WO2021218733A1

WO2021218733A1 - Signal processing method and device

Info

Publication number: WO2021218733A1
Application number: PCT/CN2021/088638
Authority: WO
Inventors: 赵曼; 邓邦梅; 柴胜钧
Original assignee: 华为技术有限公司
Priority date: 2020-04-28
Filing date: 2021-04-21
Publication date: 2021-11-04
Also published as: CN113572534B; CN113572534A

Abstract

Disclosed in the present application are a signal processing method and device. The method comprises: performing coded modulation processing on first data to obtain a first signal, wherein the coded modulation processing comprises probabilistic constellation shaping (PCS) processing, and a first probability density function (PDF) is obtained after the first data is subject to the PCS processing; determining a first quantization parameter set according to the first PDF, the first quantization parameter set comprising at least one quantization parameter, the quantization parameter being a difference between magnitude values of two adjacent level signals, and each level signal corresponding to one voltage value within a voltage measurement range of a DAC; and performing, according to the first quantization parameter set, fixed-point quantization processing on the first signal to obtain a second signal. On the basis of probability distribution characteristics of signal magnitudes subject to PCS, the method adds an ENOB-augmented coding function at a sending end, and converts original uniform quantization parameters into a non-uniform first quantization parameter set, thereby improving the accuracy of ENOB.

Description

Signal processing method and device

This application claims the priority of a Chinese patent application filed with the State Intellectual Property Office of China, the application number is 202010347348.3, and the invention title is "a signal processing method and device" on April 28, 2020. The entire content is incorporated herein by reference. Applying.

Technical field

This application relates to the field of optical communications, and in particular to a signal processing method and device.

Background technique

With the evolution of the fifth generation of mobile communication technology (5th generation, 5G) and Internet technology (Internet Technology, IT) to cloud platforms, as well as high-definition virtual reality (VR) or augmented reality (AR) ) The explosive growth of video has promoted the construction of optical fiber transmission across the entire network, such as data center Inter-connect (DCI) and optical access, metropolitan area, backbone network, long-distance land cable and ocean cable, etc. , And had a positive impact on the optical network market.

At present, 100G (Kyrgyzstan) is the mainstream technology for optical transmission networks. my country has built the world's largest 100G wavelength division multiplexing (WDM)/optical transport network (OTN) commercial transmission network. Driven by high-definition video quality requirements such as 5G and cloud platformization, the optical transmission capacity is evolving rapidly from 100G to the ultra-high-speed and large-capacity optical transmission network of 200G, 400G, 800G/1T+. With the rapid increase of optical transmission capacity, the optical transmission network, which is the foundation of the network, is facing greater pressure. On the one hand, it is necessary to use a higher baud rate to change the channel spacing from 50 gigahertz (GHz) to 75GHz, 100GHz, and 150GHz. , 200GHz and so on. On the other hand, it is also necessary to continuously improve the spectral efficiency (SE), and continuously evolve the modulation mode from quadrature amplitude modulation (QAM) to 16QAM, 64QAM, and 256QAM.

In order to increase the baud rate, it is necessary to increase the sampling frequency of the analog-to-digital converter (DAC) and the digital-to-analog converter (ADC), and upgrade the modulation mode to a higher-order modulation mode. The higher the accuracy requirements, the correspondingly higher requirements for the bit width of the DAC or ADC, which brings huge challenges to the design of the DAC/ADC. Therefore, based on the existing solution, in the case of meeting the needs of transmission equipment, Reducing the sampling frequency and bit width requirements of the DAC/ADC has become the key.

Summary of the invention

In order to reduce the requirements for the sampling frequency and bit width of the DAC/ADC while meeting the requirements of the transmission equipment, this application provides the following technical solutions:

In the first aspect, the present application provides a signal processing method, which can be applied to a network device, such as a transmitter. The transmitter includes an ODSP transmitting TX module and a digital-to-analog converter DAC. The method includes: A data is encoded and modulated to obtain a first signal, where the encoding and modulation processing includes: probability constellation shaping PCS processing, and the first data is processed by PCS to obtain a first probability density function PDF; determining the first signal according to the first PDF A set of quantization parameters, and a second signal is obtained by performing fixed-point quantization processing on the first signal according to the first set of quantization parameters.

Wherein, the first quantization parameter set includes at least one quantization parameter, each quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to a voltage value in the voltage range of the DAC ; The voltage range of this DAC is 0～1Vp2p, Vp2p is the voltage peak-to-peak value.

The method provided in this aspect, based on the probability distribution characteristics of the signal amplitude after PCS, adds the ENOB enhanced coding function at the origin of the ODSP, transforms the original uniform quantization parameter into an uneven first quantization parameter set, and uses a limited number of fixed points. The quantized sampling points are concentrated on the amplitude value of the high probability level signal, thereby improving the accuracy of ENOB.

With reference to the first aspect, in a possible implementation of the first aspect, determining the first quantization parameter set according to the first PDF includes: obtaining a first probability distribution function corresponding to the first PDF, and calculating the first probability distribution function according to The fixed-point quantization number N is distributed at equal intervals to obtain N values, where N is a positive integer and N≥1; and determine the amplitude value of the level signal corresponding to each of the N values, and calculate each phase The difference between the amplitude values of two adjacent level signals; finally, the first quantization parameter set is determined according to the difference between the amplitude values of every two adjacent level signals.

With reference to the first aspect, in another possible implementation of the first aspect, the above method further includes: sending a second signal to the DAC.

With reference to the first aspect, in another possible implementation of the first aspect, the above method further includes: sending the first quantization parameter set to the DAC.

In this implementation, the ODSP transmitter sends the first set of quantization parameters to the DAC, so that the DAC also uses the first set of quantization parameters to perform digital-to-analog conversion on the fixed-point quantized data. The enhanced encoding function of ENOB is greater than that of the DAC due to the reduction of oversampling multiples. Therefore, after the two phases are compensated, the performance of the signal output by the ODSP transmitter will not deteriorate, but has been improved, so that the low sampling frequency DAC can be used for the high baud rate ODSP.

With reference to the first aspect, in yet another possible implementation of the first aspect, the performing fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain the second signal includes: converting the first quantization parameter set to the second signal. Two quantization parameter sets, the number of types of quantization parameters included in the second quantization parameter set is B, and the first signal is subjected to fixed-point quantization processing using B types of quantization parameters in the second quantization parameter set to obtain the second signal. The number of types of quantization parameters included in the first quantization parameter set is A, and A>B, and both A and B are positive integers.

With reference to the first aspect, in yet another possible implementation of the first aspect, before converting the first quantization parameter set to the second quantization parameter set, the method further includes: acquiring all B-type quantization parameters that meet the first preset condition Set, the first preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and the number of quantization parameters included in each quantization parameter set N is equal to 2. ^M , M is the bit width of the data; calculate the error between each quantization parameter set of type B that meets the first preset condition and the first quantization parameter set, the error is the sum of squares of the difference, and then determine The quantization parameter set with the smallest error is the second quantization parameter set.

In this implementation manner, the types of quantization parameters of the first quantization parameter set are simplified, so that the types of quantization parameters included in the simplified second quantization parameter are reduced, which is convenient for implementation, and also improves the DAC processing efficiency.

In the second aspect, the present application provides a signal processing method, which can be applied to a network device, such as a receiver, the receiver includes: an ODSP receiving RX module and an analog-to-digital converter ADC, the method includes: ADC After receiving the fourth signal sent by the transmitter, the fourth signal is analog-to-digital converted into a third signal, and then transmitted to the ODSP receiving RX module. The ODSP receiving RX module receives the third signal output by the ADC to obtain the third signal of the third signal. The second probability density function PDF, and the third quantization parameter set is determined according to the second PDF.

Wherein, the third quantization parameter set includes at least one quantization parameter, each quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to a voltage value in the voltage range of the ADC. The voltage range of ADC is 0～1Vp2p, Vp2p is the voltage peak-to-peak value.

In this implementation manner, an enhanced decoding function of ENOB is added at the ODSP receiving end, and an optimal quantization parameter set is provided to the ADC, and the ADC performs analog-to-digital conversion on the signal according to the optimal quantization parameter set, which improves the ENOB of the ADC. Then the ADC output data is converted into uniform quantization, and the subsequent demodulation and decoding processing is performed.

With reference to the second aspect, in a possible implementation of the second aspect, determining the third quantization parameter set according to the second PDF includes: obtaining a second probability distribution function corresponding to the second PDF; and converting the second probability distribution function according to The fixed-point quantization number N is equally spaced to obtain N values, where N is a positive integer and N≥1; determine the amplitude value of the level signal corresponding to each of the N values, and calculate each adjacent value The difference between the amplitude values of the two level signals; the third quantization parameter set is determined according to the difference between the amplitude values of each adjacent two level signals.

With reference to the second aspect, in another possible implementation of the second aspect, the above method further includes: sending a third set of quantization parameters to the ADC.

With reference to the second aspect, in another possible implementation of the second aspect, the above method further includes: the ADC receives the fifth signal sent from the transmitter, and performs analog-to-digital conversion processing on the fifth signal to output the sixth signal, and The sixth signal is sent to the ODSP receiving RX module, the ODSP receiving RX module receives the sixth signal output by the ADC, and performs fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain the seventh signal; decodes and modulates the seventh signal Get the first data. The above-mentioned decoding and modulation processing includes performing PCS inverse processing on the seventh signal.

With reference to the second aspect, in another possible implementation of the second aspect, performing fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain the seventh signal includes: converting the third quantization parameter set to the fourth quantization Parameter set, where the number of types of quantization parameters included in the third quantization parameter set is C, and the number of types of quantization parameters included in the fourth quantization parameter set is D, and C>D, C and D are both positive integers ; Using D types of quantization parameters in the fourth quantization parameter set to perform fixed-point quantization processing on the sixth signal to obtain the seventh signal.

With reference to the second aspect, in another possible implementation of the second aspect, before converting the third quantization parameter set to the fourth quantization parameter set, the method further includes: acquiring all D types of quantization that meet the second preset condition Parameter set, the second preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the ADC, and the number of quantization parameters included in each quantization parameter set N is equal to 2 ^M , M Is the bit width of the data; calculate the error between each D-type quantization parameter set that meets the second preset condition and the third quantization parameter set, and determine the quantization parameter set with the smallest error as the fourth quantization parameter gather.

In this implementation manner, the types of quantization parameters in the third quantization parameter set are simplified, so that the types of quantization parameters included in the simplified fourth quantization parameter are reduced, which is convenient for implementation and also improves the ADC processing efficiency.

In a third aspect, the present application provides a signal processing device, which includes at least one module, which may include a receiving module, a processing module, a sending module, and so on.

When the signal processing device is used as a signal transmitting device, the processing module is used to perform encoding and modulation processing on the first data to obtain the first signal, wherein the encoding and modulation processing includes PCS processing, and the first data is obtained after PCS processing The first PDF; the processing module is also used to determine the first quantization parameter set according to the first PDF, and perform fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain the second signal.

In combination with the third aspect, in a possible implementation of the third aspect, the processing module is specifically used to obtain the first probability distribution function corresponding to the first PDF; and the first probability distribution function is calculated according to the fixed-point quantization number N, etc. N values are obtained by interval allocation, where N is a positive integer and N≥1; the amplitude value of the level signal corresponding to each of the N values is determined, and every two adjacent electrical The difference between the amplitude values of the flat signal; the first quantization parameter set is determined according to the difference between the amplitude values of each adjacent two level signals.

With reference to the third aspect, in another possible implementation of the third aspect, the sending module is configured to send the second signal to the DAC.

With reference to the third aspect, in another possible implementation of the third aspect, the sending module is further configured to send the first set of quantization parameters to the DAC.

With reference to the third aspect, in another possible implementation of the third aspect, the processing module is specifically configured to convert the first quantization parameter set to the second quantization parameter set, wherein the value of the quantization parameter contained in the first quantization parameter set is The number of types is A, the number of types of quantization parameters included in the second quantization parameter set is B, and A>B, A and B are both positive integers; and the B quantization parameter pair in the second quantization parameter set is used to The first signal is subjected to fixed-point quantization processing to obtain a second signal.

With reference to the third aspect, in another possible implementation of the third aspect, the processing module is further configured to obtain all B types of data that meet the first preset condition before converting the first quantization parameter set to the second quantization parameter set. Quantization parameter sets, the first preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and the number of quantization parameters included in each quantization parameter set The number N is equal to 2 ^M , and M is the bit width of the data; calculate the error between each quantization parameter set of type B that meets the first preset condition and the first quantization parameter set, and determine that the error is the smallest One set of quantization parameters is the second set of quantization parameters.

Optionally, when the signal processing device is used as a signal receiving device, the receiving module is used to receive the third signal output by the ADC, and the processing module is used to obtain the second PDF of the third signal, and determine the third signal according to the second PDF. The quantization parameter set, wherein the third quantization parameter set includes at least one quantization parameter, the quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to the voltage range of the ADC A voltage value of.

With reference to the third aspect, in another possible implementation of the third aspect, the processing module is used to obtain a second probability distribution function corresponding to the second PDF, and the second probability distribution function is equally spaced according to the fixed-point quantization number N Allocate N values, N is a positive integer and N≥1; determine the amplitude value of the level signal corresponding to each of the N values, and calculate the magnitude of each adjacent two of the level signals The difference between the amplitude values determines the third quantization parameter set according to the difference between the amplitude values of each adjacent two level signals.

With reference to the third aspect, in another possible implementation of the third aspect, the sending module is configured to send the third set of quantization parameters to the ADC.

With reference to the third aspect, in another possible implementation of the third aspect, the receiving module is further configured to receive the sixth signal output by the ADC, and the processing module is further configured to perform fixed-point quantization processing on the sixth signal according to the third quantization parameter set Obtain the seventh signal; and perform decoding and modulation processing on the seventh signal to obtain the first data, where the decoding and modulation processing includes performing PCS inverse processing on the seventh signal.

With reference to the third aspect, in yet another possible implementation of the third aspect, the processing module is specifically configured to convert the third quantization parameter set into a fourth quantization parameter set, wherein the quantization included in the third quantization parameter set The number of types of parameters is C, the number of types of quantization parameters included in the fourth quantization parameter set is D, and C>D, C and D are both positive integers; and D types of quantization parameters in the fourth quantization parameter set are used Performing fixed-point quantization processing on the sixth signal to obtain the seventh signal.

With reference to the third aspect, in another possible implementation of the third aspect, the processing module is further configured to obtain all D types that meet the second preset condition before converting the third quantization parameter set to the fourth quantization parameter set. A quantization parameter set, calculating the error between each D type quantization parameter set that meets the second preset condition and the third quantization parameter set, and determining the quantization parameter set with the smallest error as the fourth quantization parameter gather.

Wherein, the second preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the ADC, and the number of quantization parameters included in each quantization parameter set N is equal to 2. ^M , M is the bit width of the data.

In a fifth aspect, this application provides a transmitter that includes an optical digital signal processor ODSP, a digital-to-analog converter DAC, and an electro-optical converter E/O; wherein, ODSP is each of the foregoing third aspect and the third aspect. The signal processing device described in this implementation manner is used to implement the foregoing first aspect and the signal processing methods described in various implementation manners of the first aspect.

The DAC is used to receive the digital signal sent by the ODSP and convert the digital signal into a first analog electrical signal.

The electrical-optical converter is used for receiving the first analog electrical signal output by the DAC, converting the first analog electrical signal into a first optical signal, and transmitting the first optical signal.

Optionally, the E/O can be integrated inside the transmitter or outside the transmitter, which is not limited in this application.

In a sixth aspect, the present application provides a receiver that includes an optical-to-electrical converter E/O, an optical digital signal processor ODSP, and an analog-to-digital converter ADC; wherein, the E/O is used to receive the first optical Signal, the first optical signal is converted into a second analog electrical signal, and the second analog electrical signal is transmitted to the ADC.

The ADC is used to receive the second analog electrical signal, convert the second analog electrical signal into a digital signal, and transmit the digital signal to the ODSP.

The ODSP is used to receive the digital signal sent by the ADC and execute the signal processing method described in the foregoing second aspect and various implementation manners of the second aspect. Among them, the ODSP is the signal processing device described in the foregoing third aspect and various implementation manners of the third aspect.

In a seventh aspect, the present application also provides a processing chip including a processor and a memory, and the processor is coupled with the memory. Specifically, the memory is used to store computer program instructions; the processor is used to execute the instructions stored in the memory to The processing chip is caused to execute the foregoing first aspect and various implementation manners of the first aspect, or the method in the second aspect and various implementation manners of the second aspect.

In an eighth aspect, the present application also provides a network device. The network device may be the transmitter described in the foregoing fifth aspect, or may also be the receiver described in the foregoing sixth aspect.

In a ninth aspect, the embodiments of the present application also provide a computer-readable storage medium that stores instructions in the storage medium, so that when the instructions run on a computer or a processor, they can be used to execute the aforementioned first aspect and the first aspect. The methods in various implementation manners of the aspect, or the foregoing second aspect and the methods in the various implementation manners of the second aspect may also be executed.

In addition, the embodiments of the present application also provide a computer program product. The computer program product includes a computer instruction. When the instruction is executed by a computer or a processor, it can implement various implementations of the first aspect or the second aspect. Methods.

The technical solution of the present application, by increasing the ENOB of the DAC/ADC processing constellation graph shaping PCS signal, offsets the performance cost introduced by reducing the ADC/DAC sampling frequency, thereby realizing the use of low sampling rate ADC/DAC to transmit high baud rate signals, greatly Reduce the difficulty of ADC/DAC design.

Description of the drawings

FIG. 1 is a schematic structural diagram of an optical communication system provided by an embodiment of this application;

2 is a schematic structural diagram of an optical communication system composed of two network devices provided by an embodiment of this application;

3A is a graph of the relationship between the oversampling multiple of a DAC and ENOB provided by an embodiment of the application;

FIG. 3B is a graph of the relationship between the oversampling multiple of the ADC and the ENOB provided by an embodiment of the application;

FIG. 4 is a schematic diagram of the OSNR index of ODSP that is affected by reducing the oversampling factor according to an embodiment of the application; FIG.

FIG. 5 is a structural block diagram of a transmitter provided by an embodiment of this application;

FIG. 6 is a structural block diagram of a receiver provided by an embodiment of the application;

FIG. 7 is a flowchart of a signal processing method provided by an embodiment of the application;

FIG. 8 is a schematic diagram of a probability distribution of a level signal after PCS processing provided by an embodiment of the application; FIG.

FIG. 9 is a schematic diagram of a probability distribution function provided by an embodiment of the application;

FIG. 10 is a schematic diagram of the impact of ENOB enhanced quantization on OSNR index performance provided by an embodiment of the application;

FIG. 11A is a schematic diagram of using a uniform quantization parameter set provided by an embodiment of the application; FIG.

FIG. 11B is a schematic diagram of using an optimal quantization parameter set according to an embodiment of the application;

FIG. 11C is a schematic diagram of a simplified quantization parameter set provided by an embodiment of the application;

FIG. 12 is a flowchart of another signal processing method provided by an embodiment of this application;

FIG. 13 is a schematic structural diagram of a signal processing device provided by an embodiment of the application;

FIG. 14 is a schematic structural diagram of a network device provided by an embodiment of this application.

Detailed ways

In order to enable those skilled in the art to better understand the technical solutions in the embodiments of this application, and to make the above-mentioned objectives, features, and advantages of the embodiments of the present application more obvious and understandable, the following describes the technology in the embodiments of the present application with reference to the accompanying drawings. The plan is explained in further detail.

Before describing the technical solutions of the embodiments of the present application, first, the application scenarios of the embodiments of the present application will be described with reference to the accompanying drawings.

The technical solution of the present application can be applied to an optical communication system, which includes at least two network devices, as shown in FIG. 1, including a first network device and a second network device, and each of the at least two network devices The equipment includes: transmitter, receiver, converter and at least one interface. Wherein, the converter includes: an electro-optical converter (E/O) and/or a photo-electric converter (O/E). At least one of the interfaces includes an optical fiber interface, and one network device can be optically connected to another network device by using the optical fiber interface. In addition, each network device may also include other components, such as a processor, a memory, etc., and this embodiment does not limit the form and specific structure of the network device.

In addition, optionally, the E/O or O/E may also be set in the transmitter or the receiver, which is not limited in this embodiment.

As shown in Figure 2, the first network device includes an X-polarized transmitter and a Y-polarized transmitter. Each transmitter includes: an optical digital signal processing (ODSP) transmitting TX module, referred to as " ODSP_TX" module and digital-to-analog converter (DAC). Correspondingly, the second network device includes an X polarization receiver and a Y polarization receiver, and each receiver includes an ODSP receiving RX module, referred to as "ODSP_RX" module and an analog to digital converter (ADC) . In addition, each of the aforementioned transmitters and receivers may also include other components, such as antennas, radio frequencies, and so on.

In the first network device, the X-polarized transmitter and the Y-polarized transmitter have the same operation flow except for the signal processing polarization direction. Therefore, any one of the transmitters can be used as an example for description. In this embodiment, assuming an X polarization transmitter as an example, the following processing operations are included:

At the transmitting end, the X polarization transmitter first encodes and modulates the input bit data, and then transmits it to the DAC for digital-to-analog conversion. After receiving by the DAC, the encoded and modulated data is converted into an analog electrical signal according to a uniform quantization method, and then the analog electrical signal is converted into a first optical signal through E/O. Similarly, the Y-polarized transmitter outputs the second optical signal after encoding, modulation, digital-to-analog conversion, and E/O processing of the received bit data. The first optical signal and the second optical signal are combined to form a third optical signal, and finally The third optical signal is transmitted to the second network device through the optical fiber.

At the receiving end, the second network device receives the third optical signal transmitted through the optical fiber, and separates the third optical signal into an optical signal in the X polarization state (i.e., the first optical signal) and an optical signal in the Y polarization state (i.e., the second Optical signal), and then send the two optical signals to the X polarization receiver and the Y polarization receiver respectively. Since each receiver has the same signal processing process, one of the signals can be used as an example for description. Assume that the first optical signal is input to the X polarization receiver as an example. When the O/E receives the X polarization optical signal Then, the first optical signal is converted into an electrical signal, and then transmitted to the X polarization receiver. After the X polarization receiver receives the electrical signal, it first uses ADC to sample and uniformly quantize the electrical signal, convert the electrical signal into a digital signal, then decode and demodulate the digital signal, and finally restore the first A bit of data sent by a network device.

In the signal processing process of the first network device and the second network device, both the DAC and the ADC generally use a lower sampling frequency, that is, reduce the oversampling multiple of the ADC/DAC sampling frequency relative to the signal baud rate, Thereby reducing the design difficulty of ADC/DAC and ensuring that the ADC/DAC in current network equipment can work normally. For example, taking the 180G baud rate of 200G channel interval, when the oversampling factor is reduced from 1.25 times (1.25x) to 1.125 times (1.125x), the sampling frequency of ADC/DAC can be changed from 225GSPS (Giga Sample Per Second) , Giga per second) is reduced to 202.5GSPS, which is reduced by 22.5GSPS. In this embodiment, "1.25x" represents a process sampling multiple of 1.25 times, and "1.125x" represents a process sampling multiple of 1.125 times.

Among them, oversampling refers to sampling the input signal with a frequency greater than the Nyquist sampling frequency. Assuming the signal baud rate is fb, if the sampling frequency of ADC or DAC is increased to R×fb, that is, the baud rate of R times, then "R" is the oversampling multiple, and R>1, thus the oversampling multiple is obtained The corresponding relationship with the frequency of adoption is:

Oversampling multiple = sampling frequency/baud rate

Among them, the baud rate refers to the modulation rate of the data signal to the carrier, which can be represented by the number of times the carrier modulation state changes per unit time, that is, the number of symbols modulated per second, and its unit is Baud (symbol/s). The baud rate is a measure of the bandwidth of the transmission channel.

When the oversampling factor is reduced, the ratio of the signal baud rate to the ADC/DAC sampling frequency will increase proportionally, causing the signal bandwidth to extend to the high frequency range of the ADC/DAC sampling frequency, and then clock jitter will occur in the ADC/DAC. Non-linear distortion, channel inconsistency, etc., cause the effective number of bits (ENOB) in the transmission signal bandwidth to decrease and the transmission performance to decrease. The ENOB curve of the DAC shown in FIG. 3A and the ENOB curve of the ADC shown in FIG. 3B show that after the oversampling factor is reduced from 1.25x to 1.125x, the ENOB in the signal bandwidth also decreases.

In addition, the decline in ENOB will also affect the transceiver performance of ODSP. As shown in Figure 4, under the same bit error rate (BER) threshold, the optical signal-to-noise ratio (OSNR) corresponding to 1.125x is greater than the OSNR of 1.25x. The lower the oversampling multiple under the BER threshold, the higher the OSNR and the worse the transmission performance.

The sampling frequency of ADC/DAC required for high baud rate is too high, which makes the design of ADC/DAC difficult. If the method of reducing the oversampling multiple is used to avoid the sampling frequency from being too high, the ENOB will be reduced and the transmission performance of the ODSP will be degraded. Therefore, the technical solution of the present application is to achieve a certain transmission performance under the same signal baud rate, while also reducing the sampling frequency of the ADC/DAC.

In order to reduce the sampling frequency of ADC/DAC losslessly, in the technical solution of this embodiment, an enhanced encoding function of ENOB is added to the ODSP_TX module of the transmitter, as shown in FIG. 5. Similarly, the enhanced decoding function of ENOB is added to the ODSP_RX module of the receiver, as shown in Figure 6, so that under the condition that the data bit width of ADC/DAC remains unchanged, the enhanced encoding and decoding functions of ENOB will The fixed-point quantization of data completely matches the distribution characteristics of the signal, which improves the ENOB during DAC and ADC processing, thereby offsetting the performance degradation caused by reducing the oversampling multiple, and achieving the beneficial effect of reducing the sampling frequency of the DAC/ADC without loss of performance.

The technical solution of this embodiment will be described in detail below.

This embodiment provides a signal processing method, which can be applied to the ODSP_TX module, and the ODSP_TX module is located in the first network device. As shown in FIG. 7, the method includes:

101: The ODSP_TX module encodes and modulates the first data and then outputs the first signal.

Wherein, the first data is bit data of the service to be sent. The coding and modulation processing includes a series of processing such as probability constellation shaping (PCS), feedforward error correction (FEC) and channel modulation.

Specifically, as shown in Figure 5, the ODSP_TX module processes the first data to obtain a first probability density function (probability density function, PDF), since the amplitude value of the level signal corresponding to the first PDF is close to Gaussian Distribution curve, so this embodiment takes Gaussian distribution curve as an example for quantization processing. As shown in Figure 8, it is a signal simulation diagram after PCS processing. The signal simulation diagram shows the correspondence between the amplitude value of the level signal and the probability density function PDF, where the ordinate represents PDF, and the abscissa represents electrical The amplitude value of the flat signal. Each point on the signal simulation curve represents the probability that the amplitude value of a level signal is in a certain interval. Therefore, the ODSP_TX module obtains the first PDF after PCS processing the first data. It can also be called the first probability distribution.

The amplitude value range of the level signal corresponds to the voltage full range of the DAC, and the voltage full range of the DAC can be expressed as voltage peak to peak (Vp2p). As shown in Figure 8, the PDF presents an unequal probability distribution, and the amplitude value of the level signal presents a trend of high in the middle and low on both sides. For example, the amplitude value of the level signal has a greater probability of appearing in the interval of 0.4 to 0.6Vp2p, and the corresponding PDF value is higher; the amplitude value of the level signal is in the interval of 0 to 0.2Vp2p, and the probability of appearing in the interval of 0.8 to 1Vp2p is smaller.

Optionally, the first PDF is represented as P1.

The processing process of FEC and channel modulation in step 101 includes: the ODSP_TX module performs channel coding and modulation on the first data, and then maps the FEC coded and modulated data into a quadrature amplitude modulation (quadrature amplitude modulation, QAM) signal , And perform pulse shaping processing to output the first signal. Wherein, the first signal is a signal output after quantizing the first data according to a uniform quantization parameter.

In addition, it should be noted that, in the encoding and modulation process, the first PDF is only output after the PCS processing, and the first PDF is not affected during the FEC encoding and channel modulation process.

102: Determine a first quantization parameter set according to the first PDF.

The first quantization parameter set includes at least one quantization parameter, and the quantization parameter is the difference between the amplitude values of two adjacent level signals. Optionally, the quantization parameter is also called a quantization interval (quantization interval).

Exemplarily, the first quantization parameter set is denoted as Q1. The amplitude value of the level signal is expressed as y, then the amplitude values corresponding to the N level signals are y1, y2, y3,..., yN, the first quantization parameter set Q1={△Y1,△Y2,△Y3, ..., △YN}, where △Y1=y2-y1, △Y2=y3-y2, △YN=yN-yN-1, N is a positive integer, and N≥2.

Specifically, step 102 includes:

102-1. Obtain a first probability distribution function corresponding to the first PDF.

Wherein, the first probability distribution function can be obtained by performing an integral operation on the first PDF, as shown in FIG. 9, which is a schematic diagram of a probability distribution function obtained after an integral operation. Among them, the ordinate represents the cumulative distribution function of the probability (Cumulative Distribution Function), and the abscissa represents the amplitude value of the level signal, and the range is Vp2p.

102-2. Assign the first probability distribution function at equal intervals according to the number N of fixed-point quantization to obtain N values, where N is a positive integer and N≥1.

Wherein, the number of fixed-point quantization is expressed as N, N is related to the bit width of the data, and the bit width of the data is determined by the internal processing capability of the ODSP_TX module. It is assumed that the bit width of the data is M, and M is a positive integer and M≥1, the number of fixed-point quantization is N=2 ^M. For example, when the bit width M is 4 bits, the number N of fixed-point quantization is 2 ⁴ , that is, 16 bits.

The voltage full-scale range of the DAC is Vp2p, which is determined internally by the DAC, and there is a corresponding relationship between the Vp2p of the DAC and the number N of fixed-point quantization. Specifically, N fixed-point quantization numbers correspond to N values, that is, correspond to N level signals, and each level signal corresponds to a voltage value in the voltage range of the DAC.

It should be noted that, in this embodiment, the voltage full-scale range of the DAC is taken as an example, and it may also be the current full-scale range, that is, Ip2p, where I represents current.

The equal interval refers to uniformly dividing the probability of the first probability distribution function. For example, as shown in FIG. 9, the cumulative distribution function of the probability is evenly distributed from 0 to 1 to obtain 16 values.

102-3. Determine the amplitude value of the level signal corresponding to each of the N values, and calculate the difference between the amplitude values of every two adjacent level signals.

Among them, the corresponding relationship between each value and the amplitude value of the level signal can be represented by the probability distribution function shown in Figure 9:

Wherein, F(x) is the probability distribution function, f(t) is the probability density function (PDF), x is the amplitude value of the level signal, and the unit is Vp2p.

According to the uniform distribution rule, the voltage range corresponding to the first probability distribution function is allocated N parts. For example, after the first probability distribution function is distributed at equal intervals, 16 (N=16) parts are obtained, and 16 values are obtained, and then The amplitude value of the level signal corresponding to each value is found through the relationship (1), and the amplitude value of the level signal is the abscissa corresponding to a value on the curve of the relationship (1). For example, the amplitude value of the level signal corresponding to the value 1/16 is y1, the amplitude value of the level signal corresponding to the value 2/16 is y2, and the amplitude value of the level signal corresponding to the value 16/16 is yN . The difference between the amplitude values of every two adjacent level signals is △Y1=y2-y1, △Y2=y3-y2,..., △YN=yN-yN-1.

102-4. Determine the first quantization parameter set according to the difference between the amplitude values of every two adjacent signals.

The first quantization parameter set Q1 includes N quantization parameters, and the N quantization parameters correspond to the amplitude values of N+1 level signals. If the quantization parameter is expressed as △Y, then the first quantization parameter set Q1={△Y1, △Y2, △Y3,..., △YN}.

103: Perform fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain a second signal. The second signal is an optimally quantized signal.

Specifically, the ODSP_TX module performs fixed-point quantization processing on the first signal output by the ODSP_TX module according to the first quantization parameter set Q1 to obtain the second signal, and then sends the second signal to the DAC. Among them, the fixed-point quantization process of the ODSP_TX module can refer to the general fixed-point quantization process, and the fixed-point quantization process is not described in detail in this embodiment.

In addition, the method of this embodiment further includes: the ODSP_TX module sends the first quantization parameter set to the DAC. Correspondingly, the DAC receives the first quantization parameter set.

104: The DAC performs digital-to-analog conversion on the second signal output by the ODSP_TX module according to the first quantization parameter set to obtain an analog electrical signal, such as a first analog electrical signal, and sends the first analog electrical signal to the E/O.

Among them, the DAC may perform sampling processing on the received second signal according to an oversampling multiple of 1.25x, or may also perform sampling according to an oversampling multiple of 1.125x, and perform digital-to-analog conversion on the sampled data.

The E/O receives the first analog electrical signal output by the DAC and converts it into an optical signal, such as a first optical signal, and finally transmits the first optical signal to the second network device through an optical fiber.

Exemplarily, taking the 180GB baud rate as an example, the ENOB enhanced encoding function can completely offset the ENOB loss caused when the ADC oversampling multiple is reduced from 1.25x to 1.125x, and achieve a lossless performance reduction in the ADC oversampling multiple. The sampling frequency can be reduced by 22.5GSPS. The performance simulation result is shown in Figure 10, where the bit width of ADC data is 5 bits, the signal modulation mode is 64QAM, the entropy of PCS is 5.87, and the threshold of FEC is 4E-2. When the oversampling factor is reduced from 1.25x to 1.125x, the OSNR performance deteriorates by 0.5dB. After adopting the scheme that combines the 1.125x oversampling factor and ENOB enhanced coding, the OSNR performance is improved by 1dB, compared to the original 1.25x oversampling factor scheme Not only completely eliminates the performance cost, but also increases the 0.5dB extra.

The method provided in this embodiment, based on the probability distribution characteristics of the signal amplitude after PCS, adds the ENOB enhanced coding function at the ODSP origin, transforms the original uniform quantization parameter into a non-uniform first quantization parameter set, and makes a limited number of quantization parameters. The sampling points of fixed-point quantization are concentrated on the amplitude value of the high-probability level signal, thereby improving the accuracy of ENOB.

In addition, the DAC of the ODSP transmitter also uses the first quantization parameter set to perform digital-to-analog conversion on the fixed-point quantized data. Since the enhanced encoding function of ENOB is greater than the performance loss of the DAC due to the reduction of the oversampling multiple, the ODSP transmitter will output the output after the two are compensated. The signal performance will not deteriorate, but has been improved, so as to realize the use of low sampling frequency DAC for high baud rate ODSP.

Optionally, in the above step 102, after the ODSP_TX module determines the first quantization parameter set, if there are many types of quantization parameters contained in the first quantization parameter set, it will be difficult to implement. The types of quantization parameters are simplified. For example, the type A quantization parameters included in the first quantization parameter set are simplified to type B, where A>B, and both A and B are positive integers.

For example, taking the bit width of the data as M=4, the number of quantization parameters N is 16, and the full scale of the DAC is a voltage peak-to-peak value Vp2p, as shown in Figure 11A, 16 level signals are obtained according to the uniform distribution rule The amplitude values of are respectively [1/16, 2/16,..., 15/16, 1]*Vp2p, and the difference between the amplitude values of the two adjacent level signals is 1/16*Vp2p, at this time, quantization There is one type of parameter.

If the fixed-point quantization process is performed according to the first quantization parameter set Q1 determined in the foregoing step 102, as shown in FIG. 11B, the amplitude values of the 16 level signals are respectively [21/64, 95/256, 205/512, 217 /512, 57/128, 475/1024, 247/512, 0.5,..., 43/64, 1]*Vp2p, since the two intervals from 0 to 0.5Vp2p and 0.5 to 1Vp2p are symmetrical, one side is 0 to 0.5 Take Vp2p as an example, calculate the difference between the amplitude values of two adjacent level signals as:

△Y1=21/64-0, △Y2=95/256-21/64, △Y3=205/512-95/256, △Y4=217/512-205/512, △Y5=57/128-217 /512, △Y6=475/1024-57/128, △Y7=247/512-475/1024, △Y8=0.5-247/512, then the first quantization parameter set Q1={△Y1,△Y2, ……,△YN}*Vp2p={336/1024, 44/1024, 30/1024, 24/1024, 22/1024, 19/1024, 19/1024, 18/1024, 19/1024, 19/1024, 22/1024, 24/1024, 30/1024, 44/1024, 336/1024}*Vp2p.

Assuming that the type of quantization parameter included in the first quantization parameter set Q1 is A, the number of types A of the corresponding quantization parameter in Q1 is 7, that is, A=7.

Assuming that the second quantization parameter set Q2 is a simplified optimal quantization parameter set, and the number of quantization parameter types included in Q2 is B, as shown in Fig. 11C, the process of converting category A to category B includes:

In the first step, the ODSP_TX module obtains all B-type quantization parameter sets that meet the first preset condition. The first preset condition includes condition 1 and condition 2. Further,

Condition 1: The sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and the maximum output voltage of the DAC is the maximum voltage in the full-scale range Vp2p.

Condition 2: The number of quantization parameters included in each quantization parameter set is equal to 2 ^M , where M is the bit width of the data.

In the second step, the ODSP_TX module calculates the error between each quantization parameter set of type B that meets the first preset condition and the first quantization parameter set,

In the third step, it is determined that the quantization parameter set with the smallest error is the second quantization parameter set.

Exemplarily, assuming that the simplified second quantization parameter set Q2 contains 3 types of quantization parameters, B=3, A=7, and data bit width M=4, then the number of fixed-point quantization N=16, when B=3 when satisfying the above conditions the first predetermined quantization parameter set may be six, because the quantization parameter a _{³ 3} a total of six combinations. According to the principle of error and minimum, one of the six quantization parameter sets satisfying the above-mentioned "condition 1" and "condition 2" may be determined as the second quantization parameter set.

Exemplarily, an implementation manner is that when B=3, the three quantization parameters divided are: 2 ^(-M-1) , 2 ^(-M) , and 2 ^(-M+1) . As shown in Figure 11C, when M=4, the three quantization parameters are: 2 ^(-M-1) = 2 ^(-4-1) = 2 ^-5 = ^{1/32, 2 (-M)} = 2 ^{( -4)} = 1/16, 2 ^(-M+1) = 2 ^(-4+1) = 2 ^-3 = 1/8.

The error between the quantization parameter set and the first quantization parameter set shown in FIG. 11B is calculated as:

D _error =(y1′-y1) ² +(y2′-y2) ² +(y3′-y3) ² +......+(yN′-yN) ²

Wherein, D _error is the error between the selected quantization parameter set and the first quantization parameter set, and y1', y2', y3',..., yN' are the N electrical parameters corresponding to the selected quantization parameter set. The amplitude values of the flat signal, y1, y2, y3,..., yN are the amplitude values of the N level signals corresponding to the first quantization parameter set.

In this embodiment, according to the error between the two quantization parameter sets shown in FIG. 11B and FIG. 11C,

Similarly, the errors between the remaining five quantization parameter sets and the first quantization parameter set are calculated, and then the set with the smallest error is selected as the second quantization parameter set.

Then, the ODSP_TX module performs fixed-point quantization processing on the first data according to the type B of the simplified second quantization parameter set, and the specific processing process is the same as the foregoing step 103.

It should be understood that in the aforementioned "second step", one can also be selected from a plurality of quantization parameter sets as the second quantization parameter set in other ways.

In this embodiment, the types of quantization parameters in the first quantization parameter set are simplified, so that the types of quantization parameters included in the simplified second quantization parameter are reduced, which is convenient for implementation, and also improves the DAC/ADC processing efficiency.

The method provided in this embodiment also includes a signal processing process at the receiving end, which is similar to the process at the transmitting end, as shown in FIG. 12, which specifically includes:

201: The ODSP_RX module sends the third quantization parameter set to the ADC. Correspondingly, the ADC receives the third quantization parameter set sent by the ODSP_RX module, and the third quantization parameter set is denoted as Q3.

Specifically: the second network device receives the optical signal sent by the first network device, such as the first optical signal, after O/E converts the first optical signal into an electrical signal, such as the fourth signal, and then sends it to the ADC.

201-1: After the ADC receives the electrical signal (fourth signal) from the O/E transmission, it performs analog-to-digital conversion on the received fourth signal according to uniform quantization parameters to obtain a digital signal, such as the third signal, and then The third signal is sent to the ODSP_RX module.

201-2: The ODSP_RX module receives the third signal, and obtains the second probability density function PDF corresponding to the third signal. The amplitude value of the level signal corresponding to the second PDF is distributed in a Gaussian curve.

Optionally, the second PDF may be expressed as P2.

201-3: The ODSP_RX module determines a third quantization parameter set Q3 according to the second PDF, and sends the third quantization parameter set Q3 to the ADC.

Wherein, the third quantization parameter set Q3 includes at least one quantization parameter, the quantization parameter is the difference between the amplitude values of two adjacent level signals, and each of the level signals corresponds to the voltage range range of the ADC The voltage range of the ADC is (0～1)Vp2p.

Since the optical signal sent by the first network device is transmitted through the optical fiber, the receiver of the second network device still maintains the Gaussian curve distribution characteristics, so the third quantization parameter set Q3 can be determined by the same method as in the foregoing step 102, specifically, include:

Obtain a second probability distribution function corresponding to the second PDF, and distribute the second probability distribution function at equal intervals according to the number of fixed-point quantization N to obtain N values, where N is a positive integer and N≥1; determine the N The amplitude value of the level signal corresponding to each of the two values, and calculate the difference between the amplitude values of every two adjacent level signals; and according to the The difference between the amplitude values determines the third quantization parameter set Q3. For the specific process, refer to the foregoing 102, which is not repeated here in this embodiment.

202: The ADC performs analog-to-digital conversion on another electrical signal output by the received O/E according to the third quantization parameter set Q3, for example, the fifth signal to obtain a sixth signal, and sends the sixth signal to the ODSP_RX module.

203: The ODSP_RX module receives the sixth signal output by the ADC, and performs fixed-point quantization processing on the sixth signal according to the third quantization parameter set Q3 to obtain the seventh signal.

204: The ODSP_RX module decodes and modulates the fixed-point quantized signal, such as the seventh signal. The decoding and modulation process is inverse to the coding and modulation process at the transmitting end, and may include channel demodulation, FEC decoding, and PCS inverse processing. Then get the bit data. The bit data is the first data sent by the aforementioned ODSP_TX module.

Optionally, when the third quantization parameter set Q3 is used for fixed-point quantization processing in step 203, the method further includes simplifying the third quantization parameter set Q3 to obtain a fourth quantization parameter set Q4, wherein the quantization included in Q4 The parameter type is smaller than the type of quantization parameter included in Q3. For example, if the type of quantization parameter included in Q3 is C, and the type of quantization parameter included in Q4 is D, then D<C, and both C and D are positive integers.

Among them, the ODSP_RX module obtains all C-type quantization parameter sets that meet the second preset condition. The second preset condition includes condition 3 and condition 4. Further,

Condition 3: The sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the ADC.

Condition 4: The number of quantization parameters included in each quantization parameter set is equal to 2 ^M , where M is the bit width of the data.

Then, the ODSP_RX module calculates the error between each D type quantization parameter set that meets the

conditions

3 and 4 and the third quantization parameter set Q3, and determines that the fourth quantization parameter set Q4 is all the errors The smallest set of quantization parameters. The specific method is the same as the "second step" and "third step" of the foregoing embodiment, and will not be repeated here.

In this embodiment, the ENOB enhanced decoding function is added to the ODSP receiving end, and the optimal quantization parameter set is provided to the ADC. The ADC performs analog-to-digital conversion on the signal according to the optimal quantization parameter set, which improves the ENOB of the ADC. Then the ADC output data is converted into uniform quantization, and the subsequent demodulation and decoding processing is performed.

In this embodiment, by increasing the ENOB of the DAC/ADC processing constellation to shape the PCS signal, it offsets the performance cost introduced by reducing the ADC/DAC sampling frequency, so as to realize the use of a low sampling rate ADC/DAC to transmit high baud rate signals, which greatly reduces the ADC /DAC is difficult to design.

The following describes device embodiments corresponding to the foregoing method embodiments.

FIG. 13 is a schematic structural diagram of a signal processing device provided by an embodiment of the application. The device may be an ODSP_TX module or an ODSP_RX module in the foregoing embodiment, or a chip that includes the functions of an ODSP_TX module or an ODSP_RX module.

Specifically, as shown in FIG. 13, the device may include: a receiving module 1301, a processing module 1302, and a sending module 1303. In addition, the device may also include other units or modules such as a storage module.

Wherein, when the device is used as an ODSP_TX module, the processing module 1302 is configured to perform encoding and modulation processing on the first data to obtain the first signal, and determine the first quantization parameter set according to the first PDF, and the first quantization parameter set is determined according to the first PDF. The quantization parameter set includes at least one quantization parameter, where the quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to a voltage value in the voltage range of the DAC; The first quantization parameter set performs fixed-point quantization processing on the first signal to obtain a second signal.

Wherein, the encoding and modulation processing includes PCS processing, and the first data is processed by the PCS to obtain a first probability density function PDF.

The sending module 1303 is configured to send the second signal to the DAC.

In addition, the sending module 1303 is configured to send the first quantization parameter set to the DAC.

Optionally, in a specific implementation manner, the processing module 1302 is specifically configured to obtain a first probability distribution function corresponding to the first PDF; make the first probability distribution function according to the fixed-point quantization number N N values are allocated at equal intervals, where N is a positive integer and N≥1; the amplitude value of the level signal corresponding to each of the N values is determined, and every two adjacent ones are calculated The difference between the amplitude values of the level signals, and the first quantization parameter set is determined according to the difference between the amplitude values of every two adjacent level signals.

Optionally, in another specific implementation manner, the processing module 1302 is specifically configured to convert the first quantization parameter set into a second quantization parameter set, and use B types of quantization parameters in the second quantization parameter set. Perform fixed-point quantization processing on the first signal to obtain the second signal; wherein the number of types of quantization parameters included in the first quantization parameter set is A, and the number of types of quantization parameters included in the second quantization parameter set Is B, and A>B, A and B are both positive integers.

Optionally, in another specific implementation manner, the processing module 1302 is further configured to obtain all B-type quantization parameter sets that meet the first preset condition, and calculate each B-type set that meets the first preset condition. The error between the quantization parameter set and the first quantization parameter set, and the second quantization parameter set is determined to be the smallest quantization parameter set among all the errors.

Wherein, the first preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and the number of quantization parameters included in each quantization parameter set is N Equal to 2 ^M , where M is the data bit width.

In addition, when the device is used as an ODSP_RX module, the receiving module 1301 is used to receive the third signal converted by the analog-to-digital converter ADC, and the third signal is processed by the ADC on the fourth signal according to the uniform quantization parameter. Obtained after analog-to-digital conversion; a processing module 1302, configured to obtain a second probability density function PDF of the third signal, and determine a third set of quantization parameters according to the second PDF, and the third set of quantization parameters includes at least one A quantization parameter, where the quantization parameter is the difference between the amplitude values of two adjacent level signals; the amplitude value distribution of the level signal corresponding to the second PDF is a Gaussian curve distribution; the sending module 1303 is configured to send The ADC sends the third quantization parameter set.

Optionally, in a specific implementation manner, the processing module 1302 is specifically configured to obtain a second probability distribution function corresponding to the second PDF, and make the second probability distribution function according to the fixed-point quantization number N N values are allocated at equal intervals, where N is a positive integer and N≥1; the amplitude value of the level signal corresponding to each of the N values is determined, and every two adjacent ones are calculated The difference between the amplitude values of the level signals is determined according to the difference between the amplitude values of every two adjacent level signals.

Optionally, in another specific implementation manner, the processing module 1302 is further configured to obtain all D types that meet the second preset condition before converting the third quantization parameter set to the fourth quantization parameter set. Quantization parameter set, calculate the error between each D type quantization parameter set that meets the second preset condition and the third quantization parameter set, and determine that the quantization parameter set with the smallest error is the fourth Quantization parameter set.

Optionally, in another specific implementation manner, the receiving module 1301 is further configured to receive a sixth signal sent by the ADC, where the ADC performs analog-to-digital conversion on the received fifth signal After processing, the processing module 1302 is further configured to perform fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain a seventh signal, and perform decoding and modulation processing on the seventh signal, for example, including channel demodulation, The first data is obtained by FEC decoding and probabilistic constellation shaping PCS de-shaping processing.

The device provided in this embodiment offsets the performance cost introduced by reducing the ADC/DAC sampling frequency by increasing the ENOB of the DAC/ADC processing constellation shaping PCS signal, thereby realizing the use of low sampling rate ADC/DAC to transmit high baud rate signals. This greatly reduces the difficulty of ADC/DAC design.

In addition, in specific hardware implementation, this embodiment also provides a network device, which can be used to implement the signal processing method in the foregoing embodiment.

Specifically, FIG. 14 shows a schematic structural diagram of a network device. The network device may include a transceiver 141, a processor 142, a memory 143, a transmission bus 144, and at least one interface 145. In addition, it may also include E/O, O/E, sensor modules, cameras and other components. In other embodiments of the present application, the network device may include more or fewer components than those shown in the figure, or combine certain components, or split certain components, or arrange different components. The illustrated components can be implemented in hardware, software, or a combination of software and hardware.

The transceiver 141 includes a transmitter 1411 and a receiver 1412. The transmitter 1411 is a transmitter as shown in FIG. 5 and can be used to implement all the functions of the ODSP_TX module. Wherein, the ODSP_TX module may include but is not limited to: PCS shaping module, FEC encoding module, channel modulation module, ENOB enhanced encoding module, etc.

Specifically, when the network device is used as a transmitter, the transmitter includes: ODSP, DAC, and electro-optical converter E/O. Wherein, the ODSP is the signal processing device described in the foregoing embodiment, which can be used to implement all the functions of the ODSP_TX module. The DAC is used to receive the digital signal sent by the ODSP and convert the digital signal into a first analog electrical signal; the E/O is used to receive the first analog electrical signal output by the DAC, and The first analog electrical signal is converted into a first optical signal, and the first optical signal is transmitted.

Similarly, the receiver 1412 is the receiver shown in FIG. 6, which can be used to implement all the functions of the ODSP_RX module. Wherein, the ODSP_RX module may include, but is not limited to: a PCS de-shaping module, an FEC decoding module, a channel demodulation module, an ENOB enhanced decoding module, and so on.

Specifically, when the network device is used as a receiver, the receiver includes an opto-electronic converter O/E, ODSP, and ADC. The O/E is used to receive the first optical signal sent by the transmitter, convert the first optical signal into a second analog electrical signal, and transmit the second analog electrical signal to the ADC; the ADC uses When receiving a second analog electrical signal, converting the second analog electrical signal into a digital signal, and transmitting the digital signal to the ODSP; the ODSP is the signal processing device described in the foregoing embodiment, which can be used to implement All functions of the ODSP_RX module.

The processor 142 may include one or more processing modules. For example, the processor 142 may include a modem processor, a signal processor, and a controller. Among them, different processing modules may be independent devices, or integrated in one or more processors.

The processor 142 may also be provided with a memory 143 for storing program instructions and data. In some embodiments, the memory 143 in the processor 142 is a cache memory. The memory can store instructions or data that have just been used or recycled by the processor 142. If the processor 142 needs to use the instruction or data again, it can be directly called from the memory, avoiding repeated access, reducing the waiting time of the processor 142, and improving the efficiency of the system.

In some embodiments, the at least one interface may include an optical fiber interface, an inter-integrated circuit (I2C) interface, and/or a universal serial bus (USB) interface, etc. The optical fiber interface can be used to connect the optical fiber to connect it with the receiver.

The memory 143 may be used to store computer executable program code, where the executable program code includes instructions. In addition, the memory 143 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, a flash memory device, a universal flash storage (UFS), and the like. The processor 142 executes various functional applications and signal processing by running instructions stored in the internal memory 143 and/or instructions stored in a memory provided in the processor.

Optionally, this embodiment also provides a processing chip, including a processor and a memory, the processor is connected to the memory, and the memory is used to store computer program instructions; the processor is used to execute the Instructions so that the processing chip executes the method described in FIG. 7 or FIG. 12.

Wherein, when the computer loads and executes the computer program instructions, all or part of the method flows or functions according to the foregoing embodiments of the present application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium.

In addition, an embodiment of the present application also provides an optical communication system, which includes two or more network devices for implementing the signal processing method in the foregoing embodiment. The structure of each of the network devices may be the same as the structure of the network device shown in FIG. 14.

In addition, an embodiment of the present application further provides a computer storage medium, where the computer storage medium may store a program, and the program may include part or all of the steps of the signal processing method provided in the present application when the program is executed. The storage medium includes, but is not limited to, magnetic disks, optical disks, read only memory (ROM) or random access memory (RAM), etc.

In the above-mentioned embodiments, all or part of it may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in the form of a computer program product in whole or in part.

In addition, in order to facilitate a clear description of the technical solutions of the embodiments of the present application, in the embodiments of the present application, words such as "first" and "second" are used to distinguish the same or similar items with substantially the same function and effect. Those skilled in the art can understand that words such as "first" and "second" do not limit the quantity and order of execution, and words such as "first" and "second" do not limit the difference.

The implementation manners of the application described above do not constitute a limitation on the protection scope of the application.

Claims

A signal processing method, characterized in that the method includes:

Perform coding and modulation processing on the first data to obtain a first signal, where the coding and modulation processing includes a probability constellation shaping PCS processing, and the first data is processed by the PCS to obtain a first probability density function PDF;

A first quantization parameter set is determined according to the first PDF, the first quantization parameter set includes at least one quantization parameter, and the quantization parameter is the difference between the amplitude values of two adjacent level signals. The flat signal corresponds to a voltage value in the voltage range of the digital-to-analog converter DAC;

Performing fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain a second signal.
The method according to claim 1, wherein determining a first set of quantization parameters according to the first PDF comprises;

Obtaining a first probability distribution function corresponding to the first PDF;

The first probability distribution function is distributed at equal intervals according to the number N of fixed-point quantization to obtain N values, where N is a positive integer and N≥1;

Determine the amplitude value of the level signal corresponding to each of the N values, and calculate the difference between the amplitude values of every two adjacent level signals;

The first quantization parameter set is determined according to the difference between the amplitude values of every two adjacent level signals.
The method according to claim 1 or 2, wherein the method further comprises:

Send the second signal to the DAC.
The method according to any one of claims 1-3, wherein the method further comprises:

Sending the first quantization parameter set to the DAC.
The method according to any one of claims 1 to 4, wherein the performing fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain the second signal comprises:

Convert the first quantization parameter set to a second quantization parameter set, where the number of types of quantization parameters included in the first quantization parameter set is A, and the number of types of quantization parameters included in the second quantization parameter set is B , And A>B, both A and B are positive integers;

Performing fixed-point quantization processing on the first signal by using B types of quantization parameters in the second quantization parameter set to obtain the second signal.
The method according to claim 5, wherein before converting the first quantization parameter set to the second quantization parameter set, the method further comprises:

Acquire all B-type quantization parameter sets that meet a first preset condition, where the first preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and each The number N of quantization parameters included in the quantization parameter set is equal to 2 M , and M is the bit width of the data;

Calculating the error between each quantization parameter set of type B that meets the first preset condition and the first quantization parameter set,

It is determined that the one quantization parameter set with the smallest error is the second quantization parameter set.
A signal processing method, characterized in that the method includes:

Receiving a third signal output by an analog-to-digital converter ADC, where the third signal is output after the ADC performs analog-to-digital conversion on the received fourth signal;

Obtaining a second probability density function PDF of the third signal;

A third quantization parameter set is determined according to the second PDF. The third quantization parameter set includes at least one quantization parameter. The quantization parameter is the difference between the amplitude values of two adjacent level signals. The flat signal corresponds to a voltage value in the voltage range of the ADC.
The method according to claim 7, wherein determining a third set of quantization parameters according to the second PDF comprises:

Obtaining a second probability distribution function corresponding to the second PDF;

The second probability distribution function is allocated at equal intervals according to the number N of fixed-point quantization to obtain N values, where N is a positive integer and N≥1;

Determine the amplitude value of the level signal corresponding to each of the N values, and calculate the difference between the amplitude values of every two adjacent level signals;

The third quantization parameter set is determined according to the difference between the amplitude values of every two adjacent level signals.
The method according to claim 7 or 8, wherein the method further comprises:

Sending the third quantization parameter set to the ADC.
The method according to any one of claims 7-9, wherein the method further comprises:

Receiving a sixth signal output by the ADC, where the sixth signal is output after the ADC performs analog-to-digital conversion on the received fifth signal;

Performing fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain a seventh signal;

Performing decoding and modulation processing on the seventh signal to obtain the first data, and the decoding and modulation processing includes performing probability constellation shaping PCS inverse processing on the seventh signal.
The method according to claim 10, wherein the performing fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain the seventh signal comprises:

Convert the third quantization parameter set to a fourth quantization parameter set, where the number of types of quantization parameters included in the third quantization parameter set is C, and the number of types of quantization parameters included in the fourth quantization parameter set is D , And C>D, C and D are both positive integers;

Perform fixed-point quantization processing on the sixth signal by using D types of quantization parameters in the fourth quantization parameter set to obtain the seventh signal.
The method according to claim 11, wherein before converting the third quantization parameter set to the fourth quantization parameter set, the method further comprises:

Acquire all D types of quantization parameter sets that meet a second preset condition, where the second preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the ADC, and each The number N of quantization parameters included in the quantization parameter set is equal to 2 M , and M is the bit width of the data;

Calculating the error between each D-type quantization parameter set that meets the second preset condition and the third quantization parameter set,

It is determined that the quantization parameter set with the smallest error is the fourth quantization parameter set.
A signal processing device, characterized by comprising: a processing module,

The processing module is configured to perform coding and modulation processing on first data to obtain a first signal, wherein the coding and modulation processing includes probabilistic constellation shaping PCS processing, and the first data is processed by the PCS to obtain a first probability Density function PDF;

The processing module is further configured to determine a first quantization parameter set according to the first PDF, and perform fixed-point quantization processing on the first signal according to the first quantization parameter set to obtain a second signal;

The first quantization parameter set includes at least one quantization parameter, the quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to the voltage range range of the digital-to-analog converter DAC One of the voltage values in.
The device according to claim 13, wherein:

The processing module is specifically configured to obtain a first probability distribution function corresponding to the first PDF; distribute the first probability distribution function at equal intervals according to the number N of fixed-point quantization to obtain N values, where N is positive Integer and N≥1; determine the amplitude value of the level signal corresponding to each of the N values, and calculate the difference between the amplitude values of every two adjacent level signals; according to the The difference between the amplitude values of every two adjacent level signals determines the first quantization parameter set.
The device according to claim 13 or 14, further comprising a sending module,

The sending module is configured to send the second signal to the DAC.
The device of claim 15, wherein:

The sending module is further configured to send the first quantization parameter set to the DAC.
The device according to any one of claims 13-16, characterized in that:

The processing module is specifically configured to convert the first quantization parameter set into a second quantization parameter set, where the number of types of quantization parameters included in the first quantization parameter set is A, and the number of quantization parameters included in the second quantization parameter set is A. The number of types of quantization parameters is B, and A>B, A and B are both positive integers; and using B quantization parameters in the second quantization parameter set to perform fixed-point quantization processing on the first signal to obtain Mentioned second signal.
The device of claim 17, wherein:

The processing module is further configured to obtain all B-type quantization parameter sets that meet a first preset condition before converting the first quantization parameter set to a second quantization parameter set, and the first preset condition includes: The sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the DAC, and the number of quantization parameters included in each quantization parameter set N is equal to 2 M , and M is the bit width of the data Calculate the error between each quantization parameter set of type B that meets the first preset condition and the first quantization parameter set, and determine that the quantization parameter set with the smallest error is the second quantization parameter set .
A signal processing device, characterized by comprising: a receiving module and a processing module,

The receiving module is configured to receive a third signal output by an analog-to-digital converter ADC, where the third signal is output after the ADC performs analog-to-digital conversion on the received fourth signal;

The processing module is configured to obtain a second probability density function PDF of the third signal, and determine a third quantization parameter set according to the second PDF, the third quantization parameter set including at least one quantization parameter, the The quantization parameter is the difference between the amplitude values of two adjacent level signals, and each level signal corresponds to a voltage value in the voltage range of the ADC.
The device of claim 19, wherein:

The processing module is configured to obtain a second probability distribution function corresponding to the second PDF, and distribute the second probability distribution function at equal intervals according to the number N of fixed-point quantization to obtain N values, where N is a positive integer And N≥1; determine the amplitude value of the level signal corresponding to each of the N values, and calculate the difference between the amplitude values of every two adjacent level signals, according to each The difference between the amplitude values of two adjacent level signals determines the third quantization parameter set.
The device according to claim 19 or 20, further comprising a sending module,

The sending module is configured to send the third set of quantization parameters to the ADC.
The device of claim 21, wherein:

The receiving module is further configured to receive a sixth signal output by the ADC, where the sixth signal is output after the ADC performs analog-to-digital conversion on the received fifth signal;

The processing module is further configured to perform fixed-point quantization processing on the sixth signal according to the third quantization parameter set to obtain a seventh signal; and perform decoding and modulation processing on the seventh signal to obtain first data, and the decoding The modulation processing includes performing PCS inverse processing of probabilistic constellation shaping on the seventh signal.
The device according to any one of claims 19-22, characterized in that:

The processing module is specifically configured to convert the third quantization parameter set into a fourth quantization parameter set, where the number of types of quantization parameters included in the third quantization parameter set is C, and the fourth quantization parameter set includes The number of types of quantization parameters is D, and C>D, and C and D are all positive integers; and using D quantization parameters in the fourth quantization parameter set to perform fixed-point quantization processing on the sixth signal to obtain The seventh signal.
The device of claim 23, wherein:

The processing module is further configured to obtain all D-type quantization parameter sets that meet the second preset condition before converting the third quantization parameter set to the fourth quantization parameter set, and calculate each set that meets the second preset condition. Suppose the error between the conditional D-type quantization parameter set and the third quantization parameter set, and determine the quantization parameter set with the smallest error as the fourth quantization parameter set;

Wherein, the second preset condition includes: the sum of all quantization parameters in each quantization parameter set is equal to the maximum output voltage of the ADC, and the number of quantization parameters included in each quantization parameter set is N Equal to 2 M , where M is the bit width of the data.
A transmitter, characterized in that it comprises an optical digital signal processor ODSP, a digital-to-analog converter DAC, and an electro-optical converter;

The ODSP is the signal processing device according to any one of claims 13 to 18;

The DAC is used to receive the digital signal sent by the ODSP, and convert the digital signal into a first analog electrical signal;

The electrical-optical converter is used for receiving the first analog electrical signal output by the DAC, converting the first analog electrical signal into a first optical signal, and transmitting the first optical signal.
A receiver, characterized in that it comprises an optical-to-electrical converter, an optical digital signal processor ODSP, and an analog-to-digital converter ADC;

The photoelectric converter is configured to receive a first optical signal, convert the first optical signal into a second analog electrical signal, and transmit the second analog electrical signal to the ADC;

The ADC is configured to receive the second analog electrical signal, convert the second analog electrical signal into a digital signal, and transmit the digital signal to the ODSP;

The ODSP is the signal processing device according to any one of claims 19 to 24.
A processing chip includes a processor and a memory, the processor is coupled with the memory, and is characterized in that:

The memory is used to store computer program instructions;

The processor is configured to execute the instructions stored in the memory, so that the processing chip executes the method according to any one of claims 1 to 12.
A computer-readable storage medium, characterized in that computer program instructions are stored in the computer-readable storage medium, and when the computer program instructions are executed, the computer program instructions according to any one of claims 1 to 12 are implemented. method.