PON (Passive Optical Network) modulation method and device with high signal-to-noise ratio

Abstract

The invention discloses a high-signal-to-noise-ratio PON modulation method and device, and belongs to the field of optical fiber communication. The device comprises a signal generation module, an optical modulation module, an optical transmission module and a signal processing module. The signal generation module comprises a signal mapping unit, a parallel Delta-Sigma modulation unit and a power domain superposition unit. Based on power domain layering and a Delta-Sigma modulator, a high-order modulation signal is decomposed into two low-order signals in a power domain, and the transmission efficiency of the quantized signal is optimized through power distribution. When the sampling frequency of the Delta-Sigma modulation method is far higher than the signal bandwidth, the quantization noise is uniformly distributed in the whole sampling bandwidth, and the quantization noise in the signal bandwidth is reduced; and under the condition that the noise power is relatively low, the quantized noise of the decomposed high-power signal can be ignored to realize high signal-to-noise ratio analog-to-digital conversion. According to the method, the signal can be accurately recovered by using the receiving end DSP, and the noise suppression capability is remarkably enhanced.

Description

Technical Field

[0001] This invention relates to a power domain hierarchical PON analog-to-digital conversion method and apparatus, belonging to the field of optical fiber communication. Background Technology

[0002] With the rapid development of 5G networks, the demand for signal transmission in passive optical networks (PONs) is also growing rapidly. Modern information services such as cloud computing, data centers, and IoT systems have generated huge demands for high-capacity optical fiber communication. Intensity modulation and direct detection (IM / DD) systems achieve intensity modulation of the optical carrier by converting electrical signals into changes in optical intensity in a Mach-Zehnder modulator, offering advantages such as strong anti-interference capabilities and simple hardware structure. Therefore, IM / DD systems still dominate in short-distance optical fiber communication networks. In practical communication channels, high-resolution analog-to-digital converters (ADCs) are needed to achieve high signal-to-noise ratio (SNR) transmission, ensuring reliable demultiplexing and accurate signal resolution.

[0003] In IM / DD systems, the receiver ADC resolution is set to a low value to reduce computational complexity. The output of a low-resolution ADC cannot accurately represent the original analog signal, introducing quantization errors and failing to meet high signal-to-noise ratio (SNR) requirements. Therefore, achieving high SNR using a low-resolution ADC has been a long-standing challenge. In delta-sigma modulation (DSM) systems, the oversampling operation of DSM expands the spectral range of quantization noise, thereby compressing its magnitude. Furthermore, DSM reshapes the signal spectrum by squeezing quantization noise out of the signal bandwidth, further suppressing in-band noise and reducing the system's ADC resolution requirements. However, traditional low-order DSMs have insufficient noise suppression and cannot support high-order QAM. To further improve SNR, current methods mainly focus on DSM design, such as high-order DSMs or MASH-based DSMs. These DSP designs are complex, have poor stability, and are unsuitable for fiber optic communication systems with high SNR requirements. Summary of the Invention

[0004] The purpose of this invention is to provide a high signal-to-noise ratio (SNR) PON modulation method and apparatus. Based on power domain layering and a Delta-Sigma modulator, a high-order modulation signal is decomposed into two low-order signals in the power domain, and the transmission efficiency of the quantized signal is optimized through power allocation. Under relatively low noise power conditions, the quantization noise of the decomposed high-power signal can be ignored to achieve high SNR analog-to-digital conversion.

[0005] The objective of this invention is achieved through the following technical solution:

[0006] This invention discloses a high signal-to-noise ratio PON modulation method, comprising the following steps:

[0007] Step one maps the original high-order signal into two low-order signals x1(t) and x2(t). Specifically, this is achieved by modulating the signal into a high-order modulation scheme. 4m -QAM format (m≥1), uses a comparator array to divide the input signal amplitude range, and the output centroid value is used as x1(t); the residual generation circuit calculates using a subtractor. get , where k is the amplitude scaling factor. The two low-order signals x1(t) and x2(t) after power domain layering are both formatted as 2. 2m -QAM.

[0008] Step 2: The parallel Delta-Sigma modulation is divided into two parallel channels. The first channel contains an integrator, a quantizer, and a feedback DAC. It performs noise shaping on x1(t) and quantizes it to output an OOK bitstream s1[n]. Quantization noise is generated during the quantization process. The second channel has a structure symmetrical to the first channel. It quantizes x2(t) to generate an OOK bitstream s2[n]. Quantization noise is also generated during the quantization process.

[0009] Delta-Sigma modulation is applied to the two corresponding low-order signals x1(t) and x2(t) in two parallel channels to improve the signal-to-noise ratio. The Delta-Sigma modulation method consists of quantization, oversampling, noise shaping, and filtering. In sampling optical communication input signals, quantization noise is dispersed throughout the signal bandwidth. When the sampling frequency of the Delta-Sigma modulation method is much higher than the signal bandwidth, the quantization noise is evenly distributed throughout the entire sampling bandwidth, reducing the quantization noise within the signal bandwidth.

[0010] Noise shaping improves the signal-to-noise ratio by squeezing in-band quantization noise out of the signal band through the noise transfer function (NTF). The q-order NTF of the Delta-Sigma modulation method is expressed as:

[0011] (1)

[0012] in Zero point The poles are determined by the oversampling rate (OSR), the modulator order, and the maximum radius of the poles in the noise transfer function (NTF).

[0013] Quantize the bit streams x1(t) and x2(t) (2)

[0014] Bitstream (3)

[0015] Where V_min is the minimum value of the signal, V_max is the maximum value of the signal, k is the interval index (k = 0, 1, 2, ..., L-1), and Δ is the interval between each discrete level.

[0016] Noise shaping improves the signal-to-noise ratio by adjusting the noise transfer function (NTF) to squeeze in-band quantization noise out of the signal band. A low-pass filter (LPF) is used to filter out out-of-band quantization noise at the receiver.

[0017] The sampling frequency being much higher than the signal bandwidth means that the sampling frequency is more than twice the signal bandwidth.

[0018] Step 3: Convert s1[n] into a current pulse with a weight of 4I0; convert s2[n] into a current pulse with a weight of I0, and then convert the two currents according to the weighting coefficients. and The sums generate a multi-level transmission signal s[n].

[0019] Step 4: The noise cancellation and reconstruction module uses serial interference cancellation (SIC) technology at the receiving end to separate and recover the original signal, separating the high-power signal from s[n]. and low power signals and utilize The high-power characteristics eliminate the quantization noise generated in step two, resulting in... and The original high-order signal is restored, and the digital signal is converted into an analog signal, thus realizing power domain hierarchical PON analog-to-digital conversion.

[0020] Preferably, quantization noise is concentrated in the low-power signal path through power domain layering, and quantization noise in the high-power path is eliminated through signal reconstruction at the receiver, so that the total SNR improvement of the passive optical network system meets the following requirements:

[0021] SNR Gain =10lg (P signal / P noise (4)

[0022] Among them: SNR Gain For signal-to-noise ratio gain, P signal For signal power, P noise This represents noise power.

[0023] This invention discloses a high signal-to-noise ratio (SNR) PON modulation system for implementing the aforementioned high SNR PON modulation method. The high SNR PON modulation system includes a signal generation module, an optical modulation module, an optical transmission module, and a signal processing module.

[0024] The signal generation module includes a signal mapping unit, a parallel Delta-Sigma modulation unit, and a power domain superposition unit; the signal mapping unit is used to implement step one of the method, "mapping the original high-order signal into two low-order signals". and The function includes a comparator array and a residual generation circuit; the parallel Delta-Sigma modulation unit is used to perform the parallel modulation in step two of the method, for... and Noise shaping and quantization are performed to output an OOK bitstream; the power domain superposition module is used to realize the multi-level signal synthesis in step three of the method, converting the two OOK bitstreams into multi-level transmission signals, which includes a current conversion unit and a signal superposition unit; the superimposed signal is encoded into a waveform generated by an arbitrary waveform generator, and the optical carrier is generated by an external cavity laser.

[0025] The optical modulation module is used to modulate electrical signals onto an optical carrier and convert them into transmittable optical signals. A Mach-Zehnder modulator (MZM) receives the optical carrier output from the optical generator module and the s[n] output from the power domain superposition module. By controlling the bias voltage of the MZM with an electrical signal, the amplitude variation of the electrical signal is converted into an intensity variation of the optical carrier (IM / DD modulation). The modulation bandwidth is ≥50GHz to support 40Gbaud signal transmission. An erbium-doped fiber amplifier (EDFA) amplifies the modulated optical signal, with adjustable output optical power (0~10dBm) to compensate for subsequent optical path losses. An optical coupler splits the amplified optical signal into two paths (the power ratio can be adjusted as needed) for multi-mode transmission scenarios (such as OAM mode multiplexing). A spatial optical modulator (SLM) receives the collimated beam and converts the Gaussian beam into an orbital angular momentum (OAM) mode with specific topological charges (such as l=±2, ±3) by loading a vortex phase pattern, achieving spatial multiplexing. A beam splitter splits multiple OAM signals. The beam is multiplexed, a quarter-wave plate converts linearly polarized light into circularly polarized light, and a polarization beam splitter decomposes the circularly polarized light into orthogonally linearly polarized light, ensuring the consistency of the polarization state of the multiplexed beam.

[0026] The optical transmission module is used to realize long-distance transmission of optical signals and includes: few-mode fiber (such as 2.3km RCF). Low-loss, low-mode-crosstalk few-mode fiber is used to couple the multiplexed OAM beam into the fiber. The modal dispersion of the fiber is ≤10ps / km, ensuring minimal signal distortion during transmission.

[0027] The signal processing module is used to implement signal recovery and noise cancellation in step four of the method. It includes a preprocessing unit, a synchronization unit, and a demodulation unit; the synchronization unit is the demodulation unit of the noise cancellation and reconstruction module. The preprocessing unit has a built-in low-pass filter (cutoff frequency matched to signal bandwidth) and a resampling circuit to filter out out-of-band quantization noise and adjust the electrical signal sampling rate to the signal baud rate. The synchronization unit includes a clock recovery circuit and a frequency offset estimator, achieving symbol synchronization and carrier frequency offset compensation through pilot signals, with a synchronization accuracy ≤1ppm. The noise cancellation and reconstruction module consists of a serial interference cancellation (SIC) unit, a hard decision unit, and a signal synthesizer; the SIC unit separates high-power components from the received signal. and low power components Hard decision unit Perform amplitude calibration, eliminate quantization noise, and reconstruct. The signal synthesizer will reconstruct Compared with noise-suppressed (by subtracting) The components are synthesized to restore the original high-order signal; the demodulation unit uses an Adabound MIMO equalizer to compensate for mode crosstalk and dispersion during transmission, and restores the original data through DMT demodulation (including FFT transformation and channel estimation), and finally outputs the bit error rate (BER) calculation result.

[0028] Beneficial effects:

[0029] 1. This invention discloses a high signal-to-noise ratio (SNR) PON modulation method and apparatus, which decomposes a high-order modulation signal into multiple low-order signals in the power domain and optimizes the transmission efficiency of the quantized signals through power allocation. Under conditions of relatively low noise power, the quantization noise of the decomposed high-power signals is ignored, thereby improving the overall SNR of the passive optical network system.

[0030] 2. This invention discloses a high signal-to-noise ratio (SNR) PON modulation method and apparatus. Since noise power is usually fixed, high-power signals generate a high SNR, making them less susceptible to noise. Under favorable conditions, the receiver's DSP can accurately recover the signal with negligible or even zero error, significantly enhancing noise suppression capabilities. This results in higher-order QAM modulation formats exhibiting better performance and stronger robustness, whereas traditional DSM quantization schemes struggle to provide sufficient SNR to support similar modulation formats.

[0031] 3. This invention discloses a high signal-to-noise ratio (SNR) PON modulation method and apparatus. Using multiple low-order DSM schemes to achieve the same SNR gain is more stable than using a single high-order DSM scheme. While maintaining a high quantization SNR, it reduces hardware and design complexity, improves the SNR performance of the transmission system, and significantly reduces ADC resolution requirements. Therefore, this invention is conducive to the application of ultra-high-order modulation optical fiber communication systems. Attached Figure Description

[0032] Figure 1 (a) Power domain superposition principle. (b) Addition of parallel signal power domains.

[0033] Figure 2 This is the principle of DSM noise shaping technology.

[0034] Figure 3 Diagram of an MDM-PON IM / DD transmission system at a 2.3km RCF.

[0035] Figure 4 The relationship between measured BER and received optical power (ROP) is presented in an MDM-PON IM / DD transmission system with a 2.3 km RCF for (a) 65,536 QAM with l = 2, (b) 65,536 QAM with l = 3, and (c) 1048,576 QAM with l = 2 and (d) 1048,576 QAM with l = 3.

[0036] Figure 5 The ratio of the measured signal-to-noise ratio to the received optical power of the recovered 65,536QAM signal in OAM mode l = 2;

[0037] Figure 6 This is a flowchart of a high signal-to-noise ratio PON modulation method according to the present invention. Detailed Implementation

[0038] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. The technical problems solved by the present invention and its beneficial effects are also described. It should be noted that the described embodiments are only intended to facilitate understanding of the present invention and do not constitute any limitation thereof.

[0039] Example 1: Signal Mapping and Delta-Sigma Modulation

[0040] like Figure 6 As shown in the figure, the specific implementation steps of the high signal-to-noise ratio PON modulation method disclosed in this embodiment are as follows:

[0041] Step 1: The original signal has 1048576 (2 20-QAM (m=5) constellation points, symbol rate 40 Gbaud. A comparator array is used to divide the input signal amplitude into 1024 intervals. This can be decomposed into two 1024 (2... 10 ) blocks, each block has 1024 (2 10 Point ). Output the centroid values ​​of each interval as a high-power signal to obtain the decomposed signal y1. Move the centroids of each block of the original signal to the origin position to obtain the low-power decomposed signal y2, as shown. Figure 1 As shown in (a).

[0042] Each symbol of the two obtained low-order signals can be represented as:

[0043] (5)

[0044] (6)

[0045] in

[0046] (7)

[0047] (8)

[0048] (9)

[0049] (10)

[0050] (11)

[0051] The new sign point of the decomposed signal obtained through a certain shrinkage operation is consistent with the constellation point of the standard 1024 QAM.

[0052] Here, `label` represents the vectorized form of the set. Function The operator that returns the index of the minimum value in a given vector.

[0053] Step 2: Perform 1-bit DSM quantization on the two signals y1 and y2 separately, as follows: Figure 2 As shown, with the oversampling rate OSR set to 8 and the quantization bits set to 1, the generated OOK signals are x1(t) and x2(t). After running the closed-loop analysis of the noise shaping (CLANS) function, based on the generated zeros and poles, the NTF expression can be expressed as:

[0054]

[0055] The signal-to-noise ratio (SNR) provided by a 1-bit DSM can easily meet the SNR requirements of low-order signals. Therefore, the original low-order signal y1 can be effectively reconstructed from the received, negligible error signal s1. Subsequently, the hard decision result is the amplitude-error-free recovery of y1, which also means that the quantization noise associated with s1 can be removed. Signal s1 can be regarded as a clean signal with negligible quantization noise, significantly improving the overall SNR of the system.

[0056] Step 3: The two OOK signals x1(t) and x2(t) obtained in Step 2 are superimposed in the power domain to obtain a new PAM-4 signal, such as... Figure 1 As shown in (b), it can be represented as:

[0057] (12)

[0058] For the total power of P, the powers allocated to the two signals are P1 and P2, respectively. In this scheme, a normal PAM4 signal can be obtained through appropriate power allocation (P1=0.8 and P2=0.2). Since noise power is usually fixed, high-power signals produce a higher signal-to-noise ratio, making them less susceptible to noise.

[0059] Example 2: Optical Transmission and Reception

[0060] like Figure 3 As shown, a pseudo-random binary sequence (PRBS) is generated at the transmitter, followed by standard QAM mapping. The QAM signal is converted into a DMT signal with an FFT size of 1024. OFDM modulation is then performed, with an IFFT operation applied to the signal, and a cyclic prefix of 1 / 4 IFFT length is introduced as a guard interval to prevent inter-symbol interference. An OFDM symbol consists of 256 subcarriers, of which 248 subcarriers carry data symbols, 5 subcarriers are used for pilot loading, and the remaining subcarriers are filled with zeros. After serial-to-parallel conversion, the signal is input to the AWG at a rate of 40 GSa / s for the analog-to-digital conversion method provided in step 1. It is then amplified by an electrical amplifier (EA). A Mach-Zehnder modulator (MZM) modulates the generated 40 Gbaud PAM4 electrical signal onto the optical carrier. After EDFA amplification, the beam is split into two components using an optical coupler (OC). One branch is delayed by 10 mSMF for data mode decorrelation. The two beams are then collimated and linearly polarized, and then converted into OAM beams with l = +2 and l = +3 using two spatial light modulators. The multiplexed OAM beams are converted into circularly polarized states by a quarter-wave plate (QWP) and coupled into a 2.3 km RCF.

[0061] At the receiver, the multiplexed OAM mode is split into two beams using a beam splitter (BS) and then converted into linearly polarized beams via a QWP. The beams are then converted into Gaussian beams using a vortex phase plate (VPP) with opposite topological charges. After photodiode (PD) detection, a real-time oscilloscope (OSC) records the waveform at a sampling rate of 256 GSa / s. Offline DSP includes resampling, low-pass filtering, synchronization, SiC, DMT demodulation, and bit error rate calculation, such as... Figure 3 As shown.

[0062] Figure 4 The bit error rate performance at 65,536 QAM and 1,048,576 QAM is compared with the received optical power of the power-domain hierarchical PON analog-to-digital converter, the MASH analog-to-digital converter, and the traditional 2-bit DSM analog-to-digital converter. The 6th-order NTF of the MASH analog-to-digital converter can be rewritten as:

[0063] (13)

[0064] To eliminate in-band quantization noise, the quantization noise generated by the DSM is extracted as its in-band component through a filter. The filter design must carefully consider the gain and bandwidth limitations within the signal band. Furthermore, since the in-band quantization noise power is typically low, it must be amplified to meet the input requirements of the DSM2. Correspondingly, at the receiver, this component must be attenuated using the same factor used for amplification at the transmitter. MASH DSMs also require a higher-order NTF to maintain a high signal-to-noise ratio, as shown in Equations 12 and 13, which further increases the system complexity. In contrast, our proposed scheme eliminates the need for filter design and is independent of a higher-order DSM, thus significantly reducing complexity. The measured received optical power range is... When the ROP reaches 5 dBm, the bit error rate (BER) of the 2nd-order 2-bit DSM scheme cannot be lower than the HD-FEC threshold in OAM mode (l = 2, 3). Conversely, using a 6th-order 2-bit DSM, the BER in OAM mode can be lower than the HD-FEC threshold l = 2 when the ROP reaches 5 dBm. Increasing the order of the DSM can improve the signal-to-noise ratio (SNR), but the improvement is relatively limited. In the ROP range of -3 to 5 dBm, the BER of the 65,536QAM signal transmitted using the PDL-DSM scheme with two 2nd-order 1-bit DSMs is lower than the Hard Decision Forward Error Correction (HD-FEC) threshold at both l = 2 and l = 3. Compared with the MASH DSM scheme using two 6th-order 1-bit DSMs, the PDL-DSM scheme not only reduces the hardware design complexity of using only 2nd-order DSMs, but also improves the receiver sensitivity by 3 dB in OAM mode at l = 3 with the HD-FEC threshold.

[0065] In this scheme, when transmitting 1,048,576 QAM, the bit error rate (BER) of the l=2 OAM mode is lower than the HD-FEC threshold when the ROP is 3 dBm. However, other schemes cannot meet the HD-FEC threshold. Similarly, in OAM mode, only the PDL-DSM scheme achieves a BER lower than the HD-FEC threshold when the ROP reaches 2 dBm. Experimental results show that simply increasing the order of the DSM has limited effect on improving the signal-to-noise ratio (SNR), while the PDL-DSM scheme can improve the SNR when using lower orders.

[0066] To comprehensively evaluate the signal-to-noise ratio enhancement capability of the PDL-DSM scheme, Figure 5 The signal-to-noise ratio (SNR) and relative error post (ROP) performance of the recovered 65,536 QAM signal in OAM mode l = 2 are described. When the ROP is 3 dBm, the PDL-DSM scheme achieves a SNR gain of 52.5 dB, sufficient to support 65,536 QAM transmission, with a corresponding bit error rate of 7 × 10⁻⁵, far below the HD-FEC threshold. In contrast, under the same conditions, the SNR of a 2-bit, 2nd-order DSM quantization scheme operating at the same baud rate as the PAM4 signal is only about 37.9 dB. This indicates that the PDL-DSM scheme provides a significant SNR improvement of approximately 14.6 dB compared to the 2-bit, 2nd-order DSM scheme, significantly enhancing its noise suppression capability and enabling higher-order QAM modulation formats to exhibit better performance and stronger robustness. Traditional 2-bit, 2nd-order DSM quantization schemes struggle to provide sufficient SNR to support similar modulation formats.

[0067] The above detailed description further illustrates the purpose, technical solution, and beneficial effects of the invention. It should be understood that the above description is only a specific embodiment of the present invention and does not limit the scope of protection of the present invention. Any modifications, substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A high signal-to-noise ratio PON modulation method, characterized in that: Based on power domain hierarchical structure and Delta-Sigma modulator, a high-order modulated signal is decomposed into two low-order signals in the power domain. The transmission efficiency of the quantized signal is optimized by power allocation, thereby achieving high signal-to-noise ratio analog-to-digital conversion.

2. The high signal-to-noise ratio PON modulation method as described in claim 1, characterized in that: Includes the following steps: Step one maps the original high-order signal into two low-order signals x1(t) and x2(t). Specifically, this is achieved by modulating the signal into a high-order modulation scheme. 4m -QAM format (m≥1), uses a comparator array to divide the input signal amplitude range, and the output centroid value is used as x1(t); the residual generation circuit calculates using a subtractor. We obtain x2(t), where k is the amplitude scaling factor; the two lower-order signals x1(t) and x2(t) after power domain layering are both formatted as 2. 2m -QAM; Step 2: The parallel Delta-Sigma modulation is divided into two parallel channels; the first channel contains an integrator, a quantizer and a feedback DAC, which performs noise shaping on x1(t) and quantizes it to output an OOK bit stream s1[n]. Quantization noise will be generated during the quantization process. The structure of the second channel is symmetrical to that of the first channel. It quantizes x2(t) to generate an OOK bit stream s2[n]. Quantization noise will be generated during the quantization process. Delta-Sigma modulation is applied to the two corresponding low-order signals x1(t) and x2(t) in the two parallel channels to improve the signal-to-noise ratio. The Delta-Sigma modulation method consists of quantization, oversampling, noise shaping, and filtering. In the sampling of optical communication input signals, quantization noise is dispersed throughout the signal bandwidth. When the sampling frequency of the Delta-Sigma modulation method is much higher than the signal bandwidth, the quantization noise is evenly distributed throughout the entire sampling bandwidth, reducing the quantization noise in the signal bandwidth. Step 3: Convert s1[n] into a current pulse with a weight of 4I0; convert s2[n] into a current pulse with a weight of I0, and then convert the two currents according to the weighting coefficients. and The sums generate a multi-level transmission signal s[n]; Step 4: The noise cancellation and reconstruction module uses serial interference cancellation (SIC) technology at the receiving end to separate and recover the original signal, separating the high-power signal from s[n]. and low power signals and utilize The high-power characteristics eliminate the quantization noise generated in step two, resulting in... and It restores the original high-order signal and converts the digital signal into an analog signal, thus realizing power domain hierarchical PON analog-to-digital conversion.

3. The high signal-to-noise ratio PON modulation method as described in claim 2, characterized in that: In step two, Noise shaping improves the signal-to-noise ratio by squeezing in-band quantization noise out of the signal band through the noise transfer function (NTF). The q-order noise transfer function (NTF) of the Delta-Sigma modulation method is expressed as: (1) in Zero point The poles are determined by the oversampling rate OSR, the modulator order, and the maximum radius of the poles in the noise transfer function NTF. Quantize the bit streams x1(t) and x2(t) (2) Bitstream (3) Where: V_min is the minimum value of the signal, V_max is the maximum value of the signal, k is the interval index (k = 0, 1, 2,..., L-1), and Δ is the interval between each discrete level; Noise shaping improves the signal-to-noise ratio by setting the noise transfer function (NTF) to squeeze in-band quantization noise out of the signal band; and uses a low-pass filter (LPF) to filter out out-of-band quantization noise at the receiver.

4. The high signal-to-noise ratio PON modulation method as described in claim 3, characterized in that: The sampling frequency being much higher than the signal bandwidth means that the sampling frequency is more than twice the signal bandwidth.

5. The high signal-to-noise ratio PON modulation method as described in claim 4, characterized in that: By using power domain layering to concentrate quantization noise on the low-power signal path and eliminating quantization noise on the high-power path through signal reconstruction at the receiver, the total SNR improvement of the passive optical network system meets the following requirements: SNR Gain =10lg (P signal / P noise ) (4) Among them: SNR Gain For signal-to-noise ratio gain, P signal For signal power, P noise This represents noise power.

6. A high signal-to-noise ratio (SNR) PON modulation system for implementing the high SNR PON modulation method as described in claim 2, 3, 4, or 5, characterized in that: It includes a signal generation module, an optical modulation module, an optical transmission module, and a signal processing module; The signal generation module includes a signal mapping unit, a parallel Delta-Sigma modulation unit, and a power domain superposition unit; the signal mapping unit is used to implement step one of the method, "mapping the original high-order signal into two low-order signals". and It includes a comparator array and a residual generation circuit; a parallel Delta-Sigma modulation unit is used to perform the parallel modulation in step two of the method, for... and Perform noise shaping and quantization to output an OOK bitstream; The power domain superposition module is used to realize the multi-level signal synthesis in step three of the method, converting two OOK bit streams into multi-level transmission signals. It includes a current conversion unit and a signal superposition unit. The superimposed signals are encoded into a waveform generated by an arbitrary waveform generator, and the optical carrier is generated by an external cavity laser. The optical modulation module modulates an electrical signal onto an optical carrier and converts it into a transmittable optical signal. The Mach-Zehnder modulator (MZM) receives the optical carrier output from the optical generator module and the s[n] output from the power domain superposition module. By controlling the bias voltage of the MZM with an electrical signal, it converts the amplitude change of the electrical signal into an intensity change of the optical carrier, with a modulation bandwidth ≥ 50 GHz to support 40 Gbaud signal transmission. An erbium-doped fiber amplifier amplifies the modulated optical signal, with an output optical power modulation range of 0~10 dBm to compensate for subsequent optical path losses. An optical coupler splits the amplified optical signal into two paths for multi-mode transmission scenarios. The spatial optical modulator (SLM) receives the collimated beam and converts it into an orbital angular momentum (OAM) mode with specific topological charges (e.g., l=±2, ±3) by loading a vortex phase pattern, achieving spatial multiplexing. A beam splitter splits the multiple OAM signals. The beam is multiplexed, a quarter-wave plate converts linearly polarized light into circularly polarized light, and a polarization beam splitter decomposes the circularly polarized light into orthogonally linearly polarized light, ensuring the consistency of the polarization state of the multiplexed beam. The optical transmission module is used to realize long-distance transmission of optical signals. It includes: few-mode fiber; using low-loss, low-mode crosstalk few-mode fiber, the multiplexed OAM beam is coupled into the fiber, and the mode dispersion of the fiber is ≤10ps / km to ensure that the signal distortion is minimized during transmission. The signal processing module is used to implement signal recovery and noise cancellation in step four of the method. It includes a preprocessing unit, a synchronization unit, and a demodulation unit. The synchronization unit is the demodulation unit of the noise cancellation and reconstruction module. The preprocessing unit incorporates a low-pass filter and a resampling circuit. It filters out out-of-band quantization noise and adjusts the electrical signal sampling rate to the signal baud rate. The synchronization unit includes a clock recovery circuit and a frequency offset estimator, achieving symbol synchronization and carrier frequency offset compensation through pilot signals, with a synchronization accuracy ≤1ppm. The noise cancellation and reconstruction module consists of a serial interference cancellation (SIC) unit, a hard decision unit, and a signal synthesizer. The SIC unit separates high-power components from the received signal. and low power components Hard decision unit Perform amplitude calibration, eliminate quantization noise, and reconstruct. The signal synthesizer will reconstruct Compared with noise-suppressed The original high-order signal is synthesized and restored; the demodulation unit uses an Adabound MIMO equalizer to compensate for mode crosstalk and dispersion during transmission, restores the original data through DMT demodulation, and outputs the bit error rate (BER) calculation result.

7. A high signal-to-noise ratio PON modulation system as described in claim 6, characterized in that: The preprocessing unit has a built-in low-pass filter whose cutoff frequency is matched to the signal bandwidth.

8. A high signal-to-noise ratio PON modulation system as described in claim 7, characterized in that: The DMT demodulation includes FFT transformation and channel estimation.

Images