AI-based adaptive data encryption method and system

Abstract

The invention relates to the technical field of communication, and discloses an AI-based adaptive data encryption method and system, and the method comprises the steps: carrying out the frequency domain transformation analysis of an original data flow, and generating a spectrum distribution feature set; the carrier frequencies are screened through a dynamic frequency selection algorithm to obtain an optimal carrier frequency range; performing phase modulation on the data bit stream, and embedding the data bit stream into a subcarrier in an optimal carrier frequency range to generate a first encrypted data stream; when the interference intensity exceeds an interference threshold value, performing directional adjustment on a frequency deviation parameter according to an interference type, and generating a second encrypted data stream; dynamically adjusting the frequency distribution parameters according to the transmission distance data to obtain a frequency distribution parameter set; and performing adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set, and outputting a hidden transmission data stream. The method has the following effect that the data hidden transmission capability and the receiving and restoring accuracy in a complex environment can be improved.

Description

Technical Field

[0001] The present invention relates to the field of communication technologies, and in particular to an AI-based adaptive data encryption method and system. Background Art

[0002] Data encryption and covert transmission technologies are core safeguards for modern communications security, with widespread application value in military communications, financial transactions, the Internet of Things, and smart devices. With the increasing complexity of wireless communication environments, particularly the increasing challenges of diverse electromagnetic interference and fierce competition for spectrum resources, achieving highly discreet and reliable data transmission in open channels has become a key issue in communications security.

[0003] Currently, research in the communications security field focused on achieving high-disclosure, high-reliability data transmission over open channels primarily employs a centralized encrypted transmission architecture. This approach begins by deploying an encryption module within the communication terminal to perform real-time encryption processing on transmitted data and then send the encrypted data to a central processing node via a wireless channel. At the central node, a specific decryption algorithm is used to decrypt and analyze the received data to identify potential eavesdropping or interference threats.

[0004] However, in complex electromagnetic environments or resource-constrained scenarios, existing solutions face the following key challenges: static encryption strategies and fixed transmission modes struggle to cope with dynamically changing channel conditions; frequency selection uncertainty directly impacts data confidentiality, making it impossible to dynamically adjust the carrier frequency based on channel quality; and frequency offset and phase adjustment struggle to maintain consistency under varying environmental interference, impacting demodulation accuracy. In summary, achieving the fusion of dynamic frequency selection and precise data-frequency modulation in complex electromagnetic environments has become a pressing technical challenge in this field. Summary of the Invention

[0005] The present invention provides an AI-based adaptive data encryption method and system to improve the data concealment transmission capability and reception and restoration accuracy in complex environments.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides an AI-based adaptive data encryption method, comprising: Perform frequency domain transform analysis on the original data stream to generate a spectrum distribution feature set; According to the spectrum distribution feature set, the carrier frequency is screened by a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range; Embedding the data bit stream of the original data stream into the subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; monitoring interference information in the transmission environment in real time, and when the interference intensity exceeds a preset interference threshold, performing directionally adjustment on the frequency offset parameter according to the interference type, optimizing the first encrypted data stream to generate a second encrypted data stream; According to the transmission distance data fed back by the receiving end, the frequency distribution parameters are dynamically adjusted to obtain an optimized frequency distribution parameter set; Based on the frequency distribution parameter set, adaptive modulation and spectrum shaping are performed on the second encrypted data stream to output a concealed transmission data stream.

[0007] In an optional implementation, performing frequency domain transform analysis on the original data stream to generate a spectrum distribution feature set includes: Obtaining spectrum distribution information in the original data stream through frequency domain transformation analysis to obtain first spectrum distribution data; extracting spectrum characteristic data from the first spectrum distribution data, and screening out the spectrum characteristic data whose spectrum characteristics meet a preset screening criterion, to obtain a spectrum characteristic data group; The spectrum characteristic data group is classified to obtain the spectrum distribution feature set.

[0008] In an optional embodiment, the filtering of the carrier frequencies by a preset dynamic frequency selection algorithm based on the spectrum distribution feature set to obtain an optimal carrier frequency range includes: According to the spectrum distribution feature set, real-time channel quality parameters are obtained, and a channel state assessment result is obtained by comparing the change rate of the current channel environment with a preset spectrum distribution feature database; Based on the channel state evaluation result, using the dynamic frequency selection algorithm to screen a subset of candidate frequencies that meet the channel adaptation condition from available carriers, and determining a candidate frequency range; Performing a transmission quality test on the candidate frequency subset, retaining frequencies that meet a preset quality standard, and forming an optimized frequency combination; The optimized frequency combination is calibrated according to a preset modulation strategy to determine the optimal carrier frequency range.

[0009] In an optional implementation, embedding the data bit stream of the original data stream into a subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream includes: Performing preliminary analysis on the data bit stream to obtain basic characteristic information and determine the carrier frequency range; performing initial processing on the data bit stream using a frequency modulation technique according to the carrier frequency range to generate a preliminary modulated data stream; Applying a frequency offset adjustment technology to the preliminary modulated data stream to obtain adjusted frequency distribution data to obtain first frequency distribution data; The first frequency distribution data is secondary processed by using a phase modulation technology, and the data bit stream is embedded into a subcarrier within the optimal carrier frequency range to generate the first encrypted data stream.

[0010] In an optional embodiment, the real-time monitoring of interference information in the transmission environment, when the interference intensity exceeds a preset interference threshold, directionally adjusting the frequency offset parameter according to the interference type, and optimizing the first encrypted data stream to generate the second encrypted data stream, includes: Monitor interference information in the transmission environment in real time and extract interference feature data; Comparing the interference feature data with a preset interference pattern library to obtain interference analysis results; When the interference intensity in the interference analysis result exceeds a preset interference intensity threshold, calculating an initial adjustment range of the frequency offset parameter; Using a support vector machine algorithm to optimize the initial adjustment range and output adjustment parameters; The frequency offset of the encrypted data stream is adjusted according to the adjustment parameter, and the first encrypted data stream is optimized to generate the second encrypted data stream.

[0011] In an optional implementation, dynamically adjusting the frequency distribution parameters according to the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set includes: Obtaining initial frequency distribution parameter values, and establishing a corresponding relationship model between parameters and distances based on the law of changes in the frequency distribution parameter values and transmission distances; According to the corresponding relationship model, it is determined whether the current frequency distribution parameter value exceeds the preset allowable range; if it exceeds, the parameter is compensated in combination with the influencing factor of the transmission distance to generate a frequency distribution parameter group after preliminary correction; collecting interference data in the current communication environment in real time according to the frequency distribution parameter group; Using a support vector machine algorithm to classify the interference data and determine the impact level of each type of interference on the frequency parameters; According to the impact level, the weight distribution ratio of the frequency distribution parameters is adjusted, and the frequency distribution parameter set is output.

[0012] In an optional embodiment, performing adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set to output a covert transmission data stream includes: According to the frequency distribution parameter set, matching the second encrypted data stream with a preset frequency distribution model, obtaining an adjusted parameter configuration, and generating a first covert transmission data stream; When it is detected that the signal strength of the first covert transmission data stream in the target frequency band is lower than a threshold, a signal enhancement technology is used for optimization to generate a second covert transmission data stream; Analyzing the transmission characteristics of the second covert transmission data stream using a support vector machine algorithm to obtain a stability evaluation result; When the stability evaluation result does not meet a preset stability standard, modulating the second covert transmission data stream to generate a third covert transmission data stream; extracting multi-dimensional features of the third covert transmission data stream, verifying transmission performance of the third covert transmission data stream in a target frequency band, and outputting a transmission verification result; According to the transmission verification result, the covert transmission data stream is output and relevant analysis data is stored.

[0013] In an optional embodiment, the method further includes: At the receiving end, performing demodulation frequency identification processing on the concealed transmission data stream to obtain a first demodulated data stream; performing joint frequency offset and phase correction on the first demodulated data stream in combination with the channel state information and the dynamic parameters of the transmitting end to generate a second demodulated data stream; The second demodulated data stream is verified and processed. When data loss or error occurs, a frequency parameter retrospective adjustment mechanism is triggered to output a final data stream.

[0014] Compared with the prior art, the present invention has the following beneficial effects: (1) Perform frequency domain transform analysis on the original data stream to generate a spectrum distribution feature set. This technology generates a structured spectrum distribution feature set, which will be used as input data to evaluate and determine its applicability in the frequency selection process. This step provides reliable data support for subsequent dynamic frequency selection and covert modulation through systematic spectrum feature extraction and analysis, and is a key technical foundation for achieving environmental adaptability of communication systems.

[0015] (2) Based on the spectrum distribution feature set, the carrier frequency is screened and processed using a preset dynamic frequency selection algorithm to obtain the optimal carrier frequency range. This technology uses an intelligent algorithm to deeply analyze the spectrum feature set. This step achieves dynamic optimization of the carrier frequency. As the intelligent decision-making center of the entire technical solution, its dynamic and accurate performance directly determines the overall performance upper limit of the covert transmission system.

[0016] (3) The data bit stream of the original data stream is embedded into the subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream. This process adopts an intelligent multi-level modulation strategy, and through the coordinated optimization of frequency modulation and phase modulation, it significantly improves the signal concealment while ensuring spectrum efficiency. As the core modulation link of the covert transmission system, the coding accuracy and anti-interference ability of this step directly determine the reliable transmission performance of the system in complex electromagnetic environments, achieving the best balance between security and transmission efficiency.

[0017] (4) Real-time monitoring of interference information in the transmission environment. When the interference intensity exceeds a preset interference threshold, the frequency offset parameter is adjusted according to the type of interference, optimizing the first encrypted data stream to generate the second encrypted data stream. This technology, through a closed-loop control mechanism of environmental perception and parameter adjustment, enables the system to dynamically combat complex interference, effectively improving the environmental adaptability and survivability of the covert transmission system.

[0018] (5) Based on the transmission distance data fed back by the receiving end, the frequency distribution parameters are dynamically adjusted to obtain the optimized frequency distribution parameter set. This process breaks through the limitations of traditional fixed parameter systems by constructing a dynamic distance-frequency mapping relationship and a collaborative optimization mechanism for interference response, providing a frequency optimization solution with environmental perception, intelligent decision-making, and dynamic adaptability for complex environments.

[0019] (6) Adaptive modulation and spectrum shaping are performed on the second encrypted data stream to output a covert transmission data stream. This process ensures that the output signal has low observability, high reliability, and strong environmental adaptability through a multi-level optimization mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a flow chart of an AI-based adaptive data encryption method provided by an embodiment of the present invention; Figure 2 This is a structural diagram of an AI-based adaptive data encryption system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0022] Reference Figure 1 , an embodiment of the present invention provides an AI-based adaptive data encryption method, comprising the following steps: S11, performing frequency domain transformation analysis on the original data stream to generate a spectrum distribution feature set; S12, screening the carrier frequency using a preset dynamic frequency selection algorithm according to the spectrum distribution feature set to obtain an optimal carrier frequency range; S13, embedding the data bit stream of the original data stream into the subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; S14, monitoring interference information in the transmission environment in real time, and when the interference intensity exceeds a preset interference threshold, directionally adjusting the frequency offset parameter according to the interference type, optimizing the first encrypted data stream to generate a second encrypted data stream; S15, dynamically adjusting the frequency distribution parameters according to the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set; S16 , performing adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set, and outputting a concealed transmission data stream.

[0023] In step S11, frequency domain transformation analysis is performed on the original data stream to generate a spectrum distribution feature set.

[0024] In one embodiment, frequency spectrum distribution information in the original data stream is obtained by frequency domain transformation analysis to obtain first frequency spectrum distribution data; extracting spectrum characteristic data from the first spectrum distribution data, and screening out the spectrum characteristic data whose spectrum characteristics meet a preset screening criterion, to obtain a spectrum characteristic data group; The spectrum characteristic data group is classified to obtain the spectrum distribution feature set.

[0025] Specifically, using an audio signal with a sampling rate of 1000Hz and a duration of 5 seconds (containing 5000 samples) as an example, the system generates a set of spectral distribution features through the following sequential steps: First, a fast Fourier transform (FFT) is performed on the original time-domain signal, using a window size of 1024 points to obtain a spectral resolution of approximately 0.976Hz. Normalized spectral data is then calculated. Frequency-domain transform analysis reveals that the signal's primary energy is concentrated in the 50-200Hz band, with a peak at 100Hz and an amplitude of 12.5 (normalized value), resulting in the first spectral distribution data. Spectral characteristic data is then extracted from this first spectral distribution data and filtered according to pre-set criteria (frequency range 0-500Hz, normalized amplitude ≥ 3.0, frequency resolution ≥ 0.9Hz). The data is then divided into three frequency bands: low frequency (0-100Hz), mid frequency (100-300Hz), and high frequency (300-500Hz). The energy contributions of each band are calculated to be 60%, 30%, and 10%, respectively, forming a spectral characteristic data set. Finally, key parameters such as low-frequency proportion, medium-frequency proportion, high-frequency proportion, peak frequency, and peak amplitude are constructed into a five-dimensional feature vector [60, 30, 10, 100, 12.5]. This feature vector, together with other similar data, constitutes a spectrum distribution feature set, providing a comprehensive feature basis for subsequent frequency selection.

[0026] In step S12, based on the spectrum distribution feature set, the carrier frequency is screened by a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range; In one embodiment, a real-time channel quality parameter is obtained based on the spectrum distribution feature set, and a channel state assessment result is obtained by comparing the change rate of the current channel environment with a preset spectrum distribution feature database; Based on the channel state evaluation result, using the dynamic frequency selection algorithm to screen a subset of candidate frequencies that meet the channel adaptation condition from available carriers, and determining a candidate frequency range; Performing a transmission quality test on the candidate frequency subset, retaining frequencies that meet a preset quality standard, and forming an optimized frequency combination; The optimized frequency combination is calibrated according to a preset modulation strategy to determine the optimal carrier frequency range.

[0027] Specifically, a spectrum analyzer is first used to scan the 10MHz-100MHz frequency band at 1MHz intervals, collecting signal strength at 90 sampling points (e.g., -50dBm at 20MHz and -60dBm at 30MHz). Using a fast Fourier transform algorithm, the main frequency component and noise level are extracted, and the signal-to-noise ratio (SNR) at 20MHz and 30MHz is calculated to be 20dB and 15dB, respectively. This creates a pre-defined database of spectrum distribution characteristics. Next, channel quality parameters are obtained in real time from the base station (e.g., a bit error rate of 0.001% and a fading coefficient of 0.8 at 20MHz; a bit error rate of 0.01% and a fading coefficient of 0.7 at 30MHz). By comparing the fluctuations in the SNR, BER, and fading coefficient at different times, the rate of change of the current channel environment is assessed, resulting in a channel status assessment result. Based on the channel state assessment results, a dynamic frequency selection algorithm was used to screen candidate frequencies from available carriers. Using a weighted scoring model (weighting signal-to-noise ratio (50%), bit error rate (BER) (30%), and fading coefficient (20%)), a score was calculated for each frequency (e.g., 20MHz scored 12.6, 30MHz scored 10.4). The top 10% of frequencies (scoring no less than 12 points) were selected, and the 20MHz-25MHz range was determined as the candidate frequency range. Transmission quality testing was then performed on this subset of candidate frequencies, retaining those with packet loss rates below 1% and transmission rates that met the required standards. This resulted in an optimized frequency combination. Finally, the candidate frequencies were analyzed for spectral flatness and interference levels, and calibrated according to a pre-set modulation strategy. 20MHz was selected as the optimal carrier frequency and adapted for QPSK (Quadrature Phase Shift Keying) modulation. This in turn determined the optimal carrier frequency range encompassing this frequency, ensuring stable data transmission.

[0028] In step S13, the data bit stream of the original data stream is embedded into the subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; In one embodiment, a preliminary analysis is performed on the data bit stream to obtain basic characteristic information and determine a carrier frequency range; performing initial processing on the data bit stream using a frequency modulation technique according to the carrier frequency range to generate a preliminary modulated data stream; Applying a frequency offset adjustment technology to the preliminary modulated data stream to obtain adjusted frequency distribution data to obtain first frequency distribution data; The first frequency distribution data is secondary processed by using a phase modulation technology, and the data bit stream is embedded into a subcarrier within the optimal carrier frequency range to generate the first encrypted data stream.

[0029] Specifically, the first encrypted data stream is generated through the following coherent process: first, the data bit stream is preliminarily analyzed to determine that 10kHz-15kHz is the carrier frequency range; then, the input bit stream (such as 0101) is frequency modulated using a frequency shift keying algorithm, mapping bit 0 to 10kHz (for 1ms) and bit 1 to 15kHz (for 1ms), generating a sine wave signal. , where A=1, φ=0, and under the condition of 20dB signal-to-noise ratio, the bit error rate is about 0.001%, and the preliminary modulated data stream is obtained. Subsequently, the presence of channel noise at 12kHz is detected, and the frequency offset adjustment technology is used to adjust the mapping frequency of bit 1 from 15kHz to 14.5kHz. After adjustment, the signal becomes After fast Fourier transform analysis, the spectrum energy is concentrated at 10kHz and 14.5kHz, and the side lobe is attenuated to -30dB, obtaining the first frequency distribution data. Then, the binary phase shift keying algorithm is used to perform secondary processing on the first frequency distribution data, and bit 0 corresponds to phase 0° and bit 1 corresponds to phase 180° to generate a composite signal. , realize the embedding of data bit stream into subcarriers in the carrier frequency range of 10kHz-15kHz. Finally, use Logistic mapping A chaotic sequence is generated and XOR-encrypted with the modulated composite signal. Spectral analysis shows that the signal entropy increases from 4.2 to 7.8, approaching an ideal random signal. This ultimately generates the first encrypted data stream. The entire process is executed by a digital signal processor, with each step seamlessly integrated, effectively embedding data while ensuring the security and efficiency of the communication system.

[0030] In step S14, interference information in the transmission environment is monitored in real time. When the interference intensity exceeds a preset interference threshold, the frequency offset parameter is directionally adjusted according to the interference type to optimize the first encrypted data stream to generate a second encrypted data stream. In one embodiment, interference information in the transmission environment is monitored in real time, and interference feature data is extracted; Comparing the interference feature data with a preset interference pattern library to obtain interference analysis results; When the interference intensity in the interference analysis result exceeds a preset interference intensity threshold, calculating an initial adjustment range of the frequency offset parameter; Using a support vector machine algorithm to optimize the initial adjustment range and output adjustment parameters; The frequency offset of the encrypted data stream is adjusted according to the adjustment parameter, and the first encrypted data stream is optimized to generate the second encrypted data stream.

[0031] Specifically, the signal acquisition module first monitors the transmission environment in real time and extracts interference feature data. The collected signal-to-noise ratio is 15.2dB (the standard interference-free environment is 20dB), and the interference intensity is calculated to be 4.8dB. Using the fast Fourier transform algorithm to decompose the spectrum, it is found that there is periodic pulse interference with a peak value of 3.5mV and a duration of 2ms in the 2.4GHz frequency band. Subsequently, the extracted interference feature is compared with the preset interference pattern library. Because the peak value exceeds the preset threshold of 2.0mV, it is determined that the current interference poses a threat to data transmission, triggering the frequency offset adjustment mechanism. Based on the difference between the interference peak value and the threshold, a linear mapping algorithm is used to calculate the frequency offset adjustment amount: , adjusting the initial frequency offset parameter from 5kHz to 8kHz. Next, the support vector machine (SVM) algorithm was used to optimize the initial adjustment range. Parameters such as interference intensity of 4.8dB, interference frequency of 2.4GHz, and initial offset of 3kHz were input. The support vector machine model output the optimal adjustment parameter as 8.2kHz. Finally, the optimized frequency offset parameter was applied to the modulation module to remodulate the first encrypted data stream to generate the second encrypted data stream. The signal-to-noise ratio of the adjusted signal was improved to 18.5dB, and the bit error rate was reduced from 0.1% to 0.01%, ensuring transmission stability in interference environments. The system simultaneously records the adjustment log for subsequent optimization reference, forming a complete closed loop from interference detection to parameter adjustment.

[0032] In step S15, the frequency distribution parameters are dynamically adjusted according to the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set; In one embodiment, an initial frequency distribution parameter value is obtained, and a corresponding relationship model between the parameter and the distance is established according to a law of change of the frequency distribution parameter value and the transmission distance; According to the corresponding relationship model, it is determined whether the current frequency distribution parameter value exceeds the preset allowable range; if it exceeds, the parameter is compensated in combination with the influencing factor of the transmission distance to generate a frequency distribution parameter group after preliminary correction; collecting interference data in the current communication environment in real time according to the frequency distribution parameter group; Using a support vector machine algorithm to classify the interference data and determine the impact level of each type of interference on the frequency parameters; According to the impact level, the weight distribution ratio of the frequency distribution parameters is adjusted, and an optimized frequency distribution parameter set is output.

[0033] Specifically, the initial frequency distribution parameter values were first obtained. Assuming the initial encrypted data stream contained 1000 packets, its frequency distribution parameters were based on a normal distribution with a mean of 50 Hz and a standard deviation of 5 Hz. Using the Advanced Encryption Standard (AES) encryption algorithm, a pseudo-random frequency offset ranging from -10 Hz to +10 Hz was generated. The adjusted mean was calculated to be 52 Hz with a standard deviation of 5.2 Hz, indicating a slight right skewness. Next, a correlation model was established based on the frequency distribution parameter values and the variation in transmission distance. For a transmission distance of 10 km, a linear attenuation model with a frequency offset of -0.1 Hz per kilometer was used to calculate a total offset of -1 Hz. The mean was adjusted to 51 Hz, while the standard deviation remained unchanged. Analysis revealed that long-distance transmission reduced signal concealment by approximately 5%, requiring further optimization. Subsequently, the current parameter values were determined based on the correspondence model. Since the standard deviation exceeded the preset allowable range, the Kalman filter adaptive algorithm, taking into account the influencing factors of transmission distance, was used. Using real-time monitored ambient noise with a mean of 2 Hz as input, the frequency distribution parameters were dynamically adjusted. The optimized mean was 51.5 Hz, and the standard deviation was reduced to 4.8 Hz. The detection probability of packets in complex environments was reduced to 3%, and the concealment capability was improved by 8%. Based on this preliminary correction, the system then collected interference data from the current communication environment in real time. A support vector machine algorithm was used to classify this interference data, determining the impact of each type of interference on the frequency parameters. The weighting of the frequency distribution parameters was then adjusted accordingly, resulting in the final optimized frequency distribution parameter set (mean 51.5 Hz, standard deviation 4.8 Hz). This parameter set was applied to data stream reconstruction. Spectral analysis verified that the distinguishability of data streams in noisy environments was reduced to 0.02, significantly improving the concealed transmission capability. Using a border surveillance data transmission scenario as an example, this parameter adjustment enabled the system to automatically adapt to changes in complex electromagnetic environments, reducing the system false alarm rate from 10% to 2%, ensuring transmission reliability.

[0034] In step S16, adaptive modulation and spectrum shaping are performed on the second encrypted data stream based on the frequency distribution parameter set, and a concealed transmission data stream is output.

[0035] In one embodiment, according to the frequency distribution parameter set, the second encrypted data stream is matched through a preset frequency distribution model to obtain an adjusted parameter configuration and generate a first covert transmission data stream; When it is detected that the signal strength of the first covert transmission data stream in the target frequency band is lower than a threshold, a signal enhancement technology is used for optimization to generate a second covert transmission data stream; Analyzing the transmission characteristics of the second covert transmission data stream using a support vector machine algorithm to obtain a stability evaluation result; When the stability evaluation result does not meet a preset stability standard, modulating the second covert transmission data stream to generate a third covert transmission data stream; extracting multi-dimensional features of the third covert transmission data stream, verifying transmission performance of the third covert transmission data stream in a target frequency band, and outputting a transmission verification result; According to the transmission verification result, the covert transmission data stream is output and relevant analysis data is stored.

[0036] Specifically, the initial encrypted data stream undergoes first-level encryption based on the optimized frequency distribution parameter set. Assuming the parameter set includes a spectrum allocation scheme with a center frequency of 2.4 GHz and a bandwidth of 20 MHz, the data is divided into 64 subcarriers using orthogonal frequency division multiplexing (OFDM) with a subcarrier spacing of 312.5 kHz. The data is mapped to the frequency domain using an inverse fast Fourier transform (IFFT), with the peak-to-average ratio (PAPR) controlled to no more than 6 dB to reduce nonlinear distortion. This results in a spectral efficiency of 4 bits / s / Hz, generating the first covert transmission data stream. When the signal strength of the first covert transmission data stream in the target frequency band (2.38 GHz-2.42 GHz) falls below a preset threshold, the second-level encryption optimization is initiated. A pseudorandom sequence (PN) code (PN code length 127) is used for spread spectrum processing, extending the signal bandwidth to 40 MHz and maintaining a signal-to-background noise correlation coefficient of less than 0.1 to enhance anti-interception capabilities. Signal enhancement technology is also applied, with a power control algorithm increasing the transmit power by 3 dB to generate the second covert transmission data stream. The support vector machine algorithm is used to analyze the transmission characteristics of the second covert transmission data stream, constructing a multidimensional feature vector (including spectral flatness, phase jitter, and intersymbol interference). Under a Rayleigh fading channel model, the signal is corrected using a minimum mean square error equalization algorithm, achieving a bit error rate of 0.001% and an improved signal-to-noise ratio of 15dB. If the evaluation results indicate a stability of less than 99.9% (a preset standard), the system enters the third-level encryption process. This third-level encryption utilizes adaptive modulation technology, dynamically adjusting the modulation order based on real-time channel status information. When signal fluctuations caused by multipath fading are detected, 64QAM modulation is automatically downgraded to 16QAM (the communication system automatically adjusts the modulation method based on channel conditions). Turbo codes are also introduced for forward error correction, with a coding rate set to 3 / 4. Constellation rotation and power injection algorithms are used to further optimize the signal distribution, generating the third covert transmission data stream. Multidimensional features of the third covert transmission data stream (such as time-frequency domain features, high-order statistical features, and wavelet transform coefficients) are extracted, and transmission performance is tested in the target frequency band. A spectrum sensing algorithm detects adjacent channel interference. If WiFi signal interference is detected in the 2.4 GHz band, it automatically switches to 2.41 GHz and resynchronizes the PN code. Tests show that signal interruption caused by frequency switching is less than 1ms, meeting real-time requirements. Ultimately, the system outputs a covert transmission data stream optimized with three levels of encryption. Key processing parameters (such as modulation mode, spreading gain, and error correction coding rate) are stored in a historical database. When similar channel conditions arise in the future, the AI model can directly recall the historically optimized parameter configuration for rapid adaptive adjustment.

[0037] In one embodiment, the AI-based adaptive data encryption method further includes: At the receiving end, performing demodulation frequency identification processing on the concealed transmission data stream to obtain a first demodulated data stream; performing joint frequency offset and phase correction on the first demodulated data stream in combination with the channel state information and the dynamic parameters of the transmitting end to generate a second demodulated data stream; The second demodulated data stream is verified and processed. When data loss or error occurs, a frequency parameter retrospective adjustment mechanism is triggered to output a final data stream.

[0038] Specifically, the receiver first demodulates the concealed transmission data stream to identify the modulation frequency. Using a 44100Hz sampling rate, the fast Fourier transform algorithm converts the time-domain signal into a frequency-domain signal. Spectral analysis reveals that the primary frequency components are concentrated between 500Hz and 2000Hz, with a peak frequency at 1200Hz. Further analysis of the spectral energy distribution reveals that the modulation frequency range is 800Hz to 1600Hz, with an error range of ±50Hz. Subsequently, a coherent demodulation method is employed, using a 1200Hz carrier frequency. A 400Hz bandpass filter (800Hz to 1600Hz) is used to filter out noise. The signal is then mixed with a local 1200Hz sine wave and processed through a low-pass filter with a cutoff frequency of 200Hz. Binary data with a bit rate of 9600bps is extracted, resulting in a bit error rate of approximately 2.3%. To optimize the performance, a Hamming code error correction mechanism is introduced to reduce the bit error rate to below 0.5%, resulting in the first demodulated data stream. Next, the system combines channel state information with dynamic transmitter parameters to perform joint frequency offset and phase correction on the first demodulated data stream. Assuming the initial demodulated data stream center frequency is 10.5MHz and the actual received signal center frequency is 10.52MHz, a fast Fourier transform algorithm with 0.001MHz frequency resolution is used to accurately calculate a 0.02MHz frequency offset. This frequency shift is then achieved using digital down-conversion technology multiplied by the frequency correction factor. For an initial phase deviation of 0.3 radians, a phase-locked loop algorithm is used with an iteration step of 0.01 radians. The number of iterations is dynamically adjusted based on the signal-to-noise ratio (SNR): 10 iterations for a SNR of 20dB, increasing to 15 for a SNR below 15dB. This reduces the phase deviation to within 0.01 radians, resulting in a second demodulated data stream with a bit error rate of 0.001% (below the 0.01% threshold). Finally, the second demodulated data stream is verified. Assuming the demodulated data stream is "101101001011", by calculating the total number and characteristic value of "1" in the original data stream and the demodulated data stream, it is found that the difference between the two is 1, which exceeds the preset threshold of 0.5, and the data is determined to be abnormal. The frequency parameter retrospective adjustment mechanism is then triggered. The system extracts the 10.5MHz frequency value of the most recent successful demodulation from the historical frequency parameter database, and attempts to decrease the frequency value in steps of 0.1MHz. When it is adjusted to 10.4MHz and demodulated again, the characteristic value of the new data stream "101101001010" is consistent with the original data, and the verification is determined to be passed, and the final data stream is output. If the verification still fails after adjustment to 9.5MHz, a log is recorded and an alarm is sent to the system monitoring module.

[0039] refer to Figure 2 The second embodiment of the invention provides an AI-based adaptive data encryption system, comprising: The frequency domain scanning module performs frequency domain transformation analysis on the original data stream to generate a set of spectrum distribution features; A frequency screening module, which screens the carrier frequency according to the spectrum distribution feature set using a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range; A first encryption module, which embeds the data bit stream of the original data stream into a subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; a second encryption module that monitors interference information in the transmission environment in real time, and when the interference intensity exceeds a preset interference threshold, adjusts the frequency offset parameter in a directionally appropriate manner according to the interference type, thereby optimizing the first encrypted data stream to generate a second encrypted data stream; The dynamic adjustment module dynamically adjusts the frequency distribution parameters according to the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set; The concealed transmission module performs adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set, and outputs a concealed transmission data stream.

[0040] It should be noted that the AI-based adaptive data encryption system provided in an embodiment of the present invention is used to execute all the process steps of the AI-based adaptive data encryption method in the above embodiment. The working principles and beneficial effects of the two correspond one to one, so they will not be repeated here.

[0041] An embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a data acquisition program. When the processor executes the computer program, the steps in the above-mentioned embodiments of the AI-based adaptive data encryption method are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are realized, such as the data acquisition module.

[0042] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device.

[0043] The electronic device may be a computing device such as a desktop computer, notebook, PDA, or smart tablet. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will appreciate that the aforementioned components are merely examples of electronic devices and do not constitute a limitation of the electronic device. The electronic device may include more or fewer components than those described above, or a combination of certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, and the like.

[0044] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, connecting various parts of the entire electronic device using various interfaces and lines.

[0045] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and accessing data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area may store data generated based on the use of the mobile phone (such as audio data, a phone book, etc.). In addition, the memory may include high-speed random access memory and non-volatile memory, such as a hard disk, internal memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one disk storage device, a flash memory device, or other volatile solid-state storage device.

[0046] If the module / unit integrated into the electronic device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the process steps in the above-mentioned method embodiments by using a computer program to instruct the relevant hardware. The computer program can be stored in a computer-readable storage medium. When executed by a processor, the computer program can implement the steps of each of the above-mentioned method embodiments. The computer program includes computer program code, which can be in source code form, object code form, executable file, or some intermediate form. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard drive, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium. It should be noted that the content of the computer-readable medium can be appropriately increased or decreased based on the requirements of legislation and patent practice in a jurisdiction. For example, in some jurisdictions, based on legislation and patent practice, computer-readable media does not include electric carrier signals and telecommunication signals.

[0047] It should be noted that the device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed across multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement the present invention without inventive effort.

[0048] The specific embodiments described above further illustrate the objectives, technical solutions, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. An AI-based adaptive data encryption method, characterized in that: include: Perform frequency domain transform analysis on the original data stream to generate a spectrum distribution feature set; According to the spectrum distribution feature set, the carrier frequency is screened by a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range; Embedding the data bit stream of the original data stream into the subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; monitoring interference information in the transmission environment in real time, and when the interference intensity exceeds a preset interference threshold, performing directionally adjustment on the frequency offset parameter according to the interference type, optimizing the first encrypted data stream to generate a second encrypted data stream; According to the transmission distance data fed back by the receiving end, the frequency distribution parameters are dynamically adjusted to obtain an optimized frequency distribution parameter set; Based on the frequency distribution parameter set, adaptive modulation and spectrum shaping are performed on the second encrypted data stream to output a concealed transmission data stream.

2. The AI-based adaptive data encryption method according to claim 1, characterized in that: The frequency domain transform analysis of the original data stream to generate a spectrum distribution feature set includes: Obtaining spectrum distribution information in the original data stream through frequency domain transformation analysis to obtain first spectrum distribution data; extracting spectrum characteristic data from the first spectrum distribution data, and screening out the spectrum characteristic data whose spectrum characteristics meet a preset screening criterion, to obtain a spectrum characteristic data group; The spectrum characteristic data group is classified to obtain the spectrum distribution feature set.

3. The AI-based adaptive data encryption method according to claim 1, characterized in that: The method of screening the carrier frequency according to the spectrum distribution feature set by using a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range includes: According to the spectrum distribution feature set, real-time channel quality parameters are obtained, and a channel state assessment result is obtained by comparing the change rate of the current channel environment with a preset spectrum distribution feature database; Based on the channel state evaluation result, using the dynamic frequency selection algorithm to screen a subset of candidate frequencies that meet the channel adaptation condition from available carriers, and determining a candidate frequency range; Performing a transmission quality test on the candidate frequency subset, retaining frequencies that meet a preset quality standard, and forming an optimized frequency combination; The optimized frequency combination is calibrated according to a preset modulation strategy to determine the optimal carrier frequency range.

4. The AI-based adaptive data encryption method according to claim 1, characterized in that: The step of embedding the data bit stream of the original data stream into a subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream includes: Performing preliminary analysis on the data bit stream to obtain basic characteristic information and determine the carrier frequency range; performing initial processing on the data bit stream using a frequency modulation technique according to the carrier frequency range to generate a preliminary modulated data stream; Applying a frequency offset adjustment technology to the preliminary modulated data stream to obtain adjusted frequency distribution data to obtain first frequency distribution data; The first frequency distribution data is secondary processed by using a phase modulation technology, and the data bit stream is embedded into a subcarrier within the optimal carrier frequency range to generate the first encrypted data stream.

5. The AI-based adaptive data encryption method according to claim 1, characterized in that: The real-time monitoring of interference information in the transmission environment, and when the interference intensity exceeds a preset interference threshold, directionally adjusting the frequency offset parameter according to the interference type, and optimizing the first encrypted data stream to generate a second encrypted data stream, includes: Monitor interference information in the transmission environment in real time and extract interference feature data; Comparing the interference feature data with a preset interference pattern library to obtain interference analysis results; When the interference intensity in the interference analysis result exceeds a preset interference intensity threshold, calculating an initial adjustment range of the frequency offset parameter; Using a support vector machine algorithm to optimize the initial adjustment range and output adjustment parameters; The frequency offset of the encrypted data stream is adjusted according to the adjustment parameter, and the first encrypted data stream is optimized to generate the second encrypted data stream.

6. The AI-based adaptive data encryption method according to claim 1, characterized in that: The frequency distribution parameters are dynamically adjusted based on the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set, including: Obtaining initial frequency distribution parameter values, and establishing a corresponding relationship model between parameters and distances based on the law of changes in the frequency distribution parameter values and transmission distances; According to the corresponding relationship model, it is determined whether the current frequency distribution parameter value exceeds the preset allowable range; if it exceeds, the parameter is compensated in combination with the influencing factor of the transmission distance to generate a frequency distribution parameter group after preliminary correction; collecting interference data in the current communication environment in real time according to the frequency distribution parameter group; Using a support vector machine algorithm to classify the interference data and determine the impact level of each type of interference on the frequency parameters; According to the impact level, the weight distribution ratio of the frequency distribution parameters is adjusted, and the frequency distribution parameter set is output.

7. The AI-based adaptive data encryption method according to claim 1, characterized in that: The step of performing adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set to output a covert transmission data stream includes: According to the frequency distribution parameter set, matching the second encrypted data stream with a preset frequency distribution model to obtain an adjusted parameter configuration and generate a first covert transmission data stream; When it is detected that the signal strength of the first covert transmission data stream in the target frequency band is lower than a threshold, a signal enhancement technology is used for optimization to generate a second covert transmission data stream; Analyzing the transmission characteristics of the second covert transmission data stream using a support vector machine algorithm to obtain a stability evaluation result; When the stability evaluation result does not meet a preset stability standard, modulating the second covert transmission data stream to generate a third covert transmission data stream; extracting multi-dimensional features of the third covert transmission data stream, verifying transmission performance of the third covert transmission data stream in a target frequency band, and outputting a transmission verification result; According to the transmission verification result, the covert transmission data stream is output and relevant analysis data is stored.

8. The AI-based adaptive data encryption method according to any one of claims 1 to 7, characterized in that: After outputting the covert transmission data stream, it also includes: At the receiving end, performing demodulation frequency identification processing on the concealed transmission data stream to obtain a first demodulated data stream; performing joint frequency offset and phase correction on the first demodulated data stream in combination with the channel state information and the dynamic parameters of the transmitting end to generate a second demodulated data stream; The second demodulated data stream is verified and processed. When data loss or error occurs, a frequency parameter retrospective adjustment mechanism is triggered to output a final data stream.

9. An AI-based adaptive data encryption system, characterized in that: include: The frequency domain scanning module performs frequency domain transformation analysis on the original data stream to generate a set of spectrum distribution features; A frequency screening module, which screens the carrier frequency according to the spectrum distribution feature set using a preset dynamic frequency selection algorithm to obtain an optimal carrier frequency range; A first encryption module, which embeds the data bit stream of the original data stream into a subcarrier within the optimal carrier frequency range through phase modulation to generate a first encrypted data stream; a second encryption module that monitors interference information in the transmission environment in real time, and when the interference intensity exceeds a preset interference threshold, adjusts the frequency offset parameter in a directionally appropriate manner according to the interference type, thereby optimizing the first encrypted data stream to generate a second encrypted data stream; The dynamic adjustment module dynamically adjusts the frequency distribution parameters according to the transmission distance data fed back by the receiving end to obtain an optimized frequency distribution parameter set; The concealed transmission module performs adaptive modulation and spectrum shaping on the second encrypted data stream based on the frequency distribution parameter set, and outputs a concealed transmission data stream.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the AI-based adaptive data encryption method according to any one of claims 1 to 8.

Images