Data encryption method based on multistage dynamic iteration mechanism

Abstract

The invention discloses a data encryption method based on a multi-stage dynamic iteration mechanism, and solves the problem that hidden dangers are brought to system security due to the fact that secret key information is repeatedly transmitted in pseudo noise information in the existing chaotic encryption technology. According to the method, encryption information in a 16 QAM modulation format and a key sequence in a QPSK modulation format are superposed into 64QAM signals for transmission by adopting different powers, encryption dimensions and a chaos model are embedded into small-power QPSK signals at the same time, and a multi-stage dynamic iteration mechanism is combined, so that more dynamic and flexible key information updating is realized, the safety performance of a system is improved, and the method is suitable for large-scale popularization and application. And potential safety hazards caused by a fixed encryption mode and encrypted information are avoided.

Description

Technical Field

[0001] This invention relates to data encryption methods, specifically to a data encryption method based on a multi-level dynamic iteration mechanism. Background Technology

[0002] The exponential growth of data traffic generated by network applications has laid the foundation for the development of next-generation optical networks, which aim to achieve higher speeds, flexibility, and reliability. As one of the key enabling technologies for the physical layer of next-generation optical communication, superposition coding technology has received widespread attention from academia and industry. This method reduces the need for additional spectrum resources and makes the decoding process more flexible. Furthermore, superposition coding is one of the key technologies for realizing Non-Orthogonal Multiple Access (NOMA). Before signal superposition, NOMA allocates different power levels to different users in the power domain. When combined with Orthogonal Frequency Division Multiplexing (OFDM), NOMA-OFDM systems can achieve multi-dimensional multiplexing in both the frequency and power domains. NOMA-OFDM systems offer advantages such as higher spectral efficiency, more flexible access capabilities, and lower latency. As the backbone of optical fiber communication, it effectively addresses the limitations of system resource utilization and network user capacity. Therefore, it allows the system to reach a better operating point, thereby optimizing spectral efficiency and energy efficiency.

[0003] While optical communication systems have made some progress in improving spectral efficiency, the increasing number of users and transmission capacity has made signal transmission security a growing focus for researchers. There are generally two main methods to enhance optical communication security: protocol-based upper-layer encryption schemes and physical layer encryption schemes. Data Encryption Standard (DES) and Advanced Encryption Standard (AES), as classic upper-layer algorithms, have received extensive research. However, with the rapid development of quantum computing, the keys used in these algorithms can now be cracked in a short time through brute force. Furthermore, protocol-based upper-layer encryption schemes have gradually revealed problems such as header information leakage and difficulties in protocol management. Therefore, increasing attention has turned to physical layer encryption methods to enhance system security. To date, researchers have proposed various physical layer encryption technologies, including chaotic lasers, quantum key distribution, and digital chaotic perturbations. However, chaotic laser encryption schemes are limited by laser bandwidth and fiber dispersion effects, restricting the overall performance improvement of the system. While quantum key distribution technology can detect unauthorized attacks, its key generation rate is too low to meet the demands of high-speed optical communication. Digital chaotic perturbation technology has gained widespread attention due to its advantages such as high initial value sensitivity, good randomness, and large key space. In NOMA-OFDM systems, signal generation, modulation, encryption, and power distribution can all be completed through digital signal processing modules, which provides convenient conditions for improving system security using digital chaotic perturbation technology. However, chaotic encryption technology based on digital signal processing (DSP) has several inherent security flaws—the chaotic model, encryption dimension, and initial key used to generate the key must all be shared in advance between the transmitter and receiver and cannot be changed once transmitted. Because it relies on fixed security information shared between the transmitter and receiver, the lack of key flexibility makes the system vulnerable to attacks. To address this problem, a chaotic encryption scheme based on a key-following mechanism has been proposed. By embedding the key information in a pseudo-noise signal and superimposing it on the encrypted signal for transmission, synchronous key updates are achieved. However, since the key information is repeatedly transmitted within the pseudo-noise signal, if the signal is obtained by an unauthorized receiver, it still poses a security risk to the system. Furthermore, when system security requirements change, the security level of the encryption module cannot be dynamically adjusted, limiting the practical application of this technology. Summary of the Invention

[0004] To address the technical problem in existing chaotic encryption technologies where key information is repeatedly transmitted within pseudo-noise information, posing a security risk to the system, this invention provides a data encryption method based on a multi-level dynamic iteration mechanism.

[0005] To achieve the above objectives, the present invention adopts the following technical solution:

[0006] A data encryption method based on a multi-level dynamic iteration mechanism is characterized by the following steps:

[0007] Step A:

[0008] A1. A three-dimensional perturbation sequence is generated by processing the dynamic key sequence through a multi-level dynamic iteration mechanism; the three-dimensional perturbation sequence includes perturbation sequence X, perturbation sequence Y, and perturbation sequence Z;

[0009] A2. Perform serial-to-parallel conversion, convolutional coding, and QPSK modulation on the three-dimensional perturbation sequence in sequence to obtain the QPSK signal;

[0010] A3. Perform a fast inverse Fourier transform on the QPSK signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a QPSK signal of length T.

[0011] A4. Power distribution is performed on a QPSK signal of length T to obtain a low-power QPSK signal;

[0012] Step B:

[0013] B1. Bit encryption is performed on the data to be encrypted using the perturbation sequence X, and serial-to-parallel conversion, convolution numbering, and 16QAM modulation are performed sequentially to obtain a 16QAM signal;

[0014] B2. Perform two-dimensional joint scrambling encryption of the 16QAM signal using perturbation sequences Y and Z, involving subcarriers and symbols.

[0015] B3. Perform a fast inverse Fourier transform on the encrypted 16QAM signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a 16QAM signal of length T.

[0016] B4. Perform power allocation on a 16QAM signal of length T to obtain a high-power 16QAM signal; the power allocated to the high-power 16QAM signal is greater than the power allocated to the low-power QPSK signal.

[0017] Step C:

[0018] The low-power QPSK signal obtained in step A4 is superimposed and electro-optically modulated with the high-power 16QAM signal obtained in step B4 to generate a NOMA-OFDM optical transmission signal, thus completing the data transmission performance optimization.

[0019] Further, step A1 specifically includes:

[0020] The CMI bit data of the data to be encrypted is identified. Based on the correspondence between Chen's chaotic model and Lorenz chaotic model and 0 and 1, and the CMI bit data, Chen's chaotic model or Lorenz chaotic model is switched in real time to process the dynamic key sequence and generate a three-dimensional perturbation sequence.

[0021] Furthermore, in step A1, Chen's chaos model is as follows:

[0022]

[0023] Where α1, β1, and γ1 are the control parameters of Chen's chaotic model, and x1, y1, and z1 are the initial values ​​of the perturbation sequence of Chen's chaotic model. Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

[0024] Further, in step A1, the Lorenz chaotic model is as follows:

[0025]

[0026] Where α2, β2, and γ2 are the control parameters of the Lorenz chaotic model, and x2, y2, and z2 are the initial values ​​of the perturbation sequence of the Lorenz chaotic model. Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

[0027] Furthermore, in step A1:

[0028] The values ​​of α1, β1, and γ1 are respectively (-40, 40), (-10, 10), and (-30, 30);

[0029] The values ​​of x1, y1, and z1 are respectively (-20, 20), (-20, 20), and (-20, 20).

[0030] Furthermore, in step A1:

[0031] The values ​​of α2, β2, and γ2 are respectively (-10, 10), (-10, 10), and (-30, 30);

[0032] The ranges of x2, y2, and z2 are (-20, 20), (-20, 20), and (0, 50), respectively.

[0033] When using the Lorenz chaotic model, it is also necessary to subtract 25 from z2 beforehand to ensure that z2(0)∈(-25,25).

[0034] Further, in step A2, the convolutional encoding specifically includes:

[0035] Channel coding is performed using a convolutional code with a constraint length L=4 and a generator polynomial of [13,17]. The code rate is configured to be 1 / 2, the number of encoder states is 8, and the trace depth is set to 10 clock cycles.

[0036] Furthermore, step A3 specifically involves:

[0037] Perform a 1024-point inverse fast Fourier transform on the QPSK signal, then insert a cyclic prefix and perform parallel-to-serial conversion.

[0038] Furthermore, step B3 specifically involves:

[0039] The 16QAM signal after symbol encryption is subjected to a 1024-point fast inverse Fourier transform, followed by the insertion of a cyclic prefix and parallel-to-serial conversion.

[0040] Furthermore, in step A1:

[0041] The control parameters α1, β1, and γ1 of the Chen's chaotic model are specifically: α1 = 35, β1 = 3, and γ1 = 28.

[0042] The initial values ​​of the perturbation sequence x1, y1, and z1 in the Chen's chaotic model are specifically: x1(0) = 0.995, y1(0) = 0.995, z1(0) = 0.995;

[0043] The control parameters α2, β2, and γ2 of the Lorenz chaotic model are specifically: α2 = 10, β2 = 2.667, and γ2 = 28;

[0044] The initial values ​​of the perturbation sequence x2, y2, z2 of the Lorenz chaotic model are specifically: x2(0) = 12, y2(0) = 2, z2(0) = 9.

[0045] The beneficial effects of this invention are:

[0046] 1. The present invention provides a data encryption method based on a multi-level dynamic iteration mechanism, which superimposes 16QAM modulation format encryption information and QPSK modulation format key sequence into a 64QAM signal with different powers for transmission, and embeds encryption dimension and chaotic model into low-power QPSK signal at the same time. Combined with the multi-level dynamic iteration mechanism, it realizes more dynamic and flexible key information update, improves the security performance of the system, and avoids the security risks caused by fixed encryption methods and encryption information.

[0047] 2. This invention provides a data encryption method based on a multi-level dynamic iteration mechanism. This mechanism updates the key sequence in real time. When the transmission sequence at the current moment is cracked, the actual parameter values ​​still require the transmission information from the current moment and all previous moments. This overcomes the security vulnerabilities caused by fixed encryption mechanisms, chaotic models, and perturbation sequences. The system's resistance to brute-force attacks can be increased by up to 10%. 95 times. Attached Figure Description

[0048] Figure 1 These are schematic diagrams of resource block distribution for OFDM and NOMA-OFDM, where (a) is a schematic diagram of resource block distribution for OFDM and (b) is a schematic diagram of resource block distribution for NOMA-OFDM.

[0049] Figure 2 These are constellation distribution feature diagrams of high-power 16QAM signals, low-power QPSK signals, and superimposed 64QAM signals in embodiments of the present invention; wherein, (a) is a constellation distribution feature diagram of high-power 16QAM signals, (b) is a constellation distribution feature diagram of low-power QPSK signals, and (c) is a constellation distribution feature diagram of superimposed 64QAM signals.

[0050] Figure 3 This is a schematic diagram illustrating the specific implementation principle of the multi-level dynamic iteration mechanism in this embodiment of the invention.

[0051] Figure 4 This is a flowchart of an embodiment of a data encryption method based on a multi-level dynamic iteration mechanism according to the present invention;

[0052] Figure 5 These are the xyz phase space trajectory diagrams of the attractors in Chen's chaotic model and Lorenz chaotic model in the embodiments of the present invention; wherein, (a) is the xyz phase space trajectory diagram of the attractor in Chen's chaotic model, and (b) is the xyz phase space trajectory diagram of the attractor in Lorenz chaotic model.

[0053] Figure 6 This is a schematic diagram illustrating the key sequence configuration and construction principle in an embodiment of the present invention;

[0054] Figure 7 This is an experimental test system diagram used to test a data encryption method based on a multi-level dynamic iteration mechanism in an embodiment of the present invention;

[0055] Figure 8 This is a bit error rate curve of the seven-core optical fiber performance test in an embodiment of the present invention;

[0056] Figure 9 This is a comparison chart of the bit error rate curves of the encryption method proposed in this embodiment of the invention and an illegal receiver. Detailed Implementation

[0057] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] like Figure 1 As shown, multiple signals in an OFDM-NOMA system are assigned different powers and superimposed at the transmitter. The power-division multiplexed signal q can be mathematically represented as:

[0059]

[0060] In the formula, And q i and p i This represents the signal and signal power of the i-th user. In the process of forming the transmitted NOMA-OFDM signal in this embodiment of the invention, high-power and low-power signals can be superimposed using different modulation formats to form multiplexed signals with different modulation formats. In this way, different transmission spectral efficiencies can be flexibly achieved according to channel characteristics. This embodiment of the invention takes a scheme where the low-power signal uses QPSK modulation, the high-power signal uses 16QAM modulation, and the NOMA-OFDM signal forms a 64QAM modulation format as an example. In this 64QAM NOMA-OFDM signal, each symbol carries 6 effective bits of information; the first two bits contain the effective information from the low-power QPSK signal, and the last four bits contain the effective information from the high-power 16QAM signal.

[0061] To prevent eavesdroppers from simultaneously intercepting and cracking the key and data during transmission, this invention employs a multi-level dynamic iterative mechanism. In the signal preprocessing stage, the high-power user data stream undergoes bit-level encryption, while the low-power user stream carries the initial parameter information for the encryption module. First, the key sequence at this moment is updated to the transmission sequence of the low-power signal through a multi-level dynamic iterative mechanism. Then, both signals undergo serial-to-parallel conversion and are modulated using convolutional coding and 16QAM and QPSK to generate OFDM symbols in the frequency domain. To improve the spectral efficiency of signal transmission, 16QAM and QPSK modulation schemes are used for transmitting high-power and low-power signals, respectively. During signal encryption, the high-power signal is subjected to two-dimensional joint scrambling encryption of subcarriers and symbols based on the key sequence. Subsequently, both signals are converted to the time domain using a 1024-point inverse Fast Fourier Transform (IFFT), and a cyclic prefix is ​​inserted to suppress inter-symbol interference caused by multipath effects. After parallel-to-serial conversion and power allocation, the 16QAM and QPSK signals are superimposed to generate a NOMA-OFDM signal, ultimately resulting in a multiplexed signal with a 64QAM modulation format. A training sequence is embedded before the IFFT processing of the high-power signal for channel state estimation and equalization at the receiver. Figure 2 The constellation distribution characteristics of high-power 16QAM signals, low-power QPSK signals, and superimposed 64QAM signals are shown.

[0062] In this method, the encryption dimension and Chen's and Lorenz chaotic models are switched in real time based on binary chaotic map identifiers (CMIs), and the parameter set of the key sequence is updated iteratively through an XOR operation. Switching chaotic models achieves the interleaving of chaotic models, similar in principle to the dynamic iteration of the key sequence, with the main purpose of further improving system security. Although high-dimensional chaotic models already possess extremely high complexity, with the development of technologies such as quantum computing, a single chaotic model still faces the risk of being deciphered. In this embodiment, through model switching and dynamic iteration operations, the key is generated by the interleaving of two chaotic models. Even when the chaotic model and parameters are deciphered, it is still difficult for an unauthorized receiver to crack the chaotic model and key information used at each moment, achieving a significant improvement in security performance.

[0063] like Figure 3 As shown, at time t, the model to which the parameter set belongs is first determined based on the received CMI: if the CMI is the same as at time t-1, there is no need to switch the chaotic model. Otherwise, it switches to another chaotic model. Similarly, the three-bit encryption dimension identifiers indicate the encryption operations in the three dimensions of bits, subcarriers, and symbols, and the security level is updated according to the changes in the identifiers. Initially, all dimension identifiers are set to 0, and a three-dimensional encryption scheme using Chen's chaotic model is adopted by default.

[0064] At time t, the set of parameters in the key sequence is compared with the actual parameter value key output at time t-1. t-1 Perform an XOR operation to generate the updated real parameter value key. t At this point, the actual parameter value is key. t It is obtained by iteratively XORing the parameter sets in the key sequence at each previous time step. Let u be the parameter set in the key sequence at time t. t key t The specific formula is as follows:

[0065]

[0066] This operation is then used to obtain the true parameter set key for the encrypted sequence generated by the chaotic model. t It exhibits strong temporal correlation. Specifically, it requires obtaining the key sequence received at all times up to time t in order to acquire the true parameter set and the chaotic model at the current time, thereby successfully decrypting the information.

[0067] At the transmitting end, assume the perturbation sequence r used at time t for encrypting bits, subcarriers, and symbol data. t Let there be three sets of sequences, denoted as (r 1t ,r 2t ,r 3t The sender, based on (r) 1t ,r 2t ,r 3t The high-power signal's bits, subcarrier, and symbols are input into the encryption module for encryption. During bit encryption, each bit of the high-power signal is encrypted with the perturbation sequence r. 1t The XOR operation is performed, and the specific encryption process is as follows:

[0068]

[0069] In the formula, X is the perturbation sequence, mod(·) is the remainder function, and floor(·) is the floor function. s represents the original binary bit stream, and S represents the encrypted binary bit stream. Subsequently, the subcarriers of the OFDM signal and the symbols on each subcarrier can be viewed as a two-dimensional matrix. Each column of this two-dimensional matrix represents each subcarrier, and each element represents a symbol on the corresponding subcarrier. Finally, based on the chaotic sequence r... 2t and r 3t The order of elements in the middle is used to shuffle the order of subcarriers and symbols.

[0070] At the receiving end, the key at time t is obtained. t Then, the parameters are input into the specified chaotic model to obtain the perturbation sequence (r) used for encryption operations. 1t ,r 2t ,r3t The decryption process is used to recover bit information, subcarriers, and sequence order. In this case, the decryption steps are the same as the encryption principle, but in reverse order. The received signal, after continuous recovery of symbols, subcarriers, and bits, undergoes S / P transformation to finally obtain the decrypted original data.

[0071] Specifically, this invention provides a data encryption method based on a multi-level dynamic iteration mechanism, such as... Figure 4 As shown, the data encryption method includes the following steps:

[0072] Step A:

[0073] A1. A three-dimensional perturbation sequence is generated by processing the dynamic key sequence through a multi-level dynamic iteration mechanism; the three-dimensional perturbation sequence includes perturbation sequence X, perturbation sequence Y, and perturbation sequence Z; specifically:

[0074] The CMI bit data of the data to be encrypted is identified. Based on the correspondence between Chen's chaotic model and Lorenz chaotic model and 0 and 1, and the CMI bit data, Chen's chaotic model or Lorenz chaotic model is switched in real time to process the dynamic key sequence and generate a three-dimensional perturbation sequence.

[0075] Traditional chaos-based encryption schemes employ chaotic models with higher dimensionality, resulting in higher model complexity and better performance of the generated chaotic sequences. Therefore, the choice of chaotic model is a crucial factor in determining the security of an encryption strategy. Selecting a three-dimensional chaotic model to generate sequences can improve system security and match three-dimensional (bit, subcarrier, symbol) encryption schemes, meaning one sequence is used for encryption of one dimension of information. This invention is theoretically compatible with any chaotic model. In this embodiment, Chen's chaotic model and Lorenz chaotic model are used as examples to generate three-dimensional perturbation sequences, and the model is switched and updated in real time based on the CMI bit data of the data to be encrypted.

[0076] Chen's chaos model is as follows:

[0077]

[0078] Where α1, β1, and γ1 are the control parameters of Chen's chaotic model, and x1, y1, and z1 are the initial values ​​of the perturbation sequence of Chen's chaotic model. Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

[0079] To ensure that the parameter values ​​can achieve the chaotic characteristics of the model, the ranges of α1, β1, and γ1 are set to (-40, 40), (-10, 10), and (-30, 30), respectively; and the ranges of x1, y1, and z1 are set to (-20, 20), (-20, 20), and (-20, 20), respectively.

[0080] In this embodiment, the initial values ​​of the perturbation sequence of Chen's chaotic model are specifically set as x1(0) = 0.995, y1(0) = 0.995, z1(0) = 0.995, and the control parameters are specifically set as α1 = 35, β1 = 3, γ1 = 28, resulting in the following... Figure 5 The trajectory of the chaotic attractor in the xyz phase space is shown in (a).

[0081] The Lorenz chaos model is as follows:

[0082]

[0083] Where α2, β2, and γ2 are the control parameters of the Lorenz chaotic model, and x2, y2, and z2 are the initial values ​​of the perturbation sequence of the Lorenz chaotic model; Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

[0084] Similarly, to ensure that the parameter values ​​can achieve the chaotic characteristics of the model, the ranges of α2, β2, and γ2 are set to (-10,10), (-10,10), and (-30,30), respectively; the ranges of x2, y2, and z2 are set to (-20,20), (-20,20), and (0,50), respectively.

[0085] In this embodiment, the control parameters of the Lorenz chaotic model are specifically set as α2 = 10, β2 = 2.667, and γ2 = 28, and the initial values ​​of the perturbation sequence are specifically set as x2(0) = 12, y2(0) = 2, and z2(0) = 9. The trajectory of the chaotic attractor in the xyz phase space of this model is as follows: Figure 5 As shown in (b).

[0086] When using the Lorenz chaotic model, it is also necessary to subtract 25 from z2 beforehand to ensure that z2(0)∈(-25,25) to ensure that the model is in a chaotic state. During the first information transmission, the dimension identifiers are all set to 0, and CMI=0, α1=5.123235792165831, β1=3.473217412334834, γ1=28.945623124845713, x1(0)=0.995736152473421, y1(0)=0.956577344138231, z1(0)=0.972837456272828.

[0087] like Figure 6 As shown, due to the good sensitivity of Chen's chaotic model and Lorenz chaotic model to initial conditions, the output sequence can be within 10% of the initial value of the perturbation sequence and the control parameter. -15 Significant changes occur under these conditions. During transmission, the initial value structure of this scheme includes the initial values ​​of the perturbation sequence from the chaotic model and control parameters. Each decimal number is converted into four binary bits. Therefore, each initial parameter retains 77 binary bits, including a sign bit, a 4×4 bit integer part, and a 15×4 bit fractional part. Specifically, for positive and negative numbers, the sign bit of each parameter is represented by "1" and "0", respectively. Furthermore, a 3-bit binary encryption dimension identifier and a 1-bit binary chaos map identifier (CMI) are added at the beginning of the key sequence, indicating whether the encryption dimension and the chaotic model used at the current moment are switched, respectively. This scheme uses two different chaotic models for cross-switching, thus requiring only one binary identifier. In practical applications, the chaotic model and encryption dimension can be adjusted according to security requirements.

[0088] After the key sequence configuration is completed, the initial setup segment is repeated multiple times to construct a low-power pseudo-noise key signal with the same length as the high-power signal. This repetition can further assist the decoding process of the low-power signal at the receiver: by comparing and correcting errors in the repeated sequence, the initial value segment can be recovered more effectively.

[0089] A2. Perform serial-to-parallel conversion, convolutional coding, and QPSK modulation on the three-dimensional perturbation sequence in sequence to obtain the QPSK signal;

[0090] The specific steps of convolutional coding are as follows: channel coding is performed using a convolutional code with a constraint length L=4 and a generator polynomial of [13,17]. The code rate is configured to be 1 / 2, and the number of encoder states is 8. The trace depth is set to 10 clock cycles to balance decoding performance and latency.

[0091] A3. Perform an inverse Fast Fourier Transform on the QPSK signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a QPSK signal of length T; specifically:

[0092] Perform a 1024-point inverse fast Fourier transform on the QPSK signal, then insert a cyclic prefix and perform parallel-to-serial conversion.

[0093] A4. Power distribution is performed on a QPSK signal of length T to obtain a low-power QPSK signal;

[0094] Step B:

[0095] B1. Bit encryption is performed on the data to be encrypted using the perturbation sequence X, and serial-to-parallel conversion, convolution numbering, and 16QAM modulation are performed sequentially to obtain a 16QAM signal;

[0096] B2. Perform two-dimensional joint scrambling encryption of the 16QAM signal using perturbation sequences Y and Z, involving subcarriers and symbols.

[0097] B3. Perform an inverse Fast Fourier Transform on the encrypted 16QAM signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a 16QAM signal of length T; specifically:

[0098] The 16QAM signal after symbol encryption is subjected to a 1024-point fast inverse Fourier transform, followed by the insertion of a cyclic prefix and parallel-to-serial conversion.

[0099] B4. Perform power distribution on a 16QAM signal of length T to obtain a high-power 16QAM signal; the power allocated to the high-power 16QAM signal is greater than the power allocated to the low-power QPSK signal.

[0100] Step C:

[0101] The low-power QPSK signal obtained in step A4 is superimposed and electro-optically modulated with the high-power 16QAM signal obtained in step B4 to generate a NOMA-OFDM optical transmission signal, thus completing the data transmission performance optimization.

[0102] like Figure 7As shown, the experiment verifies the use of a seven-core fiber optic transmission system based on intensity modulation / direct detection (IM / DD). At the transmitting end, the encrypted NOMA-OFDM signal is first generated through offline digital signal processing, with 256 effective subcarriers. The system employs a 1024-point IFFT transform, with a training sequence length of 1 / 101 of the OFDM symbol. Subsequently, a cyclic prefix of 1 / 4 of the OFDM symbol length is added as a guard interval for each symbol. The modulated signal is converted into an RF signal by an arbitrary waveform generator with a sampling rate of 10 GSa / s, amplified by an electrical amplifier, and then injected into a Mach-Zehnder modulator for intensity modulation. The information-carrying optical signal is split into seven paths by an optical coupler and injected into a two-kilometer seven-core fiber through a fan-in device. At the receiving end, the optical signal's received optical power is adjusted by an adjustable optical attenuator, and after photoelectric conversion by a photodiode, analog-to-digital conversion is performed by a mixed-signal oscilloscope with a sampling rate of 50 GSa / s. Finally, signal decryption and recovery are completed by an offline DSP.

[0103] In the experiment, the transmission performance of the seven-core optical fiber used in the system was first tested. The results are as follows: Figure 8 As shown, the bit error rate of the system decreases regularly with the increase of received optical power, and the seven curves almost overlap. At a bit error rate of 3.9 × 10⁻⁶, the system's bit error rate decreases. -5 At the same time, compared with core 4 which has the maximum received optical power, the power loss of core 6 is only 0.45dB. Experimental results show that the performance of the seven-core optical fiber is very stable, and inter-core crosstalk does not affect the transmission effect of the system.

[0104] In the experiment, an incorrect key was used to decrypt the transmitted signal to simulate an unauthorized receiver, and its performance was compared with the proposed scheme to verify the system's security performance. The bit error rate of the proposed scheme and the unauthorized receiver was calculated under the same received optical power. The unauthorized receiver attempts to use an incorrect pseudo-noise key to decrypt the transmitted information, thus generating a chaotic sequence that cannot be used for decryption. Figure 9 As shown, even with low received optical power, high-power signals cannot be recovered in an unauthorized receiver because the chaotic model and perturbation sequence cannot be reconstructed, proving that the present invention can effectively guarantee the security of user data transmission.

[0105] A brute-force attack involves trying every possible decryption key in turn until the correct key is found. Furthermore, the security performance of different encryption systems is measured by the time taken to execute a brute-force attack (BAT). In the comparison, three traditional high-level encryption algorithms were selected as the comparison objects, and it was assumed that the computer could execute 10 [unclear] per nanosecond. 4 This involves several decryption operations. The key length for Advanced Encryption Standard (AES-2) is twice that of Advanced Encryption Standard (AES-1). Table 1 lists the brute-force attack time results for different schemes.

[0106] Table 1 Comparison of security metrics for different security algorithms

[0107]

[0108] The results show that the decryption time required by this invention is increased to 10. 151 2019 was nearly 10 years since the implementation of the Advanced Encryption Standard (AES-2) scheme. 95 Furthermore, by employing dimension-switching identifiers to dynamically update the system's security level, brute-force attacks can be mitigated within 10 seconds. 29 ~10 151 Flexible adjustments within a year's timeframe allow for a better balance between the model's safety and complexity.

[0109] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present invention should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A data encryption method based on a multi-level dynamic iteration mechanism, characterized in that, Includes the following steps: Step A: A1. A three-dimensional perturbation sequence is generated by processing the dynamic key sequence through a multi-level dynamic iteration mechanism; the three-dimensional perturbation sequence includes perturbation sequence X, perturbation sequence Y, and perturbation sequence Z; A2. Perform serial-to-parallel conversion, convolutional coding, and QPSK modulation on the three-dimensional perturbation sequence in sequence to obtain the QPSK signal; A3. Perform a fast inverse Fourier transform on the QPSK signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a QPSK signal of length T. A4. Power distribution is performed on a QPSK signal of length T to obtain a low-power QPSK signal; Step B: B1. Bit encryption is performed on the data to be encrypted using the perturbation sequence X, and serial-to-parallel conversion, convolution numbering, and 16QAM modulation are performed sequentially to obtain a 16QAM signal; B2. Perform two-dimensional joint scrambling encryption of the 16QAM signal using perturbation sequences Y and Z, involving subcarriers and symbols. B3. Perform a fast inverse Fourier transform on the encrypted 16QAM signal, then insert a cyclic prefix and perform a parallel-to-serial conversion to obtain a 16QAM signal of length T. B4. Perform power allocation on a 16QAM signal of length T to obtain a high-power 16QAM signal; the power allocated to the high-power 16QAM signal is greater than the power allocated to the low-power QPSK signal. Step C: The low-power QPSK signal obtained in step A4 is superimposed and electro-optically modulated with the high-power 16QAM signal obtained in step B4 to generate a NOMA-OFDM optical transmission signal, thus completing the data transmission performance optimization.

2. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 1, characterized in that, Step A1 is as follows: The CMI bit data of the data to be encrypted is identified. Based on the correspondence between Chen's chaotic model and Lorenz chaotic model and 0 and 1, and the CMI bit data, Chen's chaotic model or Lorenz chaotic model is switched in real time to process the dynamic key sequence and generate a three-dimensional perturbation sequence.

3. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 2, characterized in that, In step A1, Chen's chaos model is as follows: Where α1, β1, and γ1 are the control parameters of Chen's chaotic model, and x1, y1, and z1 are the initial values ​​of the perturbation sequence of Chen's chaotic model. Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

4. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 3, characterized in that, In step A1, the Lorenz chaotic model is as follows: Where α2, β2, and γ2 are the control parameters of the Lorenz chaotic model, and x2, y2, and z2 are the initial values ​​of the perturbation sequence of the Lorenz chaotic model. Let x be the derivative of the perturbation sequence x with respect to time t. Let be the derivative of the perturbation sequence y with respect to time t. Let z be the derivative of the perturbation sequence z with respect to time t.

5. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 4, characterized in that, In step A1: The values ​​of α1, β1, and γ1 are respectively (-40, 40), (-10, 10), and (-30, 30); The values ​​of x1, y1, and z1 are respectively (-20, 20), (-20, 20), and (-20, 20).

6. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 5, characterized in that, In step A1: The values ​​of α2, β2, and γ2 are respectively (-10, 10), (-10, 10), and (-30, 30); The ranges of x2, y2, and z2 are (-20, 20), (-20, 20), and (0, 50), respectively. When using the Lorenz chaotic model, it is also necessary to subtract 25 from z2 beforehand to ensure that z2(0)∈(-25,25).

7. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 6, characterized in that, In step A2, the convolutional coding specifically involves: Channel coding is performed using a convolutional code with a constraint length L=4 and a generator polynomial of [13,17]. The code rate is configured to be 1 / 2, the number of encoder states is 8, and the trace depth is set to 10 clock cycles.

8. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 7, characterized in that, Step A3 is as follows: Perform a 1024-point inverse fast Fourier transform on the QPSK signal, then insert a cyclic prefix and perform parallel-to-serial conversion.

9. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 8, characterized in that, Step B3 specifically involves: The 16QAM signal after symbol encryption is subjected to a 1024-point fast inverse Fourier transform, followed by the insertion of a cyclic prefix and parallel-to-serial conversion.

10. The data encryption method based on a multi-level dynamic iteration mechanism according to claim 9, characterized in that, In step A1: The control parameters α1, β1, and γ1 of the Chen's chaotic model are specifically: α1 = 35, β1 = 3, and γ1 = 28. The initial values ​​of the perturbation sequence x1, y1, and z1 in the Chen's chaotic model are specifically: x1(0) = 0.995, y1(0) = 0.995, z1(0) = 0.995; The control parameters α2, β2, and γ2 of the Lorenz chaotic model are specifically: α2 = 10, β2 = 2.667, and γ2 = 28; The initial values ​​of the perturbation sequence x2, y2, z2 of the Lorenz chaotic model are specifically: x2(0) = 12, y2(0) = 2, z2(0) = 9.

Images