阅读说明：本技术 用于光通信的方法、设备、装置和计算机可读介质 (Method, apparatus, device and computer readable medium for optical communication ) 是由 叶晨晖 胡小锋 张凯宾 于 2019-10-25 设计创作，主要内容包括：本公开的实施例涉及用于光通信的方法、设备、装置和计算机可读介质。该方法包括在光线路终端处,在从光网络单元到光线路终端的上行链路信道上接收畸变信号序列。该畸变信号序列由原始信号序列在上行链路信道上的传输过程中畸变而产生。该方法还包括将畸变信号序列编码化,以确定用于表征上行链路信道的属性的特征参数以及基于畸变信号序列和特征参数,确定畸变信号序列和所述原始信号序列之间的关联关系,以用于恢复原始信号序列。以此方式,一方面实现针对不同的上行链路信道的畸变信号恢复,另一方面,由于与发生畸变相关联的信道特征是通过畸变信号来提取的,这提高了恢复信号的准确性,降低了系统的复杂程度。(Embodiments of the present disclosure relate to methods, devices, apparatuses and computer-readable media for optical communication. The method includes receiving, at an optical line terminal, a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal. The distorted signal sequence results from distortion of the original signal sequence during transmission on the uplink channel. The method further includes encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel and determining an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameter for recovering the original signal sequence. In this way, on the one hand, distorted signal recovery for different uplink channels is achieved, and on the other hand, since the channel characteristics associated with the occurrence of distortion are extracted by the distorted signal, the accuracy of the recovered signal is improved, and the complexity of the system is reduced.)

1. A method for optical communication, comprising:

receiving, at an optical line terminal, a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal, the distorted signal sequence resulting from distortion of an original signal sequence during transmission on the uplink channel;

encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel; and

and determining the association relation between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameters so as to recover the original signal sequence.

2. The method of claim 1, wherein determining the feature parameter comprises:

obtaining from the distorted signal sequence at least one of the following signal reference parameters:

the voltage amplitude of the distorted signal sequence, or

The power of the distorted signal sequence;

generating a histogram by counting the signal reference parameters; and

determining the feature parameter by encoding the histogram.

3. The method of claim 1, wherein determining the feature parameter comprises:

performing discrete Fourier transform on the distorted signal sequence; and

the characteristic parameter is determined by encoding the transformed distorted signal sequence.

4. The method of claim 1, wherein the attribute comprises at least one of:

the bandwidth of the uplink channel is such that,

a center wavelength of the uplink channel, an

A length of optical fiber for carrying the uplink channel.

5. The method of claim 1, wherein determining the association comprises:

determining a degree of distortion experienced by the distorted signal sequence in transmission of the uplink channel based on the characteristic parameter;

restoring the distorted signal sequence to the original signal sequence based on the degree of distortion; and

determining the correlation based on a deviation between the distorted signal sequence and the original signal sequence.

6. The method of claim 1, further comprising:

receiving another distorted signal sequence resulting from another original signal sequence being distorted during transmission on another uplink channel from another optical network unit to the optical line terminal; and

updating the correlation based on a characteristic parameter characterizing a property of the other uplink channel and the other distorted signal sequence; and

restoring the other distorted signal sequence to the other original signal sequence based on the updated correlation.

7. An apparatus for optical communication, comprising:

at least one processor; and

a memory coupled with the at least one processor, the memory containing instructions stored therein that, when executed by the at least one processing unit, cause the apparatus to:

receiving a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal, wherein the distorted signal sequence is generated by distortion of an original signal sequence during transmission on the uplink channel;

encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel; and

determining an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameters for restoring the original signal sequence.

8. The apparatus of claim 7, wherein the apparatus is caused to determine the characteristic parameter by:

obtaining from the distorted signal sequence at least one of the following signal reference parameters:

the voltage amplitude of the distorted signal sequence, or

The power of the distorted signal sequence;

generating a histogram by counting the signal reference parameters; and

determining the feature parameter by encoding the histogram.

9. The apparatus of claim 7, wherein the apparatus is caused to determine the characteristic parameter by:

performing discrete Fourier transform on the distorted signal sequence; and

the characteristic parameter is determined by encoding the transformed distorted signal sequence.

10. The apparatus of claim 7, wherein the attribute comprises at least one of:

the bandwidth of the uplink channel is such that,

a center wavelength of the uplink channel, an

A length of optical fiber for carrying the uplink channel.

11. The apparatus of claim 7, wherein the apparatus is caused to determine the association by:

determining a degree of distortion experienced by the distorted signal sequence in transmission of the uplink channel based on the characteristic parameter;

restoring the distorted signal sequence to an original signal sequence based on the distortion degree; and

determining the correlation based on a deviation between the distorted signal sequence and the original signal sequence.

12. The apparatus of claim 7, wherein the at least one memory and the instructions are further configured to, with the at least one processor, cause the apparatus to:

receiving another distorted signal sequence resulting from another original signal sequence being distorted during transmission on another uplink channel from another optical network unit to the optical line terminal; and

updating the correlation based on a characteristic parameter characterizing a property of the other uplink channel and the other distorted signal sequence; and

restoring the other distorted signal sequence to the other original signal sequence based on the updated correlation.

13. An apparatus for optical communication, comprising:

means for receiving a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal, the distorted signal sequence resulting from distortion of an original signal sequence during transmission on the uplink channel;

means for encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel; and

means for determining an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameter for recovering the original signal sequence.

14. A computer-readable medium having instructions stored thereon, which, when executed by at least one processing unit, cause the at least one processing unit to be configured to perform the method of any of claims 1-6.

Technical Field

Embodiments of the present disclosure relate to the field of optical communications, and more particularly, to methods, devices, apparatuses, and computer-readable media for optical communications.

Background

Currently, standardization of high speed passive optical networks (g.hsps) is being initiated, which aims to provide intensity modulated direct detection (IM/DD) at 50Gbps per wavelength, which is technically very challenging. As the interaction between factors can lead to a degradation in signal quality. Such factors include, for example, non-explicitly expressed channel responses (also known as non-gaussian and spectrally selective channel responses), insufficient bandwidth, fiber dispersion, and intensity-dependent nonlinear responses of the device and fiber, among others.

Therefore, much discussion and research has been conducted on signal quality management for a Passive Optical Network (PON) scale within 20km of a 50G (+) PON. Feed Forward Equalizers (FFE), Decision Feedback Equalizers (DFE), Volterra equalizers, and various types of Machine Learning (ML) have all been considered carefully for signal equalization, while Neural Networks (NN) have proven to be the most powerful and sufficiently adaptive.

Disclosure of Invention

In general, embodiments of the present disclosure relate to a method, apparatus, device, and computer-readable medium for optical communication.

In a first aspect of the disclosure, a method for optical communication is provided. The method comprises receiving, at an optical line terminal, a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal. The distorted signal sequence results from an original signal sequence being distorted during transmission on the uplink channel. The method further includes encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel and determining an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameter for recovering the original signal sequence.

In a second aspect of the disclosure, an apparatus for optical communication is provided. The apparatus includes at least one processor; and a memory coupled to the at least one processor, the memory containing instructions stored therein, which when executed by the at least one processing unit, cause the apparatus to perform the method of the first aspect.

In a third aspect of the present disclosure, an apparatus for optical communication is provided. The apparatus comprises means for receiving a distorted signal sequence on an uplink channel from an optical network unit to the optical line terminal. The distorted signal sequence results from distortion of the original signal sequence during transmission on said uplink channel. The apparatus also includes means for encoding the distorted signal sequence to determine a characteristic parameter characterizing a property of the uplink channel and means for determining an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the characteristic parameter for recovering the original signal sequence.

In a fourth aspect of the disclosure, a computer-readable medium is provided. The computer-readable medium has stored thereon instructions which, when executed by at least one processing unit, cause the at least one processing unit to be configured to perform the method of the first aspect.

It should be understood that the statements herein reciting aspects are not intended to limit the critical or essential features of the embodiments of the present disclosure, nor are they intended to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

Fig. 1 is a schematic diagram of a communication system 100 in which embodiments described in the present disclosure may be implemented;

fig. 2 illustrates a flow diagram of a method 200 for optical communication, in accordance with certain embodiments of the present disclosure;

3A-3C illustrate schematic diagrams of signal processing implemented by certain embodiments of the present disclosure;

FIG. 4 shows a schematic diagram of signal processing implemented according to certain embodiments of the present disclosure;

FIG. 5 shows a schematic diagram of implementing signal processing according to some embodiments of the present disclosure;

FIG. 6 shows a schematic of exemplary experimental results obtained using embodiments according to the present disclosure;

FIG. 7 illustrates a simplified block diagram of an electronic device suitable for implementing embodiments of the present disclosure; and

FIG. 8 illustrates a schematic diagram of a computer-readable medium suitable for implementing embodiments of the present disclosure.

Detailed Description

The principles and spirit of the present disclosure will be described with reference to a number of exemplary embodiments shown in the drawings. It is understood that these specific embodiments are described merely to enable those skilled in the art to better understand and implement the present disclosure, and are not intended to limit the scope of the present disclosure in any way.

As used herein, the terms "comprises," comprising, "and the like are to be construed as open-ended inclusions, i.e.," including, but not limited to. The term "based on" should be understood as "based at least in part on". The term "one embodiment" or "the embodiment" should be understood as "at least one embodiment". The terms "first," "second," and the like may refer to different or the same object. Other explicit and implicit definitions are also possible below.

As used herein, the term "determining" encompasses a wide variety of actions. For example, "determining" can include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, a database or another data structure), ascertaining and the like. Further, "determining" can include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), and the like. Further, "determining" may include resolving, selecting, choosing, establishing, and the like.

The term "circuitry" as used herein refers to one or more of the following: (a) hardware-only circuit implementations (such as implementations in only analog and/or digital circuitry); and (b) a combination of hardware circuitry and software, such as (if applicable): (i) a combination of analog and/or digital hardware circuitry and software/firmware, and (ii) any portion of a hardware processor and software (including a digital signal processor, software, and memory that work together to cause an apparatus, such as an OLT or other computing device, to perform various functions); and (c) hardware circuitry and/or a processor, such as a microprocessor or a portion of a microprocessor, that requires software (e.g., firmware) for operation, but may be software-free when software is not required for operation.

The definition of circuit applies to all usage scenarios of this term in this application, including any claims. As another example, the term "circuitry" as used herein also covers an implementation of merely a hardware circuit or processor (or multiple processors), or a portion of a hardware circuit or processor, or software or firmware accompanying it. For example, the term "circuitry" would also cover a baseband integrated circuit or processor integrated circuit or a similar integrated circuit in an OLT or other computing device, as applicable to the particular claim element.

The term "Neural Network (NN)" as used herein may be understood, for example, as a machine learning model that is capable of learning from training data the associations between respective inputs and outputs, such that after training is completed, a given input is processed based on a trained set of parameters to generate a corresponding output. "neural networks" may also sometimes be referred to as "learning networks", "learning models", "networks", or "models". These terms are used interchangeably herein.

Fig. 1 is a schematic diagram of a communication system 100 in which embodiments described in the present disclosure may be implemented. As shown in fig. 1, the communication system 100 includes an Optical Line Terminal (OLT)110 and Optical Network Units (ONUs) 120-1 and 120-2 (which may be collectively referred to as ONUs 120 hereinafter). As shown, communication is enabled between the OLT 110 and the ONUs 120. For example, between the OLT 110 and the ONUs 120, data transmission can be performed in the uplink from the ONUs 120 to the OLT 110, and data transmission can be performed in the downlink from the OLT 110 to the ONUs 120. It should be understood that although two ONUs 120 are shown in fig. 1, any number of ONUs 120 may be included in communication system 100.

As can be seen in fig. 1, the OLT 110 comprises a signal processing device 112, which is for example an NN-based signal processing arrangement. In the communication system 100 described herein, for the case of a plurality of ONUs 120, all ONUs 120 may share the signal processing means at the OLT 110.

As described above, much discussion and research has been conducted on signal quality management for a Passive Optical Network (PON) scale within 20km of a 50G (+) PON. Feed Forward Equalizers (FFE), Decision Feedback Equalizers (DFE), Volterra equalizers, and various types of Machine Learning (ML) have all been considered carefully for signal equalization, while NN has proven to be the most powerful and agile enough. NN-based machine learning is capable of extracting and learning certain features in a particular transmission channel and compensating for them in a supervised manner. In this way, it is possible to restore, at the OLT, a distorted signal received via the ONU-to-OLT uplink to an original signal, and determine the association relationship of the distorted signal and the original signal, thereby improving transmission reliability.

In response to the presently discussed problem of standardization of high speed passive optical networks (g.hsps), it is known that the interaction between various factors can lead to a degradation of signal quality. Such factors include, for example, ambiguous channel responses (also known as non-gaussian and selective spectral responses), insufficient bandwidth, fiber dispersion, and intensity-dependent nonlinear responses of the device and fiber. These problems related to channel loss are regarded as a level of problems in the standardization process of g.hsp.

Another aspect is the inconsistency of the uplink channels from different ONUs to the OLT. For example, signal equalization for a single ONU is only effective for a particular end-to-end transmission link and therefore lacks versatility. Taking fig. 1 as an example, signal equalization for a signal sequence transmitted on an uplink channel from the ONU 120-1 to the OLT may not be applicable to a signal sequence transmitted on an uplink channel from the ONU 120-2 to the OLT, for example because the uplink channel from the ONU 120-1 to the OLT and the uplink channel from the ONU 120-2 to the OLT have different channel characteristics.

Therefore, in order to properly handle the above problems, an ideal OLT equalizer must not only be able to cope with various channel losses, but should also use a common set of equalization configurations to account for the differences between the uplink channels to the OLT from multiple ONUs (typically 64 ONUs) based on a common or common NN equalizer.

In view of this, embodiments of the present disclosure provide a communication method for a passive optical network. Through the embodiment of the disclosure, the OLT can extract the characteristic parameters for each channel from the signals transmitted on the uplink channels between different ONUs and the OLT, so that the NN-based signal processing device can be trained based on the characteristic parameters of the channels, and the signal processing device can be applied to all ONUs in the network. In this way, not only can complex channel impairments be eliminated, but also the commonality problem of distorted signal recovery for different ONUs can be solved.

Fig. 2 shows a flow diagram of a communication method 200 according to an embodiment of the present disclosure. In some embodiments, the method 200 may be implemented by the OLT 110, e.g., may be implemented by the signal processing device 112 of the OLT 110. In other embodiments, the method 200 may also be implemented by a computing device that is separate from the OLT 110. For ease of discussion, the method 200 will be discussed in conjunction with FIG. 1.

A predetermined number of ONUs, for example, may be included in communication system 100, and may include, for example, 32 or 64 ONUs. For any one ONU, the OLT may receive a signal sequence via an uplink channel with that ONU.

At 210, the OLT 110 receives a distorted signal sequence on an uplink channel from the ONUs 120 to the OLT 110. The distorted signal sequence results from the original signal sequence from the ONU 120 being distorted during transmission on the uplink channel.

The causes of distortion in the original signal sequence, as already mentioned above, may include factors such as ambiguous channel response, insufficient bandwidth, fiber dispersion, and nonlinear response related to the strength of the equipment and fiber, for example. The distortions due to these factors will be processed in the OLT 110 to recover the original signal sequence.

At 220, OLT 110 encodes the distorted signal sequence. The encoding operation aims at determining from the distorted signal sequence characteristic parameters characterizing the properties of the uplink channel from the ONUs 120 to the OLT 110. The characteristic parameter may comprise, for example, at least one of a bandwidth of the uplink channel, a center wavelength of the uplink channel, or a length of optical fiber used to carry the uplink channel.

In some embodiments, one or more signal reference parameters may be obtained from the distorted signal sequence, which may be, for example, a voltage magnitude of the distorted signal sequence, a power of the distorted signal sequence, and so on. During the encoding operation on the distorted signal sequence, the OLT 110 may perform statistics on at least one signal reference parameter to generate a histogram. The OLT 110 determines by encoding the histogram as a characteristic parameter for characterizing the properties of the uplink channel from the ONUs 120 to the OLT 110.

In some embodiments, the OLT 110 may perform a discrete fourier transform on the distorted signal sequence to determine, by encoding the transformed distorted signal sequence, a characteristic parameter that characterizes a property of an uplink channel from the ONUs 120 to the OLT 110.

At 230, the OLT 110 determines an association between the distorted signal sequence and the original signal sequence based on the distorted signal sequence and the determined characteristic parameters for recovering the original signal sequence. The term "associative relationship" as used herein may be understood as the association between the input and the output of the neural network for signal processing described above.

In some embodiments, the OLT 110 may determine the degree of distortion experienced by the distorted signal sequence in the transmission of the uplink channel through the characteristic parameters. The distortion degree may change, for example, the weight or offset of a node of a training layer in the neural network, and the training layer can restore the distorted signal sequence to the original signal sequence according to the distortion degree reflected by the characteristic parameter by inputting the distorted signal sequence and the characteristic parameter as inputs to the neural network. The OLT 110 may thus in turn determine the correlation between the distorted signal sequence and the original signal sequence based on the deviation between the distorted signal sequence and the original signal sequence.

In some embodiments, the OLT 110 may also receive another distorted signal sequence on the uplink channel from the ONUs 120 to the OLT 110, which is generated by another original signal sequence being distorted during transmission on another uplink channel different from the original uplink channel. The OLT 110 may update the correlation based on the characteristic parameters characterizing the other uplink channels and the further distorted signal sequence. For example, the characteristic parameters characterizing the other uplink channel are taken as input to the neural network together with the further distorted signal sequence, and the previously determined correlation may be updated in this process, so that the OLT 110 may restore the further distorted signal sequence to the further original signal sequence based on the updated correlation.

The process of determining the characteristic parameter from the distorted signal sequence and determining the correlation based on the characteristic parameter and the distorted signal sequence to restore the distorted signal sequence to the original signal sequence described above can be better understood by the examples shown in fig. 3A to 3C, for example.

Fig. 3A-3C show schematic diagrams of signal processing implemented by certain embodiments of the present disclosure. The principles and processes for recovering distorted signals according to the method illustrated in connection with fig. 2 using a neural network for signal processing are set forth in further detail below in connection with fig. 3A-3C.

Fig. 3A shows an example plot 301 of a distorted signal sequence transmitted by the ONU 120-1 in fig. 3B that is distorted across the uplink channel from the ONU 120-1 to the OLT 110. In fig. 3B, the OLT 110 may convert the optical signal into an electrical signal by its signal processing equipment 112 to quantize the signal. After conversion, the distorted signal sequence may be provided to a signal equalizer 130 to restore the distorted signal sequence to the original signal sequence.

The signal equalizer 130 may be comprised within the OLT 110, e.g. may be integrated in the signal processing device 112. Furthermore, the signal equalizer 130 may also be implemented as a separate component, but controlled by the signal processing device 112.

The signal processing device 112 may sample the received distorted signal sequence. The plurality of sample points may be input as a sample distorted signal sequence to an input layer 331 of a neural network 330 in the signal equalizer 130. The plurality of sampled distorted signal sequences can be realized, for example, by selecting distorted signal sequences at a plurality of time points on the curve in fig. 3A. The time may be, for example, at time t, time t-1, time t-2, time t-3, time t-4, etc. The sampled distorted signal sequences at the respective time points may be input to the corresponding nodes of the input layer 331, respectively.

The signal equalizer 130 may also include an encoder 310. The electrical signal converted by the signal processing device 112, that is, the distorted signal sequence may be input to the encoder 310. In some embodiments, encoder 310 may include memory 320. The memory 320 may, for example, store a plurality of distorted signal sequence bursts (bursts) transmitted to the ONU 120 via an uplink channel from the ONU 120-1 to the OLT 110. The memory 320 may also store, for example, distorted signal sequences transmitted via other ONUs (e.g., ONU 120-2 in fig. 1) other than ONU 120-1.

The encoder 310 may be, for example, an NN-based encoder. The input 311 of the encoder 310 may be a distorted signal sequence, a variant of the distorted signal sequence, or some parameter associated with the distorted signal sequence, while the output is a characteristic parameter characterizing the properties of the uplink channel. The neural network of the encoder 310 can be viewed as an associative relationship between the distorted signal sequence and the properties of the uplink channel. The weights of the various nodes in the training layer 312 of the neural network of the encoder 310 may be preset and fine-tuned based on the inputs and outputs during the training process.

Taking one or more distorted signal sequence bursts from ONU 120-1 as an example, at least one signal reference parameter may be determined from the distorted signal sequence, as already mentioned above. The signal reference parameter may be, for example, the voltage amplitude or the power of the signal sequence. The histogram may be generated by counting at least one signal reference parameter. For example, the vertical axis of the histogram may represent the voltage magnitude or power level. In some embodiments, the histogram is input as an input 311 to the neural network of the encoder 310 to a training layer 312 of the neural network of the encoder 310.

Furthermore, the distorted signal sequence may be fourier transformed and the transformed distorted signal sequence may be used as an input 311 to the neural network of the encoder 310. Thus, in some embodiments, the transformed distorted signal sequence is input to the training layer 312 of the neural network of the encoder 310.

Thereby, a characteristic parameter or a characteristic code characterizing the properties of the uplink channel may be obtained at the output layer 332 of the neural network of the encoder 310. The output layer 332 of the neural network of the encoder 310 may be viewed as the input layer 332 of the neural network 330 in the signal equalizer 130.

Thus, the neural network 330 receives inputs from the input layers 331 and 332, which are samples of the distorted signal sequence and characteristic parameters characterizing the properties of the uplink channel derived based on the distorted signal sequence, respectively. These inputs are fed to the training layer 334 of the neural network 330. Although the neural network 330 in fig. 3B has only one training layer 334. It should be understood that the neural network 330 may also include other suitable numbers of training layers.

The training layer 334 in the neural network 330 can train the existing training model through the above inputs, i.e., the samples of the distorted signal sequence and the feature parameters. Existing training models may include, for example, historical training parameter samples from other ONUs. Tens or hundreds of sample sequences may be required for one ONU. As there are typically 32 or 64 ONUs in a passive optical network. To train a generic neural network, the database will select a complete data set that covers all existing ONU data.

During the training process, the neural network 330 may call historical training parameter samples in the database. Next, the neural network 330 retrains the neural network model in combination with the historical training parameter samples, the samples of the distorted signal sequence sent by ONU 120-1, and the characteristic parameters. The training process may be understood as using the characteristic parameters to adjust the weights of the nodes in the neural network model, and weighting the samples of the distorted signal sequence, so as to converge the samples of the distorted signal sequence into the samples of the original signal sequence, thereby obtaining the original signal sequence. The process of adjusting the weights of the nodes in the neural network model using the characteristic parameters can be understood as calibrating the correlation between the distorted signal sequence and the original signal sequence.

The original signal sequence may be output through the output layer 335 of the neural network 330. The recovered original signal sequence may be represented by curve 302 in fig. 3C. The neural network model, which is retrieved via the training parameters from ONU 120-1, will be used as the updated neural network model.

Similarly, the neural network 330 can also be trained according to distorted signal sequences from different uplink channels from other ONUs to the OLT and characteristic parameters characterizing the corresponding uplink channels to obtain a signal equalizer applicable to all ONUs or all uplink channels.

For distorted signal sequences from the same uplink channel, the characteristic parameters characterizing the uplink channel may be stored in the memory 320 once the characteristic parameters have been determined. When the distorted signal sequence is received again from the uplink channel, the characteristic parameters are no longer trained but are recalled directly from the memory 320.

Here, the NN-based encoder may be viewed as a sub-neural network of the neural network 330 in the signal equalizer 130. The encoder 320 is trained by using a histogram based on statistical features or a transformed distorted signal sequence, and the characteristic value or characteristic value group of the channel-specific non-explicit expression is extracted, and the OLT is guided to distinguish and provide customized equalization for different ONUs through the neural network 330 by using the characteristic value or characteristic value group. Since the parameters of the encoder 320 are globally trained, after the neural network 330 is trained, when the OLT distinguishes the ONUs and provides the customized equalization, the parameter configuration of the encoder 320 does not need to be changed, and only the statistical features of the corresponding ONUs need to be input into the encoder 320 or pre-trained feature values are directly provided to the subsequent neural network 330. Therefore, the equalization method and apparatus for providing ONU identification by the encoder 320 and providing a general equalization service by the neural network 330 can enable the OLT to provide an optimal signal equalization service for different ONUs without modifying NN parameters.

Hereinafter, an example of implementing signal equalization based on pre-trained characteristic parameters for characterizing a distorted signal sequence will be described with reference to fig. 4 and 5.

Fig. 4 shows a schematic diagram of implementing signal processing according to some embodiments of the present disclosure. As shown in fig. 4, upon receiving the distorted signal sequence from the ONU 120-1, the sampled distorted signal sequence and the feature parameters directly called from the memory 320 can be input to the training layer 334 of the neural network 330 via the input layers 331 and 332, respectively, so that the distorted signal sequence is restored to the original signal sequence and output through the output layer 335. The signal equalization process in the neural network 330 is similar to that described in conjunction with fig. 3B, and therefore is not described in detail here.

Fig. 5 shows a schematic diagram of implementing signal processing according to some embodiments of the present disclosure. Fig. 5 shows an example of receiving distorted signal sequences from a plurality of different ONUs. Likewise, once all ONUs in the communication system 100 have sent signals to the OLT 110, characteristic parameters characterizing the properties of each uplink channel from these ONUs 120 to the OLT 110 may be stored in the memory 320 of the encoder 310 of the OLT 110. For one or more bursts of distorted signal sequences from multiple ONUs, such as ONU 120-1 and ONU 120-2 shown in fig. 5, memory 320 may invoke corresponding feature codes or feature parameters for recovery of the distorted signal sequence.

In this way, not only can signal distortion caused by complex channel damage be eliminated, but also a universal signal equalizer can be provided to solve the problem of universality of distorted signal recovery for different ONUs.

Fig. 6 shows a schematic of exemplary experimental results obtained using embodiments according to the present disclosure. A simulated environment was established to test the quantization gain of embodiments of the present disclosure. The configuration of the simulated environment is shown in the table below.

Table 1: configuration of a simulation environment

As shown in fig. 6, the original signal 603 obtained according to the scheme of the present disclosure has a lower error rate than the original signal recovered by the other two equalization methods (see 601 and 602).

A characteristic encoder nested NN equalizer (EN-NNE) for an uplink channel of a PON is presented in the present disclosure. With the embedded feature encoder sub-neural network, EN-NNE may use the statistical data for deep feature extraction and encode it to help the neural network of the signal equalizer perform signal equalization. This solution can be used in the OLT of the uplink channel of future 50G + PONs to equalize the signal quality degradation of all ONUs with different responses in a unified way.

Fig. 7 is a simplified block diagram of a device 700 suitable for implementing embodiments of the present disclosure. The device 700 may be provided to implement communication devices such as the OLT 110, the ONU 120-1, and the ONU 120-2 as shown in fig. 1. As shown, device 700 includes one or more processors 710, one or more memories 740 coupled to processors 710, and one or more transmitters and/or receivers (TX/RX)740 coupled to processors 710.

TX/RX 740 is used for bi-directional communication. TX/RX 740 has at least one antenna to facilitate communication. A communication interface may represent any interface necessary to communicate with other network elements.

The processor 710 may be of any type suitable to a local technology network, and may include, but is not limited to, one or more of general purpose computers, special purpose computers, microcontrollers, digital signal controllers (DSPs), and controller-based multi-core controller architectures. The device 700 may have multiple processors, such as application specific integrated circuit chips, that are time dependent from a clock synchronized with the main processor.

Memory 720 may include one or more non-volatile memories and one or more volatile memories. Examples of non-volatile memory include, but are not limited to, Read Only Memory (ROM)724, Erasable Programmable Read Only Memory (EPROM), flash memory, a hard disk, a Compact Disc (CD), a Digital Video Disc (DVD), and other magnetic and/or optical storage. Examples of volatile memory include, but are not limited to, Random Access Memory (RAM)722 and other volatile memory that does not persist for the duration of the power down.

The computer programs 730 include computer-executable instructions that are executed by the associated processor 710. The program 730 may be stored in the ROM 720. Processor 710 may perform any suitable actions and processes by loading programs 730 into RAM 720.

Embodiments of the present disclosure may be implemented by way of the program 730 such that the device 700 may perform any of the processes of the present disclosure as discussed with reference to fig. 2-5. Embodiments of the present disclosure may also be implemented by hardware or by a combination of software and hardware.

In some embodiments, the program 730 can be tangibly embodied in a computer-readable medium, which can be included in the device 700 (such as in the memory 720) or other storage device accessible by the device 700. The program 730 may be loaded from a computer-readable medium into the RAM 722 for execution. The computer readable medium may include any type of tangible, non-volatile memory, such as ROM, EPROM, flash memory, a hard disk, a CD, a DVD, etc. Fig. 8 shows an example of a computer readable medium 800 in the form of a CD or DVD. The program 730 is stored on a computer readable medium.

In general, the various embodiments of the disclosure may be implemented in hardware or special purpose circuits, software, logic or any combination thereof. Some aspects may be implemented in hardware, while other aspects may be implemented in firmware or software, which may be executed by a controller, microprocessor or other computing device. While various aspects of the embodiments of the disclosure are illustrated and described as block diagrams, flow charts, or using some other pictorial representation, it is well understood that the blocks, apparatus, systems, techniques or methods described herein may be implemented in, as non-limiting examples, hardware, software, firmware, special purpose circuits or logic, general purpose hardware or controller or other computing devices, or some combination thereof.

The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product comprises computer executable instructions, such as instructions included in program modules, that are executed in the device on the target's real or virtual processor to perform the method 200 as described above with reference to fig. 2. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform particular tasks or implement particular abstract data types. In various embodiments, the functionality of the program modules may be combined or split between program modules as desired. Machine-executable instructions for program modules may be executed within local or distributed devices. In a distributed facility, program modules may be located in both local and remote memory storage media.

Computer program code for implementing the methods of the present disclosure may be written in one or more programming languages. These computer program codes may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the program codes, when executed by the computer or other programmable data processing apparatus, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be performed. The program code may execute entirely on the computer, partly on the computer, as a stand-alone software package, partly on the computer and partly on a remote computer or entirely on the remote computer or server.

In the context of the present disclosure, computer program code or related data may be carried by any suitable carrier to enable a device, apparatus or processor to perform various processes and operations described above. Examples of a carrier include a signal, computer readable medium, and the like. Examples of signals may include electrical, optical, radio, acoustic, or other forms of propagated signals, such as carrier waves, infrared signals, and the like.

The computer readable medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination thereof. More detailed examples of a computer-readable storage medium include an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical storage device, a magnetic storage device, or any suitable combination thereof.

Further, while the operations of the methods of the present disclosure are depicted in the drawings in a particular order, this does not require or imply that these operations must be performed in this particular order, or that all of the illustrated operations must be performed, to achieve desirable results. Rather, the steps depicted in the flowcharts may change the order of execution. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions. It should also be noted that the features and functions of two or more devices according to the present disclosure may be embodied in one device. Conversely, the features and functions of one apparatus described above may be further divided into embodiments by a plurality of apparatuses.

While the present disclosure has been described with reference to several particular embodiments, it is to be understood that the disclosure is not limited to the particular embodiments disclosed. The disclosure is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.