Encrypted transmission method and system for wireless information transmission communication base station

Abstract

The present invention provides an encryption transmission method and system for a wireless information transmission communication base station, which relates to the technical field, including the present invention discloses an encryption transmission method for a wireless information transmission communication base station, which belongs to the field of wireless communication technology. The method includes: receiving a communication signal and extracting identity authentication information and signal characteristic parameters; performing identity authentication scoring based on a deep neural network, performing signal characteristic scoring through spectrum analysis, and constructing a communication security assessment model; using an elliptic curve cryptographic algorithm to generate a key pair, performing obfuscation transformation and diffusion transformation on the data before encryption; using quantum key distribution technology to transmit the key; selecting the optimal transmission channel based on the operating status data of the communication base station, and using interleaving coding and adaptive power control technology for data transmission. The present invention improves the security and transmission reliability of wireless communications and enhances anti-interference capabilities.

Description

Technical Field

[0001] The present invention relates to wireless information technology, and in particular to an encryption transmission method and system for a wireless information transmission communication base station. Background Art

[0002] With the rapid development of wireless communication technology, communication base stations, as important infrastructure of wireless communication networks, undertake the key task of information transmission. At present, communication base stations mainly use traditional encryption algorithms and channel coding technologies to ensure the security and reliability of data transmission. These technologies include basic security protection measures such as symmetric encryption, digital signatures, and channel coding, which ensure the confidentiality and integrity of the communication process to a certain extent.

[0003] However, the existing technology still has the following problems: First, the traditional identity authentication mechanism is relatively simple and cannot effectively prevent security threats such as identity disguise and signal deception; second, the existing key distribution method has the risk of eavesdropping and man-in-the-middle attacks, and it is difficult to ensure the absolute security of the key exchange process; third, conventional channel coding and transmission technologies fail to fully consider the dynamic characteristics of the channel and have limited anti-interference capabilities in complex electromagnetic environments.

[0004] The present invention aims to solve technical problems such as insufficient security authentication, insecure key distribution and weak anti-interference ability in the process of information transmission of communication base stations, and to provide an encrypted transmission method that can comprehensively ensure communication security and improve transmission reliability. Summary of the invention

[0005] The embodiments of the present invention provide an encryption transmission method and system for a wireless information transmission communication base station, which can solve the problems in the prior art.

[0006] According to a first aspect of the embodiments of the present invention,

[0007] Provided is an encrypted transmission method for a wireless information transmission communication base station, comprising:

[0008] Receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, the identity authentication information includes user identity identification, device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; and construct a communication security assessment model based on the identity authentication score and the signal characteristic score;

[0009] Generate an asymmetric key pair based on the elliptic curve cryptography algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in blocks, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations;

[0010] Collect the operation status data of the communication base station, the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; build a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve the anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; use an integrity check mechanism to verify the reliability of data transmission.

[0011] Performing a credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score includes:

[0012] Construct an identity authentication feature vector, extract the user's historical login time distribution, operation behavior sequence, and geographic location migration trajectory from the user identity to form user behavior features, extract hardware configuration information, system environment parameters, and network connection features from the device authentication code to form device fingerprint features, and extract geographic coordinate sequence, movement speed, and activity range boundaries from location information to form spatial features; use the minimum-maximum normalization method to map the identity authentication feature vector to a unified interval, and reduce the feature dimension through principal component analysis;

[0013] The normalized identity authentication feature vector is input into a multi-layer convolutional neural network, which extracts local feature patterns through a hole convolution layer, uses an attention mechanism to highlight key features, and optimizes feature transfer through residual connections; the output features of the multi-layer convolutional neural network are input into a recurrent neural network, which uses a long short-term memory unit to process temporal features and captures contextual information through a bidirectional gated recurrent unit network;

[0014] The output features of the multi-layer convolutional neural network and the output features of the recurrent neural network are fused by using an attention fusion mechanism, feature association relationships are established through a cross-modal transformer, and feature dependency relationships are constructed by building a feature interaction graph network model; feature importance is dynamically adjusted based on a gating unit, and feature screening is achieved by using a sparse attention mechanism;

[0015] The credibility of user behavior, device features, and location trajectory are calculated separately, and the credibility of different dimensions is integrated into the identity authentication score through a hierarchical aggregation network; the temperature scaling method is used to adjust the probability distribution, and the network output is mapped to a confidence score through the platt scaling technique;

[0016] A multi-task learning framework is used to simultaneously optimize the authentication accuracy, anomaly detection rate, and scoring consistency, and the optimization goals are balanced through the adaptive adjustment mechanism of task weights. An incremental learning method is used to continuously update model parameters, and historical model experience is transferred through knowledge distillation technology. A scoring explanation model is constructed to generate scoring basis, and scoring result feedback is collected to establish scoring quality measurement indicators. Abnormal scores are identified and corrected through anomaly detection algorithms.

[0017] The signal characteristic parameters are verified for authenticity using spectrum analysis technology to generate signal characteristic scores including:

[0018] Collect the original waveform data of the communication signal, perform sampling rate conversion and bandpass filtering on the communication signal, and determine the signal frame boundary through signal synchronization technology; extract the power spectrum density characteristics, signal amplitude envelope, and power time-varying characteristics from the communication signal to form the signal strength characteristics, extract the carrier frequency offset, frequency modulation depth, and spectrum expansion coefficient to form the signal frequency characteristics, extract the spectrum occupied bandwidth, spectrum efficiency index, and sideband characteristic parameters to form the signal bandwidth characteristics, and extract the constellation diagram characteristics, modulation depth index, and phase noise characteristics to form the signal modulation characteristics;

[0019] The time-frequency distribution characteristics of the communication signal are obtained by short-time Fourier transform, multi-resolution analysis is achieved by wavelet transform, and the instantaneous frequency characteristics of the signal are extracted by Wigner distribution to construct a time-frequency energy density diagram; the high-order cumulative amount of the communication signal is calculated to extract non-Gaussian features, the phase coupling phenomenon is detected by bispectral analysis to identify nonlinear distortion, and the asymmetry characteristics are extracted by trispectral analysis to evaluate the modulation characteristics;

[0020] Collect standard signal samples of different communication formats to construct feature templates, establish a signal feature statistical distribution model to describe the range of feature parameter changes, and build a feature pattern dictionary through cluster analysis; use Euclidean distance to measure the degree of difference in feature vectors, build an anomaly detection model through Mahalanobis distance to identify feature deviations, and use dynamic time warping algorithm to compare the matching degree of time-varying feature sequences;

[0021] Verify signal spectrum compliance based on the time-frequency distribution characteristics, verify signal waveform integrity through the non-Gaussian features, and verify signal modulation standardization based on the asymmetry features; set weight coefficients for different feature dimensions, generate a comprehensive score through weighted summation, and use a fuzzy reasoning mechanism to process feature verification uncertainty to generate a credibility score;

[0022] Adaptive filtering technology is used to suppress narrowband interference, notch processing is used to eliminate single-frequency interference, and space-time adaptive processing technology is used to suppress directional interference; multi-scale analysis is used to improve the scale invariance of feature extraction, and rotation-invariant features are used to improve directional adaptability; verification results are continuously collected to feedback and update feature templates and verification rules, and feature weight configuration is optimized through parameter adaptive adjustment.

[0023] Generate an asymmetric key pair based on the elliptic curve cryptography algorithm, divide the asymmetric key pair into a public key and a private key; use block encryption to process the communication data, and perform obfuscation transformation and diffusion transformation on each data block, including:

[0024] Select elliptic curve parameters on the prime field to construct the curve equation, select the base point on the elliptic curve as the generator, generate the private key integer through the random number generator, and calculate the public key point by using the elliptic curve point multiplication operation; use the HMAC-DRBG algorithm to generate random numbers, expand them through the SHA-256 hash function, and introduce the key derivation function to derive the session key from the master key material;

[0025] Divide the communication data into blocks of fixed length, perform PKCS#7 padding on the last data block, and use the HMAC-SHA256 algorithm to calculate the MAC value of the data block; use the CBC mode to associate adjacent data blocks, generate a random initialization vector to break the association of data blocks, and establish a serialization index for the data block;

[0026] Construct a nonlinear substitution table to achieve byte-level substitution transformation, improve the nonlinearity of the substitution table through composite domain operations, and use Boolean function optimization technology to reduce the complexity of the substitution table; perform irregular cyclic displacement transformation on the data block matrix, and dynamically adjust the byte position relationship through displacement parameters; use MDS matrix coefficients to perform column vector linear transformation, and diffuse data correlation through finite field multiplication operations;

[0027] The substitution transformation, displacement transformation and linear transformation are combined into round functions, and the round constant is designed to eliminate fixed patterns; the Feistel structure is constructed to realize data block diffusion, the data correlation is transmitted through XOR operation, and the nonlinear diffusion function is designed to optimize the propagation path; the round key is extended to generate a subkey sequence, and the key correlation is broken through the nonlinear function;

[0028] SIMD instruction set is used to optimize basic operations, and pipeline technology is used to achieve parallel computing. Table lookup method is used to accelerate replacement table operations, pre-calculation technology is used to reduce operating overhead, and data cache structure is optimized to reduce memory access conflicts.

[0029] The anti-differential capability is improved by optimizing the differential characteristics of the replacement table, and a dynamic replacement table is introduced to eliminate fixed differential paths. The linear characteristics of the replacement table are optimized to reduce the probability of linear approximation, and the linear structure is broken through mixed algebraic operations. Time equalization technology is used to eliminate time side channels, and a mask scheme is used to protect energy analysis, realizing operation randomization to reduce electromagnetic leakage.

[0030] The private key is used to digitally sign the obfuscated data, and the public key is used to encrypt the diffused data; the encrypted data block is combined with the digital signature information to form an encrypted data packet; a key distribution protocol is constructed, and the quantum key distribution technology is used to securely transmit key information between communication base stations, including:

[0031] The SHA-3 hash algorithm is used to calculate the message digest of the obfuscated data. A random number is generated as a temporary key based on the ECDSA elliptic curve signature algorithm. The algebraic combination of the elliptic curve point and the private key is calculated. The RFC 6979 deterministic signature scheme is used to generate a signature value pair. A salt value is introduced to enhance the randomness of the signature.

[0032] The ECIES integrated encryption scheme is used to generate a temporary key pair for the data after diffusion transformation, the shared key point is calculated through the key derivation function, the ciphertext and authentication tag are generated using the AES-GCM mode, and the session key is established using the elliptic curve Diffie-Hellman key exchange;

[0033] The version number, algorithm identifier, key identifier, random number, ciphertext data, digital signature, and authentication tag are serialized into data packets using the ASN.1 encoding rules. The data packets are fragmented according to the maximum transmission unit. The fragment authentication code is calculated using a chain structure to prevent replay, and the fragments are reassembled through the session identifier.

[0034] The BB84 quantum key distribution protocol is used to prepare the polarization state of photons. The sender randomly selects a basis vector to encode the quantum bit, and the receiver randomly selects a measurement basis vector to obtain the measurement value. The original key is extracted by exchanging basis vector information through public communication.

[0035] Low-density parity check codes are used to correct errors in the original key, cascade check is used to improve error correction efficiency, and universal hash functions are used to amplify privacy. The quantum channel bit error rate is estimated through random sampling, and the final key length is determined based on statistical tests.

[0036] A key library system is built to implement hierarchical key management, session keys are dynamically allocated based on the quantum key pool, a two-phase commit protocol is used to ensure key consistency, and the key status is detected through a heartbeat mechanism; a lattice cryptographic algorithm is used to build a key exchange mechanism, hardware-isolated storage of keys is implemented, key operations are protected through a trusted execution environment, and a threshold key sharing scheme is used to disperse storage risks.

[0037] Constructing a communication channel evaluation model, selecting the optimal transmission channel based on the operating status data; performing channel coding on the encrypted data packet, using interleaving coding technology to improve anti-interference capability; adjusting the transmission power through adaptive power control technology, and optimizing the transmission route based on the link prediction algorithm, including:

[0038] Collect signal-to-noise ratio, bit error rate, bandwidth utilization, and delay jitter parameters, establish a channel assessment model through weighted summation, deploy distributed detection nodes to obtain channel status information, use a sliding window to calculate the channel quality trend, and combine the Kalman filter algorithm with the deep learning model to predict the evolution of the channel status;

[0039] The score of each channel is calculated based on the channel evaluation index, the load balancing factor is introduced to determine the channel priority, the channel switching threshold is set to trigger the dynamic switching mechanism, the soft switching technology is used to maintain the continuity of data transmission, and the spectrum competition strategy is optimized through the game theory model;

[0040] The data is processed by cascade coding. The outer layer uses Reed-Solomon code to correct burst errors, and the inner layer uses low-density parity check code to provide error correction capability. The coding parameters are dynamically adjusted based on the channel status, and block interleaving structure and convolution interleaving structure are used to break up burst errors.

[0041] The received signal strength and bit error rate are collected to establish a power control feedback loop, and the transmit power is adjusted through a proportional-integral control algorithm. Macro power planning is performed at the base station level, and micro power adjustment is performed at the terminal level. A collaborative control algorithm is used to optimize system power efficiency.

[0042] Build a link state database to record end-to-end transmission quality indicators, use time series analysis and deep learning to predict link state evolution, trigger route switching in advance based on the prediction results, and establish a multi-path transmission mechanism to improve transmission reliability;

[0043] Energy detection and feature matching are used to identify interference signals, interference maps are constructed to assist channel selection, spectrum expansion is achieved by controlling the frequency hopping pattern through pseudo-random sequences, beamforming is performed using adaptive antenna arrays, and transmission strategies are optimized through spatial diversity reception and hybrid automatic retransmission mechanisms.

[0044] According to a second aspect of the embodiments of the present invention,

[0045] Provides an encrypted transmission system for wireless information transmission communication base stations, including:

[0046] The first unit is used to receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, the identity authentication information includes a user identity identifier, a device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; and construct a communication security assessment model based on the identity authentication score and the signal characteristic score;

[0047] The second unit is used to generate an asymmetric key pair based on an elliptic curve cryptographic algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in a block manner, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations;

[0048] The third unit is used to collect the operation status data of the communication base station, and the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; construct a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve the anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; and use an integrity check mechanism to verify the reliability of data transmission.

[0049] According to a third aspect of the embodiments of the present invention,

[0050] An electronic device is provided, comprising:

[0051] processor;

[0052] a memory for storing processor-executable instructions;

[0053] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0054] A fourth aspect of the embodiments of the present invention is:

[0055] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0056] The beneficial effects of this application are as follows:

[0057] This method ensures the confidentiality, integrity and authenticity of communication data through multiple security mechanisms. First, a deep neural network is used to evaluate the credibility of identity authentication information, and the signal characteristic parameters are verified for authenticity in combination with spectrum analysis technology, effectively preventing fake base stations and identity fraud. Secondly, an elliptic curve cryptography algorithm is used to generate an asymmetric key pair, and digital signature technology is combined to ensure that the data is not tampered with or forged. Finally, quantum key distribution technology is used to securely transmit key information, further improving the security of key distribution.

[0058] This method improves the efficiency and reliability of data transmission by optimizing transmission strategies and channel processing technologies. On the one hand, by building a communication channel evaluation model and combining adaptive power control technology and link prediction algorithms, the optimal transmission channel and route can be dynamically selected to improve transmission efficiency. On the other hand, by using technologies such as interleaving coding, channel equalization and error correction decoding, the impact of channel interference and distortion on data transmission is effectively reduced, thereby enhancing transmission reliability.

[0059] By collecting the operating status data of the communication base station and building a communication security assessment model, this method can monitor the operating status of the base station in real time and take corresponding measures in time to ensure the stable operation of the base station. By analyzing data such as channel quality indicators, resource occupancy, signal interference intensity, and link stability indicators, potential risks can be effectively identified and early warnings can be issued, thereby improving the reliability and stability of the base station. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 A schematic diagram of a flow chart of an encryption transmission method of a wireless information transmission communication base station according to an embodiment of the present invention;

[0061] Figure 2 It is a schematic diagram of the structure of an encrypted transmission system of a wireless information transmission communication base station according to an embodiment of the present invention. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only 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.

[0063] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0064] Figure 1 FIG. 1 is a flow chart of an encryption transmission method of a wireless information transmission communication base station according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0065] S11. Receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, the identity authentication information includes user identity identification, device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; and construct a communication security assessment model based on the identity authentication score and the signal characteristic score;

[0066] S12. Generate an asymmetric key pair based on the elliptic curve cryptography algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in blocks, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations;

[0067] S13. Collect the operation status data of the communication base station, the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; construct a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; use an integrity check mechanism to verify the reliability of data transmission.

[0068] First, the communication base station receives the communication signal from the user device. The signal contains identity authentication information and signal characteristic parameters. The identity authentication information includes the user identity (e.g., a unique 128-bit number), the device authentication code (e.g., a 16-bit hexadecimal string), and the location information (e.g., longitude and latitude coordinates: 31.23°N, 121.47°E). The signal characteristic parameters include signal strength (e.g., -70dBm), signal frequency (e.g., 2.4GHz), signal bandwidth (e.g., 20MHz), and signal modulation method (e.g., OFDM).

[0069] The base station then performs a credibility assessment on the received information. Using a pre-trained deep neural network model, the authentication information is used as input and an authentication score between 0 and 1 is output. For example, if the user identity, device authentication code, and location information are all consistent with the records in the database, the output score is 0.95. At the same time, spectrum analysis techniques, such as fast Fourier transform, are used to analyze signal feature parameters, and their authenticity is judged based on a preset threshold to generate a signal feature score between 0 and 1. For example, if the signal strength, frequency, bandwidth, and modulation method are all within the normal range, the output score is 0.88.

[0070] Next, the base station constructs a communication security assessment model based on the identity authentication score and the signal feature score. For example, the weighted average method can be used to weight the two scores and obtain the final communication security score. Assuming that the weight of the identity authentication score is 0.7 and the weight of the signal feature score is 0.3, the final communication security score is 0.7*0.95+0.3*0.88=0.931.

[0071] Subsequently, the base station uses the elliptic curve cryptography algorithm to generate an asymmetric key pair, including a public key and a private key. For example, a specific elliptic curve and a base point are selected, a random number is generated as a private key, and the corresponding public key is calculated based on the point multiplication operation on the elliptic curve. The communication data is processed in blocks, for example, the data is divided into 1024-byte blocks. An obfuscation transformation is performed on each data block, for example, encryption is performed using the AES encryption algorithm. A diffusion transformation is then performed on each data block, for example, a hash value is calculated using the SHA-256 algorithm. The obfuscated data is digitally signed using the private key, for example, a signature is generated using the ECDSA algorithm. The data after the diffusion transformation is encrypted using the public key. The encrypted data block is combined with the digital signature information to form an encrypted data packet. Quantum key distribution technology is used between the base station and the receiving end to securely transmit key information for subsequent data encryption and decryption.

[0072] At the same time, the base station collects operating status data, including channel quality indicators (e.g., bit error rate: 10^-5), resource occupancy rate (e.g., CPU utilization rate: 70%), signal interference strength (e.g., -90dBm), and link stability indicators (e.g., packet loss rate: 1%). The base station builds a communication channel evaluation model, for example, based on the weighted average of channel quality indicators, resource occupancy rate, signal interference strength, and link stability indicators, to select the optimal transmission channel. Channel code the encrypted data packets, for example, using LDPC codes for encoding, and using interleaving coding technology to improve anti-interference capabilities. Adjust the transmit power through adaptive power control technology, for example, adjust the transmit power in real time according to the channel quality. Optimize the transmission route based on the link prediction algorithm, for example, select the optimal transmission path based on the network topology and real-time traffic information.

[0073] At the receiving end, the receiving end uses channel equalization technology to eliminate channel distortion, for example, using an adaptive equalizer to compensate for channel fading. The original data is restored through error correction decoding, for example, using the LDPC decoding algorithm for decoding. The receiving end uses an integrity check mechanism to verify the reliability of data transmission, for example, using the SHA-256 algorithm to calculate the hash value of the received data and compare it with the hash value provided by the sender.

[0074] The beneficial effects of this method can be summarized in the following three aspects:

[0075] 1. Enhanced security: Through deep neural network authentication, signal feature verification of spectrum analysis, encryption of elliptic curve cryptography and quantum key distribution technology, the multi-level security mechanism effectively prevents unauthorized access and data leakage, significantly improving the security of communications.

[0076] 2. Improve reliability: The application of technologies such as channel coding, interleaving coding, adaptive power control, link prediction algorithm, channel equalization and error correction decoding effectively reduces the impact of channel noise and interference, and ensures the reliability and integrity of data transmission.

[0077] 3. Optimize performance: The optimal transmission channel is selected through the communication channel evaluation model, and combined with adaptive power control and link prediction algorithms, resource utilization and transmission efficiency are optimized, improving the overall performance of the communication system.

[0078] In an optional implementation, performing credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score includes:

[0079] Construct an identity authentication feature vector, extract the user's historical login time distribution, operation behavior sequence, and geographic location migration trajectory from the user identity to form user behavior features, extract hardware configuration information, system environment parameters, and network connection features from the device authentication code to form device fingerprint features, and extract geographic coordinate sequence, movement speed, and activity range boundaries from location information to form spatial features; use the minimum-maximum normalization method to map the identity authentication feature vector to a unified interval, and reduce the feature dimension through principal component analysis;

[0080] The normalized identity authentication feature vector is input into a multi-layer convolutional neural network, which extracts local feature patterns through a hole convolution layer, uses an attention mechanism to highlight key features, and optimizes feature transfer through residual connections; the output features of the multi-layer convolutional neural network are input into a recurrent neural network, which uses a long short-term memory unit to process temporal features and captures contextual information through a bidirectional gated recurrent unit network;

[0081] The output features of the multi-layer convolutional neural network and the output features of the recurrent neural network are fused by using an attention fusion mechanism, feature association relationships are established through a cross-modal transformer, and feature dependency relationships are constructed by building a feature interaction graph network model; feature importance is dynamically adjusted based on a gating unit, and feature screening is achieved by using a sparse attention mechanism;

[0082] The credibility of user behavior, device features, and location trajectory are calculated separately, and the credibility of different dimensions is integrated into the identity authentication score through a hierarchical aggregation network; the temperature scaling method is used to adjust the probability distribution, and the network output is mapped to a confidence score through the platt scaling technique;

[0083] A multi-task learning framework is used to simultaneously optimize the authentication accuracy, anomaly detection rate, and scoring consistency, and the optimization goals are balanced through the adaptive adjustment mechanism of task weights. An incremental learning method is used to continuously update model parameters, and historical model experience is transferred through knowledge distillation technology. A scoring explanation model is constructed to generate scoring basis, and scoring result feedback is collected to establish scoring quality measurement indicators. Abnormal scores are identified and corrected through anomaly detection algorithms.

[0084] Get the user identity, device authentication code and location information. For example, the user identity can be a user name or user ID, the device authentication code can be the device's IMEI number or MAC address, and the location information can be obtained through GPS or base station positioning. Assume that the user information obtained is as follows: the user ID is 12345, the historical login time is 2023-10-26 08:00, 2023-10-26 12:00, 2023-10-27 09:00; the operation behavior sequence is login-browse products-add to shopping cart-order-pay; the geographic location migration trajectory is A-B-C. The device authentication code is abcdef, the hardware configuration information is CPU model X, the memory size is 8GB, the system environment parameter is the operating system version Y, and the network connection feature is WiFi connection. The location information is a sequence of geographic coordinates (longitude 116.3, latitude 39.9), (longitude 116.4, latitude 40.0), (longitude 116.5, latitude 40.1), the moving speed is 60 km / h, and the boundary of the activity range is area D.

[0085] Authentication features are extracted from user identification, device authentication code, and location information. User behavior features include historical login time distribution (e.g., login time interval, login frequency), operation behavior sequence (e.g., the order and number of operations such as login, browsing, and ordering), and geographic location migration trajectory (e.g., residence time, movement distance). Device fingerprint features include hardware configuration information (e.g., CPU model, memory size), system environment parameters (e.g., operating system version, browser type), and network connection features (e.g., IP address, WiFi name). Spatial features include geographic coordinate sequence, movement speed, and activity range boundaries.

[0086] Normalize the extracted authentication features. Use the minimum-maximum normalization method to map all feature values ​​to between 0 and 1. For example, assuming that the minimum value of the user login time interval is 1 hour and the maximum value is 24 hours, the login time intervals of user ID 12345 are 4 hours and 21 hours, and the normalized values ​​are (4-1) / (24-1)=0.13 and (21-1) / (24-1)=0.83. Perform similar normalization on all features.

[0087] Use principal component analysis to reduce feature dimensionality. Select principal components that can explain most of the data variance, such as selecting the first k principal components and discarding the remaining principal components to reduce computational complexity. Assume that the first three principal components are selected, and these three principal components can explain 95% of the data variance.

[0088] The normalized authentication feature vector is input into the multi-layer convolutional neural network. The dilated convolutional layer can extract local feature patterns of different scales, for example, it can extract short-term and long-term patterns in the user behavior sequence. The attention mechanism can highlight key features, for example, it can highlight high-risk login behaviors based on the user's historical behavior. The residual connection can optimize feature transfer and avoid the gradient vanishing or gradient exploding problems. Assume that the multi-layer convolutional neural network contains 3 convolutional layers, and each convolutional layer uses a different dilation rate, such as 1, 2, 4, to capture features of different scales.

[0089] The output features of the multi-layer convolutional neural network are input into the recurrent neural network. Long short-term memory units can process temporal features, such as user behavior sequences and geographic location migration trajectories. The bidirectional gated recurrent unit network can capture contextual information. For example, the user's next operation behavior can be predicted based on the user's previous operation behavior. Assume that the recurrent neural network contains 2 layers of LSTM units and uses a bidirectional GRU network to capture contextual information.

[0090] The output features of multi-layer convolutional neural networks and recurrent neural networks are fused using the attention fusion mechanism. The cross-modal transformer can establish associations between features, for example, user behavior features and device fingerprint features can be associated. The feature interaction graph network can model the dependencies between features, for example, the dependencies between user behavior features, device fingerprint features, and spatial features can be modeled. Assume that a multi-head attention mechanism is used for feature fusion, and a graph convolutional network is used to construct a feature interaction graph network.

[0091] The gating unit is used to dynamically adjust the feature importance, and the sparse attention mechanism is used for feature screening to further reduce the computational complexity and improve the generalization ability of the model.

[0092] The user behavior credibility, device feature credibility, and location track credibility are calculated respectively. For example, the user behavior credibility can be calculated based on the user behavior features, the device feature credibility can be calculated based on the device fingerprint features, and the location track credibility can be calculated based on the spatial features.

[0093] A hierarchical aggregation network is used to fuse the trustworthiness of different dimensions into an authentication score. For example, user behavior trustworthiness, device feature trustworthiness, and location trajectory trustworthiness can be used as inputs to calculate the final authentication score through a hierarchical aggregation network.

[0094] The probability distribution is adjusted using the temperature scaling method, and the network output is mapped to a confidence score using the Platt scaling technique, for example, the output value is mapped between 0 and 1, indicating the confidence of the authentication. Assuming the final authentication score is 0.8, it means that the user's authentication confidence is 80%.

[0095] The multi-task learning framework is used to optimize the authentication accuracy, anomaly detection rate, and scoring consistency at the same time. The task weight adaptive adjustment mechanism can balance the optimization objectives. For example, the task weights can be dynamically adjusted according to the importance of different tasks.

[0096] Use incremental learning methods to continuously update model parameters, and use knowledge distillation techniques to migrate historical model experience to adapt to new data and environments.

[0097] Build a rating explanation model to generate the basis for the rating, for example, explain which features contribute most to the final rating. Collect rating result feedback to establish rating quality metrics, for example, use user feedback to evaluate the accuracy and reliability of the rating. Use anomaly detection algorithms to identify abnormal ratings and make corrections. For example, if a user's rating is significantly different from the historical rating, it can be marked as an abnormal rating and further investigated.

[0098] Beneficial Effects

[0099] 1. Improve the accuracy and security of identity authentication: Through deep neural networks and multimodal feature fusion, the authenticity of user identities can be more accurately evaluated, and identity fraud can be effectively identified and prevented.

[0100] 2. Improve user experience: Through adaptive learning and anomaly detection mechanisms, the misjudgment rate can be reduced, unnecessary interference to legitimate users can be avoided, and a smoother identity authentication experience can be provided.

[0101] 3. Reduce operating costs: Through automated scoring and interpretation models, the workload of manual review can be reduced, the efficiency of identity authentication can be improved, and operating costs can be reduced.

[0102] In an optional implementation, the authenticity of the signal characteristic parameters is verified by using spectrum analysis technology, and generating a signal characteristic score includes:

[0103] Collect the original waveform data of the communication signal, perform sampling rate conversion and bandpass filtering on the communication signal, and determine the signal frame boundary through signal synchronization technology; extract the power spectrum density characteristics, signal amplitude envelope, and power time-varying characteristics from the communication signal to form the signal strength characteristics, extract the carrier frequency offset, frequency modulation depth, and spectrum expansion coefficient to form the signal frequency characteristics, extract the spectrum occupied bandwidth, spectrum efficiency index, and sideband characteristic parameters to form the signal bandwidth characteristics, and extract the constellation diagram characteristics, modulation depth index, and phase noise characteristics to form the signal modulation characteristics;

[0104] The time-frequency distribution characteristics of the communication signal are obtained by short-time Fourier transform, multi-resolution analysis is achieved by wavelet transform, and the instantaneous frequency characteristics of the signal are extracted by Wigner distribution to construct a time-frequency energy density diagram; the high-order cumulative amount of the communication signal is calculated to extract non-Gaussian features, the phase coupling phenomenon is detected by bispectral analysis to identify nonlinear distortion, and the asymmetry characteristics are extracted by trispectral analysis to evaluate the modulation characteristics;

[0105] Collect standard signal samples of different communication formats to construct feature templates, establish a signal feature statistical distribution model to describe the range of feature parameter changes, and build a feature pattern dictionary through cluster analysis; use Euclidean distance to measure the degree of difference in feature vectors, build an anomaly detection model through Mahalanobis distance to identify feature deviations, and use dynamic time warping algorithm to compare the matching degree of time-varying feature sequences;

[0106] Verify signal spectrum compliance based on the time-frequency distribution characteristics, verify signal waveform integrity through the non-Gaussian features, and verify signal modulation standardization based on the asymmetry features; set weight coefficients for different feature dimensions, generate a comprehensive score through weighted summation, and use a fuzzy reasoning mechanism to process feature verification uncertainty to generate a credibility score;

[0107] Adaptive filtering technology is used to suppress narrowband interference, notch processing is used to eliminate single-frequency interference, and space-time adaptive processing technology is used to suppress directional interference; multi-scale analysis is used to improve the scale invariance of feature extraction, and rotation-invariant features are used to improve directional adaptability; verification results are continuously collected to feedback and update feature templates and verification rules, and feature weight configuration is optimized through parameter adaptive adjustment.

[0108] In order to verify the authenticity of the communication signal and generate a signal feature score, this embodiment provides a signal feature parameter authenticity verification method based on spectrum analysis technology.

[0109] First, the original waveform data of the communication signal is collected. For example, a 2-second signal is collected at a sampling rate of 10 megasamples per second using a software radio platform.

[0110] Next, the collected communication signal is preprocessed. This includes converting the sampling rate of the signal to 20 MHz and using a bandpass filter to filter out noise and interference signals outside the target frequency band. The passband range of the filter is set to 2 MHz to 4 MHz. Then, using signal synchronization technology, such as a synchronization method based on a pilot sequence, the start and end positions of the signal frame are determined, and the continuous signal is divided into multiple frames for subsequent feature extraction. Assume that the length of each frame signal is 0.1 seconds.

[0111] Then, multi-dimensional features are extracted from the preprocessed signal. The power spectrum density, signal amplitude envelope and power time-varying characteristics are extracted to form the signal strength characteristics. By analyzing the frequency variation of the signal, the carrier frequency offset, frequency modulation depth and spectrum expansion coefficient are extracted to form the signal frequency characteristics. The spectrum occupied bandwidth, spectrum efficiency index and sideband characteristic parameters are measured to form the signal bandwidth characteristics. The signal constellation characteristics, modulation depth index and phase noise characteristics are analyzed to form the signal modulation characteristics. For example, assume that the extracted power spectrum density peak is -20dBm, the carrier frequency offset is 100Hz, and the spectrum occupied bandwidth is 2MHz.

[0112] In order to further analyze the time-frequency characteristics of the signal, short-time Fourier transform is used to obtain the time-frequency distribution characteristics of the communication signal. For example, a short-time Fourier transform with a window length of 128 sampling points is used. Through wavelet transform, for example, using Daubechies 4 wavelet for 5-level decomposition, multi-resolution analysis of the signal is achieved, thereby extracting the characteristics of the signal at different scales. The instantaneous frequency characteristics of the signal are extracted using Wigner distribution, and a time-frequency energy density diagram is constructed.

[0113] In order to extract the nonlinear characteristics of the signal, the high-order cumulants of the communication signal are calculated, such as the fourth-order cumulants of the signal to extract non-Gaussian features. The phase coupling phenomenon in the signal is detected by bispectral analysis to identify nonlinear distortion. For example, the peak position and amplitude in the bispectral graph are analyzed to determine whether nonlinear distortion exists. The asymmetric characteristics of the signal are extracted by using trispectral analysis to evaluate the modulation characteristics of the signal.

[0114] In order to establish a feature verification model, standard signal samples of different communication standards are collected, such as signal samples of GSM, WCDMA and LTE standards, to build a feature template. Based on the collected standard signal samples, a signal feature statistical distribution model is established to characterize the range of variation of feature parameters. For example, a Gaussian distribution model is used to describe the distribution of feature parameters. Through cluster analysis, such as using the K-means algorithm, a feature pattern dictionary is constructed to classify similar feature vectors into one category.

[0115] In order to verify the authenticity of the signal, the Euclidean distance is used to measure the degree of difference between the feature vectors. For example, the Euclidean distance between the feature vector of the signal to be verified and the feature template is calculated. The anomaly detection model is constructed through the Mahalanobis distance to identify feature deviations. For example, the Mahalanobis distance of the signal to be verified is calculated, and a threshold is set to determine whether there is an anomaly. The dynamic time warping algorithm is used to compare the matching degree of the time-varying feature sequence. For example, the dynamic time warping distance between the time-varying feature sequence of the signal to be verified and the time-varying feature sequence of the standard signal is calculated.

[0116] Verify the authenticity of the signal based on the extracted features. Verify the compliance of the signal spectrum based on the time-frequency distribution characteristics. Verify the integrity of the signal waveform through non-Gaussian features. Verify the standardization of signal modulation based on asymmetric features.

[0117] To generate the final score, weight coefficients are set for different feature dimensions. For example, a weight of 0.4 is set for the signal strength feature, a weight of 0.3 is set for the signal frequency feature, a weight of 0.2 is set for the signal bandwidth feature, and a weight of 0.1 is set for the signal modulation feature. A comprehensive score is generated by weighted summation. A fuzzy reasoning mechanism is used to handle the uncertainty of feature verification and generate a credibility score. For example, a fuzzy logic system is used to fuse the verification results of multiple features into a credibility score.

[0118] In order to improve the robustness of the system, adaptive filtering technology is used to suppress narrowband interference. Single-frequency interference is eliminated through notch processing. Directional interference is suppressed by space-time adaptive processing technology. The scale invariance of feature extraction is improved through multi-scale analysis. The rotation invariant feature is used to improve directional adaptability. The verification result feedback is continuously collected to update the feature template and verification rules. The feature weight configuration is optimized through parameter adaptive adjustment.

[0119] The beneficial effects of this method can be summarized in the following three aspects:

[0120] 1. Improved accuracy of signal verification: Through multi-dimensional feature extraction and analysis, the characteristics of the signal can be more comprehensively characterized, thereby improving the accuracy and reliability of signal verification. For example, by combining the features of the time domain, frequency domain, and time-frequency domain, real signals and forged signals can be distinguished more effectively.

[0121] 2. Enhanced system robustness: By adopting technologies such as adaptive filtering, notch processing, and space-time adaptive processing, various interferences can be effectively suppressed, and the robustness and anti-interference ability of the system can be improved. At the same time, the application of multi-scale analysis and rotation invariant features also enhances the adaptability of the system, enabling it to cope with signal changes in different environments.

[0122] 3. Automation and intelligence are achieved: By establishing feature templates, statistical distribution models and feature pattern dictionaries, and adopting fuzzy reasoning mechanisms and parameter adaptive adjustment methods, the automation and intelligence of signal verification are achieved, reducing the need for manual intervention and improving efficiency.

[0123] In an optional implementation, an asymmetric key pair is generated based on an elliptic curve cryptographic algorithm, and the asymmetric key pair is divided into a public key and a private key; communication data is processed in blocks using a block encryption method, and obfuscation transformation and diffusion transformation are performed on each data block, including:

[0124] Select elliptic curve parameters on the prime field to construct the curve equation, select the base point on the elliptic curve as the generator, generate the private key integer through the random number generator, and calculate the public key point by using the elliptic curve point multiplication operation; use the HMAC-DRBG algorithm to generate random numbers, expand them through the SHA-256 hash function, and introduce the key derivation function to derive the session key from the master key material;

[0125] Divide the communication data into blocks of fixed length, perform PKCS#7 padding on the last data block, and use the HMAC-SHA256 algorithm to calculate the MAC value of the data block; use the CBC mode to associate adjacent data blocks, generate a random initialization vector to break the association of data blocks, and establish a serialization index for the data block;

[0126] Construct a nonlinear substitution table to achieve byte-level substitution transformation, improve the nonlinearity of the substitution table through composite domain operations, and use Boolean function optimization technology to reduce the complexity of the substitution table; perform irregular cyclic displacement transformation on the data block matrix, and dynamically adjust the byte position relationship through displacement parameters; use MDS matrix coefficients to perform column vector linear transformation, and diffuse data correlation through finite field multiplication operations;

[0127] The substitution transformation, displacement transformation and linear transformation are combined into round functions, and the round constant is designed to eliminate fixed patterns; the Feistel structure is constructed to realize data block diffusion, the data correlation is transmitted through XOR operation, and the nonlinear diffusion function is designed to optimize the propagation path; the round key is extended to generate a subkey sequence, and the key correlation is broken through the nonlinear function;

[0128] SIMD instruction set is used to optimize basic operations, and pipeline technology is used to achieve parallel computing. Table lookup method is used to accelerate replacement table operations, pre-calculation technology is used to reduce operating overhead, and data cache structure is optimized to reduce memory access conflicts.

[0129] The anti-differential capability is improved by optimizing the differential characteristics of the replacement table, and a dynamic replacement table is introduced to eliminate fixed differential paths. The linear characteristics of the replacement table are optimized to reduce the probability of linear approximation, and the linear structure is broken through mixed algebraic operations. Time equalization technology is used to eliminate time side channels, and a mask scheme is used to protect energy analysis, realizing operation randomization to reduce electromagnetic leakage.

[0130] First, generate an asymmetric key pair. Select appropriate elliptic curve parameters on the prime field, such as curve coefficients and curve order, to construct the elliptic curve equation. Select a base point on the elliptic curve as the generator. Use the HMAC-DRBG algorithm to generate a random number as the private key. The range of this random number needs to be smaller than the order of the curve. Then, multiply the base point with the private key. This multiplication is a point multiplication operation on the elliptic curve, and the result is the public key point. For example, suppose an elliptic curve is selected on the prime field GF(p), the base point is G, and the private key is k, then the public key P=k*G.

[0131] Next, derive the session key. Use the HMAC-DRBG algorithm to generate a random number as the master key material. Use the SHA-256 hash function to extend the master key material to the required length. Then, use a key derivation function (such as HKDF) to derive the session key from the extended master key material. This session key will be used for subsequent data encryption and integrity verification. For example, assuming that the master key material is KM, the salt value is salt, and the information is info, then the session key SK=HKDF-SHA256(KM,salt,info).

[0132] Then the communication data is divided into blocks and padded. The communication data is divided into blocks according to the predetermined fixed length. If the length of the last data block is less than the predetermined length, it is padded using the PKCS#7 padding standard. For example, if the block length is 16 bytes and the last block has only 10 bytes of data, 6 bytes need to be padded, and the value of each byte is 6.

[0133] After that, the message authentication code is calculated. The HMAC-SHA256 algorithm is used to calculate the message authentication code (MAC) of each data block. The data block is used as input and the session key is used as the key to calculate the MAC value to ensure the integrity of the data block.

[0134] Next, the data blocks are encrypted. The CBC (Cipher Block Chaining) mode is used for data block encryption. First, a random number is generated as the initialization vector (IV) using the HMAC-DRBG algorithm. For the first data block, it is XORed with the IV and then encrypted using the session key. For subsequent data blocks, it is XORed with the ciphertext of the previous data block and then encrypted using the session key. A serialization index is established for each data block to facilitate subsequent decryption operations.

[0135] During the encryption process, obfuscation transformation and diffusion transformation are performed on each data block. The goal of obfuscation transformation is to make the relationship between ciphertext and plaintext as complex as possible. Nonlinear substitution tables can be used to implement byte-level substitution transformations. In order to improve the nonlinearity of the substitution table, composite domain operations can be used. In order to reduce the complexity of the substitution table, Boolean function optimization technology can be used.

[0136] The goal of diffusion transformation is to spread the influence of a single plaintext bit to multiple ciphertext bits. You can perform irregular cyclic shift transformation on the data block matrix, and dynamically adjust the byte position relationship through the shift parameter. You can also use the MDS (Maximum Distance Separable) matrix coefficients to perform column vector linear transformation and spread data correlation through finite field multiplication operations.

[0137] The substitution transformation, displacement transformation and linear transformation are combined into round functions, and the round constants are designed to eliminate fixed patterns. The Feistel structure is constructed to realize data block diffusion, the data correlation is transmitted through XOR operation, and the nonlinear diffusion function is designed to optimize the propagation path.

[0138] The round key is expanded to generate a sub-key sequence, and the correlation between the keys is broken by a nonlinear function.

[0139] To improve computing efficiency, SIMD (Single Instruction Multiple Data) instruction sets can be used to optimize basic operations, and pipeline technology can be used to achieve parallel computing. Table lookup methods can be used to accelerate replacement table operations, and pre-computation technology can be used to reduce operating overhead. Data cache structures can also be optimized to reduce memory access conflicts.

[0140] To enhance security, the differential characteristics of the substitution table can be optimized to improve the ability to resist differential attacks, and a dynamic substitution table can be introduced to eliminate fixed differential paths. The linear characteristics of the substitution table can also be optimized to reduce the probability of linear approximation attacks, and the linear structure can be broken through mixed algebraic operations. To resist side channel attacks, time equalization technology can be used to eliminate time side channels, and masking schemes can be used to protect against energy analysis attacks. Operation randomization can also be implemented to reduce electromagnetic leakage. For example, when performing key operations, random delays or random operations can be inserted to make it difficult for attackers to infer key information through side channel information.

[0141] The beneficial effects can be summarized into the following three aspects:

[0142] 1. Security enhancement: Through multiple security mechanisms such as elliptic curve cryptography, HMAC-SHA256 integrity check, CBC encryption mode, obfuscation and diffusion transformation, the security of data transmission is effectively improved, and it can effectively resist various known attacks, such as differential attacks, linear attacks and side channel attacks.

[0143] 2. Performance optimization: Through SIMD instruction set optimization, pipeline technology, table lookup, pre-calculation technology and data cache optimization and other performance optimization strategies, the efficiency of data encryption and decryption is significantly improved, the computing overhead is reduced, and it is suitable for scenarios with high performance requirements.

[0144] 3. High flexibility: The scheme supports customized parameters such as block length, key length and number of rounds, which can be flexibly adjusted according to the needs of actual application scenarios and has strong adaptability.

[0145] In an optional implementation, the private key is used to digitally sign the obfuscated transformed data, and the public key is used to encrypt the diffused transformed data; the encrypted data block is combined with the digital signature information to form an encrypted data packet; a key distribution protocol is constructed, and the quantum key distribution technology is used to securely transmit key information between communication base stations, including:

[0146] The SHA-3 hash algorithm is used to calculate the message digest of the obfuscated data. A random number is generated as a temporary key based on the ECDSA elliptic curve signature algorithm. The algebraic combination of the elliptic curve point and the private key is calculated. The RFC 6979 deterministic signature scheme is used to generate a signature value pair. A salt value is introduced to enhance the randomness of the signature.

[0147] The ECIES integrated encryption scheme is used to generate a temporary key pair for the data after diffusion transformation, the shared key point is calculated through the key derivation function, the ciphertext and authentication tag are generated using the AES-GCM mode, and the session key is established using the elliptic curve Diffie-Hellman key exchange;

[0148] The version number, algorithm identifier, key identifier, random number, ciphertext data, digital signature, and authentication tag are serialized into data packets using the ASN.1 encoding rules. The data packets are fragmented according to the maximum transmission unit. The fragment authentication code is calculated using a chain structure to prevent replay, and the fragments are reassembled through the session identifier.

[0149] The BB84 quantum key distribution protocol is used to prepare the polarization state of photons. The sender randomly selects a basis vector to encode the quantum bit, and the receiver randomly selects a measurement basis vector to obtain the measurement value. The original key is extracted by exchanging basis vector information through public communication.

[0150] Low-density parity check codes are used to correct errors in the original key, cascade check is used to improve error correction efficiency, and universal hash functions are used to amplify privacy. The quantum channel bit error rate is estimated through random sampling, and the final key length is determined based on statistical tests.

[0151] A key library system is built to implement hierarchical key management, session keys are dynamically allocated based on the quantum key pool, a two-phase commit protocol is used to ensure key consistency, and the key status is detected through a heartbeat mechanism; a lattice cryptographic algorithm is used to build a key exchange mechanism, hardware-isolated storage of keys is implemented, key operations are protected through a trusted execution environment, and a threshold key sharing scheme is used to disperse storage risks.

[0152] First, prepare the data to be transmitted. Divide the data into multiple data blocks. Each data block will be subjected to obfuscation and diffusion transformation.

[0153] Perform an obfuscation transformation on each data block. The goal of the obfuscation transformation is to hide the statistical relationship between the original data and the encrypted data. For example, the round function of a block cipher algorithm such as AES or SM4 can be used for obfuscation transformation. Suppose a data block is "12345678", after the obfuscation transformation, it may become "9AB7CD5E".

[0154] The message digest is calculated by using the SHA-3 hash algorithm on the obfuscated data. For example, the message digest of "9AB7CD5E" is calculated by using the SHA-3-256 hash algorithm to obtain "F1A2B3C4...".

[0155] Digitally sign the message digest based on the ECDSA elliptic curve signature algorithm. First, select a suitable elliptic curve and base point. Then, generate a random number as a temporary key. Multiply the temporary key with the elliptic curve base point to obtain a point on the elliptic curve. Multiply the horizontal coordinate of the point with the private key and add the temporary key to obtain the first part of the signature value. Multiply the message digest with the first part of the signature value and divide it by the temporary key to obtain the second part of the signature value. In order to enhance the randomness of the signature, a random salt value is introduced, and the salt value is concatenated with the message digest when calculating the signature value. Assuming that the private key is "00001111", the temporary key is "22223333", and the salt value is "44445555", the final signature value pair is ("AAAA5555", "BBBB6666").

[0156] Diffusion transformation is performed on each data block. The goal of diffusion transformation is to diffuse the influence of a single plaintext bit to multiple ciphertext bits to hide the statistical characteristics of the plaintext data. For example, the diffusion transformation can be performed using the Feistel network structure or the SPN network structure. Assume that the obfuscated data block "9AB7CD5E" becomes "EF5DCB79A" after diffusion transformation.

[0157] The data after diffusion transformation is encrypted based on the ECIES integrated encryption scheme. First, a temporary key pair is generated. Then, the shared key point is calculated by a key derivation function (such as HKDF). The data is encrypted using the AES-GCM mode and an authentication tag is generated. Assuming that the temporary public key is "FFFFFF00" and the temporary private key is "0000FFFF", the final generated ciphertext is "0123456789ABCDEF" and the authentication tag is "FEDCBA9876543210".

[0158] The version number, algorithm identifier, key identifier, random number (salt value), ciphertext data, digital signature, and authentication tag are serialized into a data packet according to the ASN.1 encoding rules. For example, the structure of a data packet can be expressed as:

[0159] {"version":"1.0","algorithm":"ECIES+ECDSA","key_id":"0001","nonce":"44445555","ciphert ext":"0123456789ABCDEF","signature":("AAAA5555","BBBB6666"),"tag":"FEDCBA9876543210"}.

[0160] The data packet is fragmented according to the maximum transmission unit (MTU). If the data packet size exceeds the MTU, it is divided into multiple fragments. The authentication code is calculated for each fragment, and a chain structure is used to use the authentication code of the previous fragment as the input for calculating the authentication code of the next fragment to prevent replay attacks. The fragment reassembly is achieved using the session identifier.

[0161] The BB84 quantum key distribution protocol is used to securely transmit key information between communication base stations. The sender randomly selects horizontal and vertical basis vectors or diagonal basis vectors, and encodes the quantum bits into different polarization states according to the selected basis vectors. The receiver randomly selects basis vectors for measurement. The sender and receiver exchange basis vector information through an open communication channel, discard quantum bits with mismatched basis vectors, and extract consistent measurement results as the original key.

[0162] Low-density parity check (LDPC) codes are used to correct errors in the original key, and cascaded check codes are used to improve error correction efficiency. Universal hash functions are used to amplify privacy and reduce the information that eavesdroppers may obtain. The quantum channel bit error rate is estimated through random sampling, and the final key length is determined based on statistical tests.

[0163] Build a key library system to achieve hierarchical key management. Dynamically distribute session keys based on quantum key pool. Use two-phase commit protocol to ensure key consistency. Detect key status through heartbeat mechanism. Use lattice cryptographic algorithm to build key exchange mechanism. Implement hardware-isolated key storage. Protect key operations through trusted execution environment. Use threshold key sharing scheme to disperse storage risks.

[0164] The beneficial effects can be summarized into the following three aspects:

[0165] 1. Enhanced data security: It combines multiple security technologies such as obfuscation transformation, diffusion transformation, digital signature and encryption processing to effectively prevent data leakage, tampering and forgery, and ensure data confidentiality and integrity. The use of quantum key distribution technology further enhances the security of key distribution and effectively resists quantum computing attacks.

[0166] 2. Improve transmission reliability: Adopt ASN.1 encoding rules, fragmentation mechanism, chain authentication code and other technologies to ensure the integrity and sequence of data packets, and effectively prevent data packet loss and replay attacks.

[0167] 3. Optimize key management: Build a key library system to implement key hierarchical management, dynamic allocation, consistency assurance, status detection and other functions, which improves the efficiency and security of key management. Use technologies such as lattice cryptography, hardware isolated storage, trusted execution environment and threshold key sharing to further enhance the security of keys.

[0168] In an optional implementation, constructing a communication channel evaluation model, selecting an optimal transmission channel based on the operating status data; performing channel coding on the encrypted data packet, using interleaving coding technology to improve anti-interference capability; adjusting the transmission power through adaptive power control technology, and optimizing the transmission route based on a link prediction algorithm include:

[0169] Collect signal-to-noise ratio, bit error rate, bandwidth utilization, and delay jitter parameters, establish a channel assessment model through weighted summation, deploy distributed detection nodes to obtain channel status information, use a sliding window to calculate the channel quality trend, and combine the Kalman filter algorithm with the deep learning model to predict the evolution of the channel status;

[0170] The score of each channel is calculated based on the channel evaluation index, the load balancing factor is introduced to determine the channel priority, the channel switching threshold is set to trigger the dynamic switching mechanism, the soft switching technology is used to maintain the continuity of data transmission, and the spectrum competition strategy is optimized through the game theory model;

[0171] The data is processed by cascade coding. The outer layer uses Reed-Solomon code to correct burst errors, and the inner layer uses low-density parity check code to provide error correction capability. The coding parameters are dynamically adjusted based on the channel status, and block interleaving structure and convolution interleaving structure are used to break up burst errors.

[0172] The received signal strength and bit error rate are collected to establish a power control feedback loop, and the transmit power is adjusted through a proportional-integral control algorithm. Macro power planning is performed at the base station level, and micro power adjustment is performed at the terminal level. A collaborative control algorithm is used to optimize system power efficiency.

[0173] Build a link state database to record end-to-end transmission quality indicators, use time series analysis and deep learning to predict link state evolution, trigger route switching in advance based on the prediction results, and establish a multi-path transmission mechanism to improve transmission reliability;

[0174] Energy detection and feature matching are used to identify interference signals, interference maps are constructed to assist channel selection, spectrum expansion is achieved by controlling the frequency hopping pattern through pseudo-random sequences, beamforming is performed using adaptive antenna arrays, and transmission strategies are optimized through spatial diversity reception and hybrid automatic retransmission mechanisms.

[0175] A communication method for constructing a communication channel evaluation model and selecting an optimal transmission channel based on operating status data comprises the following steps:

[0176] First, collect channel status information. On multiple distributed detection nodes, collect parameters such as signal-to-noise ratio, bit error rate, bandwidth utilization, and delay jitter in real time. For example, the data collected by detection node 1 is: signal-to-noise ratio 20dB, bit error rate 0.001, bandwidth utilization 80%, and delay jitter 2ms; the data collected by detection node 2 is: signal-to-noise ratio 15dB, bit error rate 0.002, bandwidth utilization 70%, and delay jitter 3ms.

[0177] Then, a channel evaluation model is established. The collected signal-to-noise ratio, bit error rate, bandwidth utilization, and delay jitter parameters are weighted and summed to obtain a comprehensive evaluation value of the channel quality. For example, if the weights are set to 0.4, 0.3, 0.2, and 0.1, the channel quality score of detection node 1 is: 0.4*20+0.3*(1-0.001)+0.2*80+0.1*(1-2 / 10)=24.798. Similarly, the channel quality score of detection node 2 can be calculated to be 19.697.

[0178] Next, predict the evolution of the channel state. Use a sliding window mechanism, for example, set the window size to 10, and count the channel quality evaluation values ​​of the last 10 moments. Combine the Kalman filter algorithm and the deep learning model, such as using the long short-term memory network (LSTM), to predict the future evolution trend of the channel state. For example, predict that the channel quality scores of the next 5 moments are 25, 26, 27, 26, and 25 respectively.

[0179] After that, the optimal transmission channel is selected. The score of each channel is calculated based on the channel evaluation index, and the load balancing factor is introduced, such as the amount of data currently carried by each channel, to determine the priority of the channel. For example, channel A scores 25 and the load factor is 0.8; channel B scores 24 and the load factor is 0.2. The final score is calculated as: channel score * (1-load factor), then the final score of channel A is 5, and the final score of channel B is 19.2. The channel with the highest score is selected as the optimal channel.

[0180] When the channel quality is lower than the preset threshold, for example, the threshold is set to 15, the dynamic switching mechanism is triggered. Soft switching technology is used, that is, before switching to the new channel, a short period of parallel transmission with the original channel is maintained to ensure the continuity of data transmission. Game theory models, such as Nash equilibrium, are used to optimize the spectrum competition strategy among multiple users.

[0181] The data is processed by concatenated coding. The outer layer uses Reed-Solomon code to correct burst errors, and the inner layer uses low-density parity check code to provide error correction capability. For example, the outer layer coding parameters are set to (255, 223) and the inner layer coding parameters are set to (1024, 512). The coding parameters are dynamically adjusted according to the channel status. For example, when the channel quality is poor, a stronger error correction coding parameter is selected, such as (255, 127). Block interleaving structure and convolution interleaving structure are used to break up burst errors. For example, the data is divided into data blocks of size 10x10 for interleaving.

[0182] Perform adaptive power control. Collect received signal strength and bit error rate to establish a power control feedback loop. For example, when the bit error rate is higher than 0.01, increase the transmit power; when the bit error rate is lower than 0.001, reduce the transmit power. Adjust the transmit power through the proportional integral control algorithm. The base station performs macro power planning, such as the power cap allocated to each user. The terminal performs micro power adjustment, such as fine-tuning the transmit power according to the real-time channel status. Use a collaborative control algorithm to optimize system power efficiency.

[0183] Optimize transmission routes. Build a link state database to record end-to-end transmission quality indicators, such as latency, packet loss rate, etc. Use time series analysis and deep learning models, such as recurrent neural networks (RNNs), to predict link state evolution. For example, predict the trend of link latency changes in the next 10 minutes. Trigger route switching in advance based on the prediction results. For example, if it is predicted that the latency of a certain link will increase significantly in the future, switch to the backup link in advance. Establish a multi-path transmission mechanism to improve transmission reliability.

[0184] Interference suppression. Use energy detection and feature matching to identify interference signals, such as narrowband interference and broadband interference. Construct interference maps to assist channel selection. For example, select channels with less interference for data transmission. Use pseudo-random sequences to control frequency hopping patterns to achieve spectrum expansion. Use adaptive antenna arrays to perform beamforming to focus signal energy in the direction of the target user and reduce the impact of interference. Optimize transmission strategies through spatial diversity reception and hybrid automatic retransmission mechanisms.

[0185] Beneficial effects:

[0186] 1. Improve transmission reliability: Through technologies such as channel assessment, coding, power control and routing optimization, the bit error rate and packet loss rate of data transmission are effectively reduced, and the reliability of transmission is improved.

[0187] 2. Improve transmission efficiency: Through technologies such as adaptive power control, load balancing and spectrum contention optimization, the spectrum utilization and system throughput are improved, and the transmission efficiency is improved.

[0188] 3. Enhanced anti-interference capability: Through interference identification, frequency hopping, beam forming and other technologies, the influence of various interference signals is effectively suppressed, and the system's anti-interference capability is enhanced.

[0189] Figure 2 FIG. 1 is a schematic diagram of the structure of an encrypted transmission system of a wireless information transmission communication base station according to an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0190] The first unit is used to receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, the identity authentication information includes a user identity identifier, a device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; and construct a communication security assessment model based on the identity authentication score and the signal characteristic score;

[0191] The second unit is used to generate an asymmetric key pair based on an elliptic curve cryptographic algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in a block manner, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations;

[0192] The third unit is used to collect the operation status data of the communication base station, and the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; construct a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve the anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; and use an integrity check mechanism to verify the reliability of data transmission.

[0193] According to a third aspect of the embodiments of the present invention,

[0194] An electronic device is provided, comprising:

[0195] processor;

[0196] a memory for storing processor-executable instructions;

[0197] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0198] According to a fourth aspect of the embodiments of the present invention,

[0199] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0200] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0201] 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 replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An encrypted transmission method for a wireless information transmission communication base station, characterized in that: include: Receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, the identity authentication information includes user identity identification, device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; and construct a communication security assessment model based on the identity authentication score and the signal characteristic score; Generate an asymmetric key pair based on the elliptic curve cryptography algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in blocks, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations; Collect the operation status data of the communication base station, the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; build a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve the anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; use an integrity check mechanism to verify the reliability of data transmission.

2. The method according to claim 1, characterized in that Performing a credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score includes: Construct an identity authentication feature vector, extract the user's historical login time distribution, operation behavior sequence, and geographic location migration trajectory from the user identity to form user behavior features, extract hardware configuration information, system environment parameters, and network connection features from the device authentication code to form device fingerprint features, and extract geographic coordinate sequence, movement speed, and activity range boundaries from location information to form spatial features; use the minimum-maximum normalization method to map the identity authentication feature vector to a unified interval, and reduce the feature dimension through principal component analysis; The normalized identity authentication feature vector is input into a multi-layer convolutional neural network, which extracts local feature patterns through a hole convolution layer, uses an attention mechanism to highlight key features, and optimizes feature transfer through residual connections; the output features of the multi-layer convolutional neural network are input into a recurrent neural network, which uses a long short-term memory unit to process temporal features and captures contextual information through a bidirectional gated recurrent unit network; The output features of the multi-layer convolutional neural network and the output features of the recurrent neural network are fused by using an attention fusion mechanism, feature association relationships are established through a cross-modal transformer, and feature dependency relationships are constructed by building a feature interaction graph network model; feature importance is dynamically adjusted based on a gating unit, and feature screening is achieved by using a sparse attention mechanism; The credibility of user behavior, device features, and location trajectory are calculated separately, and the credibility of different dimensions is integrated into the identity authentication score through a hierarchical aggregation network; the temperature scaling method is used to adjust the probability distribution, and the network output is mapped to a confidence score through the platt scaling technique; A multi-task learning framework is used to simultaneously optimize the authentication accuracy, anomaly detection rate, and scoring consistency, and the optimization goals are balanced through the adaptive adjustment mechanism of task weights. An incremental learning method is used to continuously update model parameters, and historical model experience is transferred through knowledge distillation technology. A scoring explanation model is constructed to generate scoring basis, and scoring result feedback is collected to establish scoring quality measurement indicators. Abnormal scores are identified and corrected through anomaly detection algorithms.

3. The method according to claim 1, characterized in that The signal characteristic parameters are verified for authenticity using spectrum analysis technology to generate signal characteristic scores including: Collect the original waveform data of the communication signal, perform sampling rate conversion and bandpass filtering on the communication signal, and determine the signal frame boundary through signal synchronization technology; extract the power spectrum density characteristics, signal amplitude envelope, and power time-varying characteristics from the communication signal to form the signal strength characteristics, extract the carrier frequency offset, frequency modulation depth, and spectrum expansion coefficient to form the signal frequency characteristics, extract the spectrum occupied bandwidth, spectrum efficiency index, and sideband characteristic parameters to form the signal bandwidth characteristics, and extract the constellation diagram characteristics, modulation depth index, and phase noise characteristics to form the signal modulation characteristics; The time-frequency distribution characteristics of the communication signal are obtained by short-time Fourier transform, multi-resolution analysis is achieved by wavelet transform, and the instantaneous frequency characteristics of the signal are extracted by Wigner distribution to construct a time-frequency energy density diagram; the high-order cumulative amount of the communication signal is calculated to extract non-Gaussian features, the phase coupling phenomenon is detected by bispectral analysis to identify nonlinear distortion, and the asymmetry characteristics are extracted by trispectral analysis to evaluate the modulation characteristics; Collect standard signal samples of different communication formats to construct feature templates, establish a signal feature statistical distribution model to describe the range of feature parameter changes, and build a feature pattern dictionary through cluster analysis; use Euclidean distance to measure the degree of difference in feature vectors, build an anomaly detection model through Mahalanobis distance to identify feature deviations, and use dynamic time warping algorithm to compare the matching degree of time-varying feature sequences; Verify signal spectrum compliance based on the time-frequency distribution characteristics, verify signal waveform integrity through the non-Gaussian features, and verify signal modulation standardization based on the asymmetry features; set weight coefficients for different feature dimensions, generate a comprehensive score through weighted summation, and use a fuzzy reasoning mechanism to process feature verification uncertainty to generate a credibility score; Adaptive filtering technology is used to suppress narrowband interference, notch processing is used to eliminate single-frequency interference, and space-time adaptive processing technology is used to suppress directional interference; multi-scale analysis is used to improve the scale invariance of feature extraction, and rotation-invariant features are used to improve directional adaptability; verification results are continuously collected to feedback and update feature templates and verification rules, and feature weight configuration is optimized through parameter adaptive adjustment.

4. The method according to claim 1, characterized in that: Generate an asymmetric key pair based on an elliptic curve cryptography algorithm, and divide the asymmetric key pair into a public key and a private key; The communication data is processed in blocks using block encryption, and the obfuscation transformation and diffusion transformation are performed on each data block, including: Select elliptic curve parameters on the prime field to construct the curve equation, select the base point on the elliptic curve as the generator, generate the private key integer through the random number generator, and calculate the public key point by using the elliptic curve point multiplication operation; use the HMAC-DRBG algorithm to generate random numbers, expand them through the SHA-256 hash function, and introduce the key derivation function to derive the session key from the master key material; Divide the communication data into blocks of fixed length, perform PKCS#7 padding on the last data block, and use the HMAC-SHA256 algorithm to calculate the MAC value of the data block; use the CBC mode to associate adjacent data blocks, generate a random initialization vector to break the association of data blocks, and establish a serialization index for the data block; Construct a nonlinear substitution table to achieve byte-level substitution transformation, improve the nonlinearity of the substitution table through composite domain operations, and use Boolean function optimization technology to reduce the complexity of the substitution table; perform irregular cyclic displacement transformation on the data block matrix, and dynamically adjust the byte position relationship through displacement parameters; use MDS matrix coefficients to perform column vector linear transformation, and diffuse data correlation through finite field multiplication operations; The substitution transformation, displacement transformation and linear transformation are combined into round functions, and the round constant is designed to eliminate fixed patterns; the Feistel structure is constructed to realize data block diffusion, the data correlation is transmitted through XOR operation, and the nonlinear diffusion function is designed to optimize the propagation path; the round key is extended to generate a subkey sequence, and the key correlation is broken through the nonlinear function; SIMD instruction set is used to optimize basic operations, and pipeline technology is used to achieve parallel computing. Table lookup method is used to accelerate replacement table operations, pre-calculation technology is used to reduce operating overhead, and data cache structure is optimized to reduce memory access conflicts. The anti-differential capability is improved by optimizing the differential characteristics of the replacement table, and a dynamic replacement table is introduced to eliminate fixed differential paths. The linear characteristics of the replacement table are optimized to reduce the probability of linear approximation, and the linear structure is broken through mixed algebraic operations. Time equalization technology is used to eliminate time side channels, and a mask scheme is used to protect energy analysis, realizing operation randomization to reduce electromagnetic leakage.

5. The method according to claim 1, characterized in that The private key is used to digitally sign the obfuscated data, and the public key is used to encrypt the diffused data; the encrypted data block is combined with the digital signature information to form an encrypted data packet; Constructing a key distribution protocol and using quantum key distribution technology to securely transmit key information between communication base stations includes: The SHA-3 hash algorithm is used to calculate the message digest of the obfuscated data. A random number is generated as a temporary key based on the ECDSA elliptic curve signature algorithm. The algebraic combination of the elliptic curve point and the private key is calculated. The RFC 6979 deterministic signature scheme is used to generate a signature value pair. A salt value is introduced to enhance the randomness of the signature. The ECIES integrated encryption scheme is used to generate a temporary key pair for the data after diffusion transformation, the shared key point is calculated through the key derivation function, the ciphertext and authentication tag are generated using the AES-GCM mode, and the session key is established using the elliptic curve Diffie-Hellman key exchange; The version number, algorithm identifier, key identifier, random number, ciphertext data, digital signature, and authentication tag are serialized into data packets using the ASN.1 encoding rules. The data packets are fragmented according to the maximum transmission unit. The fragment authentication code is calculated using a chain structure to prevent replay, and the fragments are reassembled through the session identifier. The BB84 quantum key distribution protocol is used to prepare the polarization state of photons. The sender randomly selects a basis vector to encode the quantum bit, and the receiver randomly selects a measurement basis vector to obtain the measurement value. The original key is extracted by exchanging basis vector information through public communication. Low-density parity check codes are used to correct errors in the original key, cascade check is used to improve error correction efficiency, and universal hash functions are used to amplify privacy. The quantum channel bit error rate is estimated through random sampling, and the final key length is determined based on statistical tests. A key library system is built to implement hierarchical key management, session keys are dynamically allocated based on the quantum key pool, a two-phase commit protocol is used to ensure key consistency, and the key status is detected through a heartbeat mechanism; a lattice cryptographic algorithm is used to build a key exchange mechanism, hardware-isolated storage of keys is implemented, key operations are protected through a trusted execution environment, and a threshold key sharing scheme is used to disperse storage risks.

6. The method according to claim 1, characterized in that Constructing a communication channel evaluation model, selecting the optimal transmission channel based on the operating status data; performing channel coding on the encrypted data packet, using interleaving coding technology to improve anti-interference capability; adjusting the transmission power through adaptive power control technology, and optimizing the transmission route based on the link prediction algorithm, including: Collect signal-to-noise ratio, bit error rate, bandwidth utilization, and delay jitter parameters, establish a channel assessment model through weighted summation, deploy distributed detection nodes to obtain channel status information, use a sliding window to calculate the channel quality trend, and combine the Kalman filter algorithm with the deep learning model to predict the evolution of the channel status; The score of each channel is calculated based on the channel evaluation index, the load balancing factor is introduced to determine the channel priority, the channel switching threshold is set to trigger the dynamic switching mechanism, the soft switching technology is used to maintain the continuity of data transmission, and the spectrum competition strategy is optimized through the game theory model; The data is processed by cascade coding. The outer layer uses Reed-Solomon code to correct burst errors, and the inner layer uses low-density parity check code to provide error correction capability. The coding parameters are dynamically adjusted based on the channel status, and block interleaving structure and convolution interleaving structure are used to break up burst errors. The received signal strength and bit error rate are collected to establish a power control feedback loop, and the transmit power is adjusted through a proportional-integral control algorithm. Macro power planning is performed at the base station level, and micro power adjustment is performed at the terminal level. A collaborative control algorithm is used to optimize system power efficiency. Build a link state database to record end-to-end transmission quality indicators, use time series analysis and deep learning to predict link state evolution, trigger route switching in advance based on the prediction results, and establish a multi-path transmission mechanism to improve transmission reliability; Energy detection and feature matching are used to identify interference signals, interference maps are constructed to assist channel selection, spectrum expansion is achieved by controlling the frequency hopping pattern through pseudo-random sequences, beamforming is performed using adaptive antenna arrays, and transmission strategies are optimized through spatial diversity reception and hybrid automatic retransmission mechanisms.

7. An encrypted transmission system for a wireless information transmission communication base station, used to implement the method described in any one of claims 1 to 6, characterized in that: include: The first unit is used to receive a communication signal sent by a communication base station, extract identity authentication information and signal characteristic parameters from the communication signal, wherein the identity authentication information includes a user identity identifier, a device authentication code, and location information, and the signal characteristic parameters include signal strength, signal frequency, signal bandwidth, and signal modulation mode; perform credibility assessment on the identity authentication information based on a deep neural network to generate an identity authentication score; and use spectrum analysis technology to verify the authenticity of the signal characteristic parameters to generate a signal characteristic score; Constructing a communication security assessment model based on the identity authentication score and the signal feature score; The second unit is used to generate an asymmetric key pair based on an elliptic curve cryptographic algorithm, and divide the asymmetric key pair into a public key and a private key; use a block encryption method to process the communication data in a block manner, and perform an obfuscation transformation and a diffusion transformation on each data block; use the private key to digitally sign the obfuscated data, and use the public key to encrypt the diffusion transformed data; combine the encrypted data block with the digital signature information to form an encrypted data packet; construct a key distribution protocol, and use quantum key distribution technology to securely transmit key information between communication base stations; The third unit is used to collect the operation status data of the communication base station, and the operation status data includes channel quality indicators, resource occupancy rate, signal interference intensity, and link stability indicators; construct a communication channel evaluation model, and select the optimal transmission channel based on the operation status data; perform channel coding on the encrypted data packet, and use interleaving coding technology to improve the anti-interference capability; adjust the transmission power through adaptive power control technology, and optimize the transmission route based on the link prediction algorithm; use channel equalization technology at the receiving end to eliminate channel distortion, and restore the original data through error correction decoding; and use an integrity check mechanism to verify the reliability of data transmission.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images