Method and system for online registration of ONU in P2MP system

Abstract

This invention provides a method and system for ONU online registration in a P2MP system. The ONU randomly selects a sequence from a sequence code set and performs analog intensity modulation on it. The registration signal is then shifted to the frequency band guard interval and sent to the OLT along with the service signal. Upon receiving these signals, the OLT first performs frequency shifting and filtering to extract the registration signal. For the registration signal, a resampling operation is performed, followed by analog square detection to remove the DC component. Then, the two polarization signals are combined and repeatedly averaged periodically. The processed signal is then cyclically cross-correlated with the local sequence code set. The resulting peak value is detected; if the detected peak value exceeds a preset decision threshold, the sequence code is considered activated, and its delay is estimated. This invention achieves high accuracy in frequency offset estimation and delay estimation, and the designed registration signal has minimal impact on the performance of the service signal; it also allows the registration signal power to be much smaller than the service signal power.

Description

Technical Field

[0001] This invention relates to the field of optical communication technology, and more specifically, to a method and system for online registration of optical network units (ONUs) in point-to-multipoint (P2MP) systems. Background Technology

[0002] In today's era of rapid technological advancement, cutting-edge technologies such as the Internet of Things (IoT), edge computing, and 5G / 6G are emerging at an alarming pace, bringing new demands and challenges to various fields. Against this backdrop, passive optical networks (PONs) have also ushered in a period of rapid development.

[0003] With the continuous evolution of technology, service providers have not only placed unprecedented demands on passive optical networks in terms of network capacity and speed, but are also actively seeking innovative solutions that can effectively reduce transmission costs, power consumption and latency, while also hoping to further enhance the scalability and flexibility of services.

[0004] Currently, as coherent optics technology gradually penetrates into lower-end applications, coherent passive optical networks (PONs) are replacing traditional direct-modulation and direct-detection passive optical networks with their own advantages, becoming the mainstream development trend in the industry. At the same time, the flexible, scalable, and cost-effective P2MP transmission mode is also gradually emerging, becoming an important evolution direction for traditional point-to-point transmission optical networks.

[0005] One effective way to achieve P2MP coherent PON is to use Digital Subcarrier Multiplexing (DSCM) technology. This technology divides a single carrier signal with a large bandwidth into multiple independent subcarrier signals with small bandwidths in the digital domain for transmission. This innovative approach successfully eliminates the electrical aggregation process in traditional point-to-point (P2P) networks, significantly reduces the number of transceivers, and effectively lowers equipment costs.

[0006] DSCM transmission offers several advantages over single-carrier transmission, including larger dispersion compensation tolerance, higher nonlinearity tolerance, and more flexible bandwidth allocation. It is precisely because of these unique advantages that the DSCM-based P2MP coherent PON architecture has received widespread attention and research. However, in this new architecture, the fundamental change in channel multiplexing methods has led to a series of new problems that need to be addressed.

[0007] A patent search revealed patent CN107342973B, which discloses a registration method, apparatus, and system for a passive optical network (PON). This patent includes: an optical line terminal (OLT) receiving continuously transmitted registration information from an optical network unit (ONU) via a preset wavelength; the registration information is randomly delayed before being continuously transmitted by the ONU to the OLT; and the OLT performing registration processing based on the received registration information. This patent does not address the issues of effectively distinguishing between multiple users accessing the network simultaneously, or how to quickly complete uplink synchronization.

[0008] In summary, given the problems of the existing technologies, researching a method and system for ONU online registration in a P2MP system has become a critical task that urgently needs to be addressed. Summary of the Invention

[0009] To address the shortcomings of existing technologies, the present invention aims to provide a method and system for online registration of optical network units (ONUs) in point-to-multipoint (P2MP) systems.

[0010] A method for ONU online registration in a P2MP system according to the present invention includes the following steps:

[0011] Step S1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, thus obtaining an ONU with completed frequency offset correction;

[0012] Step S2, Sequence Code Set Construction: Construct the registered sequence code set using the sequence;

[0013] Step S3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence;

[0014] Step S4, Modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC sampling rate, then applies DC, and then performs a square root operation to generate the registration signal;

[0015] Step S5, Frequency shifting and transmission of the registration signal: Shift the registration signal to the frequency band guard interval and send it to the OLT along with other service signals;

[0016] Step S6, extraction of registration signal: For the received signal, the OLT first performs IQ front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering.

[0017] Step S7, Service signal processing: For the remaining service signals after extracting the registration signal, perform signal recovery and quality assessment, and finally calculate the bit error rate and signal-to-noise ratio;

[0018] Step S8, processing of the registration signal: the registration signal is squared and the DC component is removed. The registration signal is then subjected to analog intensity detection. The signals on the two polarizations are merged. The signal-to-noise ratio of the received registration signal is improved by averaging multiple signal segments. Finally, the signal is cyclically cross-correlated with the local code set to detect the registration sequence and the corresponding time delay information.

[0019] Preferably, in step S2, the sequence is an m-sequence or a Gold sequence.

[0020] Preferably, step S4 includes:

[0021] The registration signal from the transmitting end is a registration sequence randomly selected from the sequence code set, denoted as x. u (n), n=0,…,N-1. Then upsampled to the DAC sampling rate to obtain signal x. ′ u (n), by adding DC and square root operations to simulate intensity modulation, and then shifting the frequency to the guard interval of the frequency band, the signal x is transmitted. T (n) is represented by equation (1), u dc For the added DC component, f c N is the carrier frequency of the registered signal. T For the length of the transmitted signal,

[0022]

[0023] Preferably, step S6 includes:

[0024] At the receiving end, the OLT receives signals across the entire frequency domain via a coherent receiver, including the received X and Y polarized signals. This is expressed as equation (2), where Δt represents the transmission delay and Δf represents the residual frequency offset. This represents the transmitted X-polarized and Y-polarized service signals, and n0(n) represents the system noise.

[0025]

[0026] The registered signal at the guard interval is selected through frequency shifting and filtering operations. Represented by equation (3), n′0(n) represents the influence of service signals and system noise on the registration signal, u dc For the added DC component, N T For the length of the transmitted signal,

[0027]

[0028] Preferably, in step S7, signal recovery and quality assessment includes: dispersion compensation, multi-input multi-output equalization based on multi-mode algorithm, fourth-power frequency offset estimation and compensation, symbol synchronization, and carrier phase recovery based on phase-locked loop.

[0029] Preferably, step S8 includes the following sub-steps:

[0030] Step S8.1, estimate the residual frequency offset Δf: Square the signal to remove the DC component, complete the analog strength detection of the registered signal, and obtain the signal. Represented by Equation (4), ignoring signal amplitude impairment, n″0(n) represents the cumulative noise impact, including the impact of the service signal on the registration signal.

[0031]

[0032] Step S8.2, for the signal Peak detection is performed to obtain latency information;

[0033] Step S8.3: Combine the signals on the two polarizations to obtain signal x. r′ (n), expressed as equation (8).

[0034]

[0035] Step S8.4: In the time domain, the registered signal is averaged by multiple segments, and finally, it is cyclically cross-correlated with the local code set through cyclic cross-correlation detection, as shown in equation (9), x i (n) represents the i-th registration sequence code in the code set, thereby detecting the registration sequence selected by the registration signal, and calculating the time delay based on the position corresponding to the maximum correlation peak, thus completing the processing and analysis of the registration signal.

[0036] x e (n) = IFFT{FFT(x) r′ (n))FFT(x i (n))} (9)

[0037] Preferably, in step S8.2, there are two methods for peak detection: normalized absolute peak detection and peak-to-average ratio detection.

[0038] Preferably, in the normalized absolute peak detection, let the cyclic cross-correlation result of the registered sequences a(n) and b(n) be (5), where FFT refers to Fast Fourier Transform and IFFT refers to Fast Inverse Fourier Transform.

[0039] R(n)=|{IFFT{FFT{a(n)}FFT{b(n)}}}| (5)

[0040] The normalized absolute peak value is shown in equation (6), where ||a|| represents the modulus of the registered sequence a(n).

[0041]

[0042] The position corresponding to the peak is d = argmax n M(n) represents the time delay information.

[0043] Preferably, the peak-to-average power ratio (PAPR) is detected as shown in equation (7), which is the ratio of the maximum value to the average value of the relevant results.

[0044]

[0045] The position corresponding to the peak is d = argmax n M(n) represents the time delay information.

[0046] This invention also provides a system for ONU online registration in a P2MP system, comprising:

[0047] Module M1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, resulting in an ONU with completed frequency offset correction;

[0048] Module M2, Sequence Code Set Construction: Constructs the registered sequence code set using sequences;

[0049] Module M3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence;

[0050] Module M4, modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC sampling rate, then applies DC, and then performs a square root operation to generate the registration signal;

[0051] Module M5, Frequency Shifting and Transmission of Registration Signal: Shifts the registration signal to the frequency band guard interval and transmits it to the OLT along with other service signals;

[0052] Module M6, extraction of registration signal: For the received signal, the OLT first performs IQ front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering.

[0053] Module M7, Service Signal Processing: For the remaining service signals after extracting the registration signal, perform signal recovery and quality assessment, and finally calculate the bit error rate and signal-to-noise ratio;

[0054] Module M8, processing of the registration signal: squaring the registration signal, removing the DC component, performing analog intensity detection on the registration signal, merging the signals on the two polarizations, improving the signal-to-noise ratio of the received registration signal by averaging multiple signal segments, and finally performing cyclic cross-correlation with the local code set to detect the registration sequence and corresponding time delay information.

[0055] Compared with the prior art, the present invention has the following beneficial effects:

[0056] 1. In this invention, the generation and detection of the registration signal are compatible with existing coherent transceivers, and the transceiver algorithm has low complexity.

[0057] 2. The online registration scheme of the present invention has high ranging accuracy.

[0058] 3. The registration signal of this invention has little impact on the performance of service signals. At the same time, it can support a large power difference in the Optical Distribution Network (ODN) and has high practical application value. Attached Figure Description

[0059] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0060] Figure 1 This is a schematic diagram of a method for online registration of ONU in a P2MP system according to an embodiment of the present invention;

[0061] Figure 2 This is the registration signal generation and detection scheme in the embodiments of the present invention;

[0062] Figure 3 This is a schematic diagram illustrating the spectral position relationship between the registration signal and the service signal in an embodiment of the present invention;

[0063] Figure 4 This is a schematic diagram of the experimental scheme in an embodiment of the present invention;

[0064] Figure 5 The power spectrum diagrams of the registration signal and the service signal transmitted on the X and Y polarizations in this embodiment of the invention are shown.

[0065] Figure 6 This is a DSP flowchart illustrating the processing of receiving service signals and registration signals in an embodiment of the present invention.

[0066] Figure 7 The power spectrum of the received signal when the relative service signal power of location 3 in this embodiment of the invention is (a) 15dB, (b) 20dB, (c) 25dB, (d) 30dB is smaller than that of the service signal.

[0067] Figure 8 The following graph shows the results of selecting Gaussian distribution for distribution fitting and parameter estimation in this embodiment of the invention to obtain the optimal detection threshold bestTh, false detection rate Pe, false miss rate Pmiss, and false alarm rate Pfalse (Gold sequence, using normalized absolute peak detection, when the power of the registration signal at position 3 is (a) 15dB, (b) 20dB, (c) 25dB, (d) 30dB less than the power of the traffic signal, the optimal detection threshold bestTh, false detection rate Pe, false miss rate Pmiss, and false alarm rate Pfalse results).

[0068] Figure 9 The Gold sequence in this embodiment of the invention uses peak-to-average power ratio detection. When the power of the registration signal at position 3 is (a) 15dB, (b) 20dB, (c) 25dB, and (d) 30dB less than that of the service signal, the results of the optimal detection threshold bestTh, false detection rate Pe, false miss rate Pmiss, and false alarm rate Pfalse are shown in the figure.

[0069] Figure 10 The graph shows the delay estimation error results in 200 detections when the power of the registration signal at position 3 is (a) 15dB, (b) 20dB, (c) 25dB, and (d) 30dB less than the service signal power, using normalized absolute peak detection for the Gold sequence in this embodiment of the invention. The crosses indicate that the signal was not detected or was detected incorrectly.

[0070] Figure 11 The graph shows the results of delay estimation errors in 200 detections when the peak-to-average power ratio of the Gold sequence in this embodiment of the invention is used. When the power of the registration signal at position 3 is (a) 15dB, (b) 20dB, (c) 25dB, and (d) 30dB lower than that of the service signal, the crosses indicate that the signal was not detected or was detected incorrectly.

[0071] Figure 12 This is a graph showing the X and Y polarization frequency offset estimation error results from 100 experiments in the experiment of measuring the effect of different frequency offsets on detection performance in this embodiment of the invention.

[0072] Figure 13 The graph shows the results of false detection rate Pe, false alarm rate Pmiss, and false alarm rate Pfalse under different frequency offsets in the embodiments of the present invention.

[0073] Figure 14 The diagram shows the time delay estimation error results under different frequency offsets in the embodiments of the present invention (time delay estimation error in 200 detections when the Gold sequence is detected using normalized absolute peak value, and the frequency offset of the registered signal at position 3 is (a) -500MHz, (b) -250MHz, (c) 250MHz, (d) 500MHz).

[0074] Figure 15 The curves showing the variation of the polarization BER of ONU(ONU-2)X and Y at position 2 with ROP when the frequency offset is -500MHz, -250MHz, 0MHz, 250MHz, and 500MHz, representing the unregistered signal in this embodiment of the invention.

[0075] Figure 16 The curves showing the variation of the polarization BER of ONU(ONU-3)X and Y at position 3 with ROP are given for the unregistered signal (w / o register signal) at frequency offsets of -500MHz, -250MHz, 0MHz, 250MHz, and 500MHz. Detailed Implementation

[0076] Currently, as coherent optics technology continues to expand into wider fields, the DSCM-based P2MP coherent PON architecture has attracted much attention and in-depth research due to its unique advantages. Under this new architecture, the channel multiplexing method has undergone a significant transformation, changing from traditional time-division multiple access to hybrid time-frequency division multiple access. This change necessitates the exploration of a new online registration mechanism to ensure efficient and orderly data transmission within the system.

[0077] The present invention will be described in detail below with reference to specific embodiments. The following examples will help those skilled in the art to further understand the present invention, but are not intended to limit the present invention in any form. It should be noted that, for those skilled in the art, several changes and improvements can be made without departing from the scope of the present invention. These all fall within the scope of protection of the present invention.

[0078] This invention provides a method and system for ONU online registration in a P2MP system. The ONU randomly selects a sequence from a sequence code set and performs analog intensity modulation on it. The registration signal is then shifted to the frequency band guard interval and sent to the Optical Line Terminal (OLT) along with the service signal. Upon receiving these signals, the OLT first performs frequency shifting and filtering to extract the registration signal. For the registration signal, a resampling operation is performed, followed by analog square detection to remove the DC component. Then, the two polarization signals are combined and repeatedly averaged periodically. The processed signal is then cyclically cross-correlated with the local sequence code set. The resulting peak value is detected; if the detected peak value exceeds a preset decision threshold, the sequence code is considered activated, and its delay is estimated. This invention achieves high accuracy in frequency offset estimation and delay estimation, and the designed registration signal has minimal impact on the performance of the service signal; it also allows the registration signal power to be much smaller than the service signal power.

[0079] Example 1:

[0080] Figure 1 This is a schematic diagram of a method for online registration of ONU in a P2MP system according to an embodiment of the present invention.

[0081] like Figure 1 As shown, this embodiment provides a method for ONU online registration in a P2MP system, including the following steps:

[0082] Step S1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, thus obtaining an ONU with completed frequency offset correction.

[0083] Step S2, Sequence Code Set Construction: Construct the registered sequence code set using the sequence.

[0084] Specifically, the sequence is an m-sequence or a Gold sequence.

[0085] Step S3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence.

[0086] Step S4, Modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC (digital to analogue counter) sampling rate, then applies DC, and then performs a square root operation to generate the registration signal.

[0087] Figure 2 This is the registration signal generation and detection scheme in the embodiments of the present invention.

[0088] like Figure 2 As shown, step S4 includes:

[0089] The registration signal from the transmitting end is a registration sequence randomly selected from the sequence code set, denoted as x. u (n), n=0,...,N-1, then upsampled to the DAC sampling rate to obtain the signal x′u(n), simulated intensity modulation by adding DC and square root operations, and then shifted to the guard interval of the frequency band.

[0090] Figure 3 This is a schematic diagram showing the spectral position relationship between the registration signal and the service signal in an embodiment of the present invention.

[0091] Registered signal location such as Figure 3 As shown.

[0092] Send signal x T (n) is represented by equation (1), u dc For the added DC component, f c N is the carrier frequency of the registered signal.T To determine the signal length, intensity modulation is used to generate the registration signal, which can avoid the impact of frequency offset on the registration signal detection performance.

[0093]

[0094] Step S5, Frequency shifting and transmission of the registration signal: Shift the registration signal to the frequency band guard interval and transmit it to the OLT along with other service signals.

[0095] Step S6, extraction of registration signal: For the received signal, the OLT first performs IQ (In-phase and Quadrature) front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering.

[0096] Specifically, step S6 includes:

[0097] like Figure 2 As shown, at the receiving end, the OLT receives signals across the entire frequency domain via a coherent receiver, including the received X and Y polarized signals. This is expressed as equation (2), where Δt represents the transmission delay and Δf represents the residual frequency offset. This represents the transmitted X-polarized and Y-polarized service signals, and n0(n) represents the system noise.

[0098]

[0099] The registered signal at the guard interval is selected through frequency shifting and filtering operations. Represented by equation (3), n′0(n) represents the influence of service signals and system noise on the registration signal, u dc For the added DC component, N T For the length of the transmitted signal,

[0100]

[0101] Step S7, Service Signal Processing: For the remaining service signals after extracting the registration signal, perform signal recovery and quality assessment, and finally calculate the bit error rate (BER) and signal-to-noise ratio (SNR).

[0102] Specifically, signal recovery and quality assessment include: dispersion compensation, multiple-input multiple-output equalization based on multi-modulus algorithm (MMA-MIMO), fourth-order frequency offset estimation and compensation, symbol synchronization, and carrier phase recovery based on phase-locked loop.

[0103] Step S8, processing of the registration signal: preprocessing the registration signal, combining the signal, improving the signal-to-noise ratio and detecting key information to obtain the registration sequence and the corresponding time delay information.

[0104] Specifically, the registration signal is squared and the DC component is removed. The registration signal is then subjected to analog intensity detection. The signals on the two polarizations are merged. The signal-to-noise ratio of the received registration signal is improved by averaging multiple signal segments. Finally, the signal is cyclically cross-correlated with the local code set to detect the registration sequence and the corresponding time delay information.

[0105] More specifically, step S8 includes the following sub-steps:

[0106] Step S8.1, estimate the residual frequency offset Δf: Square the signal to remove the DC component, complete the analog strength detection of the registered signal, and obtain the signal. This can be expressed as equation (4), ignoring signal amplitude impairment, where n″0(n) represents the cumulative noise impact, including the impact of the service signal on the registration signal.

[0107]

[0108] Step S8.2, for the signal Peak detection is performed to obtain latency information.

[0109] Specifically, there are two methods for peak detection: normalized absolute peak detection and peak-to-average ratio detection.

[0110] In normalized absolute peak detection, let the cyclic cross-correlation result of the registered sequences a(n) and b(n) be (5).

[0111] R(n)=|{IFFT{FFT{a(n)}FFT{b(n)}}}|(5)

[0112] The normalized absolute peak value is shown in equation (6), where ||a|| represents the modulus of the registered sequence a(n).

[0113]

[0114] The position corresponding to the peak is d = argmax nM(n) represents the time delay information.

[0115] Peak-to-average ratio (PAPR) detection, as shown in equation (7), is the ratio of the maximum value to the average value of the relevant results.

[0116]

[0117] The position corresponding to the peak is d = argmax n M(n) represents the time delay information.

[0118] Step S8.3: Combine the signals on the two polarizations to obtain signal x. r′ (n), expressed as equation (8).

[0119]

[0120] Step S8.4 involves multi-segment superposition and averaging of the registration signal in the time domain to improve the signal-to-noise ratio of the received registration signal and enhance detection performance. Finally, cyclic cross-correlation detection is performed with the local code set, as shown in equation (9), x i (n) represents the i-th registration sequence code in the code set, thereby detecting the registration sequence selected by the registration signal, and calculating the time delay based on the position corresponding to the maximum correlation peak, thus completing the processing and analysis of the registration signal.

[0121] x e (n) = IFFT{FFT(x) r′ (n))FFT(x i (n))} (9)

[0122] Example 2:

[0123] This embodiment is a preferred example of Embodiment 1.

[0124] To verify the effectiveness and feasibility of the solution, an experiment was conducted on the ONU online registration signal of the P2MP system. The schematic diagram of the experimental solution is shown below. Figure 4 As shown, six 10GBaud DP-16QAM signals are generated in the electrical domain to simulate the service signals of different ONUs. The registration signal uses a 127 m / Gold sequence at a rate of 200MHz, modulated to the right third of the subcarrier guard interval. The relative positions of the registration signal and the service signal are shown in the figure. Figure 3As shown, the registration signal and service signal are modulated together by a coherent transmitter (CDM), with a 1550nm external cavity tunable semiconductor laser (ECL) as the modulation source. The modulated signal is amplified by an erbium-doped fiber amplifier (EDFA) and then transmitted through a 20km standard single-mode fiber (SSMF). At the receiving end, a variable optical attenuator (VOA) adjusts the received optical power, and the 1550nm ECL is used as the local oscillator (LO). An integrated coherent receiver performs coherent detection on the signal, and after sampling by an analog-to-digital converter (ADC), the registration signal is detected and the service signal is recovered through digital signal processing (DSP) at the receiving end.

[0125] At the transmitting end, the power of all six service signals is set to be the same, 2.02dBm. Therefore, the output of the EDFA is set to 9.8dBm. The power spectrum diagrams of the registration signal and service signals transmitted on X and Y polarization are shown below. Figure 5 As shown, the registration signal is located at 1 / 3 of the guard interval to the left of the service signal. For example, for the nth ONU, let its center frequency be f. n Then the center frequency f of the corresponding registered signal r：n =f n -1 / 2(R Data +α)-1 / 3F Guard , where R Data Where F is the service signal baud rate, i.e., 10 GBuad, α is the root-raised cosine roll-off factor, i.e., 0.0625, and F... Guard The protection interval is 600MHz. Figure 3 In the test registration signal at position 3, the power is 15dB lower than that of the service signal.

[0126] The receiving-end DSP process for registration signal detection and service signal recovery is as follows: Figure 6 As shown, for the received signal, IQ front-end correction is first performed to compensate for IQ skew and IQ imbalance. Then, the registration signal is filtered out separately. For the service signal, after resampling to 2 samples per symbol (2sps), dispersion compensation, MMA-MIMO equalization, fourth-order frequency offset estimation and compensation, symbol synchronization, carrier phase recovery based on phase-locked loop, and finally, BER and SNR are calculated. For the registration signal, after resampling to 2sps, analog square detection is performed to remove the DC component, the two polarization signals are merged, the signal is averaged 5 times, and cyclically cross-correlated with the local sequence code set. The peak value is detected, and if it exceeds the decision threshold, it indicates that the sequence code is activated, and the time delay is estimated.

[0127] During the experiment, we randomly selected a registration sequence code for the registration signal each time and conducted the experiment in the following manner, testing each case 200 times, and finally performing statistical analysis:

[0128] (1) Send a registration signal at position 3, change its relative power value, and make the power of the registration signal different dB values ​​smaller than the power of the service signal. Observe the changes in its false detection rate, false alarm rate and missed detection rate, as well as the time delay estimation.

[0129] (2) Add different frequency offsets to the registration signal at position 3 and observe the changes in its detection performance and its impact on the service signal;

[0130] When testing the impact of different signal powers on detection performance, a registration signal was transmitted at position 3, with its power set to be 15dB, 20dB, 25dB, and 30dB lower than the service signal. The false detection rate, missed detection rate, false alarm rate, and delay estimation were compared. The power spectrum of the received signal when the registration signal transmitted at position 3 was 15dB, 20dB, 25dB, and 30dB lower than the service signal power is as follows: Figure 7 As shown.

[0131] To verify the detection performance of the Gold sequence, the registration signal used was the Gold sequence, and the receiving end employed normalized absolute peak detection. The power of the registration signal at position 3 was set to be 15dB, 20dB, 25dB, and 30dB lower than the service signal power, respectively. 200 experiments were conducted. Based on the distribution of peak values ​​related to the correct sequence code and other sequence codes in the registration signal at position 3, a Gaussian distribution was used for distribution fitting and parameter estimation to obtain the optimal detection threshold bestTh, false positive rate Pe, false negative rate Pmiss, and false alarm rate Pfalse. The results are as follows: Figure 8 As shown. When peak-to-average power ratio (PAPR) detection is used at the Gold sequence receiver, the registered signal power is 15dB, 20dB, 25dB, and 30dB lower than the service signal power, respectively. After 200 experiments, based on the distribution of peak values ​​related to the correct sequence code and other sequence codes in the registered signal at position 3, a Gaussian distribution was selected for distribution fitting and parameter estimation to obtain the optimal detection threshold bestTh, false positive rate Pe, false negative rate Pmiss, and false alarm rate Pfalse. The results are as follows. Figure 9 As shown in the figure, it can be seen that the performance of peak-to-average power ratio (PAPR) detection for the Gold sequence is much worse than that of normalized absolute peak power (NEP) detection. When the registered signal is -15 dB smaller than the traffic signal, the false detection rate Pe of PAPR detection is 0.0138, and the corresponding optimal detection threshold bestTh is 3.9718. For delay estimation, when the registered signal at position 3 is -15 dB, -20 dB, -25 dB, and -30 dB smaller than the traffic signal power, respectively, the delay estimation errors using normalized absolute peak power (NEP) detection and PAPR detection are as follows: Figure 10 and Figure 11As shown in the figure, the crosses indicate that the registration signal was not detected or an incorrect sequence was detected. This is because the correlation peak between the registration signal and the correct sequence is less than the detection threshold, or the correlation peak with other sequences exceeds the threshold. Figure 10 and Figure 11 It can be seen that when the correct registration sequence code is detected, the estimation error is less than or equal to ±1 sample, that is, less than ±2.5 ns.

[0132] To verify the impact of different frequency offsets on detection performance, a dual-fiber bidirectional architecture was adopted. The ONU's transmitted light and local oscillator light both originate from the same laser. Therefore, frequency synchronization was first performed in the downlink direction. This was achieved by sweeping the downlink broadcast signal to adjust the ONU's receiving laser to the target frequency position. Then, a registration signal was sent for ranging, completing the time delay estimation. Therefore, for frequency offset estimation, two different subcarriers were selected for 4th power FFT frequency offset estimation, and the relative error was calculated. The X and Y polarization frequency offset estimation errors from 100 experiments are as follows: Figure 12 As shown, the error is less than ±0.5MHz. When the registration signal has different frequency offsets, it will overlap with the service signal in the spectrum, resulting in a decrease in the signal-to-noise ratio of the registration signal and a deterioration in detection performance. The registration signal uses a Gold sequence, and the detection scheme uses normalized absolute peak detection. The power of the registration signal at position 3 is set to be -15dB lower than that of the service signal. Different frequency offsets are set for the registration signal in the electrical domain: -500MHz, -250MHz, 250MHz, and 500MHz, to evaluate the detection performance. The results of the false detection rate Pe, missed detection rate Pmiss, and false alarm rate Pfalse under different frequency offsets are shown below. Figure 13 As shown, because the registered signal is closer to the right-hand service signal at the 600MHz guard interval, the false detection rate and other indicators will significantly deteriorate when there are 250MHz and 500MHz frequency offsets, decreasing from around 1e-74 to around 1e-44 and 1e-31 respectively, but still meeting the false detection rate requirement of 1e-15. The corresponding time delay estimation error results under different frequency offsets are shown below. Figure 14 As shown, the error range is still less than or equal to ±2.5ns.

[0133] During the online registration process, we do not want the registration signal to significantly affect the communication quality of the service signal. Therefore, we evaluate the impact of the registration signal on the service signal. During the experiment, we changed the frequency offset of the registration signal and compared the changes in the bit error rate of adjacent service signals with the case where no registration signal is sent.

[0134] In the experiment, six service signals of equal power were generated in the electrical domain. Each service signal was a DP-16QAM signal with 65536 symbols. The EDFA output was set to 9.8 dBm. The received optical power ROP was varied from -16 dBm to -10 dBm, and the polarization BER of ONUs at positions 2 and 3 was recorded. The registration signal was a Gold sequence, with its power set to be 15 dB lower than that of the service signals. Then, the ROP was adjusted, and the X and Y polarization BER curves of ONUs at positions 2 (ONU-2) and 3 (ONU-3) were compared with ROP at frequency offsets of -500 MHz, -250 MHz, 0 MHz, 250 MHz, and 500 MHz, respectively. Figure 15 and Figure 16 As shown in the figure, the impact of different frequency offset registration signals on the service signal is within 0.2dB.

[0135] Example 3:

[0136] This embodiment provides a system for ONU online registration in a P2MP system. The system for ONU online registration in a P2MP system can be implemented by executing the process steps of the method for ONU online registration in a P2MP system. That is, those skilled in the art can understand the method for ONU online registration in a P2MP system as a preferred embodiment of the system for ONU online registration in a P2MP system.

[0137] Specifically, the system includes:

[0138] Module M1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, resulting in an ONU with completed frequency offset correction;

[0139] Module M2, Sequence Code Set Construction: Constructs the registered sequence code set using sequences;

[0140] Module M3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence;

[0141] Module M4, modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC sampling rate, then applies DC, and then performs a square root operation to generate the registration signal;

[0142] Module M5, Frequency Shifting and Transmission of Registration Signal: Shifts the registration signal to the frequency band guard interval and transmits it to the OLT along with other service signals;

[0143] Module M6, extraction of registration signal: For the received signal, the OLT first performs IQ front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering.

[0144] Module M7, Service Signal Processing: For the remaining service signals after extracting the registration signal, perform signal recovery and quality assessment, and finally calculate the bit error rate and signal-to-noise ratio;

[0145] Module M8, processing of the registration signal: squaring the registration signal, removing the DC component, performing analog intensity detection on the registration signal, merging the signals on the two polarizations, improving the signal-to-noise ratio of the received registration signal by averaging multiple signal segments, and finally performing cyclic cross-correlation with the local code set to detect the registration sequence and corresponding time delay information.

[0146] Those skilled in the art will appreciate that, in addition to implementing the system and its various devices, modules, and units provided by the present invention in purely computer-readable program code, it is entirely possible to implement the same functions of the system and its various devices, modules, and units provided by the present invention in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system and its various devices, modules, and units provided by the present invention can be considered a hardware component, and the devices, modules, and units included therein for implementing various functions can also be considered as structures within the hardware component; the devices, modules, and units for implementing various functions can also be considered as both software modules implementing the method and structures within the hardware component.

[0147] The above describes specific embodiments of the present invention. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art may make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. The embodiments of this application and the features in the embodiments may be combined with each other in any manner unless there is a conflict.

Claims

1. A method for ONU online registration in a P2MP system, characterized in that, Includes the following steps: Step S1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, thus obtaining an ONU with completed frequency offset correction; Step S2, Sequence Code Set Construction: Construct the registered sequence code set using the sequence; Step S3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence; Step S4, Modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC sampling rate, then applies DC, and then performs a square root operation to generate the registration signal; Step S5, Frequency shifting and transmission of the registration signal: The registration signal is shifted to the frequency band guard interval and transmitted to the OLT terminal along with other service signals; Step S6, extraction of registration signal: For the received signal, the OLT first performs IQ front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering. Step S7, Service signal processing: For the remaining service signals after extracting the registration signal, perform signal recovery and quality assessment, and finally calculate the bit error rate and signal-to-noise ratio; Step S8, processing of the registration signal: the registration signal is squared and the DC component is removed, the registration signal is simulated for intensity detection, and the signals on the two polarizations are merged. The signal-to-noise ratio of the received registration signal is improved by averaging multiple signal segments. Finally, the registration sequence and the corresponding time delay information are obtained by cyclic cross-correlation with the local code set.

2. The method for ONU online registration in a P2MP system according to claim 1, characterized in that, In step S2, the sequence is an m-sequence or a Gold sequence.

3. The method for ONU online registration in a P2MP system according to claim 1, characterized in that, Step S4 includes: The registration sequence randomly selected from the sequence code set by the transmitting end is denoted as . , Then, the signal is upsampled to the DAC sampling rate to obtain the signal. By adding DC and square root operations to simulate intensity modulation, and then shifting the frequency to the guard interval of the frequency band, the signal is transmitted. Represented as Equation (1), For the added DC component, For the carrier frequency of the registered signal, For the length of the transmitted signal, (1)。 4. The method for ONU online registration in a P2MP system according to claim 3, characterized in that, Step S6 includes: At the receiving end, the OLT receives signals across the entire frequency domain via a coherent receiver, including the received X and Y polarized signals. It is expressed as equation (2), where Indicates transmission delay. Indicates residual frequency offset. This indicates the transmitted X-polarized and Y-polarized service signals. Indicates system noise. (2) The registered signal at the guard interval is selected through frequency shifting and filtering operations. , represented as equation (3), This indicates the impact of service signals and system noise on the registration signal. For the added DC component, For the length of the transmitted signal, (3)。 5. The method for ONU online registration in a P2MP system according to claim 1, characterized in that, In step S7, the signal recovery and quality assessment includes: dispersion compensation, multi-input multi-output equalization based on multi-mode algorithm, fourth-power frequency offset estimation and compensation, symbol synchronization, and carrier phase recovery based on phase-locked loop.

6. The method for ONU online registration in a P2MP system according to claim 4, characterized in that, Step S8 includes the following sub-steps: Step S8.1, for the residual frequency offset Estimation: The signal is squared to remove the DC component, and the analog strength detection of the registered signal is completed to obtain the signal. This is expressed as equation (4), ignoring signal amplitude impairment. This indicates the cumulative noise impact, including the effect of traffic signals on registration signals. (4) Step S8.2, for the signal Peak detection is performed to obtain latency information; Step S8.3: Combine the signals on the two polarizations to obtain the signal. , expressed as equation (8) (8) Step S8.4: In the time domain, the registered signal is averaged by multiple superpositions, and finally, it is cyclically cross-correlated with the local code set by cyclic cross-correlation detection, as shown in equation (9). For the code set The system uses a registration sequence code to detect the selected registration sequence of the registration signal, and calculates the time delay based on the position corresponding to the maximum correlation peak, thus completing the processing and analysis of the registration signal. (9)。 7. The method for ONU online registration in a P2MP system according to claim 6, characterized in that, In step S8.2, the peak detection has two detection methods: normalized absolute peak detection and peak-to-average ratio detection.

8. The method for ONU online registration in a P2MP system according to claim 7, characterized in that, In the normalized absolute peak detection, the registered sequence is set. and The cyclic cross-correlation result is (5), where FFT refers to Fast Fourier Transform and IFFT refers to Inverse Fast Fourier Transform. The normalized absolute peak value is shown in equation (6). Indicates the registration sequence The model, (6) Position corresponding to the peak This refers to time delay information.

9. The method for ONU online registration in a P2MP system according to claim 8, characterized in that, The peak-to-average power ratio (PAPR) detection, as shown in equation (7), is the ratio of the maximum value to the average value of the relevant results. (7) Position corresponding to the peak This refers to time delay information.

10. A system for ONU online registration in a P2MP system, characterized in that, include: Module M1, ONU initialization: The ONU to be connected completes the basic configuration and completes the frequency offset correction of the local laser through downlink synchronization, resulting in an ONU with completed frequency offset correction; Module M2, Sequence Code Set Construction: Constructs the registered sequence code set using sequences; Module M3, ONU selects registration sequence: The ONU that has completed frequency offset correction randomly selects a sequence from the sequence code set as the registration sequence; Module M4, Modulation of the registration signal: The ONU that has completed frequency offset correction upsamples the registration sequence to the DAC sampling rate, then applies DC, and then performs a square root operation to generate the registration signal; Module M5, Frequency Shifting and Transmission of Registration Signal: Shifts the registration signal to the frequency band guard interval and transmits it to the OLT along with other service signals; Module M6, extraction of registration signal: For the received signal, the OLT first performs IQ front-end correction to compensate for IQ skew and IQ imbalance, and then filters out the registration signal by frequency shifting and filtering. Module M7, Service Signal Processing: For the remaining service signals after extracting the registered signals, perform signal recovery and quality assessment, and finally calculate the bit error rate and signal-to-noise ratio; Module M8, processing of the registration signal: squaring the registration signal, removing the DC component, performing analog intensity detection on the registration signal, merging the signals on the two polarizations, improving the signal-to-noise ratio of the received registration signal by averaging multiple signal segments, and finally performing cyclic cross-correlation with the local code set to detect the registration sequence and the corresponding time delay information.

Images