PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption

Abstract

The present invention belongs to the field of electronic information technology, and specifically relates to a PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption. The system of the present invention is composed of a PAM8 mapping module, a fast Fourier transform module, a chaotic phase scrambling encryption module, a conjugate frequency hopping encryption module, an inverse Fourier transform module, a waveform-level minimum mean square filter module, a conjugate frequency hopping removal module, a chaotic phase scrambling removal module, a symbol-level minimum mean square filter module, and a PAM8 inverse mapping module, which are connected in sequence. The present invention encrypts the frequency spectrum obtained after Fourier transform of the PAM8 signal in two dimensions: frequency and phase; waveform-level classification is achieved through minimum mean square algorithm filtering; the codebook at the transmitting end removes the conjugate frequency hopping encryption and phase scrambling, and then achieves symbol-level classification through a second-stage LMS filter, and finally maps the symbol information back to a bit sequence. The system of the present invention has high transmission reliability, security, and transmission quality under the premise of low complexity.

Description

Technical Field

[0001] The present invention belongs to the field of electronic information technology, and particularly relates to a PAM8 underwater visible light communication system. Background Art

[0002] Underwater visible light communication (UVLC) has a high research value in the development and utilization of new spectrum resources due to its excellent performance, such as no available bandwidth restrictions, high speed, large capacity, being far away from traditional radio frequency wireless communication bands, and abundant spectrum resources [1,2]. Visible light has a short wavelength and cannot penetrate walls during transmission, which has high security performance. However, it is still impossible to prevent the signal from being eavesdropped by eavesdroppers in the same space. Therefore, it is necessary to study signal encryption technology applicable to UVLC systems to improve transmission security.

[0003] In 2018, Fudan University (Zhou Yingjun et al.) first proposed a physical layer spectrum scrambling security strategy for the UVLC system [3]. The signal was encrypted by padding the positive and negative half axes of the spectrum of the PAM (Pulse Amplitude Modulation) signal with the same number of zeros. However, this encryption method may be cracked by analyzing the spectral characteristics of the signal at the receiving end. Therefore, it is necessary to adopt higher-dimensional encryption technology to protect the signal to be transmitted. In 2021, Zhejiang University (He Sailing et al.) first verified the feasibility of using chaotic encryption in high-speed UVLC systems [4]. They used two-level chaotic encryption at the bit level and subcarrier level, and verified that chaotic encryption had no negative impact on the performance of the system. In 2022, they used a three-layer chaotic system of bit stream diffusion, in-phase / orthogonal scrambling, and time-frequency scrambling to encrypt the UVLC system [5]. However, these encryption algorithms are relatively complex to implement, have high system requirements, and do not take into account the system's spectral characteristics. Since most devices used in the system have bandpass characteristics, and the spectrum of the PAM signal includes a DC component and abundant low-frequency components, the signal spectrum needs to be shifted.

[0004] This paper proposes a dual-layer encryption technology based on conjugate frequency hopping (CFH) and chaotic phase scrambling (PS). This technology encrypts the spectrum obtained by Fourier transforming the PAM8 signal in both frequency and phase dimensions. At the receiving end, the received signal and the original transmitted signal are correlated for synchronization. Waveform-level classification is achieved through least mean square (LMS) filtering. The conjugate frequency hopping encryption and phase scrambling are removed using a codebook at the transmitting end. A second-stage LMS filter is then used for symbol-level classification, and finally, the symbol information is mapped back to a bit sequence. This system implements PAM8 signal encryption and underwater visible light transmission, achieving high transmission reliability, security, and quality while maintaining low system complexity.

[0005] [References]

[0006] [1]M.Chen, P.Zou, L.Zhang and N.Chi, "Demonstration of a 2.34Gbit / sReal-Time Single Silicon-Substrate Blue LED-Based Underwater VLC System," inIEEE Photonics Journal, vol.12, no.1, pp.1-11, Feb.2020, Art no.7900211

[0007] [2]N.Chi, Y.Zhou, Y.Wei and F.Hu, "Visible Light Communication in 6G: Advances, Challenges, and Prospects," in IEEE Vehicular Technology Magazine, vol.15, no.4, pp.93-102, Dec.2020

[0008] [3] Y.Zhou, J.Shi, J.Zhang, and N.Chi, "Spectral Scrambling for High-security PAM-8Underwater Visible Light Communication System," in AsiaCommunications and Photonics Conference (ACP) 2018, OSA Technical Digest (OpticaPublishing Group, 2018), paper Su1G.4.

[0009] [4] Ji Du, Yuan Wang, Chao Fei, Ruilin Chen, Guowu Zhang, Xiaojian Hong, and Sailing He, "Experimental demonstration of 50-m / 5-Gbps underwater optical wireless communication with low-complexity chaotic encryption," Opt.Express29,783-796(2021).

[0010] [5]Huan Deng,Zihao Du,Jianmin Xiong,Xingqi Yang,Yan Hua,and Jing Xu,"Security enhancement for OFDM-UWOC system using three-layer chaotic encryption and chaotic DFT precoding,"Chin.Opt.Lett.20,110601-(2022). Summary of the Invention

[0011] The object of the present invention is to provide a PAM8 underwater visible light communication system with high transmission reliability, security and transmission quality.

[0012] For underwater visible light communication using PAM8 signals, the components used in the system have bandpass characteristics, and the PAM signal spectrum includes a DC component and abundant low-frequency components. Therefore, the signal spectrum needs to be shifted. To maintain the conjugate symmetry of the signal spectrum to obtain a real signal in the time domain, the present invention needs to maintain the symmetry of the positive and negative half axes when performing spectrum shifting, which is rarely considered in existing technologies.

[0013] In addition, underwater visible light communication still faces the risk of signal eavesdropping, requiring effective and easy-to-implement multi-dimensional encryption of the signal. This invention combines the system's need for spectrum shifting and implements encryption in the frequency dimension through conjugate frequency hopping (CFH), which pads the positive and negative half axes of the signal spectrum with the same number of zeros, with different groups of signals having different numbers of zero padding. By utilizing the initial value sensitivity and randomness of chaotic signals, the chaotic signal can be used to rotate the phase of the signal spectrum (i.e., chaotic phase scrambling (PS)), thereby achieving encryption in the phase dimension.

[0014] The proposed PAM8 underwater visible light communication system is based on a dual-layer encryption technology using conjugate frequency hopping (CFH) and chaotic phase scrambling. This involves encrypting the spectrum obtained by Fourier transforming the PAM8 signal in both frequency and phase dimensions. At the receiving end, the received signal is synchronized with the original transmitted signal through correlation operations. Waveform-level classification is achieved through least mean square (LMS) filtering. The conjugate frequency hopping encryption and phase scrambling are removed using a codebook at the transmitting end. A second-stage LMS filter is then used to achieve symbol-level classification, and the symbol information is finally mapped back into a bit sequence. This system achieves PAM8 signal encryption and underwater visible light transmission, achieving high transmission reliability, security, and quality while maintaining low system complexity.

[0015] The present invention proposes a PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption, such as Figure 1 As shown, it is composed of a PAM8 mapping module, a fast Fourier transform module, a chaotic phase scrambling encryption module, a conjugate frequency hopping encryption module, an inverse Fourier transform module, a waveform-level minimum mean square filter module, a conjugate frequency hopping removal module, a chaotic phase scrambling removal module, a symbol-level minimum mean square filter module, and a PAM8 inverse mapping module, which are connected in sequence.

[0016] The PAM8 mapping module is used to perform PAM mapping on the bit information stream, mapping every 3 bits into a symbol, and then grouping the PAM8 signal;

[0017] The fast Fourier transform module is used to transform the PAM8 real signal in the time domain into the frequency domain, where the frequency domain is conjugate symmetric;

[0018] The chaotic phase scrambling encryption module is used to perform phase rotation on the frequency domain signal using the chaotic signal generated by the Logistic mapping to achieve encryption;

[0019] The conjugate frequency hopping encryption module is used to randomly generate a frequency hopping code, the number of which is the same as the number of groups of PAM8 signals, conjugately add a corresponding number of zeros to the zero-frequency and low-frequency parts of the frequency domain of each group of signals to make the low-frequency component of the signal zero, and perform upsampling;

[0020] The inverse Fourier transform module is used to restore the twice-encrypted signal to the time domain to obtain a time domain real signal for transmission;

[0021] The waveform-level minimum mean square filtering module is used to adaptively filter the received waveform signal to eliminate noise and crosstalk;

[0022] The conjugate frequency hopping removal module is used to remove the added zeros after transforming the filtered signal into the frequency domain to achieve decryption;

[0023] The chaotic phase scrambling code removal module is used to remove the chaotic phase rotation added to the frequency domain signal to achieve decryption;

[0024] The symbol-level minimum mean square filtering module is used to perform a second-level adaptive filtering on the decrypted signal to obtain a signal with less interference for decoding.

[0025] The PAM8 demapping module is used to demodulate the decrypted PAM8 signal to obtain a received bit sequence for bit error rate testing and analysis.

[0026] Further:

[0027] In the present invention, the PAM8 mapping module is used to perform PAM mapping on the bit information stream, mapping every 3 bits into a symbol, and then grouping the PAM8 signal; specifically, assuming that the length of each group is N, divided into M groups, an M×N dimensional time domain PAM8 signal matrix D is obtained. M×N , all elements of this matrix are real numbers. After PAM8 mapping, the experimental signal is in the time-frequency domain.

[0028] In the present invention, the fast Fourier transform module is used to transform the time domain PAM8 real signal into the frequency domain:

[0029] D M×N =FFT(d M×N ), (1)

[0030]

[0031] Due to the conjugate symmetry of the Fourier transform, real signals in the time domain have conjugate symmetry after being transformed into the frequency domain. Taking one of the signal groups as an example (i.e., any value of g in the range [1, M] satisfies this), we have:

[0032]

[0033] The positive and negative half axes of the spectrum are:

[0034] D + =[D(0),D(1),…,D(N / 2)] T , (4)

[0035] D - =[D(0),D(-1),…,D(-N / 2+1)] T , (5)

[0036] Then it satisfies:

[0037]

[0038] Indicates D k The conjugate of Indicates D + The conjugate of

[0039] Since the subsequent chaotic encryption and conjugate frequency hopping encryption are both performed in the frequency domain, we can only align the positive semi-axis D + Perform the operation and then conjugate symmetry to the negative semi-axis D - That's it.

[0040] In the present invention, the chaotic phase scrambling encryption module is used to perform phase rotation on the frequency domain signal generated by the logistic map to achieve encryption. Specifically, the state of the logistic chaotic system is uniquely determined by the initial value and the bifurcation factor. The chaotic signal based on the logistic map can be generated using one of the following three models. After determining the initial value θ0 and the bifurcation factor μ of the system, a set of logistic chaotic signals can be generated by continuously iterating the following formula to generate the next value:

[0041]

[0042] θ k+1 =F(θ k )=μθ k (1-θ k ),μ∈(0,4), (8)

[0043]

[0044] Among them, the bifurcation factor μ determines whether the system is in a stable or chaotic state. The state of the system can be measured by the Lyapunov index λ. When λ>0, the system is in a chaotic state; when λ≤0, the system is in a stable or periodic state:

[0045]

[0046] For the first two models of generating chaotic signals, the initial values of the parameters are set to θ0 and μ respectively, and two sets of chaotic signals are generated, denoted as θ1 and θ2. In order to maintain the conjugate symmetry of the positive and negative half axes of the signal spectrum, only the chaotic signal of the positive half axis is generated, and the negative half axis is conjugate symmetric with the positive half axis. Next, the two generated chaotic signals are linearly combined, multiplied by a weight factor w, and then applied to the phase of the spectrum of the PAM8 signal to obtain the chaotic encrypted signal D. PS :

[0047] D PS+ =D + exp[jw(aΘ1+bΘ2)], (11)

[0048] D PS- =D - exp[-jw(aΘ1+bΘ2)], (12)

[0049] Here, a and b are the coefficients for linear weighting of the two chaotic signals, w is the weight of the chaotic phase on the signal phase, j is the imaginary unit, and the whole term is placed inside exp() to represent the signal D + and D - Do phase rotation, D PS+ Indicates the positive half axis, D PS- Represents the positive half axis.

[0050] In the present invention, the conjugate frequency hopping encryption module is used to randomly generate M frequency hopping codes z for M groups of signals. hop , conjugately add the corresponding number of zeros to the zero-frequency and low-frequency parts of the frequency domain of each group of signals, so that the low-frequency components of the signal are zero and encryption is achieved. Set the signal D to be 4 times the length of the original spectrum signal UP , whose dimension is M×(4N). Since the components of the system have bandpass characteristics, the low-frequency components of the signal are susceptible to attenuation. The DC component and the low-frequency component are first z-ed. ｐad Zero padding of points, that is:

[0051] [D UP (-(z pad +z hop )),...,D UP (-1), D UP (0),...,D UP (z pad +z hop )] T =[0,0,...,0] T , (13)

[0052] Then the phase encrypted signal is moved to the upsampled signal D UP middle:

[0053] [D UP (z pad +z hop +1),...,D UP (z pad +z hop +N / 2+1)] T =D PS+ , (14)

[0054] [D UP (-(z pad +z hop +N / 2)),...,D UP (-(z pad +z hop +1))] T =D PS- , (15)

[0055] D UP The other elements of are filled with zeros, thereby realizing phase encryption and conjugate frequency hopping encryption of the signal in the frequency domain and realizing 4 times upsampling of the signal.

[0056] In the present invention, the inverse Fourier transform module is used to restore the twice-encrypted signal to the time domain:

[0057] d UP M×4N =IFFT(D UP M×4N ), (16)

[0058]

[0059] The obtained time domain signal d UP M×4N They are all real signals and can be used for transmission directly.

[0060] In the present invention, the waveform-level minimum mean square filter module is used to filter the received signal r M×4N Adaptive filtering is performed to obtain a signal with less crosstalk and noise.

[0061] In the present invention, the conjugate frequency hopping removal module is used to remove the added zeros after transforming the filtered signal into the frequency domain to achieve decryption. hop , we only need to perform fast Fourier transform on the received signal to get R M×4N , then extract:

[0062] [R(z pad +z hop +1), ..., R(z pad +z hop+N / 2+1)] T ,and

[0063] [R(-(z pad +z hop +N / 2)),...,R(-(z pad +z hop +1))] T

[0064] As R PS+ and R PS- It can be used to decrypt the chaotic phase.

[0065] In the present invention, the chaotic phase scrambling removal module is used to remove the chaotic phase rotation added to the frequency domain signal to achieve decryption. Using the phase encrypted codebook used by the transmitter, decryption can be achieved through the following formula:

[0066] R + =R PS+ exp[-jw(aΘ1+bΘ2)], (18)

[0067] P - =R PS- exp[jw(aΘ1+bΘ2)], (19)

[0068] In the present invention, the symbol-level minimum mean square filtering module is used to perform a second-level adaptive filtering on the decrypted signal to obtain a signal with less interference.

[0069] In the present invention, the PAM8 demapping module is used to demodulate the decrypted PAM8 signal to obtain a received bit sequence, and reshape it into a one-dimensional sequence with a length of 3×N×M for bit error rate testing and analysis.

[0070] This invention proposes a method for phase rotation of the PAM8 signal spectrum during signal transmission in a PAM8 underwater visible light communication system using a logistic chaotic map to generate a scrambling code. Conjugate frequency hopping is then used to encrypt the signal in the frequency dimension and remove DC and low-frequency components. The signal is then transmitted through the underwater visible light communication system. At the receiving end, waveform-level LMS filtering is used to sequentially remove the conjugate frequency hopping and chaotic phase scrambling codes. Symbol-level LMS filtering is then used to achieve signal equalization, resulting in a cleaner signal for demapping and bit error rate testing. This invention achieves effective signal encryption through easy-to-implement technology, achieving high transmission reliability, security, and quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 This is a diagram of the PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption of the present invention.

[0072] Figure 2 This is a schematic diagram of the PAM8 underwater visible light communication principle based on conjugate frequency hopping and chaotic phase scrambling encryption of the present invention.

[0073] Figure 3 The bit error rate of the system of the present invention increases with the bias current I b and AWG voltage peak to peak value V pp Research on changes.

[0074] Figure 4 This is a study on how the bit error rate of the system of the present invention changes with the transmission rate.

[0075] Figure 5 This is a study on how the bit error rate of the system of the present invention changes with the phase scrambling code weight w.

[0076] The numbers in the figure are: 101 is the PAM8 mapping module, 102 is the fast Fourier transform module, 103 is the chaotic phase scrambling encryption module, 104 is the conjugate frequency hopping encryption module, 105 is the inverse Fourier transform module, 106 is the waveform-level minimum mean square filter module, 107 is the conjugate frequency hopping removal module, 108 is the chaotic phase scrambling removal module, 109 is the symbol-level minimum mean square filter module, and 110 is the PAM8 inverse mapping module. DETAILED DESCRIPTION

[0077] In the present invention, at the transmitting end, a set of bit sequences is generated as the signal to be transmitted. When implementing the hardware experiment, the selected sequence length is 76800.

[0078] In the present invention, the PAM8 mapping module is used to perform PAM mapping on the bit information stream, mapping every 3 bits into a symbol, that is, {000, 001, 010, 011, 100, 101, 110, 111} are respectively mapped to the eight levels {-7, -5, -3, -1, 1, 3, 5, 7}, and then the PAM8 signal is grouped, and the length of each group is N. When implementing the hardware experiment, N = 128, divided into 200 groups, and a 200×128-dimensional time domain PAM8 signal matrix d is obtained. 200×128 , all elements of this matrix are real numbers. After PAM8 mapping, the time-frequency domain of the experimental signal is as follows Figure 2 (c) shown.

[0079] In the present invention, the fast Fourier transform module is used to transform the time domain PAM8 real signal into the frequency domain:

[0080] D 200×128 =FFT(d 200×128 ), (1)

[0081]

[0082] Due to the conjugate symmetry of the Fourier transform, real signals in the time domain have conjugate symmetry after being transformed into the frequency domain. Taking one of the signal groups as an example (i.e., any value of g in the range [1,200] satisfies this), we have:

[0083]

[0084] The positive and negative half axes of the spectrum are:

[0085] D + =[D(0),D(1),…,D(N / 2)] T , (4)

[0086] D - =[D(0),D(-1),…,D(-N / 2+1)] T , (5)

[0087] Then it satisfies:

[0088]

[0089] Since the subsequent chaotic encryption and conjugate frequency hopping encryption are both performed in the frequency domain, we can only align the positive semi-axis D + Perform the operation and then conjugate symmetry to the negative semi-axis D - That's it.

[0090] In the present invention, the chaotic phase scrambling encryption module is used to perform phase rotation on the frequency domain signal generated by the Logistic map to achieve encryption. Due to the initial value sensitivity and randomness of the chaotic system, when the system starts with very close initial values, there will be significant differences after a period of time, thus achieving better encryption performance. The state of the Logistic chaotic system is uniquely determined by the initial value and the bifurcation factor. The chaotic signal based on the Logistic map can be generated by one of the following three models. After determining the system's initial value θ0 and bifurcation factor μ, a set of Logistic chaotic signals can be generated by continuously iterating the following formula to generate the next value:

[0091]

[0092] θ k+1 =F(θ k )=μθ k (1-θ k ),μ∈(0,4), (8)

[0093]

[0094] Among them, the bifurcation factor μ determines whether the system is in a stable or chaotic state. The state of the system can be measured by the Lyapunov index λ. When λ>0, the system is in a chaotic state; when λ≤0, the system is in a stable or periodic state:

[0095]

[0096] When implementing the hardware experiment, we used the first two chaotic signal generation models, setting the parameters to Θ0 = 0.7, μ = 2.0 and Θ0 = 0.1, μ = 3.8, respectively, to generate two sets of chaotic signals Θ1 and Θ2. To maintain the conjugate symmetry of the positive and negative half axes of the signal spectrum, we only generated chaotic signals on the positive half axis, and the negative half axis was conjugate symmetric with the positive half axis. Therefore, the dimensions of Θ1 and Θ2 are both 200 × (N / 2 + 1), or 200 × 65. Next, the two generated chaotic signals are linearly combined, multiplied by a weight factor w, and applied to the phase of the PAM8 signal spectrum to obtain the chaotic encrypted signal D. PS :

[0097] D PS+ =D + exp[jw(aΘ1+bΘ2)], (11)

[0098] D PS- =D - ·exp[-jw(aΘ1+bΘ2)], (12).

[0099] In the present invention, the conjugate frequency hopping encryption module is used to randomly generate 200 frequency hopping codes z for 200 groups of signals. hｏp , conjugately add the corresponding number of zeros to the zero-frequency and low-frequency parts of the frequency domain of each group of signals, so that the low-frequency components of the signal are zero and encryption is achieved. When implementing the hardware experiment, we set the signal D to be 4 times the length of the original spectrum signal UP , whose dimension is 200×(4N). Since the components of the system have bandpass characteristics, the low-frequency components of the signal are susceptible to attenuation. Therefore, we first perform z on the DC component and the low-frequency component. pad Zero padding of points, in our hardware experiments z pad =5, then add z according to the frequency hopping code hop The zero of point is:

[0100] [D UP (-(z pad +z hop )),...,D UP (-1), D UP (0),...,D UP (z pad +z hop )]T =[0, 0, ..., 0] T , (13)

[0101] Then the phase encrypted signal is moved to the upsampled signal D UP middle:

[0102] [D UP (z pad +z hop +1), ..., D UP (z pad +z hop +N / 2+1)] T =D PS+ , (14)

[0103] [D UP (-(z pad +z hop +N / 2)),...,D UP (-(z pad +z hop +1))] T =D PS- , (15)

[0104] D UP The other elements of are filled with zero, thereby realizing the phase encryption and conjugate frequency hopping encryption of the signal in the frequency domain, and realizing the 4-fold upsampling of the signal. The principle of this technology can be seen Figure 2 (b).

[0105] In the present invention, the inverse Fourier transform module is used to restore the twice-encrypted signal to the time domain:

[0106] d UP 200×512 =IFFT(D UP 200×512 ), (16)

[0107]

[0108] The obtained time domain signal d UP 200×512 They are all real signals and can be used for transmission directly.

[0109] In the present invention, the waveform-level minimum mean square filter module is used to filter the received signal r 200×512 Adaptive filtering is performed to obtain a signal with less crosstalk and noise. In the hardware experiment, the LMS filter tap coefficient is 59, the step size is 0.01, and the training sequence length is 2000. After that, the signal is decrypted.

[0110] In the present invention, the conjugate frequency hopping removal module is used to remove the added zeros after transforming the filtered signal into the frequency domain to achieve decryption.hop , we only need to perform fast Fourier transform on the received signal to get R 200×512 , then extract:

[0111] [R(z pad +z hop +1),…,R(z pad +z hop +N / 2+1)] T ,and

[0112] [R(-(z pad +z hop +N / 2)),…,R(-(z pad +z hop +1))] T

[0113] As R PS+ and R PS- It can be used to decrypt the chaotic phase.

[0114] In the present invention, the chaotic phase scrambling removal module is used to remove the chaotic phase rotation added to the frequency domain signal to achieve decryption. Using the phase encrypted codebook used by the transmitter, decryption can be achieved through the following formula:

[0115] R + =R PS+ exp[-jw(aΘ1+bΘ2)], (18)

[0116] R - =R PS- ·exp[jw(aΘ1+bΘ2)], (19).

[0117] In the present invention, the symbol-level least mean square filtering module is used to perform a second-level adaptive filtering on the decrypted signal to obtain a signal with less interference, wherein the LMS filter tap coefficient is 43, the step size is 0.003, and the training sequence length is 2000, and then the signal is decoded.

[0118] In the present invention, the PAM8 demapping module is used to demodulate the decrypted PAM8 signal to obtain a received bit sequence, and reshape it into a one-dimensional sequence with a length of 3×128×200 for bit error rate testing and analysis.

[0119] The specific process of the PAM8 underwater visible light communication system based on the above module is as follows:

[0120] Step 101: Through the PAM8 mapping module, every 3 bits are mapped into a symbol, and then the PAM8 signal is divided into 200 groups, each with a length of 128, to achieve the mapping of the bit stream to the PAM8 signal. The time-frequency domain of the signal obtained by the experiment is as follows Figure 2 (c)

[0121] Step 102: Using a fast Fourier transform module, transform the PAM8 real signal in the time domain into the frequency domain, where the frequency domain is conjugate symmetric.

[0122] Step 103: Through the chaotic phase scrambling encryption module, the chaotic signal generated by the Logistic map is phase-rotated on the frequency domain signal to achieve the first encryption of the signal. The time-frequency domain of the signal obtained by the experiment is as follows: Figure 2 (d)

[0123] Step 104: 200 frequency hopping codes are randomly generated through the conjugate frequency hopping encryption module. A corresponding number of zeros are conjugately added to the zero-frequency and low-frequency parts of the frequency domain of each group of signals to make the DC and low-frequency components of the signal zero. Then, 4 times upsampling is performed to achieve the second encryption. The time-frequency domain of the signal obtained by the experiment is as follows: Figure 2 (e)

[0124] Step 105: Using an inverse Fourier transform module, the twice-encrypted signal is restored to the time domain to obtain a time-domain real signal for transmission;

[0125] Step 106: Using a waveform-level least mean square filtering module to adaptively filter the received waveform signal to reduce noise and crosstalk;

[0126] Step 107: The filtered signal is transformed into the frequency domain by the conjugate frequency hopping removal module, and the zeros added by the low-frequency and DC components are removed to achieve the first decryption;

[0127] Step 108: Using the chaotic phase scrambling removal module, the chaotic phase rotation added to the frequency domain signal is removed to achieve a second decryption.

[0128] Step 109: Through the symbol-level least mean square filter module, the second-level adaptive filtering is implemented on the decrypted signal to obtain a signal with less interference for decoding. The signal constellation diagram after LMS filtering obtained in the experiment is as follows: Figure 2 (f);

[0129] Step 110: Using the PAM8 demapping module, the decrypted PAM8 signal is demodulated to obtain a received bit sequence for bit error rate testing and analysis. Figure 3 The bit error rate of the system of the present invention increases with the bias current I b and AWG voltage peak to peak value V ppStudy of changes; when V pp When the bias current I b As the bias current I b When fixed, as the AWG voltage peak-to-peak value V pp As V pp or I b When it is too high, the performance of the system will decrease due to nonlinear effects; pp or I b If it is too low, the signal will be submerged in the noise, the signal-to-noise ratio at the receiving end will decrease, and the BER will increase. Therefore, the system has an optimal working point. According to the experimental results, when I b =130mA and V pp =0.5V, the system performance reaches the best, and the BER is 1.96×10 -3 ; In addition, between the black lines in the figure, the transmission BER is lower than the 7% HD-FEC threshold.

[0130] In the present invention, the length, order, number of groups of the PAM signal, parameters of the Logistic chaotic signal, tap coefficients and step length of the LMS filter, and the means of receiving end signal equalization can all be adjusted according to actual needs to obtain the best effect.

[0131] The PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption proposed in the present invention can achieve high transmission reliability, security and transmission quality. On the premise that the system principle is simple and easy to implement, it has good encryption effect and high transmission rate.

[0132] According to the simulation and implemented experiments, at the optimal working point, Figure 4 This is a study of the bit error rate of the system of the present invention as the transmission rate changes. When the receiving end does not remove the conjugate frequency hopping and phase scrambling ( Figure 4 The system BER is basically around 0.5, which means that even if the receiver eavesdrops on the signal, it cannot obtain the correct transmission signal without the correct key for decryption, which proves the security of the experimental system. Figure 4 The blue line shows that as the transmission rate increases, the BER also gradually increases. This is because as the transmission speed increases, the inter-symbol interference at the receiving end increases due to the limited sampling rate and bandwidth of the system, thus increasing the possibility of errors in sampling decisions. When the BER is below the 7% HD-FEC threshold, the maximum achievable transmission rate is 2.1 Gbps, and the BER at this time is 2.62×10 -3The constellation diagram of the PAM8 signal correctly decrypted by the receiving end is as follows Figure 4 (iii) shows that the constellation diagram that is not correctly decrypted is as follows Figure 4 As shown in (i), the eavesdropping end cannot obtain effective information of the transmitted signal.

[0133] At the optimal working point and the highest transmission rate, the influence of the phase scrambling weight w on the system performance is studied. Figure 5 As shown, when the receiver does not correctly remove the conjugate frequency hopping and phase scrambling ( Figure 5 The blue line in the middle), and the receiver only removes the phase scrambling ( Figure 5 In the case of the orange line in the middle, it is impossible to correctly decode and judge to obtain useful information, and the BER is around 0.5. The constellation diagram is as follows Figure 5 (i). When the receiving end only removes the conjugate frequency hopping ( Figure 5 When the scrambling weight w increases, the transmission BER also gradually increases, which means that a high-weight scrambling code requires higher accuracy in decryption at the receiving end and has better encryption effect. A low-weight scrambling code has less impact on the signal and can be removed as system noise by the receiving end, resulting in poor encryption effect. When the weight is 50%, the receiving end constellation diagram is as follows: Figure 5 (ii), it can be seen that there is almost no boundary between adjacent symbols, and it is extremely difficult to correctly classify and judge. When the receiver correctly removes the conjugate frequency hopping and phase scrambling ( Figure 5 When the BER is lower than 7% under HD-FEC, the highest weight w = 50%. The constellation diagram of the receiving end is as follows: Figure 5 As shown in (iii), the BER is 2.84×10 -3 .

[0134] Finally, the encryption performance of the system is analyzed. First, the chaotic phase scrambling code is analyzed. Since the parameters to be determined are the initial values θ0 and bifurcation factors μ of the two chaotic signals Θ1 and Θ2, as well as the coefficients a and b of their linear combination, there are a total of six unknown parameters. Since the precision of the AWG used in the experiment is 8 bits, the phase scrambling code provides 2 8×6 =2.81×10 14 The key space of conjugate frequency hopping is then analyzed. Since the number of frequency hopping codes is 200 (one for each PAM8 signal), and its value is an integer between 1 and 5, 5 200 =6.22×10 139 Therefore, the key space of the system can reach 2.81×10 14 ×6.22×10 139 =1.75×10 154, the system has a high level of security. In addition, although a large part of the key space is provided by conjugate frequency hopping, the role of chaotic phase scrambling cannot be denied. Because if only a single encryption method is used, an eavesdropper may be able to observe the encryption method of conjugate frequency hopping by analyzing the signal spectrum.

[0135] The division of each step in this embodiment is only for the purpose of clear description. When implemented, they can be combined into one step or some steps can be split and decomposed into multiple steps. As long as they contain the same logical relationship, they are within the scope of protection of the present invention.

[0136] Those skilled in the art will appreciate that the aforementioned embodiments are specific examples of the present invention. In actual applications, various modifications may be made to the embodiments and details without departing from the spirit and scope of the present invention. For example, the length, order, and number of groups of the PAM signal, the parameters of the logistic chaotic signal, the tap coefficients and step size of the LMS filter, and the means of signal equalization at the receiving end may all be adjusted as needed to achieve optimal results.

Claims

1. A PAM8 underwater visible light communication system based on conjugate frequency hopping and chaotic phase scrambling encryption, characterized in that: It consists of a PAM8 mapping module, a fast Fourier transform module, a chaotic phase scrambling encryption module, a conjugate frequency hopping encryption module, an inverse Fourier transform module, a waveform-level minimum mean square filter module, a conjugate frequency hopping removal module, a chaotic phase scrambling removal module, a symbol-level minimum mean square filter module, and a PAM8 inverse mapping module connected in sequence; wherein: The PAM8 mapping module is used to perform PAM mapping on the bit information stream, mapping every 3 bits into a symbol, and then grouping the PAM8 signal; The fast Fourier transform module is used to transform the PAM8 real signal in the time domain into the frequency domain, where the frequency domain is conjugate symmetric; The chaotic phase scrambling encryption module is used to perform phase rotation on the frequency domain signal using the chaotic signal generated by the Logistic mapping to achieve encryption; The conjugate frequency hopping encryption module is used to randomly generate a frequency hopping code, the number of which is the same as the number of groups of PAM8 signals, conjugately add a corresponding number of zeros to the zero-frequency and low-frequency parts of the frequency domain of each group of signals to make the low-frequency component of the signal zero, and perform upsampling; The inverse Fourier transform module is used to restore the twice-encrypted signal to the time domain to obtain a time domain real signal for transmission; The waveform-level minimum mean square filtering module is used to adaptively filter the received waveform signal to eliminate noise and crosstalk; The conjugate frequency hopping removal module is used to remove the added zeros after transforming the filtered signal into the frequency domain to achieve decryption; The chaotic phase scrambling code removal module is used to remove the chaotic phase rotation added to the frequency domain signal to achieve decryption; The symbol-level minimum mean square filtering module is used to perform a second-level adaptive filtering on the decrypted signal to obtain a signal with less interference for decoding; The PAM8 demapping module is used to demodulate the decrypted PAM8 signal to obtain a received bit sequence for bit error rate testing and analysis.

2. The PAM8 underwater visible light communication system according to claim 1, characterized in that: The PAM8 mapping module is used to perform PAM mapping on the bit information stream, mapping every 3 bits into a symbol, and then grouping the PAM8 signal; specifically, assuming that the length of each group is N, it is divided into M groups to obtain an M×N dimensional time domain PAM8 signal matrix D M×N , all elements of this matrix are real numbers; after PAM8 mapping, the time-frequency domain of the signal is obtained.

3. The PAM8 underwater visible light communication system according to claim 2, characterized in that: The Fast Fourier Transform module is used to transform the time domain PAM8 real signal into the frequency domain: D M×N =FFT(d M×N ), (1) Due to the conjugate symmetry of Fourier transform, the real signal in the time domain has conjugate symmetry after being transformed into the frequency domain; g takes any value in the range [1, M], and we have: The positive and negative half axes of the spectrum are: D + =[D(0),D(1),…,D(N / 2)] T , (4) D - =[D(0),D(-1),…,D(-N / 2+1)] T , (5) Then it satisfies: Indicates D k The conjugate of Indicates D + conjugation of; Since both chaotic encryption and conjugate frequency hopping encryption are performed in the frequency domain, only the positive semi-axis D + Perform the operation and then conjugate symmetry to the negative semi-axis D - That's it.

4. The PAM8 underwater visible light communication system according to claim 3, characterized in that: The chaotic phase scrambling encryption module is used to perform phase rotation on the frequency domain signal generated by the Logistic map to achieve encryption. Specifically, the state of the Logistic chaotic system is uniquely determined by the initial value and the bifurcation factor. The chaotic signal based on the Logistic map is generated by one of the following three models. After determining the initial value θ0 and the bifurcation factor μ of the system, the next value is generated by continuously iterating the following formula to generate a set of Logistic chaotic signals: i k+1 =F(θ k )=μθ k (1-θ k ),μ∈(0,4), (8) Among them, the bifurcation factor μ determines whether the system is in a stable or chaotic state. The state of the system can be measured by the Lyapunov exponent λ. When λ>0, the system is in a chaotic state; when λ≤0, the system is in a stable or periodic state: For the first two models of generating chaotic signals, the initial values of the parameters are set to θ0 and μ respectively, and two sets of chaotic signals are generated, denoted as θ1 and θ2. In order to maintain the conjugate symmetry of the positive and negative half axes of the signal spectrum, only the chaotic signal of the positive half axis is generated, and the negative half axis is conjugate symmetric with the positive half axis. Next, the two generated chaotic signals are linearly combined, multiplied by a weight factor w, and then applied to the phase of the PAM8 signal spectrum to obtain the chaotic encrypted signal D. PS : D PS+ =D + ·exp[jw(aΘ1+bΘ2)], (11) D PS- =D - ·exp[-jw(aΘ1+bΘ2)], (12) Here, a and b are the coefficients for linear weighting of the two chaotic signals, w is the weight of the chaotic phase on the signal phase, j is the imaginary unit, and the whole term is placed inside exp() to represent the signal D + and D - Do phase rotation, D PS+ Indicates the positive half axis, D PS- Represents the positive half axis.

5. The PAM8 underwater visible light communication system according to claim 4, characterized in that: The conjugate frequency hopping encryption module is used to randomly generate M frequency hopping codes z for M groups of signals. hop , conjugately add a corresponding number of zeros to the zero-frequency and low-frequency parts of the frequency domain of each group of signals, so that the low-frequency components of the signal are zero and encryption is achieved. Specifically, set the signal D to be 4 times the length of the original spectrum signal UP , whose dimension is M×(4N); Since the components of the system have bandpass characteristics, the low-frequency components of the signal are susceptible to attenuation, the DC component and the low-frequency component are first z pad Zero padding of points, that is: [D UP (-(z pad +z hop )),...,D UP (-1),D UP (0),...,D UP (z pad +z hop )] T =[0,0,...,0] T , (13) Then the phase encrypted signal is moved to the upsampled signal D UP middle: [D UP (z pad +z hop +1),...,D UP (z pad +z hop +N / 2+1)] T =D PS+ , (14) [D UP (-(z pad +z hop +N / 2)),...,D UP (-(z pad +z hop +1))] T D PS- , (15) D UP The other elements of are filled with zeros, thereby realizing phase encryption and conjugate frequency hopping encryption of the signal in the frequency domain and realizing 4 times upsampling of the signal.

6. The PAM8 underwater visible light communication system according to claim 5, characterized in that: The inverse Fourier transform module is used to restore the twice-encrypted signal to the time domain: d UP M×4N =IFFT(D UP M×4N ), (16) The obtained time domain signal d UP M×4N They are all real signals and are used directly for transmission.

7. The PAM8 underwater visible light communication system according to claim 6, characterized in that: The waveform level minimum mean square filter module is used to filter the received signal r M×4N Adaptive filtering is performed to obtain a signal with less crosstalk and noise.

8. The PAM8 underwater visible light communication system according to claim 7, characterized in that: The conjugate frequency hopping removal module is used to remove the added zeros after transforming the filtered signal into the frequency domain to achieve decryption. Specifically, the code book z used by the transmitter is used to hop , perform fast Fourier transform on the received signal to get R M×4N , then extract: [R(z pad +z hop +1),...,R(z pad +z hop +N / 2+1)] T , and [R(-(z pad +z hop +N / 2)),...,R(-(z pad +z hop +1))] T As R PS+ and R PS- It can be used to decrypt the chaotic phase.

9. The PAM8 underwater visible light communication system according to claim 8, characterized in that: The chaotic phase scrambling removal module is used to remove the chaotic phase rotation added to the frequency domain signal to achieve decryption. Specifically, the phase encrypted codebook used by the transmitter is used to achieve decryption through the following formula: R + =R PS+ ·exp[-jw(aΘ1+bΘ2)], (18) R - =R PS- ·exp[jw(aΘ1+bΘ2)], (19)。 10. The PAM8 underwater visible light communication system according to claim 9, characterized in that: The symbol-level minimum mean square filtering module is used to perform a second-level adaptive filtering on the decrypted signal to obtain a signal with less interference. The PAM8 demapping module is used to demodulate the decrypted PAM8 signal to obtain a received bit sequence, and reshape it into a one-dimensional sequence with a length of 3×N×M for bit error rate testing and analysis.

Images