Key generation method and system based on Internet of Vehicles channel phase reciprocity compensation

Abstract

The invention provides a key generation method and system based on Internet of Vehicles channel phase reciprocity compensation, and belongs to the technical field of communication, and the method comprises the steps: obtaining first channel state information of a sender in an Internet of Vehicles channel based on channel detection, and carrying out the preprocessing of the first channel state information, and obtaining a real part feature and an imaginary part feature of the first channel state information; respectively inputting the real part features and the imaginary part features to a pre-trained network model, wherein the network model compensates the real part features and the imaginary part features to realize phase feature compensation of the first channel state information; and generating a communication key according to the compensated first channel state information. According to the invention, the real part feature and the imaginary part feature of the first channel state information of the sender are compensated through the pre-trained network model, so that phase feature compensation of the first channel state information is realized, and the influence of Doppler frequency shift caused by two communication parties in the moving process on the phase is reduced; therefore, the reciprocity of channel characteristics of two communication parties is improved, and the inconsistency rate of communication keys is reduced.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a key generation method and system based on phase reciprocity compensation of vehicle networking channels. Background Art

[0002] In the Internet of Vehicles scenario with a large number of users, the Physical Layer Key Generation (PKG) technology utilizes the short-term reciprocity of the wireless channel to generate keys without the involvement of a third party, saving the overhead of key management and distribution, and is a lightweight solution.

[0003] In a real vehicle network communication environment, since the communicating parties usually have a high moving speed, causing a high Doppler frequency shift, the change in frequency will cause the phase to shift in time, and the reciprocity of the channel characteristics of the communicating parties will be significantly reduced, which will lead to a high inconsistency rate in generating keys using this channel characteristic. Summary of the invention

[0004] The present invention provides a key generation method and system based on phase reciprocity compensation of vehicle networking channels, which are used to solve the defect in the prior art that the reciprocity of channel characteristics of both communicating parties is low, resulting in a high key inconsistency rate.

[0005] The present invention provides a key generation method based on phase reciprocity compensation of a vehicle networking channel, which is applicable to a sender, and includes: Obtaining first channel state information of a sender in a vehicle networking channel based on channel detection; Preprocessing the first channel state information to obtain real and imaginary features of the first channel state information; Inputting the real feature and the imaginary feature into a pre-trained network model respectively, the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, the phase feature is determined based on the real feature and the imaginary feature; A communication key of the sender is generated according to the compensated first channel state information.

[0006] According to a key generation method based on phase reciprocity compensation of a vehicle networking channel provided by the present invention, generating a communication key of a sender according to the compensated first channel state information includes: Performing parameter estimation on the compensated first channel state information to obtain a quantization reference point of the compensated first channel state information; The compensated first channel state information is angle quantized according to the quantization reference point to obtain a communication key of the sender.

[0007] According to a key generation method based on phase reciprocity compensation of a vehicle networking channel provided by the present invention, the parameter estimation of the compensated first channel state information to obtain a quantized reference point of the compensated first channel state information includes: Parameter estimation is performed on the compensated first channel state information based on a nonlinear least squares optimization algorithm to obtain a quantization reference point of the compensated first channel state information.

[0008] According to a key generation method based on phase reciprocity compensation of a vehicle networking channel provided by the present invention, the compensated first channel state information is angle quantized according to the quantization reference point to obtain a communication key of the sender, including: Determining a quantization phase according to the quantization reference point; The complex plane corresponding to the compensated first channel state information is divided into at least one angle quantization interval according to the quantization phase, the channel state data of each angle quantization interval is encoded, and all the encoded channel state data are used as the communication key of the sender.

[0009] According to a key generation method based on phase reciprocity compensation of a vehicle networking channel provided by the present invention, the training step of the network model includes: Sending a first signal frame having multiple demodulation reference symbols to a receiving party, and acquiring a second signal frame having multiple demodulation reference symbols sent by the receiving party, wherein the first signal frame is used for the receiving party to determine a second channel state information sample; Determine a first channel state information sample according to the second signal frame; Taking the first channel state information sample as an input variable, and taking the second channel state information sample or any column of data of the second channel state information sample as an output variable; Preprocessing the input variable and the output variable respectively, and obtaining a training sample according to a first real feature and a first imaginary feature corresponding to the input variable and a second real feature and a second imaginary feature corresponding to the output variable; The neural network is trained according to the training samples to obtain the network model.

[0010] According to a key generation method based on phase reciprocity compensation of vehicle network channels provided by the present invention, the neural network includes a feature learning module, and the feature learning module is composed of a composite layer of the same network structure, and the composite layer includes 1 BiLSTM layer, 1 Dropout layer and 1 fully connected layer, which is used to learn the change characteristics between the first real feature and the second real feature and the change characteristics between the first imaginary feature and the second imaginary feature according to the first real feature, the first imaginary feature, the second real feature and the second imaginary feature.

[0011] According to a key generation method based on phase reciprocity compensation of vehicle network channels provided by the present invention, the first signal frame and the second signal frame adopt a physical direct link shared channel signal, and the demodulation reference symbol is a symbol at a preset position in the physical direct link shared channel signal.

[0012] According to a key generation method based on phase reciprocity compensation of vehicle networking channels provided by the present invention, the first channel state information, the first channel state information sample and the second channel state information sample are obtained by adopting the same channel estimation method.

[0013] According to a key generation method based on phase reciprocity compensation of vehicle network channels provided by the present invention, the preprocessing includes smoothing filtering, downsampling, real number conversion, normalization and vector rearrangement; wherein the step of vector rearrangement includes: taking the modulus of the downsampled information length according to the subscript of the real data in the vector to obtain a modulus result; arranging the modulus results in ascending order, and grouping the values ​​corresponding to the subscripts with the same results into one group during the arrangement process.

[0014] The present invention also provides a key generation system based on phase reciprocity compensation of vehicle networking channels, comprising: A channel detection module, used to obtain first channel state information of a sender in a vehicle networking channel based on channel detection; a preprocessing module, configured to preprocess the first channel state information to obtain a real feature and an imaginary feature of the first channel state information; a phase compensation module, configured to input the real feature and the imaginary feature into a pre-trained network model respectively, wherein the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, wherein the phase feature is determined based on the real feature and the imaginary feature; A key generation module is used to generate a communication key of the sender according to the compensated first channel state information.

[0015] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the processor implements a key generation method based on phase reciprocity compensation of a vehicle networking channel as described in any one of the above-mentioned methods.

[0016] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for generating a key based on phase reciprocity compensation of a vehicle networking channel as described in any one of the above is implemented.

[0017] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the above-mentioned key generation methods based on phase reciprocity compensation of vehicle network channels.

[0018] The key generation method and system based on phase reciprocity compensation of vehicle network channels provided by the present invention compensate the real and imaginary characteristics of the first channel state information of the sender through a pre-trained network model to achieve phase characteristic compensation of the first channel state information, reduce the influence of Doppler frequency shift on the phase caused by the movement of the communicating parties, thereby improving the reciprocity of the channel characteristics of the communicating parties and reducing the inconsistency rate of communication keys. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 It is a flow chart of a key generation method based on phase reciprocity compensation of an Internet of Vehicles channel provided by the present invention.

[0021] Figure 2 It is a flowchart of the key generation step provided by the present invention.

[0022] Figure 3 It is a structural schematic diagram of a key generation system based on phase reciprocity compensation of vehicle networking channels provided by the present invention.

[0023] Figure 4 It is a schematic diagram of the process of generating a key by a key generation system based on phase reciprocity compensation of an Internet of Vehicles channel provided by the present invention.

[0024] Figure 5 It is a statistical histogram of correlation coefficients between the frequency domain CSI of the receiver and the frequency domain CSI of the transmitter before compensation provided by the present invention.

[0025] Figure 6 It is a statistical histogram of correlation coefficients between the frequency domain CSI of the receiver and the frequency domain CSI of the transmitter after compensation provided by the present invention.

[0026] Figure 7 It is a schematic diagram of quantization angle interval coding of third-order angle quantization provided by the present invention.

[0027] Figure 8 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Figure 1 is a flow chart of a key generation method based on phase reciprocity compensation of an Internet of Vehicles channel provided by the present invention, such as Figure 1 As shown, the present invention provides a key generation method based on phase reciprocity compensation of vehicle network channels, which is applicable to the sender, the sender can be represented as Alice, and the receiver can be represented as Bob. The sender and the receiver are both legal communication vehicles, and encrypted communication is performed in a direct manner. In order to improve the reciprocity of channel characteristics between the communicating parties and reduce the inconsistency rate of communication keys, the present invention proposes a method for compensating the sender, which specifically includes the following steps.

[0030] Step S110, based on channel detection, first channel state information of the sender in the IoV channel is obtained, and the first channel state information is preprocessed. Channel state information (CSI) is used to characterize the knowledge of the current state and characteristics of the channel, including the gain, fading, noise level and interference of the channel.

[0031] Step S120: preprocess the first channel state information to obtain real and imaginary features of the first channel state information.

[0032] No. i The CSI can be expressed as a complex number ,in, Indicates i The complex value of the CSI, Indicates i The complex real part of the CSI, Indicates i The complex imaginary part of the CSI, j Represents an imaginary unit.

[0033] Step S130, respectively inputting the real feature and the imaginary feature into a pre-trained network model, wherein the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, wherein the phase feature is determined based on the real feature and the imaginary feature.

[0034] The network model is obtained by training a neural network using a first channel state information sample of a sender and a second channel state information sample of a receiver. Specifically, the first channel state information sample is an input variable, and the second channel state information sample is an output variable.

[0035] Step S140: Generate a communication key of the sender according to the compensated first channel state information.

[0036] It can be understood that the present invention compensates for the real and imaginary characteristics of the first channel state information of the sender through a pre-trained network model to achieve phase characteristic compensation for the first channel state information, thereby reducing the impact of Doppler frequency shift on the phase caused by the movement of the communicating parties, thereby improving the reciprocity of the channel characteristics of the communicating parties and reducing the inconsistency rate of communication keys.

[0037] Based on the above embodiment, as an optional embodiment, the sender and the receiver may use the Physical Sidelink Shared Channel (PSSCH) signal frame related to the direct communication technology Sidelink in TS36.213 for message transmission. At the same time, the communicating parties extract the demodulation reference signal (DMRS) in the PSSCH signal frame for channel estimation to obtain the channel CSI at the current moment.

[0038] Optionally, a PSSCH signal frame contains 14 single carrier frequency division multiple access (SC-FDMA) symbols. Given the high speed of vehicle movement, the 2nd, 5th, 8th, and 11th symbols (starting from 0) in the SC-FDMA symbols in each PSSCH signal frame can be set as DMRS signals based on LTE-V2X technology. LTE-V2X is a communication standard based on LTE technology, designed to support communication between vehicles and other vehicles, infrastructure, pedestrians, and networks.

[0039] The sender sends a PSSCH signal frame with multiple DMRS signals to the receiver, and the receiver responds within the coherence time. After receiving the PSSCH signal frame, the sender demodulates it and extracts the DMRS signal from the 2nd, 5th, 8th, and 11th symbols, and uses the least squares (LS) channel estimation method or the minimum mean square error (MMSE) channel estimation method to obtain the frequency domain channel state information CSI at 4 corresponding times.

[0040] It can be understood that the present invention transmits messages through PSSCH signal frames, and setting the 2nd, 5th, 8th and 11th symbols in the SC-FDMA symbols as DMRS signals can provide a reliable reference so that the receivers of the communicating parties can accurately estimate the channel status.

[0041] Based on the above embodiment, as an optional embodiment, preprocessing the first channel state information includes smoothing filtering, downsampling, real number conversion, normalization and vector rearrangement. The preprocessing can improve the compensation accuracy.

[0042] The real number calculation formula is as follows:

[0043] in, represents the result of real numbering, represents the result of downsampling, and Represents the real and imaginary data respectively.

[0044] The normalization method used is the maximum and minimum value normalization method.

[0045] The step of rearranging the vector includes: taking a modulus of the length of the downsampled CSI information according to the subscript of the real data in the vector to obtain a modulus result; arranging the modulus results in ascending order, and grouping the values ​​corresponding to the subscripts with the same results into one group during the arrangement process.

[0046] The network model can be written as , based on the trained network model, the first channel state information after preprocessing is compensated, and the process of predicting the channel state information of the receiver can be described by the following formula:

[0047] in, and They represent the CSI value obtained by Alice and the predicted CSI value respectively. Represents the set of all training hyperparameters in the network model.

[0048] It can be understood that the present invention facilitates the extraction of real and imaginary features of the first channel state information and improves the accuracy of the real and imaginary features by preprocessing the first channel state information such as smoothing filtering, downsampling, real number conversion, normalization, and vector rearrangement.

[0049] Based on the above embodiment, as an optional embodiment, the training step of the network model includes the following steps.

[0050] Step S010: sending a first signal frame having multiple demodulation reference symbols to a receiver, and acquiring a second signal frame having multiple demodulation reference symbols sent by the receiver, wherein the first signal frame is used for the receiver to determine a second channel state information sample.

[0051] Optionally, the first signal frame and the second signal frame use a physical direct link shared channel signal, and the demodulation reference symbol is a symbol at a preset position in the physical direct link shared channel signal. The first signal frame and the second signal frame having multiple demodulation reference symbols can be PSSCH signals related to the direct communication technology Sidelink in TS 36.213, a PSSCH signal frame contains 14 SC-FDMA symbols, and the 2nd, 5th, 8th, and 11th symbols of the SC-FDMA symbol are used to carry demodulation reference symbols, or it can be a signal frame with a similar structure designed by the sender Alice and the receiver Bob.

[0052] The receiver Bob extracts the DMRS from the first signal frame to perform frequency domain channel estimation, and obtains frequency domain channel state information CSI, ie, second channel state information samples, in different scenarios.

[0053] Step S020: Determine a first channel state information sample according to the second signal frame. The sender Alice extracts DMRS from the second signal frame to perform frequency domain channel estimation, and obtains frequency domain channel state information CSI, ie, first channel state information samples, in different scenarios.

[0054] It should be noted that the first channel state information, the first channel state information sample and the second channel state information sample are obtained using the same channel estimation method, and can use the least squares method LS channel estimation, or the minimum mean square error method MMSE channel estimation to ensure consistency.

[0055] Step S030, taking the first channel state information sample as an input variable, and taking the second channel state information sample or any column data of the second channel state information sample as an output variable. Specifically, taking all CSI of the sender Alice as the input variable of the network model, and taking the first column CSI of the receiver Bob as the target variable, or taking CSI data of other columns.

[0056] Step S040, preprocessing the input variable and the output variable respectively, and obtaining training samples according to the first real feature and the first imaginary feature corresponding to the input variable and the second real feature and the second imaginary feature corresponding to the output variable.

[0057] Optionally, preprocessing the input variable and the output variable includes smoothing filtering, downsampling, real numbering, normalization, and vector rearrangement; wherein the step of vector rearrangement includes: taking a modulus of the downsampled information length according to the subscript of the real data in the vector to obtain a modulus result; arranging the modulus results in ascending order, and grouping the values ​​corresponding to the subscripts with the same results into one group during the arrangement process. The preprocessing of the input variable and the output variable is the same as the preprocessing principle of the first channel state information, which will not be repeated.

[0058] Step S050, training the neural network according to the training samples to obtain the network model. During training, in order to have better performance and interpretability, the loss function can use mean square error (MSE) or a custom function.

[0059] It can be understood that the present invention provides a training scheme for the network model. According to the changing characteristics of the phase of the channel characteristics caused by the movement of the communicating parties in the Internet of Vehicles channel, the changing laws of the channel characteristics are learned through machine learning, so that the communicating parties in a mobile state can obtain channel characteristics with higher reciprocity.

[0060] On the basis of the above embodiments, as an optional embodiment, the neural network includes an input data processing module, a feature learning module and an output data integration module, the input data processing module includes 1 sequence input layer, 1 convolution layer and 1 average pooling layer; the feature learning of the real part and the imaginary part in the feature learning module is composed of a composite layer with the same network structure, and the composite layer includes 1 BiLSTM layer, 1 Dropout layer and 1 fully connected layer, which is used to learn the change characteristics between the first real feature and the second real feature and the change characteristics between the first imaginary feature and the second imaginary feature according to the first real feature, the first imaginary feature, the second real feature and the second imaginary feature; the output data integration module includes 1 deep connection layer, 1 composite layer and 1 regression layer.

[0061] It can be understood that the present invention provides a specific construction scheme of the neural network, which consists of a feature learning module composed of several composite layers to learn the change characteristics between the first real feature and the second real feature and the change characteristics between the first imaginary feature and the second imaginary feature. The first real feature and the first imaginary feature can be used to determine the phase change law of the sender during the movement, and the second real feature and the second imaginary feature can be used to determine the phase change law of the receiver during the movement, thereby being able to learn the change characteristics and change laws of the channel characteristic phase during the movement of the sender and the receiver.

[0062] Figure 2 It is a flowchart of the key generation step provided by the present invention, such as Figure 2 As shown, based on the above embodiment, as an optional embodiment, generating the communication key of the sender according to the compensated first channel state information includes the following steps.

[0063] Step S141, perform parameter estimation on the compensated first channel state information to obtain a quantization reference point of the compensated first channel state information. Parameter estimation can use a least squares method LS channel estimation, or use a minimum mean square error method MMSE channel estimation. The quantization reference point refers to the complex real part and imaginary part of the CSI data in the compensated first channel state information that minimizes the objective function.

[0064] Step S142, performing angle quantization on the compensated first channel state information according to the quantization reference point to obtain the communication key of the sender. Angle quantization refers to dividing the complex plane into multiple angle intervals according to the phase determined by the quantization reference point, and then encoding the CSI data in the angle interval based on a preset encoding method (such as Gray code) to obtain the communication key of the sender.

[0065] It can be understood that the present invention uses an angle quantization method to generate the sender's communication key based on phase compensation. The phase can be easily converted into an angle, which facilitates the implementation of angle quantization and reduces computational complexity.

[0066] Based on the above embodiment, as an optional embodiment, the performing parameter estimation on the compensated first channel state information to obtain a quantization reference point of the compensated first channel state information includes: Parameter estimation is performed on the compensated first channel state information based on a nonlinear least squares optimization algorithm to obtain a quantization reference point of the compensated first channel state information.

[0067] Specifically, the Levenberg-Marquardt algorithm can be used to solve the parameter estimation problem, using the CSI initial value The objective function when solving is:

[0068] Among them, each CSI data is a complex number , , as well as Represent the mean of the real part, imaginary part and modulus length of the complex number, and minimize the objective function. and as a quantitative benchmark.

[0069] It is understandable that the present invention uses a nonlinear least squares optimization algorithm to perform parameter estimation, which can improve the accuracy of the quantization reference point based on the high-precision characteristics of the algorithm itself.

[0070] Based on the above embodiment, as an optional embodiment, the method of performing angle quantization on the compensated first channel state information according to the quantization reference point to obtain the communication key of the sender includes the following steps.

[0071] Step S1421, determine the quantization phase according to the quantization reference point. The calculation formula of the quantization phase is as follows:

[0072] in, is the quantized phase.

[0073] Step S1422: divide the complex plane corresponding to the compensated first channel state information into at least one angle quantization interval according to the quantization phase, encode the channel state data of each angle quantization interval, and use all the encoded channel state data as the communication key of the sender.

[0074] Optionally, the compensated first channel state information is subjected to third-order angle quantization, and the complex plane is divided into 8 uniform sector intervals, namely, angle quantization intervals. The CSI data in each angle quantization interval is quantized into 3 bits respectively, and all the quantized CSI data are combined to obtain the sender's communication key.

[0075] It can be understood that the present invention can improve the randomness of the communication key by determining multiple angle quantization intervals.

[0076] The key generation system based on phase reciprocity compensation of vehicle networking channel provided by the present invention is described below. The key generation system based on phase reciprocity compensation of vehicle networking channel described below and the key generation method based on phase reciprocity compensation of vehicle networking channel described above can be referenced to each other.

[0077] Figure 3 is a structural schematic diagram of a key generation system based on phase reciprocity compensation of an Internet of Vehicles channel provided by the present invention, such as Figure 3 As shown, the present invention also provides a key generation system based on phase reciprocity compensation of vehicle network channels, including a sender and a receiver, and the sender includes the following modules.

[0078] The channel detection module 310 is used to obtain the first channel state information of the sender in the Internet of Vehicles channel based on the channel detection.

[0079] The preprocessing module 320 is used to preprocess the first channel state information to obtain real and imaginary features of the first channel state information.

[0080] The phase compensation module 330 is used to input the real feature and the imaginary feature into a pre-trained network model respectively, and the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, and the phase feature is determined based on the real feature and the imaginary feature.

[0081] The key generation module 340 is used to generate a communication key of the sender according to the compensated first channel state information.

[0082] As an optional embodiment, the present invention provides a key generation system based on phase reciprocity compensation of vehicle networking channels, and the receiving side includes the following modules.

[0083] A receiving module is used to obtain second channel state information of a receiver in a vehicle networking channel based on channel detection, and preprocess the second channel state information; perform parameter estimation on the preprocessed second channel state information to obtain a quantization reference point for the preprocessed second channel state information; and perform angle quantization on the preprocessed second channel state information according to the quantization reference point to obtain a communication key for the receiver.

[0084] The key generation system based on phase reciprocity compensation of vehicle networking channels provided by the present invention has the same technical effect as the key generation method based on phase reciprocity compensation of vehicle networking channels, and will not be repeated here.

[0085] Figure 4 : is a schematic diagram of a process of generating a key by a key generation system based on phase reciprocity compensation of a vehicle networking channel provided by the present invention, such as Figure 4 As shown, the following steps are included.

[0086] (1) During training, the sender Alice sends a signal frame with multiple demodulation reference signals (DMRS) to the receiving terminal Bob based on the transmitting terminal.

[0087] (2) The receiving terminal extracts the DMRS to perform frequency domain channel estimation and obtains the frequency domain channel state information CSI in different scenarios. The sending terminal on Alice’s side is similar.

[0088] (3) The CSI of the sending terminal on Alice’s side is used as the input variable of the network model, and the CSI of the receiving terminal on Bob’s side is used as the target variable of the model.

[0089] (4) Perform preprocessing operations on CSI to obtain training samples, including smoothing filtering, downsampling, real number conversion, normalization, and vector rearrangement.

[0090] (5) Input the samples into the constructed network model for training.

[0091] (6) During compensation, the transmitting terminal Alice and the receiving terminal Bob obtain the original channel CSI through channel detection and then perform the CSI preprocessing operation mentioned above.

[0092] (7) The sending terminal Alice uses the trained network model to compensate for the channel characteristics, while the receiving terminal Bob does not need to compensate.

[0093] (8) The transmitting terminal Alice and the receiving terminal Bob respectively use parameter estimation on the compensated channel CSI characteristics to obtain appropriate quantization reference points.

[0094] (9) The communicating parties use the reference point to quantize the angle. During quantization, Gray code is used to encode different quantization intervals to obtain the final communication key.

[0095] Figure 5 is a statistical histogram of correlation coefficients between the frequency domain CSI of the receiver and the frequency domain CSI of the transmitter before compensation provided by the present invention, Figure 6 It is a statistical histogram of the correlation coefficient between the frequency domain CSI of the receiver and the frequency domain CSI of the sender after compensation provided by the present invention. In the scenario where the signal-to-noise ratio is 30dB and the relative speed of the two communicating parties is 30km / h, the correlation coefficients of the channel CSIs of 1500 senders and the channel CSIs of the receiver are statistically analyzed after compensation by the network model. The statistical histogram of the correlation coefficients before compensation is as follows: Figure 5 As shown, the statistical histogram of the correlation coefficient after compensation is as follows Figure 6 As shown, it can be seen that compensation can improve the reciprocity of channel characteristics of both communicating parties.

[0096] Figure 7 It is a schematic diagram of quantization angle interval coding of third-order angle quantization provided by the present invention. By adopting third-order quantization, the complex plane is divided into 8 uniform sector-shaped intervals, so that the quantized key has higher randomness.

[0097] Figure 8 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 8 As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830 and a communication bus 840, wherein the processor 810, the communication interface 820 and the memory 830 communicate with each other through the communication bus 840. The processor 810 may call the logic instructions in the memory 830 to execute the key generation method based on the phase reciprocity compensation of the vehicle network channel, the method comprising: obtaining the first channel state information of the sender in the vehicle network channel based on channel detection; preprocessing the first channel state information to obtain the real feature and the imaginary feature of the first channel state information; respectively inputting the real feature and the imaginary feature into the pre-trained network model, the network model compensates the real feature and the imaginary feature to realize the phase feature compensation of the first channel state information, the phase feature is determined based on the real feature and the imaginary feature; and generating the communication key of the sender according to the compensated first channel state information.

[0098] In addition, the logic instructions in the above-mentioned memory 830 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0099] On the other hand, the present invention also provides a computer program product, which includes a computer program, and the computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the key generation method based on the phase reciprocity compensation of the vehicle network channel provided by the above methods, and the method includes: obtaining the first channel state information of the sender in the vehicle network channel based on channel detection; preprocessing the first channel state information to obtain the real feature and imaginary feature of the first channel state information; respectively inputting the real feature and the imaginary feature into a pre-trained network model, and the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, and the phase feature is determined based on the real feature and the imaginary feature; based on the compensated first channel state information, generating the sender's communication key.

[0100] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the key generation method based on phase reciprocity compensation of the vehicle network channel provided by the above methods is implemented, and the method includes: obtaining the first channel state information of the sender in the vehicle network channel based on channel detection; preprocessing the first channel state information to obtain the real and imaginary features of the first channel state information; respectively inputting the real and imaginary features into a pre-trained network model, and the network model compensates the real and imaginary features to achieve phase feature compensation of the first channel state information, and the phase feature is determined based on the real and imaginary features; and generating the communication key of the sender according to the compensated first channel state information. .

[0101] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0102] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A key generation method based on phase reciprocity compensation of vehicle networking channels, characterized in that: Applicable to senders, including: Obtaining first channel state information of a sender in a vehicle networking channel based on channel detection; Preprocessing the first channel state information to obtain real and imaginary features of the first channel state information; Inputting the real feature and the imaginary feature into a pre-trained network model respectively, the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, the phase feature is determined based on the real feature and the imaginary feature; A communication key of the sender is generated according to the compensated first channel state information.

2. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 1 is characterized in that: The generating a communication key of the sender according to the compensated first channel state information includes: Performing parameter estimation on the compensated first channel state information to obtain a quantization reference point of the compensated first channel state information; The compensated first channel state information is angle quantized according to the quantization reference point to obtain a communication key of the sender.

3. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 2 is characterized in that: The performing parameter estimation on the compensated first channel state information to obtain a quantization reference point of the compensated first channel state information includes: Parameter estimation is performed on the compensated first channel state information based on a nonlinear least squares optimization algorithm to obtain a quantization reference point of the compensated first channel state information.

4. The phase reciprocity compensation based on the Internet of Vehicles channel according to claim 2 is characterized in that: The step of performing angle quantization on the compensated first channel state information according to the quantization reference point to obtain a communication key of the sender includes: Determining a quantization phase according to the quantization reference point; The complex plane corresponding to the compensated first channel state information is divided into at least one angle quantization interval according to the quantization phase, the channel state data of each angle quantization interval is encoded, and all the encoded channel state data are used as the communication key of the sender.

5. The key generation method based on phase reciprocity compensation of vehicle networking channel according to any one of claims 1 to 4, characterized in that: The training steps of the network model include: Sending a first signal frame having multiple demodulation reference symbols to a receiving party, and acquiring a second signal frame having multiple demodulation reference symbols sent by the receiving party, wherein the first signal frame is used for the receiving party to determine a second channel state information sample; Determine a first channel state information sample according to the second signal frame; Taking the first channel state information sample as an input variable, and taking the second channel state information sample or any column data of the second channel state information sample as an output variable; Preprocessing the input variable and the output variable respectively, and obtaining a training sample according to a first real feature and a first imaginary feature corresponding to the input variable and a second real feature and a second imaginary feature corresponding to the output variable; The neural network is trained according to the training samples to obtain the network model.

6. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 5 is characterized in that: The neural network includes a feature learning module, which is composed of composite layers with the same network structure. The composite layers include 1 BiLSTM layer, 1 Dropout layer and 1 fully connected layer, and is used to learn the change characteristics between the first real feature and the second real feature and the change characteristics between the first imaginary feature and the second imaginary feature according to the first real feature, the first imaginary feature, the second real feature and the second imaginary feature.

7. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 5 is characterized in that: The first signal frame and the second signal frame use a physical direct link shared channel signal, and the demodulation reference symbol is a symbol at a preset position in the physical direct link shared channel signal.

8. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 5 is characterized in that: The first channel state information, the first channel state information sample and the second channel state information sample are obtained by using the same channel estimation method.

9. The key generation method based on phase reciprocity compensation of vehicle networking channel according to claim 5 is characterized in that: The preprocessing includes smoothing filtering, downsampling, real number conversion, normalization and vector rearrangement; wherein the step of vector rearrangement includes: taking the modulus of the downsampled information length according to the subscript of the real data in the vector to obtain the modulus result; arranging the modulus results in ascending order, and grouping the values ​​corresponding to the subscripts with the same results into one group during the arrangement process.

10. A key generation system based on phase reciprocity compensation of vehicle networking channels, characterized in that: include: A channel detection module, used to obtain first channel state information of a sender in a vehicle networking channel based on channel detection; a preprocessing module, configured to preprocess the first channel state information to obtain a real feature and an imaginary feature of the first channel state information; a phase compensation module, configured to input the real feature and the imaginary feature into a pre-trained network model respectively, wherein the network model compensates the real feature and the imaginary feature to achieve phase feature compensation of the first channel state information, wherein the phase feature is determined based on the real feature and the imaginary feature; A key generation module is used to generate a communication key of the sender according to the compensated first channel state information.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, it implements the key generation method based on phase reciprocity compensation of the Internet of Vehicles channel as described in any one of claims 1-9.

12. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the key generation method based on phase reciprocity compensation of the Internet of Vehicles channel as described in any one of claims 1 to 9 is implemented.

13. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the key generation method based on phase reciprocity compensation of the Internet of Vehicles channel as described in any one of claims 1 to 9 is implemented.

Images