SP-2048-QAM (Quadrature Amplitude Modulation) method based on staged truncation shaping

Abstract

The invention discloses an SP-2048-QAM (Quadrature Amplitude Modulation) modulation method based on staged truncation shaping, which specifically comprises the following steps of: preprocessing a signal at a transmitting end, and performing staged truncation shaping on a pseudo-random bit sequence from the angle of gradually intercepting signal points so as to realize gradual adjustment of module values and distribution of the signal points; then, shaping and four-dimensional modulation are combined to form a series of intermediate modulation formats; at a receiving end, the optical signal is amplified again and then is subjected to coherent detection, a detected analog signal is converted into a digital signal through an analog-to-digital converter, and finally, corresponding signal equalization, compensation and demapping judgment operations are performed through a digital processing algorithm module to recover original data. According to the method, the transmission performance of the system is improved, the bit error rate is reduced, and the self-adaptive capability of the system is enhanced, so that the signal entropy can be flexibly adjusted according to changing channel conditions and noise environments, and the stability and reliability of communication are ensured.

Description

Technical Field

[0001] The invention belongs to the field of optical fiber communication, and in particular relates to an SP-2048-QAM modulation method based on staged truncation and shaping. Background Art

[0002] In recent years, with the rapid development of emerging services such as big data, the Internet of Things, and virtual reality, the demand for large-capacity and high-spectral-efficiency optical transmission has also increased. Probabilistic shaping technology is a modulation optimization technology. With its advantages such as high transmission capacity, it can improve the spectrum efficiency of the system without increasing the transmission power, so that the system can obtain a higher transmission capacity. It has been widely studied in recent years and has gradually become a new technology with broad application prospects. With the gradual deepening of the research on probabilistic shaping technology, researchers have found that in probabilistic shaping systems with large shaping depths, the digital signal processing algorithm at the receiving end has insufficient ability in damage compensation, which makes the signal quality after compensation worse or even unrecoverable. In order to solve this problem, academia and industry have contributed two different solutions. One idea is to modify the digital signal processing (DSP) algorithm itself to improve the robustness of the DSP algorithm in probabilistic shaping systems with different shaping depths as much as possible. For example, Beijing University of Posts and Telecommunications proposed a quasi-MB distributed modulation technology. By changing the symbol mapping rules, the number of constellation points used for phase noise estimation is limitedly increased, thereby realizing a carrier phase recovery-friendly probabilistic shaping technology. Huazhong University of Science and Technology proposed a probability-guided blind phase search algorithm, which improved the carrier phase recovery capability of the traditional blind phase search algorithm in large-shaping depth probabilistic shaping systems.

[0003] Another idea is to develop advanced probability shaping modulation schemes based on probability shaping technology, and use new probability shaping modulation schemes to improve the compensation capability of the DSP algorithm at the receiving end. Following the second research idea, the academic community has proposed several probability shaping schemes, among which a typical technology is truncated probability shaping (TPS) technology, which reduces the modulation dimension of the constellation points by forcing the probability of the outer constellation points of the high-order modulation format to 0, so that the shaping depth is effectively reduced and the working ability of the DSP algorithm is improved. In recent years, TPS technology has been widely studied and applied. In 2022, Fudan University, based on TPS-64QAM modulation, used Raman-amplified G.654E optical fiber and multi-input and multi-output Volterra equalizer to achieve a single-wavelength 800-Gbps and a total capacity of 4-Tbps 2000-km optical transmission; in 2023, Bell Labs achieved a 226-GBaud truncated probability shaping 256-QAM signal, with a single-wavelength net data rate of up to 2.4-Tbps.

[0004] Although the truncated probability shaping (TPS) technology can significantly reduce the average transmission power and improve the system capacity and transmission distance, this technology also brings about a significant reduction in signal entropy, which has an adverse effect on system performance.

[0005] In the process of pursuing the balance between spectral efficiency and asymptotic power efficiency, the four-dimensional modulation format has attracted great attention from researchers at home and abroad due to its unique advantages. This modulation format forms a four-dimensional modulation space by simultaneously modulating the in-phase component and the orthogonal component on two independent orthogonal polarization optical carriers, which not only simplifies the implementation process, but also increases the minimum Euclidean distance between constellation points, thus having important research value in the field of optical fiber communication. Therefore, combining the staged truncation shaping scheme with the four-dimensional modulation technology has far-reaching significance for the research and development of optical fiber communication systems. Summary of the invention

[0006] In view of the above problems, the present invention provides a SP-2048-QAM modulation method based on staged truncation and shaping.

[0007] The present invention provides a SP-2048-QAM modulation method based on staged truncation and shaping, comprising the following steps:

[0008] Step 1: Perform four-dimensional modulation on the information bits.

[0009] The initial 11 information bits are XORed to generate the 12th parity check bit; through parity diversity, the X polarization and the Y polarization are correlated, and the original signal points are divided into two parts, so that the minimum Euclidean distance between constellation points is increased to √2 times of the original, thereby realizing four-dimensional modulation.

[0010] Step 2: intercept signal points within a specific modulus circle layer by layer, and form a series of intermediate modulation formats between the SP-2048-QAM signal and the TPS-SP-2048-QAM signal.

[0011] The SP-2048-QAM signal consists of nine modulus circles, each of which corresponds to a set of specific signal points. Starting from the outermost modulus circle, the signal points are intercepted layer by layer inward to generate TPS-4, TPS-12 and TPS-20 signals.

[0012] Step 3: Shape the generated signal based on the constant parameter distribution matcher (CCDM). The specific formula for constellation point distribution is as follows:

[0013]

[0014] Where: X is the set of constellation points, x i is the complex coordinate of the i-th constellation point, λ≥0 is a free parameter, and M is the number of constellation points.

[0015] When λ = 0, the distribution is uniform. The larger λ is, the greater the difference in probability between different energy levels is, that is, the smaller the energy of the constellation point is, the greater the probability of being assigned. Given λ, the corresponding P is obtained according to formula (1): X (x i ), and then calculate the information entropy H. The specific formula is as follows:

[0016]

[0017] The shaping depth is used to measure the unevenness of the probability distribution. The specific formula is as follows:

[0018]

[0019] Step 4: Upsample the modulated signal by four times, and then perform pulse shaping using a root raised cosine filter with a roll-off factor of 0.01. The specific filter formula is as follows:

[0020]

[0021] Where: T is the symbol period, α is the roll-off factor, and its value range is between 0 and 1.

[0022] Step 5: The optical modulator modulates the signal into the optical domain and transmits it through a standard single-mode optical fiber (SSMF). To enhance the signal, it passes through an erbium-doped fiber amplifier (EDFA) with a noise figure of 4.5 dB; to suppress out-of-band noise, it passes through an optical filter (OBPF).

[0023] Step 6: At the receiving end, the optical signal is amplified again and then coherently detected. The detected analog signal is converted into a digital signal through an analog-to-digital converter. Finally, the digital processing algorithm module performs corresponding signal equalization, compensation, and demapping judgment operations to restore the original data.

[0024] The beneficial technical effects of the present invention compared with the prior art are:

[0025] The present invention improves the transmission performance of the system by combining staged truncation shaping with four-dimensional modulation: on the one hand, the bit error rate is reduced by increasing the Euclidean distance of the signal point in the constellation diagram; on the other hand, the adaptive ability of the system is enhanced, so that it can flexibly adjust the signal entropy value according to the changing channel conditions and noise environment, thereby ensuring the stability and reliability of communication. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a block diagram of the transmission system of the present invention.

[0027] Figure 2 It is a schematic diagram of the process structure of the present invention.

[0028] Figure 3 It is the constellation diagram of the intermediate modulation format of the present invention.

[0029] Figure 4 It is the bit error rate performance of each modulated signal at different transmission powers when transmitted 1440 kilometers. DETAILED DESCRIPTION

[0030] The present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0031] The present invention provides a SP-2048-QAM modulation method based on staged truncation and shaping, comprising the following steps:

[0032] Step 1: Perform four-dimensional modulation on the information bits.

[0033] The initial 11 information bits are XORed to generate the 12th parity check bit; through parity diversity, the X polarization and the Y polarization are correlated, and the original signal points are divided into two parts, so that the minimum Euclidean distance between constellation points is increased to √2 times of the original, thereby realizing four-dimensional modulation.

[0034] Step 2: intercept signal points within a specific modulus circle layer by layer, and form a series of intermediate modulation formats between the SP-2048-QAM signal and the TPS-SP-2048-QAM signal.

[0035] The SP-2048-QAM signal consists of nine modulus circles, each of which corresponds to a set of specific signal points. Starting from the outermost modulus circle, the signal points are intercepted layer by layer inward to generate TPS-4, TPS-12 and TPS-20 signals.

[0036] Step 3: Shape the generated signal based on the constant parameter distribution matcher (CCDM). The specific formula for constellation point distribution is as follows:

[0037]

[0038] Where: X is the set of constellation points, x i is the complex coordinate of the i-th constellation point, λ≥0 is a free parameter, and M is the number of constellation points.

[0039] When λ = 0, the distribution is uniform. The larger λ is, the greater the difference in probability between different energy levels is, that is, the smaller the energy of the constellation point is, the greater the probability of being assigned. Given λ, the corresponding P is obtained according to formula (1): X (x i ), and then calculate the information entropy H. The specific formula is as follows:

[0040]

[0041] The shaping depth is used to measure the unevenness of the probability distribution. The specific formula is as follows:

[0042]

[0043] Step 4: Upsample the modulated signal by four times, and then perform pulse shaping using a root raised cosine filter with a roll-off factor of 0.01. The specific filter formula is as follows:

[0044]

[0045] Where: T is the symbol period, α is the roll-off factor, and its value range is between 0 and 1.

[0046] Step 5: The optical modulator modulates the signal into the optical domain and transmits it through a standard single-mode optical fiber (SSMF). To enhance the signal, it passes through an erbium-doped fiber amplifier (EDFA) with a noise figure of 4.5 dB; to suppress out-of-band noise, it passes through an optical filter (OBPF).

[0047] Step 6: At the receiving end, the optical signal is amplified again and then coherently detected. The detected analog signal is converted into a digital signal through an analog-to-digital converter. Finally, the digital processing algorithm module performs corresponding signal equalization, compensation, and demapping judgment operations to restore the original data.

[0048] Figure 1 The block diagram of the transmission system of the SP-2048-QAM modulation method based on staged truncation and shaping of the present invention mainly consists of three parts: 1) SP-2048-QAM transmitter; the signal source mainly includes signal preprocessing (101), digital-to-analog converter (102), signal light source (103), and IQ modulator (104); 2) optical fiber transmission link part; firstly, the signal is transmitted through N sections of standard single-mode optical fiber (105 1 ~105 N ) for transmission, and then by the optical amplifier (106 1 ~106 N ) is used for compensation, and then a tunable optical filter (107 1 ~107 N ) to perform optical filtering to suppress the spontaneous radiation noise generated outside the band; 3) the receiving end coherent detection digital signal processing part; finally entering the coherent receiver (108) to perform the corresponding analog-to-digital conversion to obtain a digital signal, and finally performing the corresponding signal equalization, compensation, and demapping judgment operations through the digital processing algorithm module (109). The present invention combines staged truncation shaping with four-dimensional modulation in signal preprocessing.

[0049] The specific process of the present invention is as follows Figure 2As shown, in the signal preprocessing stage (101), the first 11 bits of information are generated by XOR operation to generate the 12th check bit, and then the 12 bits of information are converted into a four-dimensional SP-2048QAM signal through bit mapping. Then, according to the modulus circle, specific points are selected from the X polarization and the Y polarization for truncation to form a truncated SP-2048QAM signal. Finally, these truncated signals are input into the CCDM module together with the MB probability distribution to obtain a truncated and shaped SP-2048QAM signal.

[0050] Figure 3 It is the constellation diagram of the intermediate modulation format of the present invention. By intercepting the signal points in the specific modulus value circle layer by layer, a series of intermediate modulation formats are gradually formed between the 64QAM signal and the TPS-64QAM signal, including TPS-4, TPS-12 and TPS-20 signals. Specifically, the 64QAM signal consists of nine modulus value circles, each of which corresponds to a set of specific signal points. The scheme starts from the outermost modulus value circle and intercepts the signal points layer by layer to generate TPS-4, TPS-12 and TPS-20 signals. Figure 3 (a), (b), and (c) show the constellation diagrams of TPS-4, TPS-12, and TPS-20 signals, respectively, where X represents the intercepted signal and modulus circle, and the dots represent the retained signal and modulus circle. Through this multi-stage truncation method, the present invention can achieve a smooth transition of entropy values ​​from 64QAM to TPS-64QAM. Each stage of truncation only affects a small number of signal points, thereby effectively avoiding a sharp drop in system entropy.

[0051] Figure 4 The bit error rate performance of each modulated signal at different transmission powers when the transmission distance is 1440 kilometers and the transmission information rate is the same. It is worth noting that in order to ensure that the transmission information rate of each modulated signal is the same, the symbol rate is set to 28G, 30.5G, 31.1G, 32.3G, 33.9G, and 36G respectively. The results show that with the increase of the degree of truncation, the entropy value of the signal gradually decreases, and then the transmission performance is improved. Specifically, when the transmission power is in the range of -1dbm to 3dbm, the signal with a smaller entropy value exhibits a smaller bit error rate and better transmission performance. It is worth noting that the optimal transmission power of the four-dimensional signal based on truncated shaping is about 1.0dBm higher than that of PDM-64QAM, which further highlights its advantage in nonlinear performance.

Claims

1. A SP-2048-QAM modulation method based on staged truncation and shaping, characterized in that: The following steps are involved: Step 1: Perform four-dimensional modulation on the information bits; The initial 11 information bits are XORed to generate the 12th parity check bit. Through parity diversity, the X polarization and the Y polarization are correlated, and the original signal points are divided into two parts, so that the minimum Euclidean distance between constellation points is increased to √2 times of the original, thereby realizing four-dimensional modulation. Step 2: intercept signal points within a specific modulus circle layer by layer, and form a series of intermediate modulation formats between the SP-2048-QAM signal and the TPS-SP-2048-QAM signal; The SP-2048-QAM signal consists of nine modulus circles, each of which corresponds to a set of specific signal points. Starting from the outermost modulus circle, signal points are intercepted layer by layer to generate TPS-4, TPS-12 and TPS-20 signals. Step 3: Shape the generated signal based on the constant parameter distribution matcher CCDM. The specific formula for constellation point distribution is as follows: Where: X is the set of constellation points, x i is the complex coordinate of the i-th constellation point, λ≥0 is a free parameter, and M is the number of constellation points; When λ = 0, the distribution is uniform. The larger λ is, the greater the difference in probability between different energy levels is, that is, the smaller the energy of the constellation point is, the greater the probability of being assigned. Given λ, the corresponding P is obtained according to formula (1): X (x i ), and then calculate the information entropy H. The specific formula is as follows: The shaping depth is used to measure the unevenness of the probability distribution. The specific formula is as follows: Step 4: Upsample the modulated signal by four times, and then perform pulse shaping using a root raised cosine filter with a roll-off factor of 0.

01. The specific filter formula is as follows: Where: T is the symbol period, α is the roll-off factor, which ranges between 0 and 1; Step 5: The optical modulator modulates the signal into the optical domain and transmits it through a standard single-mode optical fiber. To enhance the signal, it passes through an erbium-doped fiber amplifier with a noise figure of 4.5 dB; to suppress out-of-band noise, it passes through an optical filter. Step 6: At the receiving end, the optical signal is amplified again and then coherently detected. The detected analog signal is converted into a digital signal through an analog-to-digital converter. Finally, the digital processing algorithm module performs corresponding signal equalization, compensation, and demapping judgment operations to restore the original data.

Images