0044] In the formula, is the variability of the gradient data, is the variability of the Hessian data, is the gradient error threshold;

[0045] The basic restrictions on sample screening variables are as follows:

[0046] 0≤α i =α i,1 +α i,2 +α i,3 ≤1

[0047] In the formula, α i,1 , α i,2 , α i,3 Represent the screening variables of three independent small batches of samples.

[0048] As a preferred example, in step 4.3, when φ>1, the expected convergence bound of the minimized global training loss t→∞ diverges;

[0049] When φ<1, the t→∞ expected convergence bound converges,

[0050]

[0051] The minimized global training loss is calculated as:

[0052]

[0053] In the formula, α represents the sample screening variable set, and β represents the client screening variable set.

[0054] In a second aspect, the present invention further discloses a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program enables a computer to execute the air-ground collaborative online federated learning method for unreliable communication as described above.

[0055] In a third aspect, the present invention further discloses an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the air-ground collaborative online federated learning method for unreliable communication as described above is implemented.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] In the air-ground collaborative online federated learning method for unreliable communication of the present invention, the drone collects real-time image data according to the preset track and selects samples according to the dynamic data distribution for updating the local model. During the uploading process of the local model, model parameter errors occur due to the influence of unstable channels such as path loss and shadow fading. The central server selects drones to participate in global aggregation based on this. At the same time, according to the idea of ​​alternating optimization, the minimization problem is decomposed into client scheduling problem and sample screening problem, and the global training loss is minimized through iterative optimization. The method proposed by the present invention is efficient and intelligent, and greatly improves the training effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 This is a scene diagram of air-ground collaborative online federated learning under unreliable communication conditions of the present invention.

[0059] Figure 2 It is a flow chart of the air-ground collaborative online federated learning method for unreliable communication of the present invention.

[0060] Figure 3 It is a schematic diagram of the convergence numerical simulation results of the method of the present invention under different numbers of clients and different parameter error rates.

[0061] Figure 4 It is a schematic diagram showing the change of the convergence gap between the method of the present invention and the comparison algorithm as the number of iterations increases.

[0062] Figure 5 This is a schematic diagram of the impact of channel bandwidth on model performance. DETAILED DESCRIPTION

[0063] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings.

[0064] The present invention provides an air-ground collaborative online federated learning method for unreliable communication. Considering a complex dynamic environment, multiple drones equipped with embedded intelligent computing devices perform collaborative target detection tasks under the coordination of a central server. Each drone acts as an independent client and jointly trains a model without sharing data with others. The drone collects real-time image data according to a preset track and selects samples according to dynamic data distribution for updating the local model. In the process of uploading the local model, model parameter errors occur due to the influence of unstable channels such as path loss and shadow fading. The central server selects drones to participate in global aggregation based on this. By jointly optimizing client scheduling and sample screening, the problem of minimizing training loss is achieved under the conditions of satisfying delay conditions, energy consumption constraints and basic restrictions on decision variables. The client in the present invention is a drone equipped with an embedded intelligent computing device in the Internet of Things, and the central server is a ground base station. In practice, the real-time data collection by the drone leads to a continuous increase in the computing resources required for training. At the same time, in order to prevent the model parameter errors caused by unstable channels such as path loss and shadow fading during the upload of the local model from affecting the accuracy of the global model, it is necessary to schedule the drone to participate in aggregation effectively. Therefore, combining online federated learning with unreliable air-ground channels has the following challenges: 1) The design of air-ground collaborative online federated learning should consider the computational overhead brought by real-time samples to resource-constrained drones and screen samples; 2) The design of air-ground collaborative online federated learning should strictly consider the impact of unreliable air-ground channels; 3) The design of air-ground collaborative online federated learning should strictly consider timeliness and energy saving to ensure that computing and communication are completed within the specified time and energy consumption thresholds.

[0065] The air-ground collaborative online federated learning method for unreliable communication provided by the present invention specifically includes the following steps:

[0066] (1) The UAV screens sample images collected in real time in a dynamic environment and generates a local personalized model using meta-gradient update training. The Bernoulli distribution is introduced to describe the probability that errors will occur when the UAV uploads model parameters in an unstable channel, thereby affecting the accuracy of the global model. The convergence of air-ground collaborative online federated learning with parameter errors is analyzed, and the expected convergence bound is used to characterize the implicit global training loss.

[0067] The expected convergence bound G of the air-ground collaborative online federated learning with parameter errors in the present invention loss The calculation process is as follows:

[0068]

[0069] in,

[0070]

[0071] In the formula, η1 is the learning rate, L is the Lipschitz constant, and D i is the size of the local dataset of drone i, D is the size of the training set after filtering samples, N is the number of drones, α i ∈(0,1) is the sample screening variable, β i ∈{0,1} is the client screening variable, e i is the parameter error rate during the transmission process of UAV i, ε is a positive parameter, η2 is the meta-learning rate, t is the training round, θ is the waterfall threshold, E|h i | 2 is the channel gain, B U is the uplink bandwidth, p i is the transmission power of UAV i, N0 is the power spectrum density of Gaussian noise, φ is the intermediate variable expressed in the simplified formula, t is the global training round, φ t is the result of recursively summing the intermediate variable φ in the global round

[0072] (2) The communication delay of the UAV in a round of global training is calculated based on the UAV uplink / downlink rate and the model parameter quantity. The computing delay of the UAV in a round of global training is calculated based on the computing frequency of the UAV's onboard embedded device and the size of the training data set.

[0073] The uplink communication delay T in the present invention i U , downlink communication delay T i D and calculation delay T i C The calculation process is as follows:

[0074]

[0075] in,

[0076]

[0077] In the formula, S i (ω) is the number of model parameter bits uploaded by UAV i, S(Ω) is the number of global model parameter bits broadcasted, B U , B D are uplink / downlink bandwidth, respectively, i 、p B are the transmission powers of drone i and base station respectively, N0 is the power spectrum density of Gaussian noise, c i The number of CPU cycles required to process a single data sample for drone i, g i is the CPU frequency of drone i, s is the speed of light, f c is the carrier frequency, v i is the flight speed of drone i, d iis the distance from drone i to the base station, are the line-of-sight link / non-line-of-sight link losses, ζ i is the probability of line-of-sight link between UAV i and the base station.

[0078] (3) The communication energy consumption of the user in a round of global training is calculated based on the drone's transmission power and communication delay. The computing energy consumption of the drone in a round of global training is calculated based on the computing frequency of the drone's onboard embedded device and the size of the training data set.

[0079] The communication energy consumption in the present invention and calculate energy consumption The calculation process is as follows:

[0080]

[0081] In the formula, Calculate the chip effective capacitance coefficient for drone i.

[0082] (4) Decompose the original problem into two optimization problems: drone scheduling and sample screening according to the nature of the decision variables, and repeat the alternating iteration until the expected convergence bound of the minimum is obtained, and the optimal solution of client scheduling and sample screening is obtained. The specific sub-steps include the following:

[0083] Step 4.1, select the initial feasible solution of the sample screening variable, and transform the original problem into a mixed integer linear programming problem to solve the optimal solution of client scheduling, which greatly reduces the difficulty of solving the original problem. Therefore, according to the first delay constraint, the first energy consumption constraint and the basic restrictions of the client screening variables, the client scheduling is optimized through the branch and bound method to obtain the optimal solution of the current client screening variables.

[0084] The first delay constraint involved in this process is specifically:

[0085]

[0086] Where, T th represents the upper limit of the delay of a round of training, Screen the initial feasible solutions for the variables for the sample.

[0087] The first energy consumption constraint is:

[0088]

[0089] In the formula, E th Indicates the upper limit of energy consumption for one round of training.

[0090] The basic restrictions on client filter variables are:

[0091] β i ∈{0,1}

[0092] Step 4.2, substitute the optimal solution of the client screening variable obtained in step 4.1 into the original problem and transform it into a fractional linear programming problem. Combined with the second delay constraint, the second energy consumption constraint, the gradient error constraint and the basic restrictions of the sample screening variables, the sample screening variables are optimized through the Dinkelbach algorithm to obtain the optimal solution of the current sample screening variables.

[0093] The second delay constraint involved in this process is specifically:

[0094]

[0095] In the formula, β i * Optimal solution for the client screening variables found in step 4.1.

[0096] The second energy consumption constraint is:

[0097]

[0098] The gradient error is:

[0099]

[0100] In the formula, is the variability of the gradient data, is the variability of the Hessian data, is the gradient error threshold.

[0101] The basic restrictions on sample screening variables are:

[0102] 0≤α i =α i,1 +α i,2 +α i,3 ≤1

[0103] In the formula, α i,1 , α i,2 , α i,3 Represent the screening variables of three independent small batches of samples.

[0104] Step 4.3, repeat the alternating iterations of steps 4.1 to 4.2 until the relative error of the trade-off objectives of the two iterations is less than the predefined threshold, the global training loss is minimized, and the optimal solution for UAV scheduling and sample screening is obtained.

[0105] The global training loss involved in the process of minimizing is when t→∞, the expected convergence bound G loss The expression is:

[0106] When φ>1, the t→∞ expected convergence bound diverges, and this situation is not considered;

[0107] When φ<1, the t→∞ expected convergence bound converges,

[0108]

[0109] In the formula, both the numerator and the denominator contain And the signs are different, so the problem can be directly optimized. Finally, the minimized global training loss is:

[0110]

[0111] In the formula, α represents the sample screening variable set, and β represents the client screening variable set.

[0112] Figure 1 The air-ground collaborative online federated learning scenario under unreliable communication conditions is described. This embodiment considers a total of N drones equipped with embedded intelligent computing devices to perform collaborative target detection tasks under the coordination of a central server. Each drone acts as an independent client, with the goal of collaboratively training a target detection model and does not share its own dataset. In the scenario considered, drones move to different locations to collect image data in real time. During the local model training process, whenever new samples are added, the drones screen the samples based on the local dataset distribution and use meta-gradients to update the local model. Subsequently, the drones transmit their trained local models to the base station. Due to the unreliability of the air-ground wireless channel, model transmission may encounter varying degrees of parameter errors. We use Bernoulli distribution to describe the parameter error rate. In order to prevent the decrease in global model accuracy due to the aggregation of models with high parameter error rates, the base station must screen drones to perform aggregation.

[0113] Figure 2 This is a flowchart of the air-ground collaborative online federated learning method for unreliable communication. ① The base station broadcasts the initial model. ② The drone collects image data in real time and adds new samples to the original data set. ③ The drone screens samples. ④ The drone performs local training using the screened training set. ⑤ The drone uploads the local model to the base station. ⑥ Client screening. ⑦ The base station performs global aggregation. ⑧ Repeat the above steps. It should be noted that the screening of samples and clients is a joint decision solved by the optimization problem, which is performed by the drone and the base station respectively.

[0114] Figure 3The convergence of the method of the present invention was numerically simulated. In general, the algorithm converges after about 6 to 8 iterations, and the loss value after convergence is about 70% lower than the initial value. The method takes into account online learning and wireless factors, optimizes client selection and sample screening, reduces the impact of wireless transmission errors on federated learning training, and improves the performance of federated learning. In addition, we also simulated the convergence under different numbers of clients and different model parameter errors. As can be seen from the figure, with lower parameter error rates (that is, when the drone is closer to the base station to perform training tasks), the final loss value decreases as the number of drones increases, while the opposite is true for higher error rates. This is because the increase in the number of drones provides more data for federated learning training, thereby improving the approximate accuracy of the gradient of the loss function at low parameter error rates. On the contrary, at high parameter error rates, the number of drones that cannot participate in aggregation increases.

[0115] Figure 4 The vertical axis represents the expected convergence bound, from Figure 4 It can be seen that the simulation results obtained by the method of the present invention differ by less than 7% from the brute force search, which is better than the other three baselines. This confirms the effectiveness of the present invention. Figure 4 The increasing trend of the two curves also shows that as the number of drones increases, the number of drones that cannot participate in aggregation will also increase.

[0116] Figure 5 All the curves in the above figure show a common feature, that is, the target value increases with the increase of bandwidth due to the negative impact of the increase of parameter error rate on model performance. Their difference lies in the slope of each curve. When the bandwidth is high, the parameter error rate is also high. As the number of drones increases, the number of drones that do not participate in the aggregation will also increase, resulting in a decrease in the target value, which is consistent with Figure 3 The results in are consistent.

[0117] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of complete hardware embodiments, complete software embodiments, or embodiments in combination with software and hardware. Moreover, the present application can 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. The scheme in the embodiments of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0118] 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 1 A device that provides the functions specified in a block or multiple blocks.

[0119] 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.

[0120] 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 in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0121] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0122] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

Claims

1. A method for online federated learning of air-ground collaboration for unreliable communication, characterized in that: The following steps are involved: Step 1: The drone screens sample images collected in real time in a dynamic environment, uses meta-gradient update training to generate a local personalized model, introduces the Bernoulli distribution to describe the probability that errors will affect the accuracy of the global model when the drone uploads model parameters in an unstable channel, and analyzes the convergence of air-ground collaborative online federated learning with parameter errors. The expected convergence bound of air-ground collaborative online federated learning with parameter errors is used to characterize the global training loss. Step 2: Calculate the uplink communication delay and downlink communication delay of the UAV in one round of global training according to the uplink / downlink link rate of the UAV and the model parameter quantity, and calculate the calculation delay of the UAV in one round of global training according to the calculation frequency of the UAV's onboard embedded device and the size of the training data set; Step 3: Calculate the communication energy consumption of the user in one round of global training based on the transmission power and communication delay of the drone, and calculate the computing energy consumption of the drone in one round of global training based on the computing frequency of the drone's onboard embedded device and the size of the training data set; Step 4: Based on the idea of ​​alternating optimization, the problem of minimizing the global training loss is decomposed into the client scheduling problem and the sample screening problem. According to the energy consumption constraints, delay constraints and basic restrictions on optimization variables, the optimal solution for drone scheduling and sample screening is obtained through iterative optimization.

2. The air-ground collaborative online federated learning method for unreliable communication according to claim 1, characterized in that: In step 1, the expected convergence bound G of air-ground collaborative online federated learning with parameter errors is loss The calculation formula is: in, In the formula, η1 is the learning rate, L is the Lipschitz constant, and D i is the size of the local dataset of drone i, D is the size of the training set after filtering samples, N is the number of drones, α i ∈(0,1) is the sample screening variable, β i ∈{0,1} is the client screening variable, e i is the parameter error rate during the transmission process of UAV i, ε is a positive parameter, η2 is the meta-learning rate, t is the training round, θ is the waterfall threshold, E|h i | 2 is the channel gain, B U is the uplink bandwidth, p i is the transmission power of UAV i, N0 is the power spectrum density of Gaussian noise, φ is the intermediate variable, t is the global training round, φ t is the result of recursively summing the intermediate variable φ in the global round.

3. The air-ground collaborative online federated learning method for unreliable communication according to claim 1, characterized in that: In step 2, the uplink communication delay T i U , Downlink communication delay T i D and calculation delay T i C The calculation formula is: in, In the formula, S i (ω) is the number of model parameter bits uploaded by UAV i, S(Ω) is the number of global model parameter bits broadcasted, B U , B D are uplink / downlink bandwidth, respectively, i 、p B are the transmission powers of drone i and base station respectively, N0 is the power spectrum density of Gaussian noise, c i The number of CPU cycles required to process a single data sample for drone i, g i is the CPU frequency of drone i, s is the speed of light, f c is the carrier frequency, v i is the flight speed of drone i, d i is the distance from drone i to the base station, are the line-of-sight link / non-line-of-sight link losses, ζ i is the probability of line-of-sight link between UAV i and the base station, α i ∈(0,1) is the sample screening variable.

4. The air-ground collaborative online federated learning method for unreliable communication according to claim 1, characterized in that: In step 3, communication energy consumption and calculate energy consumption The calculation formula is: In the formula, Calculate the chip effective capacitance coefficient, S, for drone i i (ω) is the number of bits of model parameters uploaded by UAV i, p i is the transmission power of UAV i, B U is the uplink bandwidth, E|h i | 2 is the channel gain, α i ∈(0,1) is the sample screening variable, g i is the CPU frequency of drone i, β i ∈{0,1} is the client screening variable, and N0 is the power spectral density of Gaussian noise.

5. The air-ground collaborative online federated learning method for unreliable communication according to claim 2, characterized in that: Step 4 specifically includes the following sub-steps: Step 4.1: According to the first delay constraint, the first energy consumption constraint, the basic restrictions of the client screening variables and the initial feasible solution of the sample screening variables, the client scheduling is optimized by the branch and bound method to obtain the optimal solution of the current client screening variables; Step 4.2: Based on the optimal solution of the client screening variable obtained in step 4.1, the sample screening variable is optimized by the Dinkelbach algorithm in combination with the second delay constraint, the second energy consumption constraint, the gradient error constraint and the basic restrictions of the sample screening variable to obtain the optimal solution of the current sample screening variable; Step 4.3: Repeat the alternating iterations of steps 4.1 to 4.2 until the relative error of the optimization target of two iterations is less than the predefined threshold, the global training loss is minimized, and the optimal solution for client scheduling and sample screening is obtained.

6. The air-ground collaborative online federated learning method for unreliable communication according to claim 5, characterized in that: In step 4.1, the first delay constraint is calculated as follows: Where, T th represents the upper limit of the delay of a round of training, Screening the initial feasible solutions of variables for the sample; The first energy consumption constraint is calculated as follows: In the formula, E th Indicates the upper limit of energy consumption for one round of training; The basic restrictions on client-side filter variables are as follows: β i ∈{0,1}。 7. The air-ground collaborative online federated learning method for unreliable communication according to claim 5, characterized in that: In step 4.2, the second delay constraint is calculated as follows: In the formula, Filter the optimal solution of variables for the client; The second energy consumption constraint is calculated as follows: The gradient error is calculated as follows: In the formula, is the variability of the gradient data, is the variability of the Hessian data, is the gradient error threshold; The basic restrictions on sample screening variables are as follows: 0≤α i ＝α i,1 +α i,2 +α i,3 ≤1 In the formula, α i,1 , α i,2 , α i,3 Represent the screening variables of three independent small batches of samples.

8. The air-ground collaborative online federated learning method for unreliable communication according to claim 5, characterized in that: In step 4.3, when φ>1, the expected convergence bound of the minimized global training loss t→∞ diverges; When φ<1, the t→∞ expected convergence bound converges, The minimized global training loss is calculated as: In the formula, α represents the sample screening variable set, and β represents the client screening variable set.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which enables the computer to execute the air-ground collaborative online federated learning method for unreliable communication as described in any one of claims 1-8.

10. An electronic device, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the air-ground collaborative online federated learning method for unreliable communication as described in any one of claims 1 to 8 is implemented.

Images