An asynchronous federated learning method for realizing lightweight privacy protection in Internet of Vehicles

Abstract

The present application belongs to the technical field of federated learning and Internet of Vehicles, and particularly relates to an asynchronous federated learning method for realizing lightweight privacy protection in Internet of Vehicles. Public key exchange is performed between stations and base stations to calculate a shared secret key; vehicles update local models by using global model parameters and train by using local data; random numbers are generated based on the shared secret key and training rounds; random numbers are added to the gradients of the trained local model parameters to realize encryption of the local model parameters; RSU forwards vehicle information and training rounds to the base station, the base station finds the shared secret key corresponding to the vehicle information, and calculates the sum of the random numbers in combination with the training rounds; RSU decrypts and aggregates the encrypted gradient information based on the sum of the random numbers, and updates the global model parameters; RSU broadcasts the global model parameters, the vehicles update the local models by using the global model parameters, and then continue the next round of training until the training is completed.

Description

Technical Field

[0001] The present invention belongs to the interdisciplinary field of federated learning and Internet of Vehicles, and specifically relates to an asynchronous federated learning method for achieving lightweight privacy protection in the Internet of Vehicles. Background Art

[0002] In recent years, leveraging IoT technology, the Internet of Vehicles (IoV) has become a key component of smart cities, making people's lives more convenient. In IoV, various intelligent vehicles and transportation infrastructure can collect vast amounts of data, including traffic flow, driving behavior, and road conditions. Federated learning, a distributed machine learning algorithm, allows participants to jointly train machine learning models without compromising personal privacy. To improve IoV performance while protecting the privacy of vehicle data, federated learning has been applied to the IoV sector. It has broad application prospects, including predicting and optimizing traffic flow, assisting autonomous driving systems, improving vehicle safety, and achieving energy conservation and emission reduction.

[0003] In traditional federated learning, roadside computing units (RSUs) must wait for all vehicles to upload their local models before they can aggregate the global model. However, due to the varying local data size and computing power of each vehicle, their local training time also varies. Furthermore, differences in vehicle location and speed also lead to varying transmission latency, resulting in varying impacts on global aggregation from vehicle to vehicle. To address this issue, asynchronous federated learning has been introduced in the IoV. However, most existing IoV federated learning solutions still focus on synchronous aggregation, with little discussion of asynchronous aggregation. The few IoV asynchronous aggregation solutions also fail to consider vehicle model privacy protection and communication optimization. For example, in typical IoV federated learning solutions, models uploaded by vehicles are not encrypted. After receiving the model, RSUs can easily launch inference attacks against the vehicles and obtain local data features. Solutions designed to protect vehicle privacy in IoV often employ differential privacy or homomorphic encryption. The former compromises RSU learning efficiency, while the latter increases the system's computational and communication overhead. Therefore, a lightweight, privacy-preserving asynchronous aggregation solution is essential for the IoV environment. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a lightweight privacy-preserving asynchronous aggregation method in an Internet of Vehicles environment, which can efficiently and securely implement federated learning in the Internet of Vehicles.

[0005] The technical solutions for implementing the present invention are as follows:

[0006] A lightweight, privacy-preserving asynchronous aggregation method in an Internet of Vehicles (IoV) environment includes the following steps:

[0007] Step 1: The station and the base station exchange public keys and calculate the shared secret key based on the public keys exchanged between the two parties;

[0008] Step 2: The vehicle obtains the latest global model parameters from the roadside computing unit (RSU) for local model update and uses local data to train the local model.

[0009] Step 3: Generate a random number based on the shared secret key and the training round. The vehicle adds a random number to the gradient of the trained local model parameters to encrypt the local model parameters, and sends the vehicle information, encrypted gradient information, and training round to the RSU.

[0010] Step 4: The RSU forwards the acquired vehicle information and training rounds to the base station. The base station searches for the shared key corresponding to the vehicle information and calculates the sum of random numbers based on the training rounds. The RSU decrypts and aggregates the encrypted gradient information based on the sum of random numbers and updates the global model parameters.

[0011] Step 5: The RSU broadcasts the global model parameters, and the vehicle uses the global model parameters to update the local model, and then continues the next round of training until the training is completed.

[0012] Furthermore, the specific process of step 1 of the present invention is:

[0013] First, when a vehicle enters the communication range of a base station, it sends its identity information and public key to the base station. After receiving the information sent by the vehicle, the base station also sends its public key to the vehicle.

[0014] Secondly, the base station and the vehicle each calculate a shared key using the received public key.

[0015] Furthermore, the secret key sent by the vehicle of the present invention is The secret key sent by the base station is The calculated shared key is

[0016] Furthermore, the random number described in the present invention is generated by using a shared secret key and the current training round as seeds and using a random number generator to generate a random number.

[0017] Furthermore, the seed of the present invention is key i +t i ,key i is the shared key, t i is the training round.

[0018] Furthermore, the specific process of step 4 of the present invention is:

[0019] Step 4.1: Determine the aggregation condition: At the end of a time slot, the RSU aggregates the gradient information received in that time slot. If the number of gradient information does not exceed 1, the current round of aggregation is skipped and the received gradient information participates in the next round of aggregation. Otherwise, the RSU proceeds to step 4.2 to continue the current round of aggregation.

[0020] Step 4.2: Calculate the sum of random numbers: The RSU organizes the information of the models in this time slot and sends the round number of each model and its corresponding vehicle identity information to the base station. The base station uses the same shared key and training round number as the vehicle as the seed of the random number generator to generate random numbers and sends the sum of the random numbers of the models in the same training round to the RSU.

[0021] Step 4.3: Calculate aggregation weight: RSU calculates outdated weight based on the training round of the model uploaded by the vehicle, and RSU calculates training progress weight based on the global training progress. Both weights will be used in global aggregation;

[0022] Step 4.4: Update the global model: The RSU aggregates the local model according to the different training rounds, subtracts the sum of the random numbers received from the base station for that round from the aggregation result, and then adds the aggregation results of different rounds according to the corresponding outdated weights and training weights to update the global model of this round.

[0023] Furthermore, the training progress weight γ of the present invention is:

[0024]

[0025] Among them, α represents the progress parameter, t global Indicates the global training round for RSU calculation.

[0026] Furthermore, the outdated weight β of the present invention is:

[0027]

[0028] Among them, T limit represents the expiration threshold, t i represents the global training round during vehicle local training, t global Indicates the global training round for RSU calculation.

[0029] Furthermore, the global model of the present invention is updated as follows:

[0030]

[0031] Among them, w t Indicates the global model parameters that are not updated in this round, w t+1 represents the updated global model parameters, N newN represents the number of vehicle gradients that arrive normally without being affected by the delay. last represents the number of outdated vehicle gradients, |U j | represents the number of outdated models in round j, represents the sum of random numbers of the model trained in round t, Δw n ' ew represents the encrypted vehicle gradient sent to the RSU on time in this round of iteration, Δw i ′ represents the encrypted vehicle gradient of training round i received by RSU in this round of iteration, Represents the sum of random numbers in the latest training round.

[0032] Furthermore, the local model described in the present invention is a convolutional neural network with two 5×5 convolution layers, the first layer has 32 output channels, the second layer has 64 output channels, and each convolution layer is followed by a 2×2 maximum pooling layer.

[0033] Beneficial effects:

[0034] The present invention proposes a lightweight asynchronous federated learning solution that does not bring additional communication and computing load to federated learning in the Internet of Vehicles. To address the privacy leakage problem in existing Internet of Vehicles, the present invention proposes a lightweight encryption and decryption method that can achieve secure aggregation without increasing the computing and communication consumption of vehicles and RSUs. To address the problems of unstable communication and extended transmission time in the Internet of Vehicles, the present invention proposes an asynchronous federated learning solution that can asynchronously aggregate vehicle models, cope with the situation where vehicles are disconnected during the learning process, and improve the efficiency of federated learning.

[0035] After the vehicle of the present invention is encrypted, the subsequent decryption and global aggregation by the RSU do not require the participation of the vehicle, allowing the vehicle to leave the training midway. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 This is a flowchart of the asynchronous federated learning of the Internet of Vehicles of the present invention. DETAILED DESCRIPTION

[0038] The embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0039] It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments may be combined with each other; and, based on the embodiments in this disclosure, all other embodiments obtained by persons of ordinary skill in the art without creative work are within the scope of protection of this disclosure.

[0040] It should be noted that various aspects of the embodiments within the scope of the appended claims are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on this disclosure, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects described herein can be used to implement an apparatus and / or practice a method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this apparatus and / or practice this method.

[0041] In order to solve the problem of inconsistent arrival time of vehicle models in the Internet of Vehicles, while achieving privacy protection of vehicle data and reducing communication and computing overhead, the embodiment of the present application proposes an asynchronous federated learning method for achieving lightweight privacy protection in the Internet of Vehicles. This method uses lightweight encryption primitives to encrypt the vehicle local model, and realizes decryption and aggregation at the RSU end through collaboration with the base station.

[0042] The federated learning method in the embodiment of the present application includes three types of entities: base stations, vehicles, and roadside computing units (RSUs). The system assumes that there are several vehicles, a single base station, and a single RSU; among them, the base station can capture information of the entire system, the vehicle is the client participating in the federated learning of the Internet of Vehicles, and the RSU is responsible for processing and aggregating vehicle models in federated learning.

[0043] The embodiment of the present application provides an asynchronous federated learning method for achieving lightweight privacy protection in the Internet of Vehicles, which mainly includes global initialization, local training encryption, and global model aggregation. The specific process is as follows:

[0044] Step 1: The station and the base station exchange and share keys, and calculate the shared key based on the secret key exchanged between the two parties; the process of this step is:

[0045] Step 1.1: Vehicle sends key and identity information ID i To the base station, the base station sends the key to vehicles;

[0046] Step 1.2: The vehicle and base station each calculate a shared key: The base station stores the vehicle's keys and identity information.

[0047] In this step, the keys calculated by the vehicle and the base station are the same, thereby achieving key exchange.

[0048] Step 2: The vehicle obtains the latest global model parameters from the roadside computing unit (RSU) for local model update and uses local data to train the local model.

[0049] Local model initialization: Before local model training begins, the RSU needs to initialize the global model for the vehicle's first round of learning. The vehicle obtains the latest global model parameters from the RSU to initialize the local model.

[0050] Local model training: Vehicles use the stochastic gradient descent algorithm and local data for model training. The vehicles here refer to all vehicles within the communication range of the RSU, and multiple vehicles perform local training in parallel.

[0051] In this embodiment, the number of vehicles is assumed to be 100, and the local data of the vehicles are sampled from the MNIST dataset. The model used for training is a convolutional neural network with two 5×5 convolutional layers. The first layer has 32 output channels and the second layer has 64 output channels. Each convolutional layer is followed by a 2×2 maximum pooling layer. 450 samples are taken for each vehicle, and local training is performed using the stochastic gradient descent algorithm. The training learning rate is 0.01, the training momentum is 0.9, and the weight decay is 5e-3. After one round of local training, the local model is obtained.

[0052] In this embodiment, multiple vehicles can also train the local model simultaneously. The number of vehicles selected to participate in the aggregation in each round is 20, and the transmission delay of the vehicles is 5 to 1000ms, which obeys the Gaussian distribution.

[0053] Step 3: The vehicle adds random numbers to the gradient of the local model parameters to encrypt the local model parameters, and sends the vehicle information, encrypted gradient information, and training rounds to the RSU for global model aggregation.

[0054] In this embodiment, the vehicle adds random numbers to the gradient of the local model parameters. The random numbers are generated using a random number generator, and the seed used is the sum of the vehicle shared key and the current training round, i.e., key i +t i ;t i is the current training round.

[0055] Step 4: RSU decrypts and aggregates the received gradients.

[0056] Step 4.1: Each RSU time slot is 200ms long. At the end of the time slot, the RSU determines the aggregation condition. If no more than one gradient information is received, the current round of aggregation is skipped and the gradient information of the received model is used for the next round of aggregation. Otherwise, the current round of aggregation is continued in step 4.2.

[0057] Step 4.2: Encrypted random number calculation: RSU organizes the information of the models in this time slot, and sends the round number of each model and its corresponding vehicle identity information to the base station. The base station finds its corresponding shared key based on the vehicle information. i , use the random number generator to generate a random number r i , the seed is the key i +t i , calculate the sum of random numbers in the same training round Return to RSU, |U j | represents the number of outdated models in round j. Outdated models are updated models. For example, in round j training, the training round t i The model corresponding to a value less than j is called an outdated model.

[0058] Step 4.3: Calculate aggregation weight: RSU calculates outdated weight based on the training round of the model uploaded by the vehicle, and RSU calculates training progress weight based on the global training progress. Both weights will be used in global aggregation;

[0059] The specific expression of the training progress weight γ is as follows:

[0060]

[0061] Among them, γ represents the training progress weight, α represents the progress parameter, which is 0.2 in this example, and t global Indicates the global training round for RSU calculation.

[0062] The specific expression of outdated weight β is as follows:

[0063]

[0064] Among them, β represents the outdated weight, T limit Represents the outdated threshold, models exceeding this threshold will not be adopted into the aggregation, t i represents the global training round during vehicle local training, t global Indicates the global training round for RSU calculation.

[0065] Step 4.4: Global model update: The RSU eliminates the encrypted information and aggregates the local model parameters of the vehicle to complete the global model update. The specific expression is as follows:

[0066]

[0067] Among them, w t Indicates the global model parameters that are not updated in this round, w t+1 represents the updated global model parameters, N new N represents the number of vehicle gradients that arrive normally without being affected by the delay. last represents the number of outdated vehicle gradients, |U j | represents the number of outdated models in round j, Indicates the tth i The sum of random numbers of the model trained in rounds, Δw n ' ew represents the encrypted vehicle gradient sent to the RSU on time in this round of iteration, Δw i ′ represents the encrypted vehicle gradient of training round i received by RSU in this round of iteration, Represents the sum of random numbers in the latest training round.

[0068] Step 5: RSU broadcasts the global model. The RSU sends the global model to all vehicles within the communication range. The vehicles use the global model parameters to update the local model and then continue the next round of training until the training is completed.

[0069] In summary, the above are only preferred embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A lightweight, privacy-preserving asynchronous aggregation method in an Internet of Vehicles environment, characterized by: The steps include: Step 1: The station and the base station exchange public keys and calculate the shared secret key based on the public keys exchanged between the two parties; Step 2: The vehicle obtains the latest global model parameters from the roadside computing unit (RSU) for local model update and uses local data to train the local model. Step 3: Generate a random number based on the shared secret key and the training round. The vehicle adds a random number to the gradient of the trained local model parameters to encrypt the local model parameters, and sends the vehicle information, encrypted gradient information, and training round to the RSU. Step 4: The RSU forwards the acquired vehicle information and training rounds to the base station. The base station searches for the shared key corresponding to the vehicle information and calculates the sum of random numbers based on the training rounds. The RSU decrypts and aggregates the encrypted gradient information based on the sum of random numbers and updates the global model parameters. Step 5: The RSU broadcasts the global model parameters, and the vehicle uses the global model parameters to update the local model, and then continues the next round of training until the training is completed; The specific process of step 4 is as follows: Step 4.1: Determine the aggregation condition: At the end of a time slot, the RSU aggregates the gradient information received in that time slot. If the number of gradient information does not exceed 1, the current round of aggregation is skipped and the received gradient information participates in the next round of aggregation. Otherwise, the RSU proceeds to step 4.2 to continue the current round of aggregation. Step 4.2: Calculate the sum of random numbers: The RSU organizes the information of the models in this time slot and sends the round number of each model and its corresponding vehicle identity information to the base station. The base station uses the same shared key and training round number as the vehicle as the seed of the random number generator to generate random numbers and sends the sum of the random numbers of the models in the same training round to the RSU. Step 4.3: Calculate aggregation weight: RSU calculates outdated weight based on the training round of the model uploaded by the vehicle, and RSU calculates training progress weight based on the global training progress. Both weights will be used in global aggregation; Step 4.4: Update the global model: The RSU aggregates the local model according to the different training rounds, subtracts the sum of the random numbers received from the base station in the aggregation result, and then adds the aggregation results of different rounds according to the corresponding outdated weights and training weights to update the global model of this round; The training progress weight γ: Among them, α represents the progress parameter, t global Indicates the global training round when calculating RSU; The outdated weight β: Among them, T limit represents the expiration threshold, t i represents the global training round during vehicle local training, t global Indicates the global training round for RSU calculation.

2. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 1 is characterized in that: The specific process of step 1 is: First, when a vehicle enters the communication range of a base station, it sends its identity information and public key to the base station. After receiving the information sent by the vehicle, the base station also sends its public key to the vehicle. Secondly, the base station and the vehicle each calculate a shared key using the received public key.

3. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 2 is characterized in that: The secret key sent by the vehicle is The secret key sent by the base station is The calculated shared key is 4. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 3 is characterized in that: The random number is generated by using a shared secret key and the current training round as a seed and using a random number generator to generate a random number.

5. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 4 is characterized in that: The seed is key i +t i ,key i is the shared key, t i is the training round.

6. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 1 is characterized in that: The global model is updated as follows: Among them, w t Indicates the global model parameters that are not updated in this round, w t+1 represents the updated global model parameters, N new N represents the number of vehicle gradients that arrive normally without being affected by the delay. last represents the number of outdated vehicle gradients, |U j | represents the number of outdated models in round j, The sum of random numbers of the model trained in round t, Δw′ new represents the encrypted vehicle gradient sent to the RSU on time in this round of iteration, Δw′ i Indicates the encrypted vehicle gradient of training round i received by RSU in this iteration, Represents the sum of random numbers in the latest training round.

7. The lightweight privacy-preserving asynchronous aggregation method in the Internet of Vehicles environment according to claim 1 is characterized in that: The local model is a convolutional neural network with two 5×5 convolutional layers. The first layer has 32 output channels and the second layer has 64 output channels. Each convolutional layer is followed by a 2×2 maximum pooling layer.

Images