A biometric-based quantum-resistant data encryption system and method

Abstract

The application discloses a kind of quantum resistance data encryption systems and methods based on biological characteristics, including the real-time acquisition of multiple biological characteristics of user and environmental data by data acquisition module, after filtering by pre-processing module, the optimal fusion of multidimensional biological information is realized using multimodal fusion module, and then the biological characteristic template is generated by the feature extraction module dynamically adjusting extraction strategy.Key generation module generates dynamic key in real time according to the template, and the original data is quantum resistance encrypted by data encryption module Fragmentation.Reorganization module completes data restoration after verifying the consistency of biological characteristics and key.The present application fully integrates multi-source biological and environmental information, realizes highly personalized, dynamic data encryption and security reorganization, effectively improves the data confidentiality and attack resistance in the context of quantum computing, and is suitable for high security data protection scenarios.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data encryption, and in particular to a quantum-resistant data encryption system and method based on biological characteristics. BACKGROUND

[0002] With the rapid development of information technology, data transmission and storage in the fields of Internet, Internet of Things, mobile communication, etc. are increasingly frequent, and data security problems are increasingly prominent. Traditional data encryption methods mainly rely on symmetric encryption or asymmetric encryption algorithms such as AES, RSA, etc. However, these encryption algorithms still have strong security under current computing power, but with the continuous advancement of quantum computing technology, their cracking efficiency will be significantly improved, making existing encryption algorithms face the risk of being broken.

[0003] In order to cope with the threat of quantum computing, researchers have proposed quantum-resistant cryptographic systems based on lattice cryptography, etc., and tried to apply them to actual data encryption scenarios. However, the current quantum-resistant encryption technology still has problems such as complex key management, poor user experience, and lack of flexibility. In addition, a single-dimensional key generation mechanism (such as relying only on a single factor such as fingerprint, iris, or password) is vulnerable to attacks such as forgery, counterfeiting, and static leakage, and cannot meet the dynamic authentication and encryption needs in high-security scenarios.

[0004] In recent years, biometric recognition technology has developed rapidly, and its application in identity recognition and access control is increasingly widespread, with advantages such as uniqueness, stability, and difficulty in counterfeiting. However, most existing schemes only use a single biometric feature as the basis for identity authentication, and do not combine multi-modal biometric features and environmental factors for collaborative analysis, making it difficult to achieve higher-dimensional identity security protection.

[0005] Therefore, there is an urgent need for an encryption system that integrates multiple biometric features and environmental perception information, has a dynamic key generation mechanism, and is quantum-resistant, to improve the security and practicality of data encryption and meet the data security needs in the future quantum era. SUMMARY

[0006] In view of the deficiencies in the prior art, the purpose of the present application is to provide a quantum-resistant data encryption system and method based on biological characteristics, for improving the security and practicality of data encryption.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solution: a quantum-resistant data encryption system based on biological characteristics, comprising:

[0008] a data acquisition module for acquiring multiple biometric data of a user and multiple environmental data in the environment where the user is located in real time;

[0009] The preprocessing module is connected with the data acquisition module, and is configured to perform data preprocessing on each of the biological feature data and the environmental data according to a preset preprocessing algorithm, to obtain filtered biological data and filtered environmental data, the filtered biological data including an iris image, a fingerprint image, a heart rate and blood oxygen, and the filtered environmental data including illumination intensity, temperature, humidity and background noise.

[0010] The multi-modal fusion module is connected with the preprocessing module, and is configured to perform data fusion on the filtered biological data, to find an optimal matching relationship between the filtered biological data based on a preset optimization algorithm, and to obtain multi-modal fusion biological data through the data fusion.

[0011] The feature extraction module is connected with the multi-modal fusion module, and is configured to perform feature extraction on the multi-modal fusion biological data according to a preset feature extraction algorithm, and to dynamically adjust parameters of a feature extractor in the feature extraction algorithm according to the filtered environmental data during the feature extraction, to obtain a biological feature template.

[0012] The key generation module is connected with the feature extraction module, and is configured to generate a biological dynamic key in real time according to the biological feature template.

[0013] The data encryption module is connected with the key generation module, and is configured to dynamically divide original data into a plurality of data fragments according to the biological dynamic key, and to perform data encryption on each of the data fragments by using a quantum-resistant algorithm based on lattice cryptography, to obtain encrypted fragments.

[0014] The fragment recombination module is connected with the data encryption module, and is configured to provide a quantum-resistant fragment recombination strategy, so that the user restores the original data from the encrypted fragments when providing correct biological feature data and the dynamic key.

[0015] Further, the preprocessing module comprises:

[0016] The adaptive filtering unit is configured to suppress random noise in each of the biological feature data according to an adaptive median filtering algorithm to obtain first filtered data.

[0017] The multi-resolution analysis unit is configured to decompose an image in each of the biological feature data according to a multi-resolution method, and to restore detailed information in the image according to a threshold denoising technique to obtain second filtered data.

[0018] The deep denoising unit is configured to perform image denoising and enhancement processing on the image in each of the biological feature data according to a pre-trained deep learning algorithm to obtain third filtered data, and the first filtered data, the second filtered data and the third filtered data constitute the filtered biological data.

[0019] Further, the multi-modal fusion module comprises:

[0020] a feature alignment unit, configured to extract iris vector features and fingerprint vector features of the iris image and the fingerprint image according to a ResNet algorithm, and extract heart rate time sequence features and blood oxygen time sequence features of the heart rate and the blood oxygen according to an LSTM network algorithm, and further unify the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features and the fingerprint vector features to the same latitude according to a principal component analysis algorithm;

[0021] a feature standardization unit, connected to the feature alignment unit, configured to perform normalization processing on the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features and the fingerprint vector features according to a Z-score standardization method to obtain a plurality of normalized features, the normalized features comprising heart rate normalized features, blood oxygen normalized features, iris normalized features and fingerprint normalized features;

[0022] a fusion unit, connected to the feature standardization unit, configured to configure an initial weight for each of the normalized features, introduce an attention mechanism to perceive feature qualities of each of the normalized features, and dynamically adjust the initial weight corresponding to each of the normalized features according to the feature quality to obtain an adjusted weight, and finally calculate the multi-modal fusion biological data according to each of the normalized features and the corresponding adjusted weight.

[0023] Further, the calculation formula of the multi-modal fusion biological data is configured as:

[0024]

[0025] wherein, denotes the multi-modal fusion biological data, denotes the normalized heart rate feature, denotes the normalized blood oxygen feature, denotes the normalized iris feature, denotes the normalized fingerprint feature, denotes the adjusted weight of the heart rate feature, denotes the adjusted weight of the blood oxygen feature, denotes the adjusted weight of the iris feature, denotes the adjusted weight of the fingerprint feature, denotes a preset offset constant, denotes a Gamma function, denotes an error function, denotes a first kind of Bessel function, denotes a natural logarithm function, denotes an upper limit of summation, denotes a summation index variable, denotes a summation denominator index parameter.

[0026] Further, the multi-modal fusion module further comprises an optimization matching unit connected to the fusion unit, configured to search for a feature combination with the strongest correlation between the normalized features according to a maximum mutual information algorithm as the optimal matching relationship, and adjust a calculation formula of the multi-modal fusion biological data according to the normalized features in the optimal matching relationship and the corresponding adjustment weights, to obtain the multi-modal fusion biological data after optimization.

[0027] Further, the feature extraction module comprises:

[0028] a feature extraction unit configured to introduce a multi-layer residual convolutional network, and extract features from the multi-modal fusion biological data according to the multi-layer residual convolutional network to obtain an intermediate feature tensor;

[0029] a score modeling unit configured to construct an environmental score model based on a shallow neural network, and input the illumination intensity, the temperature, the humidity and the background noise into the environmental score model to obtain a parameter adjustment factor;

[0030] a parameter adjustment unit connected to the score modeling unit, configured to dynamically adjust a convolution kernel coefficient of a feature extractor in the feature extraction algorithm according to the parameter adjustment factor to obtain an optimized extraction algorithm model, and the optimized extraction algorithm model is configured to output an optimized feature tensor;

[0031] a template generation unit connected to the parameter adjustment unit and the feature extraction unit, configured to generate the biological feature template according to the optimized feature tensor.

[0032] Further, the calculation formula of the parameter adjustment factor is configured as:

[0033]

[0034] wherein, denotes the parameter adjustment factor, and denote two weight matrices of the environmental score model, respectively, and denote two bias terms of the environmental score model, respectively, denote the illumination intensity, the temperature, the humidity and the background noise, respectively, denotes an activation function, denotes a hyperbolic tangent function;

[0035] the convolution kernel coefficient adjustment formula is configured as:

[0036]

[0037] wherein, denotes the adjusted convolution kernel coefficient, denotes the convolution kernel parameter before adjustment, denotes the number of parameters to be adjusted, denotes the first parameter adjustment factor.

[0038] Further, the filtered environmental data further comprises a sensor angle, a physical occlusion amount and a motion jitter amount, and the data encryption module comprises:

[0039] a disturbance generation unit, configured to generate a disturbance vector based on the sensor angle, the physical occlusion amount and the motion jitter amount, the disturbance vector being used to control parameter setting of a data encryption process;

[0040] a function generation unit, connected to the disturbance generation unit, configured to generate a disturbance adjustment function according to the disturbance vector;

[0041] a data encryption unit, connected to the function generation unit, configured to dynamically adjust a number of fragments, an encryption round and a password parameter configuration in a data encryption process according to the disturbance encryption function.

[0042] Further, the disturbance adjustment function is configured as:

[0043]

[0044] wherein, denotes the disturbance adjustment function, denotes the disturbance vector, and denotes two adjustable initial constants, denotes the normalized sensor angle, the physical occlusion amount and the motion jitter amount, respectively, denotes three preset weight coefficients.

[0045] A quantum-resistant data encryption method based on biological characteristics, applied to the quantum-resistant data encryption system, comprising:

[0046] Step S1, the data acquisition module acquires a plurality of biological characteristic data of a user and a plurality of environmental data in an environment where the user is located in real time;

[0047] In step S2, the preprocessing module performs data preprocessing on each of the biological feature data and the environmental data according to a preset preprocessing algorithm to obtain filtered biological data and filtered environmental data, the filtered biological data including iris images, fingerprint images, heart rates and blood oxygen, and the filtered environmental data including illumination intensity, temperature, humidity and background noise.

[0048] In step S3, a multi-modal fusion module performs data fusion on the filtered biological data, finds an optimal matching relationship between the filtered biological data based on a preset optimization algorithm, and obtains multi-modal fusion biological data through fusion.

[0049] In step S4, a feature extraction module extracts features from the multi-modal fusion biological data according to a preset feature extraction algorithm, and dynamically adjusts parameters of a feature extractor in the feature extraction algorithm according to the filtered environmental data to obtain a biological feature template.

[0050] In step S5, a key generation module generates a biological dynamic key in real time according to the biological feature template, and a data encryption module dynamically divides original data into a plurality of data fragments according to the biological dynamic key, and encrypts each of the data fragments using a quantum-resistant algorithm based on lattice cryptography to obtain encrypted fragments.

[0051] In step S6, a fragment recombination module provides a quantum-resistant fragment recombination strategy, so that the user restores each of the encrypted fragments to the original data when providing correct biological feature data and the dynamic key.

[0052] The beneficial effects of the present application are as follows:

[0053] The present application dynamically generates a biological dynamic key with strong uniqueness and high timeliness by collecting multiple biological feature data of a user such as iris, fingerprint, heart rate and blood oxygen, and combining with external environmental data such as environmental illumination, temperature and humidity, noise, etc., which effectively prevents the key from being forged or replayed, and improves the overall security and reliability of the data encryption system.

[0054] Meanwhile, the present application uses an encryption algorithm based on lattice cryptography to encrypt the data fragments, which has a natural quantum computing attack resistance, solves the problem of being cracked in a quantum computing environment for traditional encryption algorithms, and significantly enhances the foresight and adaptability of the system.

[0055] The present application fuses multiple biological feature information of different dimensions, matches and integrates through an optimization algorithm, which not only improves the accuracy of feature recognition, but also enhances the robustness of the system to abnormal data or attack behavior, avoiding system paralysis due to single feature failure.

[0056] In addition, the application can adapt to different environmental conditions such as light, humidity and temperature by dynamically adjusting the parameters of the extractor in the feature extraction algorithm, effectively reducing the influence of external interference on the recognition result, and improving the adaptability and usability of the system in practical application.

[0057] The original data is divided into multiple data fragments and encrypted respectively, and only under the premise of providing correct biological characteristics and a key, restoration can be completed through the fragment recombination module, the anti-tampering and anti-leakage ability of the data in the transmission and storage process is enhanced, and the security of data transmission is improved. BRIEF DESCRIPTION OF DRAWINGS

[0058] Figure 1 is a structural schematic diagram of the quantum resistance data encryption system in the application;

[0059] Figure 2 is a structural schematic diagram of the preprocessing module in the application;

[0060] Figure 3 is a structural schematic diagram of the multi-modal fusion module in the application;

[0061] Figure 4 is a structural schematic diagram of the feature extraction module in the application;

[0062] Figure 5 is a structural schematic diagram of the data encryption module in the application;

[0063] Figure 6 is a step schematic diagram of the quantum resistance data encryption method in the application.

[0064] The drawings are as follows: 1, data acquisition module; 2, preprocessing module; 21, adaptive filtering unit; 22, multi-resolution analysis unit; 23, deep denoising unit; 3, multi-modal fusion module; 31, feature alignment unit; 32, feature standard unit; 33, fusion unit; 34, optimization matching unit; 4, feature extraction module; 41, feature extraction unit; 42, score modeling unit; 43, parameter adjustment unit; 44, template generation unit; 5, key generation module; 6, data encryption module; 61, disturbance generation unit; 62, function generation unit; 63, data encryption unit; 7, fragment recombination module. DETAILED DESCRIPTION

[0065] The application will be further described in detail below in combination with the drawings and examples. Wherein the same parts are represented by the same reference signs. It should be noted that the words "front", "back", "left", "right", "up" and "down" used in the following description refer to the directions in the drawings, and the words "bottom surface" and "top surface", "inner" and "outer" refer to the directions towards or away from the geometric center of a particular part.

[0066] Embodiment 1, Reference Figure 1 As a first embodiment of the present application, the embodiment provides a quantum-resistant data encryption system based on biometric features, which can improve the security and practicability of data encryption, comprising:

[0067] A data acquisition module 1 is configured to acquire multiple biometric data of a user and multiple environmental data in an environment where the user is located in real time.

[0068] A preprocessing module 2 is connected to the data acquisition module 1 and configured to perform data preprocessing on each of the biometric data and the environmental data according to a preset preprocessing algorithm to obtain filtered biometric data and filtered environmental data, wherein the filtered biometric data includes iris images, fingerprint images, heart rate and blood oxygen, and the filtered environmental data includes illumination intensity, temperature, humidity and background noise.

[0069] A multi-modal fusion module 3 is connected to the preprocessing module 2 and configured to perform data fusion according to each of the filtered biometric data, find an optimal matching relationship between each of the filtered biometric data based on a preset optimization algorithm during the fusion process, and obtain multi-modal fusion biometric data through fusion.

[0070] A feature extraction module 4 is connected to the multi-modal fusion module 3 and configured to perform feature extraction on the multi-modal fusion biometric data according to a preset feature extraction algorithm, and dynamically adjust parameters of a feature extractor in the feature extraction algorithm according to each of the filtered environmental data during the feature extraction process to obtain a biometric feature template.

[0071] A key generation module 5 is connected to the feature extraction module 4 and configured to generate a biometric dynamic key in real time according to the biometric feature template.

[0072] A data encryption module 6 is connected to the key generation module 5 and configured to dynamically divide original data into multiple data fragments according to the biometric dynamic key, and perform data encryption on each of the data fragments using a quantum-resistant algorithm based on lattice cryptography to obtain encrypted fragments.

[0073] A fragment recombination module 7 is connected to the data encryption module 6 and configured to provide a quantum-resistant fragment recombination strategy, so that the user restores each of the encrypted fragments to the original data when providing correct each of the biometric features and the dynamic key.

[0074] Specifically, in the embodiment, the data acquisition module 1 integrates the following sensors:

[0075] An iris recognition sensor: a near-infrared iris scanning module (such as IRIS ID iCAM7000) is used to acquire user iris images with high precision.

[0076] Fingerprint sensor: Real-time acquisition of fingerprint images using a capacitive fingerprint collector (e.g., FPC 1020).

[0077] Heart rate and blood oxygen sensor: Continuous measurement of user heart rate and blood oxygen saturation using a MAX30102 photoelectric blood volume pulse sensor.

[0078] Ambient light sensor: Detection of current light intensity (lux value) using a BH1750 digital light sensor.

[0079] Temperature and humidity sensor: Collection of environmental temperature and humidity using an SHT31 digital temperature and humidity sensor.

[0080] Noise sensor: Collection of background noise intensity (dB level) in the current environment using a MEMS microphone (e.g., INMP441).

[0081] Working principle of Example 1:

[0082] Data acquisition and preprocessing: Real-time acquisition of user physiological characteristics and environmental data by various sensors, and cleaning and enhancing the original data through preset algorithms such as image enhancement (CLAHE), filtering (adaptive median filtering), and denoising (deep learning algorithm), outputting high-quality filtered biological data and filtered environmental data.

[0083] Multi-modal fusion and feature extraction: Extracting features such as iris texture, fingerprint ridge direction, and pulse spectrum characteristics through fusion algorithms to form a unified biological feature vector; simultaneously introducing environmental light, temperature, humidity, and noise data as weight correction parameters to dynamically adjust the parameters of the feature extractor, making the extraction process environmentally robust.

[0084] Key generation and quantum-resistant encryption: Hash compression (e.g., using SHA3) based on the biological feature vector, combined with a timestamp or random salt value to generate a dynamic key. Subsequently, the original data is encrypted in fragments under the guidance of this key, and each fragment is encrypted through lattice cryptography algorithms (e.g., Kyber based on LWE), ensuring that even if part of the data is intercepted, the complete information cannot be restored.

[0085] Fragment recombination and data recovery: The user needs to provide real-time biological features again, and the system determines whether the same dynamic key can be generated. If verified, each data fragment can be decrypted and recombined to restore the original data.

[0086] This example generates a key based on multiple dimensions of biological features and environmental data, with strong individual uniqueness and temporal dynamics, significantly improving the system's ability to prevent forgery and replay attacks.

[0087] At the same time, the encryption algorithm based on lattice cryptography is used for data encryption, which is quantum-resistant, and ensures that even if a general quantum computer appears in the future, the key cannot be effectively cracked, thereby improving the security of data encryption.

[0088] In addition, the biological information cross verification in the embodiment reduces the recognition failure caused by single sensor failure or data anomaly, and improves the stability and reliability of the system in a complex environment. The dynamic introduction of the final environmental data enables the feature extraction process to have adaptive ability, and effectively reduces the recognition error caused by factors such as light, humidity and noise.

[0089] Preferably, the reference Figure 2 The preprocessing module 2 comprises:

[0090] An adaptive filtering unit 21 is configured to suppress random noise in each biological feature data according to an adaptive median filtering algorithm to obtain first filtered data.

[0091] A multi-resolution analysis unit 22 is configured to decompose an image in each biological feature data according to a multi-resolution method, and then restore the detail information in the image according to a threshold denoising technique to obtain second filtered data.

[0092] A deep denoising unit 23 is configured to perform image denoising and enhancement processing on the image in each biological feature data according to a pre-trained deep learning algorithm to obtain third filtered data. The first filtered data, the second filtered data and the third filtered data constitute filtered biological data.

[0093] Specifically, in the embodiment, the biological feature data includes an iris image and a fingerprint image. The resolution of the iris image is 640x480, and the format is grayscale image. The resolution of the fingerprint image is 512x512, and the format is binary image. The heart rate and blood oxygen are waveform / time series signals.

[0094] The adaptive filtering unit 21 uses an improved adaptive median filtering algorithm (AMF) for filtering: first, analyze the local pixel variance of the image, automatically adjust the filter window size, and then suppress Gaussian noise and salt and pepper noise; the output of the algorithm is the first filtered data (image noise reduction, edge blur), which can improve the basic clarity of the image and suppress burst noise. After applying this processing, the filter window size of the iris image is dynamically adjusted to 3x3 to 5x5, and the noise level is reduced by about 42%.

[0095] The multi-resolution analysis unit 22 adopts a discrete wavelet transform algorithm (DWT) + threshold denoising algorithm (hard threshold value): first, the filtered image is decomposed into 3 layers of wavelet, then the detail coefficients are denoised using threshold compression, and finally the second filtered data (image detail recovery, removal of high-frequency noise) is obtained by image reconstruction. After applying this processing, the fuzzy details such as ridge intersections in the fingerprint image are clearer, and the image signal-to-noise ratio is improved by about 18%. This processing can restore image edge details and improve structure fidelity.

[0096] The deep denoising unit 23 selects a model based on a DnCNN (Denoising Convolutional Neural Network) convolutional neural network; the training method of the model includes: using 3000 synthetic noise training images for supervised training; using an MSE loss function and an Adam optimizer; the deep denoising unit 23 inputs the second filtered image into the network; and outputs the third filtered data (clear structure, texture retention). Tests on iris images show that after using DnCNN processing, the image SSIM value is improved from 0.68 to 0.91. The final filtered biological data is a fusion representation of the pre-processing results of the three images.

[0097] Preferably, with reference to Figure 3 The multi-modal fusion module 3 comprises:

[0098] The feature alignment unit 31 is configured to extract iris vector features and fingerprint vector features of the iris image and the fingerprint image according to a ResNet algorithm, and extract heart rate time sequence features and blood oxygen time sequence features of the heart rate and the blood oxygen according to an LSTM network algorithm, and then unify the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features and the fingerprint vector features to the same latitude according to a principal component analysis algorithm.

[0099] The feature standardization unit 32 is connected to the feature alignment unit 31 and is configured to perform normalization processing on the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features and the fingerprint vector features according to a Z-score standardization method to obtain a plurality of normalized features, the normalized features including heart rate normalized features, blood oxygen normalized features, iris normalized features and fingerprint normalized features.

[0100] The fusion unit 33 is connected to the feature standardization unit 32 and is configured to configure an initial weight for each normalized feature, introduce an attention mechanism to perceive the feature quality of each normalized feature, and dynamically adjust the initial weight of each corresponding normalized feature according to the feature quality to obtain an adjusted weight, and finally calculate the multi-modal fusion biological data according to each normalized feature and the corresponding adjusted weight.

[0101] Specifically, in this embodiment, the iris image is a gray image with a resolution of 640x480, the fingerprint image is a binary image with a resolution of 512x512, the heart rate signal is a photoplethysmogram waveform within 10 seconds, the sampling rate is 250Hz, and the blood oxygen signal is collected synchronously with the heart rate, with a unit of %.

[0102] The feature alignment unit 31 converts the features of different modalities (images / timing) into vectors in the same latitude space, facilitating fusion.

[0103] Image modality processing (iris + fingerprint): using a pre-trained ResNet-34 to extract image features; the output iris vector feature and fingerprint vector feature have a numerical range of.

[0104] Timing modality processing (heart rate + blood oxygen): using a double-layer LSTM (128 hidden units) to extract timing dynamic features; each signal outputs a 128-dimensional timing vector feature; the numerical range of the output heart rate timing feature and blood oxygen timing feature is.

[0105] Feature alignment: using principal component analysis (PCA) to uniformly compress / expand all features to the same dimension, and outputting a feature vector of the same dimension.

[0106] The feature standardization unit 32 is used to unify the numerical scale of different feature dimensions, avoid certain modality features from causing dominant bias to the fusion result, and ensure the uniformity of different feature numerical scales. The central processing unit uses the Z-score standardization method to normalize each feature vector, and outputs four normalized features: heart rate normalized feature, blood oxygen normalized feature, iris normalized feature, and fingerprint normalized feature.

[0107] The fusion unit 33 combines the feature quality of each modality, realizes dynamic weighted fusion through attention mechanism, dynamically perceives the quality of each modality through the use of attention mechanism, enhances the robustness of the fusion result, and further improves the stability of the biological key, the individual uniqueness and the anti-tampering ability through multi-modal fusion.

[0108] Preferably, the calculation formula of the multi-modal fusion biological data is configured as:

[0109]

[0110] Among them, represents the multi-modal fusion biological data, represents the normalized heart rate feature, represents the normalized blood oxygen feature, represents the normalized iris feature, represents the normalized fingerprint feature, represents the adjustment weight of the heart rate feature, represents the adjustment weight of the blood oxygen feature, adjustment weight representing iris feature, adjustment weight representing fingerprint feature, preset offset constant, Gamma function, error function, first kind Bessel function, natural logarithm function, upper limit of summation, summation index variable, denominator index parameter of summation.

[0111] Specifically, in the embodiment, the central processor is embedded with a specially designed software program which operates according to the calculation formula of multi-modal fusion biological data, wherein the parameters , , , are positive real numbers for adjusting the weight of each biological feature in the fusion process. is a preset offset constant for ensuring that the input value of Gamma function is greater than zero. The program uses mathematical library to realize the operation of Gamma function , first kind Bessel function , error function and natural logarithm function , and calculates the fusion biological data , which is a positive real number, and the greater the value, the higher the overall matching or fusion degree of the four normalized biological features; at the same time, in the embodiment, the central processor can be used with a quantum resistance data encryption module to perform subsequent quantum resistance encryption processing on the fused data , and store the operation result in a secure database or transmit it to an authorized terminal, so as to realize identity authentication, access control, data encryption protection and other applications; in the device, each group of sensors is equipped with an adaptive calibration module for real-time compensation of the interference of external light, temperature, noise and motion and other environmental factors on the collected data, so as to ensure the accuracy and stability of the normalized feature values , , , The modules in the whole system are interconnected through a high-speed data bus, and self-checking and synchronous calibration are performed when the system starts, so as to ensure that the collected data, normalization processing, mathematical operation, and subsequent encryption and data transmission can be accurately completed in a short time in actual application. Through the above embodiment, the heart rate, blood oxygen, iris and fingerprint normalized features are fused through a complex multi-function mathematical model, and the output multi-modal fusion data can be used for precise identity recognition and quantum resistance data encryption application, and the value domain is a positive real number, theoretically ∈(0,+∞), wherein a larger value represents a high fusion degree and a high consistency, and is suitable for high security level biological authentication and encryption requirements.

[0112] Preferably, the multi-modal fusion module 3 further comprises an optimization matching unit 34 connected to the fusion unit 33, for searching for a feature combination with the strongest correlation between the normalized features as the optimal matching relationship according to a maximum mutual information algorithm, and adjusting the calculation formula of the multi-modal fusion biological data according to the normalized features in the optimal matching relationship and the corresponding adjustment weight, to obtain the optimized multi-modal fusion biological data.

[0113] Specifically, in the embodiment, the optimization matching unit 34 uses the maximum mutual information algorithm to calculate the information correlation between any two normalized features, and constructs a mutual information matrix, and then selects the feature combination with the largest mutual information in all feature pairs, and sets it as the optimal matching relationship, and then dynamically adjusts the calculation formula of the above multi-modal fusion biological data according to the specific normalized feature value 、 、 、 and the corresponding adjustment weight 、 、 、 to realize the optimization of the multi-modal fusion biological data. Through mutual information analysis to identify the feature combination with the strongest coupling relationship, more accurate data fusion can be realized, the discriminability of the fused biological features is enhanced, and the uniqueness and stability of the user individual recognition are effectively improved. Compared with the traditional fixed fusion strategy, the present scheme can dynamically select the fusion path according to the information intensity between the actual collected features, has stronger self-adaptation and generalization ability, and is suitable for feature changes under different physiological states or collection environments.

[0114] Since the fusion result is more consistent with the actual coupling structure between individual biological features, the generated biological dynamic key has higher entropy and unpredictability, effectively enhancing the security of the quantum-resistant password system.

[0115] The information theory method (maximum mutual information) and the lightweight deep fusion mechanism used in the embodiment do not depend on complex network structures, and can ensure efficient operation in edge devices or embedded platforms, and have good engineering implementation.

[0116] Preferably, with reference to Figure 4 , the feature extraction module 4 comprises:

[0117] a feature extraction unit 41, configured to introduce a multi-layer residual convolutional network, and perform feature extraction on the multi-modal fusion biological data according to the multi-layer residual convolutional network to obtain an intermediate feature tensor;

[0118] The scoring modeling unit 42 is configured to construct an environmental scoring model based on a shallow neural network, and input the illumination intensity, temperature, humidity and background noise into the environmental scoring model to obtain a parameter adjustment factor.

[0119] The parameter adjustment unit 43 is connected to the scoring modeling unit 42, and is configured to dynamically adjust a convolution kernel coefficient of a feature extractor in the feature extraction algorithm according to the parameter adjustment factor to obtain an optimized extraction algorithm model, and the optimized extraction algorithm model is configured to output an optimized feature tensor.

[0120] The template generation unit 44 is connected to the parameter adjustment unit 43 and the feature extraction unit 41, and is configured to generate a biological feature template according to the optimized feature tensor.

[0121] Specifically, in the embodiment, the feature extraction unit 41 introduces a multi-layer residual convolutional neural network, and the basic structure thereof is as follows: a plurality of residual modules are stacked, and each residual module includes two 3*3 convolution layers, a batch normalization (Batch Normalization) layer and a ReLU activation function. The input multi-modal fusion biological data (including the fusion results of heart rate, blood oxygen, iris image and fingerprint image) is first subjected to input layer standardization processing, and then sequentially subjected to a plurality of residual modules to extract spatial and temporal correlation features, and finally outputs an intermediate feature tensor.

[0122] The scoring modeling unit 42 constructs a shallow neural network model as an environmental scoring model, and the input of the environmental scoring model is the real-time collected illumination intensity, temperature, humidity and background noise. The structure includes an input layer, a hidden layer (8 neurons, and the activation function is ReLU), and an output layer (1 neuron, and the activation function is Sigmoid).

[0123] The scoring model outputs a parameter adjustment factor ∈ [0,1], and the calculation formula is as follows:

[0124]

[0125] wherein, represents the parameter adjustment factor, and respectively represent two weight matrices of the environmental scoring model, and respectively represent two bias terms of the environmental scoring model, respectively represent the illumination intensity, temperature, humidity and background noise, represents an activation function, represents a hyperbolic tangent function.

[0126] The parameter adjustment unit 43 dynamically adjusts the convolution kernel parameters of the convolutional layers in the feature extraction network using the parameter adjustment factor output by the scoring modeling unit 42. Specifically, the method is as follows: For each convolution kernel parameter in the residual module, the following updates are performed:

[0127] The formula for adjusting the convolution kernel coefficients is configured as follows:

[0128]

[0129] in, This represents the adjusted convolution kernel coefficients. This indicates that the previous convolution kernel parameters have been adjusted. This indicates the number of parameters that need to be adjusted. Indicates the first Each parameter is an adjustment factor.

[0130] In the model inference phase, the network is guided to adaptively strengthen the importance of features based on the current environmental state, thereby improving the system's robustness. The adjusted network is called the optimized feature extraction algorithm model, and it outputs the adjusted optimized feature tensor.

[0131] The template generation unit 44 feeds the optimized feature tensor into a compression-quantization-mapping process, ultimately generating a biometric feature template that can be used in the subsequent encryption process. The template structure includes: image domain master feature hash encoding, sparse representation encoding of physiological feature sequences, and fused identifier metadata.

[0132] This embodiment introduces an environmental perception mechanism and a dynamic network adjustment strategy, enabling the extraction of stable, accurate, and significantly individual-differentiated biometric templates even in real-world environments with fluctuations in light intensity, changes in temperature and humidity, or noise interference. This greatly enhances the security and reliability of subsequent key generation and data encryption processes.

[0133] Example 2, refer to Figure 5 This is the second embodiment of the present invention. Unlike the previous embodiment, this embodiment provides a disturbance generation unit 61, a function generation unit 62, and a data encryption unit 63, which can enhance the dynamic adaptability of the data encryption process and significantly improve the security strength and environmental robustness of the data encryption system. The filtered environmental data also includes sensor angles, physical occlusion amounts, and motion jitter amounts. Therefore, the data encryption module 6 includes:

[0134] The disturbance generation unit 61 generates a disturbance vector based on the sensor angle, physical occlusion amount, and motion jitter amount. The disturbance vector is used to control the parameter settings of the data encryption process.

[0135] The function generating unit 62 is connected with the disturbance generating unit 61, and is configured to generate a disturbance adjustment function according to the disturbance vector;

[0136] The data encryption unit 63 is connected with the function generating unit 62, and is configured to dynamically adjust the number of fragments, the number of encryption rounds and the configuration of lattice parameters in the process of encrypting data according to the disturbance encryption function.

[0137] Preferably, the disturbance adjustment function is configured as:

[0138]

[0139] wherein, denotes the disturbance adjustment function, denotes the disturbance vector, and denotes two adjustable initial constants, respectively denote the normalized sensor angle, the physical occlusion amount and the motion jitter amount, denotes three preset weight coefficients.

[0140] Working principle of the embodiment 2:

[0141] The disturbance generating unit 61 is disposed on a three-axis gyroscope on a terminal device to acquire the sensor angle in real time, and cooperates with an infrared or laser ranging module to acquire the physical occlusion amount between the target and the sensor, and combines the accelerometer data to calculate the root mean square of the motion jitter of the device in a unit time, and the three indicators are normalized to construct the disturbance vector . The function generating unit generates the disturbance adjustment function according to the disturbance vector . The data encryption unit executes the following dynamic strategy:

[0142] Fragment number setting: wherein, denotes the number of fragments, denotes the size of the original data, denotes the reference fragment size.

[0143] Encryption round number adjustment: wherein, denotes the basic encryption round, denotes the adjusted encryption round;

[0144] Dynamic expansion of lattice parameters (lattice dimension):

[0145] wherein, denotes the basic lattice latitude, and the value of the basic lattice latitude is set to 512, denotes the dynamically expanded lattice latitude.

[0146] The encryption operation is performed based on a lattice cryptography (such as NTRU or LWE variant), which guarantees quantum attack resistance. The embodiment introduces a dynamic disturbance function of an environmental disturbance factor, which can effectively enhance the dynamic adaptability of the data encryption process, especially in mobile devices, wearable devices or complex environments, and significantly improve the security strength and environmental robustness of the data encryption system.

[0147] A quantum-resistant data encryption method based on biological characteristics is applied to the quantum-resistant data encryption system, and refers to Figure 6 , comprising:

[0148] Step S1, the data acquisition module 1 acquires a plurality of biological characteristic data of the user and a plurality of environmental data in the environment where the user is in real time;

[0149] Step S2, the preprocessing module 2 respectively performs data preprocessing on each biological characteristic data and environmental data according to a preset preprocessing algorithm, to obtain filtered biological data and filtered environmental data after processing, the filtered biological data including iris images, fingerprint images, heart rate and blood oxygen, and the filtered environmental data including illumination intensity, temperature, humidity and background noise;

[0150] Step S3, the multi-modal fusion module 3 performs data fusion according to each filtered biological data, finds the optimal matching relationship between each filtered biological data based on a preset optimization algorithm in the fusion process, and obtains multi-modal fusion biological data by fusion;

[0151] Step S4, the feature extraction module 4 extracts features from the multi-modal fusion biological data according to a preset feature extraction algorithm, and dynamically adjusts the parameters of the feature extractor in the feature extraction algorithm according to each filtered environmental data in the feature extraction process, to obtain a biological feature template;

[0152] Step S5, the key generation module 5 generates a biological dynamic key in real time according to the biological feature template; the data encryption module 6 dynamically divides the original data into a plurality of data fragments according to the biological dynamic key, and encrypts each data fragment by using a quantum-resistant algorithm based on lattice cryptography to obtain encrypted fragments;

[0153] Step S6, the fragment recombination module 7 provides an anti-quantum fragment recombination strategy, so that the user restores each encrypted fragment to the original data when providing correct each biological characteristic data and dynamic key.

[0154] The above is only a preferred embodiment of the present application, and the protection scope of the present application is not limited to the above-mentioned embodiments. Any technical solutions falling within the concept of the present application shall be considered as falling within the protection scope of the present application. It should be noted that, for ordinary skilled persons in the art, some improvements and refinements without departing from the principles of the present application shall also be considered as falling within the protection scope of the present application.

Claims

1. A biometric based quantum resistant data encryption system, characterized in that, The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device for generating a dynamic key based on biological characteristics. The application relates to a multi-modal fusion method and device A fragment recombination module (7) connected to the data encryption module (6) is configured to provide an anti-quantum fragment recombination strategy, so that the user restores each encrypted fragment to the original data when providing correct biological feature data and the biological dynamic key.

2. The biometric based quantum resistant data encryption system of claim 1, wherein: The preprocessing module (2) comprises: An adaptive filtering unit (21) configured to suppress random noise in each biological feature data according to an adaptive median filtering algorithm to obtain first filtered data; A multi-resolution analysis unit (22) configured to decompose an image in each biological feature data according to a multi-resolution method, and then restore detail information in the image according to a threshold denoising technique to obtain second filtered data; A deep denoising unit (23) configured to perform image denoising and enhancement processing on the image in each biological feature data according to a pre-trained deep learning algorithm to obtain third filtered data, wherein the first filtered data, the second filtered data, and the third filtered data constitute the filtered biological data.

3. The biometric based quantum resistant data encryption system of claim 1, wherein: The calculation formula of the multi-modal fusion biological data is configured as: ； wherein, denotes the multi-modal fusion biological data, denotes the normalized heart rate feature, denotes the normalized blood oxygen feature, denotes the normalized iris feature, denotes the normalized fingerprint feature, denotes the adjustment weight of heart rate feature, denotes the adjustment weight of blood oxygen feature, denotes the adjustment weight of iris feature, denotes the adjustment weight of fingerprint feature, denotes a preset offset constant, denotes a Gamma function, denotes an error function, denotes a first kind of Bessel function, denotes a natural logarithm function, denotes a summation upper limit, denotes a summation index variable, denotes a summation denominator index parameter.

4. The biometric based quantum resistant data encryption system of claim 3, wherein: The multi-modal fusion module (3) further comprises an optimization matching unit (34) connected to the fusion unit (33) and configured to search for a feature combination with the strongest correlation between each normalized feature as the optimal matching relationship according to a maximum mutual information algorithm, and adjust the calculation formula of the multi-modal fusion biological data according to each normalized feature in the optimal matching relationship and the corresponding adjustment weight to obtain the multi-modal fusion biological data after optimization.

5. The biometric based quantum resistant data encryption system of claim 1, wherein: The feature extraction module (4) comprises: A feature extraction unit (41) configured to introduce a multi-layer residual convolutional network, and extract features from the multi-modal fusion biological data according to the multi-layer residual convolutional network to obtain an intermediate feature tensor; A score modeling unit (42) configured to construct an environmental score model based on a shallow neural network, and input the illumination intensity, the temperature, the humidity, and the background noise into the environmental score model to obtain a parameter adjustment factor; A parameter adjustment unit (43) connected to the score modeling unit (42) and configured to dynamically adjust a convolution kernel coefficient of a feature extractor in the feature extraction algorithm according to the parameter adjustment factor to obtain an optimized extraction algorithm model, wherein the optimized extraction algorithm model is configured to output an optimized feature tensor; A template generation unit (44) connected to the parameter adjustment unit (43) and the feature extraction unit (41) and configured to generate the biological feature template according to the optimized feature tensor.

6. The biometric based quantum resistant data encryption system of claim 5, wherein: The calculation formula of the parameter adjustment factor is configured as: ； wherein, represents the parameter adjustment factor, and respectively represent two weight matrices of the environmental score model, and respectively represent two bias terms of the environmental score model, respectively represent the illumination intensity, the temperature, the humidity, and the background noise, represents an activation function, represents a hyperbolic tangent function; The convolution kernel coefficient adjustment formula is configured as: ； wherein, denotes the adjusted convolution kernel coefficient, denotes the convolution kernel parameter before adjustment, denotes the number of parameters to be adjusted, denotes the th parameter adjustment factor.

7. The biometric based quantum resistant data encryption system of claim 1, wherein: When the filtered environmental data further comprises a sensor angle, a physical occlusion amount, and a motion jitter amount, the data encryption module (6) comprises: A disturbance generation unit (61) configured to generate a disturbance vector based on the sensor angle, the physical occlusion amount, and the motion jitter amount, wherein the disturbance vector is configured to control parameter setting of a data encryption process; A function generation unit (62) connected to the disturbance generation unit (61) and configured to generate a disturbance adjustment function according to the disturbance vector. A data encryption unit (63) is connected to the function generation unit (62) and is configured to dynamically adjust the number of data fragments, the encryption rounds, and the configuration of the lattice password parameters in the data encryption process according to the perturbation adjustment function.

8. The biometric based quantum resistant data encryption system of claim 7, wherein: The perturbation adjustment function is configured as: ； wherein, represents the disturbance adjustment function, represents the disturbance vector, and represents two initial constants that are adjustable, respectively represent the normalized sensor angle, the physical occlusion amount, and the motion blur amount, represents three preset weight coefficients. 9.A biometric-based quantum-resistant data encryption method applied to the quantum-resistant data encryption system of any one of claims 1-8, characterized in that, It comprises: Step S1, the data acquisition module (1) acquires a plurality of biological characteristic data of a user and a plurality of environmental data in an environment where the user is in real time; Step S2, the preprocessing module (2) respectively preprocesses each of the biological characteristic data and the environmental data according to a preset preprocessing algorithm to obtain filtered biological data and filtered environmental data after processing, the filtered biological data including an iris image, a fingerprint image, a heart rate, and blood oxygen, and the filtered environmental data including illumination intensity, temperature, humidity, and background noise; Step S3, the multi-modal fusion module (3) fuses the filtered biological data according to each of the filtered biological data, finds an optimal matching relationship between each of the filtered biological data based on a maximum mutual information algorithm in the fusion process, and obtains multi-modal fusion biological data through fusion; The multi-modal fusion module (3) comprises: The feature alignment unit (31) extracts iris vector features and fingerprint vector features of the iris image and the fingerprint image according to a ResNet algorithm, extracts heart rate time sequence features and blood oxygen time sequence features of the heart rate and the blood oxygen according to an LSTM network algorithm, and then unifies the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features, and the fingerprint vector features to the same dimension according to a principal component analysis algorithm; The feature standardization unit (32) normalizes the heart rate time sequence features, the blood oxygen time sequence features, the iris vector features, and the fingerprint vector features to obtain a plurality of normalized features, including heart rate normalized features, blood oxygen normalized features, iris normalized features, and fingerprint normalized features, according to a Z-score standardization method; The fusion unit (33) configures an initial weight for each of the normalized features, introduces an attention mechanism to perceive feature qualities of each of the normalized features, dynamically adjusts the initial weight corresponding to each of the normalized features according to each of the feature qualities to obtain an adjusted weight, and finally calculates the multi-modal fusion biological data according to each of the normalized features and the adjusted weight corresponding to each of the normalized features; Step S4, the feature extraction module (4) extracts features from the multi-modal fusion biological data according to a preset feature extraction algorithm, dynamically adjusts parameters of a feature extractor in the feature extraction algorithm according to each of the filtered environmental data in the feature extraction process, and obtains a biological feature template; Step S5, the key generation module (5) generates a biological dynamic key in real time according to the biological feature template; and the data encryption module (6) divides original data into a plurality of data fragments in real time according to the biological dynamic key, and encrypts each of the data fragments by using a quantum resistance algorithm based on a lattice password to obtain encrypted fragments. Step S6, the anti-quantum fragmentation reorganization module (7) provides an anti-quantum fragmentation reorganization strategy, so that the user restores each of the encrypted fragments to the original data when providing correct each of the biometric data and the biometric dynamic key.

Images