CN108737297B

CN108737297B - Method and apparatus for estimating received signal sequence

Info

Publication number: CN108737297B
Application number: CN201710272669.XA
Authority: CN
Inventors: 叶晨晖; 张东旭; 张凯宾
Original assignee: Nokia Shanghai Bell Co Ltd
Current assignee: Nokia Shanghai Bell Co Ltd
Priority date: 2017-04-24
Filing date: 2017-04-24
Publication date: 2021-02-12
Anticipated expiration: 2037-04-24
Also published as: WO2018197953A1; CN108737297A

Abstract

The present disclosure proposes a method and apparatus for estimating a received signal sequence. The method is implemented at a receiver configured to receive a sequence of signals over a bandwidth-limited channel. The method comprises the steps of carrying out bipolar transformation on a training sequence known by a receiver to obtain a reference sequence; obtaining a distorted training sequence after the training sequence is transmitted through a channel and subjected to bipolar transformation; determining a channel coefficient sequence of a channel based on a reference sequence and a distorted training sequence; obtaining a distortion signal sequence of a signal sequence after channel transmission and bipolar transformation; and determining an estimate of the signal sequence based on the channel coefficient sequence and the distorted signal sequence.

Description

Method and apparatus for estimating received signal sequence

Technical Field

The present disclosure relates generally to the field of communications, and more particularly, to a method and apparatus for estimating a received signal sequence.

Background

Generally, through good design of the receiver bandwidth, unipolar signals can be converted to bipolar signals for transmission. Due to such signal transformation or shaping mechanisms, it is possible to carry higher rate (e.g., 25Gb/s and even 40Gb/s) data signals over conventional narrow bandwidth components (e.g., 10G bandwidth optical transmitters and receivers) to accommodate future access systems. However, since the narrow bandwidth component will produce a low pass filtering effect on the high rate signal, this results in a partial response of the signal, and thus a severely distorted signal sequence will be received at the receiver. This will degrade the performance of the communication system and even cause communication failures.

Disclosure of Invention

Embodiments of the present disclosure provide a method and apparatus implemented at a receiver, a computer program product, and a receiver.

In a first aspect of the disclosure, a method implemented at a receiver is provided. The receiver is configured to receive a signal sequence over a bandwidth-limited channel. The method comprises the following steps: carrying out bipolar transformation on a training sequence known by a receiver to obtain a reference sequence; obtaining a distorted training sequence after the training sequence is transmitted through a channel and subjected to bipolar transformation; determining a channel coefficient sequence of a channel based on a reference sequence and a distorted training sequence; obtaining a distortion signal sequence of a signal sequence after channel transmission and bipolar transformation; and determining an estimate of the signal sequence based on the channel coefficient sequence and the distorted signal sequence.

In some embodiments, performing a bipolar transformation on the training sequence to obtain the reference sequence comprises: transforming the binary sequence into a duobinary sequence; or a four-order pulse amplitude modulation (PAM4) sequence into a bipolar PAM4 sequence.

In some embodiments, determining the sequence of channel coefficients for the channel comprises performing at least once the following: obtaining an intermediate training sequence based on convolution of the distorted training sequence and a current channel coefficient sequence of the channel; comparing the difference between the intermediate training sequence and the reference sequence to a threshold; and adjusting the current channel coefficient sequence in response to determining that the difference is greater than the threshold.

In some embodiments, determining the estimate of the signal sequence comprises: an estimate of the signal sequence is determined using a maximum likelihood sequence estimation algorithm.

In some embodiments, determining an estimate of the signal sequence using a maximum likelihood sequence estimation algorithm comprises: the channel coefficient sequence and the distorted signal sequence are used as inputs to a maximum likelihood sequence estimation algorithm to determine an estimate of the signal sequence.

In some embodiments, determining an estimate of the signal sequence using a maximum likelihood sequence estimation algorithm comprises: performing distortion compensation on the distorted signal sequence based on the channel coefficient sequence; determining an undistorted channel coefficient sequence of an undistorted bipolar channel; and determining an estimate of the signal sequence using the distortion-compensated distorted signal sequence and the undistorted channel coefficient sequence as inputs to a maximum likelihood sequence estimation algorithm.

In a second aspect of the disclosure, a communication device is provided. The communication device includes a receiver configured to receive a sequence of signals over a bandwidth-limited channel. The communication device includes at least one processor and at least one memory including computer-executable instructions. The at least one memory and the computer-executable instructions are configured to, with the at least one processor, cause the communication device to: carrying out bipolar transformation on a training sequence known by a receiver to obtain a reference sequence; obtaining a distorted training sequence after the training sequence is transmitted through a channel and subjected to bipolar transformation; determining a channel coefficient sequence of a channel based on a reference sequence and a distorted training sequence; obtaining a distortion signal sequence of a signal sequence after channel transmission and bipolar transformation; and determining an estimate of the signal sequence based on the channel coefficient sequence and the distorted signal sequence.

In a third aspect of the disclosure, a computer program product is provided. The computer program product is tangibly stored on a non-volatile computer-executable medium and includes machine-executable instructions. The machine executable instructions, when executed, cause a machine to perform the steps of the method according to the first aspect.

In a fourth aspect of the disclosure, a receiver for use in a communication device is provided. The receiver comprises a bipolar transformation module, a channel estimation module and a signal estimation module. The bipolar transformation module is configured to bipolar transform a training sequence known to the receiver to obtain a reference sequence. The channel estimation module is configured to obtain a distorted training sequence after the training sequence is transmitted through the channel and subjected to bipolar transformation, and determine a channel coefficient sequence of the channel based on the reference sequence and the distorted training sequence. The signal estimation module is configured to obtain a distorted signal sequence after transmission and bipolar transformation of the signal sequence through a channel, and determine an estimate of the signal sequence based on the channel coefficient sequence and the distorted signal sequence.

Drawings

The above and other objects, features and advantages of the embodiments of the present disclosure will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present disclosure are illustrated by way of example, and not by way of limitation, in the figures of the accompanying drawings and in which:

fig. 1 illustrates an exemplary scenario for transmitting a high-rate signal using a bandwidth-limited channel.

Fig. 2 shows an example configuration of a conventional signal estimation module.

Fig. 3 shows different eye diagrams for different channel responses as simulated.

Fig. 4 shows a schematic block diagram of a receiving architecture according to an embodiment of the present disclosure.

Fig. 5 shows a schematic diagram of a spectral transformation according to an embodiment of the present disclosure.

Fig. 6 shows a flow diagram of a method implemented at a receiver in accordance with an embodiment of the disclosure.

Fig. 7 illustrates one embodiment of determining an estimate of a signal sequence in accordance with an embodiment of the disclosure.

Fig. 8 illustrates another embodiment of determining an estimate of a signal sequence according to an embodiment of the disclosure.

FIG. 9 illustrates a block diagram of a device suitable for implementing embodiments of the present disclosure.

FIG. 10 shows a graph comparing performance of simulations according to an embodiment of the present disclosure with conventional schemes.

FIG. 11 shows another performance comparison graph of a simulation according to an embodiment of the present disclosure and a conventional scheme.

Fig. 12 shows a graph of an experimental test according to an embodiment of the present disclosure.

Fig. 13 shows a graph of another experimental test according to an embodiment of the present disclosure.

Throughout the drawings, the same or similar reference numbers are used to refer to the same or similar elements.

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.

In describing embodiments of the present disclosure, the terms "include" and its derivatives should be interpreted as being inclusive, 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.

The term "terminal equipment" or "user equipment" (UE) as used herein refers to any terminal equipment capable of wireless communication with a base station or with each other. As an example, the terminal device may include a Mobile Terminal (MT), a Subscriber Station (SS), a Portable Subscriber Station (PSS), a Mobile Station (MS), or an Access Terminal (AT), and the above-described devices in a vehicle. The terminal device may be any type of mobile terminal, fixed terminal, or portable terminal including a mobile handset, station, unit, device, multimedia computer, multimedia tablet, internet node, communicator, desktop computer, laptop computer, notebook computer, netbook computer, tablet computer, Personal Communication System (PCS) device, personal navigation device, Personal Digital Assistant (PDA), audio/video player, digital camera/camcorder, positioning device, television receiver, radio broadcast receiver, electronic book device, game device, smart meter, or other smart appliance that may be used for MTC communication, or any combination of the above. In the context of the present disclosure, the terms "terminal device" and "user equipment" may be used interchangeably for purposes of discussion convenience.

The term "network device" as used herein refers to a base station or other entity or node having a particular function in a communication network. A "base station" (BS) may represent a node B (NodeB or NB), an evolved node B (eNodeB or eNB), a Remote Radio Unit (RRU), a Radio Head (RH), a Remote Radio Head (RRH), a relay, or a low power node such as a pico base station, a femto base station, or the like. The coverage area of a base station, i.e. the geographical area where it is able to provide service, is called a cell. In the context of the present disclosure, the terms "network device" and "base station" may be used interchangeably for purposes of discussion convenience, and may primarily be referred to as an eNB as an example of a network device.

Fig. 1 illustrates an exemplary scenario 100 for transmitting a high-rate signal using a bandwidth-limited channel. As shown in fig. 1, an architecture 190 for transmission/detection by transforming a conventional unipolar signal sequence (e.g., an on-off keying signal OOK or a pulse amplitude modulation PAM4 signal) into a bipolar signal sequence (e.g., a duobinary signal or a bipolar PAM4 signal) is shown that uses a 10G bandwidth transmitting device 130 and receiving device 150 to transmit 25 Gb/s/wavelength signals.

As shown, the unipolar signal sequences (e.g., OOK signals) generated by the first through fourth processing boards 110-113 may be transmitted to the dual-polarity transmit receive architecture 190 via the 25G interface 120-123, which is represented by the dashed box. The dual-polarity transceiver architecture 190 may include four 10G transmitters 130-133 that convert the signals from the first processing board to the fourth processing board 110-113 into 10G bandwidth signals for transmission to the 10G bandwidth receiver 150 via the optical link 140.

Thereafter, the signal may be physically decoded and error corrected by the direct detection module 160, the signal estimation module 161, and the forward error correction module 162, thereby recovering a 25G bandwidth unipolar signal. The 25G bandwidth unipolar signal is then transmitted through the 25G interface 170 to the fifth processing board 180 at the receiving end for further signal processing.

It follows that in the conventional scheme shown in fig. 1, a single polarity signal emitted from the transmitter 130 and 133 (e.g. in the optical network unit ONU) will be converted into a dual polarity signal in the receiver 150 (e.g. in the optical line termination OLT), resulting in a poor bit error rate BER in the direct detection module 160. To compensate for the partial response caused by the low pass filtering effect of narrow bandwidth (e.g., 10G) components on high rate signals (e.g., 25G), advanced intelligent algorithms can be used to improve the accuracy of the estimates made of the received signal sequence with severe distortion. For example, the signal estimation module 161 may use a Maximum Likelihood Sequence Estimation (MLSE) algorithm to improve the correctness of the signal sequence estimation.

It should be noted that only components or modules relevant to embodiments of the present disclosure are shown in fig. 1 for simplicity, and in other embodiments, the exemplary scenario 100 may include more or fewer components or modules. Further, although a particular number of components or modules are shown in FIG. 1, in other embodiments, a greater or lesser number of these modules or components may be present.

Fig. 2 shows an example configuration of a conventional signal estimation module 161. As shown in fig. 2, the signal estimation module 161 may include an estimator initialization module 210 and an estimator module 220. The estimator initialization module 210 may provide a preset function 230 to set a preset condition for the estimator module 220 to perform signal sequence estimation.

As shown, the estimator initialization module 210 may receive one or more pre-inputs. A first pre-input 211, a second pre-input 212 and a third pre-input 213 are exemplarily depicted in fig. 2. In general, the first pre-input 211 may be a bipolar channel coefficient sequence, which is set to an ideal value (e.g., [1,1]) in the conventional signal estimation block 161. The second pre-input 212 may be a constellation of unipolar sequences (e.g., OOK or PAM4 sequences), which is set to [01] in the example of fig. 2. The third pre-input 213 may be a backtracking depth, which is set to 20 in the example of fig. 2. Of course, the estimator initialization module 210 may receive more or less pre-inputs as well. Based on these pre-inputs, the estimator initialization module 210 enables initialization of the estimator 220.

In one aspect, the estimator 220 is initialized by the estimator initialization module 210 based on the first pre-input 211, the second pre-input 212, and the third pre-input 213 described above. Estimator 220, on the other hand, receives distorted signal sequence 214. Based on the initialization by the estimator initialization module 210 and the distorted signal sequence 214, the estimator 220 may determine an estimate 215 of the signal sequence, e.g., by an estimation algorithm. In many cases, the estimator 220 may employ an MLSE algorithm to determine the estimate 215 of the signal sequence.

It should be noted that fig. 2 only shows components or modules related to the embodiments of the present disclosure for simplicity. In other embodiments, the signal estimation module 161 may include more or fewer components or modules. Further, although a particular number of components or modules are shown in FIG. 2, in other embodiments, a greater or lesser number of these modules or components may be present. Again, although a specific sequence of values for the various inputs and outputs is shown in fig. 2, those skilled in the art will appreciate that these sequence of values are merely illustrative and that in other embodiments may have different other sequences of values.

In case the signal estimation module 161 employs the MLSE algorithm, the accuracy of the MLSE based bipolar signal sequence estimation is very dependent on the accuracy of the channel coefficients. In the most ideal case, the channel bandwidth is adapted to transform a unipolar signal into a bipolar signal without distortion, such channel coefficients being for example [1,1 ]. Thus, in the conventional signal estimation block 161, the first pre-input 211, e.g., at the estimator initialization block 210, pre-sets the channel coefficients to values that are ideal (e.g., [1,1 ]). However, it is clear that this does not accurately represent the true channel.

In fact, for example in a passive optical network PON system, the real channel conditions of many optical network units ONU may be very different from each other. In order to visually represent the impact of different real channels on the receiving system. Fig. 3 shows different eye diagrams in case of simulated responses of different channels (e.g. ONU to OLT channels).

In fig. 3, the abscissa represents normalized frequency and the ordinate represents magnitude. Reference numeral 310-. Reference numerals 311-. During the simulation, the preset channel coefficients in the signal estimation block 161 are kept constant at [1,1], while the channel response is varied from 310 to 330, and the BER is monitored.

As can be seen from fig. 3, first the

different channel responses

310 and 330 give different duobinary waveforms in a direct detection manner. Second, in the case of inaccurate channel coefficients that do not reflect the true channel, the estimation algorithm (e.g., MLSE) will not guarantee that it can help improve the accuracy of the signal sequence estimation, or it will result in worse detection results. In other words, the estimation algorithm (e.g., MLSE) cannot provide stable efficiency due to inaccurate channel coefficient estimation.

To address at least in part the above and other potential problems, embodiments of the present disclosure propose a method and apparatus implemented at a receiver to improve an estimation algorithm (e.g., MLSE) and make the estimation algorithm more efficient and powerful. With the embodiments of the present disclosure, not only a duobinary format signal, but even a bipolar PAM4 signal can be implemented in future PONs, thereby doubling the capacity per wavelength. Thus, 50Gb/s signal transmission can be achieved by using 10G class devices.

As discussed above, in a transmission system performing a bipolar transformation, the unsatisfactory recovery of a unipolar signal (e.g., a PAM4 signal) by the receiving end is mainly due to the very inaccurate channel estimation of the bipolar channel. Therefore, in embodiments of the present disclosure, a bipolar channel estimation model is introduced and assigned with accurate channel coefficients prior to using an estimation algorithm (e.g., MLSE) to help improve the estimation correctness of the final unipolar signal (e.g., PAM4 signal).

In some embodiments, a "mid-stage" is introduced to produce a standard bipolar signal (e.g., a bipolar PAM4 signal) to increase the accuracy of the final unipolar signal (e.g., a PAM4 signal) recovery and avoid noise enhancement effects. The basic idea of an embodiment of the present disclosure is described below in conjunction with fig. 4 and 5.

Fig. 4 shows a schematic block diagram 400 of a receiving architecture according to an embodiment of the present disclosure. As shown in fig. 4, one or more

10G bandwidth transmitters

410 and 411 may transmit high rate (e.g., 50Gb/s) signal sequences (e.g., OOK or PAM4 sequences) 401 onto an optical link 420.

At receiver 430, a direct detector 431 of 10G bandwidth may perform direct detection of the received signal. The detected signal may then be analog-to-digital converted by the analog-to-digital conversion module ADC 432. The resulting digital signal sequence is then passed through a channel estimation assisted signal estimation module 440 according to an embodiment of the present invention to determine an estimate 215 of the signal sequence. Thereafter, the determined estimate 215 of the signal sequence is input to a forward error correction module 436 to recover the transmitted signal sequence 401 at a high rate (e.g., 50 Gb/s).

According to an embodiment of the present disclosure, the channel estimation assisted signal estimation module 440 may include a channel estimation module 433, a bipolar transform module 434, and a signal estimation module 435. The "intermediate stage" mentioned above may include, for example, the channel estimation module 433 and the bipolar transform module 434 of these modules. Which operate in conjunction with the signal estimation module 435 such that improved estimation of the received signal sequence may be achieved.

In particular, the channel estimation module 433 may estimate channel coefficients of the bipolar channel to learn and fill in the "gap" between the true bipolar channel and the standard bipolar channel, rather than using the invariant default channel coefficients as in conventional schemes. The bipolar transform module 434 may provide a bipolar transformed training sequence for channel coefficient learning instead of using a direct unipolar training sequence as in conventional estimation schemes, thereby avoiding noise enhancement effects. Based on these functions provided by the "intermediate stage," the signal estimation module 435 may use the learned channel coefficients to determine an estimate of the transmitted unipolar signal sequence (e.g., OOK or PAM4 sequence).

It can be seen that in the embodiment shown in fig. 4, the transmitter 410-411 may transmit a single polarity signal sequence, such as an OOK or PAM4 signal. At the receiver 430, prior to signal sequence estimation, channel estimation will first be performed by the channel estimation module 433 and fed to the signal estimation module 435 (e.g., as a channel coefficient pre-input for MLSE) to replace the default values (e.g., [1,1]) in the conventional scheme. Based on the learned updated channel model, the distorted bipolar signal will be compensated into a compensated bipolar signal according to a standard bipolar reference signal sequence (e.g., PAM4 reference signal). The signal estimation module 435 will then determine the most likely estimated sequence 215 of unipolar signals (e.g., PAM4 or OOK) from the fed-in channel coefficients.

Thus, the channel estimation assisted signal estimation module 440 of embodiments of the present disclosure may intelligently learn specific channel coefficients for each individual channel (e.g., ONU), and may also adaptively update the channel coefficients in the event of thermal, aging, or other fluctuations. Furthermore, it is transparent to different signal formats (e.g., OOK or PAM 4). The general principles of embodiments of the present disclosure are described below in conjunction with the spectral transform diagram of fig. 5.

Fig. 5 shows a schematic diagram 500 of a spectral transformation according to an embodiment of the present disclosure. As shown in fig. 5, spectrum 510 represents the spectrum of the original signal prior to transmission. Spectrum 520 represents a bandwidth spectrum of a narrow bandwidth transmission device that has insufficient bandwidth compared to the rate or baud rate of the signal. Spectrum 530 represents the spectrum of a signal received by a narrow bandwidth transmission device, which shows the spectrum of the signal severely distorted after the signal has passed through the narrow bandwidth transmission device.

Spectrum 540 also represents the spectrum of a signal received by a narrow bandwidth transmission device and spectrum 541 represents the portion of the supplemental spectrum needed to recover the original signal spectrum on the basis of spectrum 540. Spectrum 550 also represents the spectrum of the signal received by the narrow bandwidth transmission device and spectrum 551 represents the spectrum of the signal after compensation by the "mid-stage" of the standard bipolar signal. Spectrum 560 represents the signal spectrum recovered after spectrum 551 has undergone MLSE estimation.

It follows that by such an arrangement of "intermediate stages" according to embodiments of the present disclosure, the performance of the estimation algorithm (e.g., MLSE estimation) of the later stages can be maximized while minimizing the impact of noise. It should be noted that although specific frequency values are indicated in fig. 5, these frequency values are merely illustrative and may have different frequency values in other embodiments. Embodiments of the present disclosure are not limited by these specific frequency values.

A method implemented at receiver 430 according to an embodiment of the disclosure is described below with reference to fig. 6 and in conjunction with fig. 7. Fig. 6 shows a flow diagram of a method 600 implemented at receiver 430, in accordance with an embodiment of the disclosure. Fig. 7 illustrates one embodiment of signal sequence estimation in accordance with an embodiment of the disclosure.

As shown in fig. 7, the channel estimation assisted signal estimation module 440 in the embodiment of fig. 7 is similar in structure to the signal estimation module 161 depicted in fig. 2. Unlike fig. 2, the estimator initialization module 210 and the estimator module 220 are modified in accordance with the concepts of embodiments of the present disclosure. In the embodiment shown in fig. 7, the various functions of the "intermediate stage" mentioned above are implemented collectively by the estimator initialization module 210. It will be appreciated that in other embodiments, these functions of the "intermediate stage" may be performed by different modules or components in the channel estimation assisted signal estimation module 440 (and even the receiver 430).

It should be noted that fig. 7 only shows components or modules relevant to embodiments of the present disclosure for simplicity, and in other embodiments, the channel estimation assisted signal estimation module 440 may include more or fewer components or modules. Further, although a particular number of components or modules are shown in FIG. 7, in other embodiments, a greater or lesser number of these modules or components may be present. Again, although a specific sequence of values for the various inputs and outputs is shown in FIG. 7, those skilled in the art will appreciate that these sequence of values are merely illustrative and that in other embodiments may have different other sequences of values.

Referring back to fig. 6, in some embodiments, the method 600 in fig. 6 may be performed by the estimator initialization module 210 and the estimator module 220 of the channel estimation assisted signal estimation module 440 implemented at the receiver 430. In other embodiments, method 600 may also be performed by any suitable module or components in receiver 430.

At block 610, the estimator initialization module 210 performs a bipolar transform on the training sequence 701 known to the receiver 430 to obtain the reference sequence 702. In some embodiments, such bipolar transformations may include transforming binary sequences (e.g., OOK sequences) into duobinary sequences (e.g., DB sequences), or transforming fourth order pulse amplitude modulation (PAM4) sequences into bipolar PAM4 sequences.

At block 615, the estimator initialization module 210 obtains the distorted training sequence 703 after transmission of the training sequence 701 through the channel and the bipolar transformation. Specifically, in some embodiments, the

transmitters

410, 411 in fig. 4 may transmit a known training sequence 701 over the optical link 420, so that the estimator initialization module 210 may receive the distorted training sequence 703 after transmission through the channel and the bipolar transform at the receiver 430. In other embodiments, training sequence 701 may be transmitted onto a channel by other modules or components for transmission to receiver 430.

At block 620, the estimator initialization module 210 determines a channel coefficient sequence 704 for the channel based on the reference sequence 702 and the distorted training sequence 703. In some embodiments, the estimator initialization module 210 may learn the channel coefficient sequence 704 for the channel through an iterative process.

Specifically, as shown in fig. 7, the estimator initialization module 210 may compare the distorted training sequence 703 with the current channel coefficient sequence (a) after processing by the delay module 710₀，a₁，a₂… …) are multiplied via multiplier 720 and summed by summer 730 to obtain an intermediate training sequence (not shown). The intermediate training sequence is then compared to the reference sequence 702 in the error function block 740. Specifically, the error function module 740 compares the difference between the intermediate training sequence and the reference sequence 702 to a threshold. It should be understood that the threshold values herein may be preset depending on the implementation environment and/or design goals.

Further, the determining module 750 can determine whether the difference between the midamble and the reference sequence 702 is less than a threshold. The decision module 750 may determine the current channel coefficient sequence (a) if the difference is less than or equal to the threshold value, indicating convergence of the learning process for the channel coefficients₀，a₁，a₂… …) as channel coefficientsThe sequence 704 is output to the estimator module 220.

If the difference is greater than the threshold, it indicates that the learning process for the channel coefficients has not converged. In this case, the estimator initialization module 210 may adjust the current channel coefficient sequence (a)₀，a₁，a₂… …). This adjustment may be made, for example, by the formula listed in fig. 7, where μ represents a preset constant value, e represents a preset threshold value, and i represents the number of iterations. It should be noted that the estimator initialization module 210 adjusts the current channel coefficient sequence (a) by the specific formula₀，a₁，a₂… …) is just one specific example. Those skilled in the art will appreciate that in other embodiments, the current channel coefficient sequence (a)₀，a₁，a₂… …) may be adjusted in any suitable manner to cause the channel coefficient learning process to converge.

Referring back to fig. 6, at block 625, the estimator module 220 obtains the distorted signal sequence 214 after transmission and bipolar transformation of the signal sequence 401 through the channel. In particular, the

transmitters

410, 411 in fig. 4 may transmit the signal sequence 401 over the optical link 420, so that the estimator module 220 may receive the distorted signal sequence 214 after transmission through the channel and the bipolar transformation in the receiver 430.

At block 630, estimator module 220 determines an estimate 215 of the signal sequence based on channel coefficient sequence 704 and distorted signal sequence 214. For example, where a MLSE algorithm is used, the estimator module 220 may determine the estimate 215 of the signal sequence using the channel coefficient sequence 704 and the distorted signal sequence 214 as inputs to the MLSE algorithm.

It should be appreciated that the estimator module 220 may also use other ways to determine the estimate 215 of the signal sequence. Another embodiment of the estimator module 220 determining the estimate 215 of the signal sequence is described below in conjunction with fig. 8. Most of the modules or components in fig. 8 are the same as or similar to those in fig. 7, and thus their description will not be repeated.

As shown, fig. 8 differs from fig. 7 mainly in that the coefficient sequence (a) at the current channel₀，a₁，a₂… …) is determined to converge by the decision block 750, the estimator initialization block 210 does not provide the estimator block 220 with the current channel coefficient sequence (a) through the decision block 750₀，a₁，a₂… …) as a sequence of channel coefficients 704, but within the estimator initialization module 210 by basing the sequence 704 on the channel coefficients, i.e. the current sequence of channel coefficients (a)₀，a₁，a₂… …) to perform distortion compensation on the distorted signal sequence 214, and then the distortion compensated distorted signal sequence 801 is input to the estimator module 220.

To this end, unlike in fig. 7, the distorted signal sequence 214 is input to the estimator initialization module 210. In addition, a fourth preset input 802 is added to the estimator module 220, which is the default bipolar channel coefficient. In other words, in the embodiment illustrated in FIG. 8, the estimator module 220 may operate similar to a conventional estimator. This is because the input to the estimator module 220 is already the compensated signal sequence 801 based on the learned channel coefficients 704.

Thus, where the estimator module 220 uses the MLSE algorithm, the estimator module 220 may determine the undistorted channel coefficient sequence 802 for the undistorted bipolar channel (e.g., as a fourth pre-input) and determine the estimate 215 of the signal sequence with the distortion-compensated distorted signal sequence 801 and the undistorted channel coefficient sequence 802 as inputs to the MLSE algorithm.

Fig. 9 illustrates a block diagram of a device 900 suitable for implementing embodiments of the present disclosure. In some embodiments, device 900 may be used to implement method 600 implemented at receiver 430 according to the present disclosure. In other embodiments, the apparatus 900 may be used to implement the channel estimation assisted signal estimation module 440 shown in fig. 4 or fig. 7, or a portion thereof.

As shown in fig. 9, the device 900 includes a controller 910. The controller 910 controls the operation and functions of the device 900. For example, in certain embodiments, the controller 910 may perform various operations by way of instructions 930 stored in a memory 920 coupled thereto. The memory 920 may be of any suitable type suitable to the local technical environment and may be implemented using any suitable data storage technology, including but not limited to semiconductor-based memory devices, magnetic memory devices and systems, optical memory devices and systems. Although only one memory module is shown in FIG. 9, multiple physically distinct memory modules may be present in device 900.

The controller 910 may be of any suitable type suitable to the local technical environment, 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 900 may also include a plurality of controllers 910. The controller 910 is coupled to a transceiver 940, and the transceiver 940 may enable receiving and transmitting information via one or more antennas 950 and/or other components.

When the apparatus 900 functions as the channel estimation assisted signal estimation module 440 depicted in fig. 4, 7, or 8, the controller 910 and the transceiver 940 may operate in cooperation to implement the method 600 described above with reference to fig. 6. All of the features described above with reference to fig. 4-8 apply to the apparatus 900 and are not described in detail herein.

FIG. 10 shows a graph comparing performance of simulations according to an embodiment of the present disclosure with conventional schemes. As shown in fig. 10, the abscissa represents the signal-to-noise ratio SNR in decibel dB, and the ordinate represents the bit error rate BER. Curve 1001 in the lower left insert of fig. 10 represents the channel impulse response given the channel coefficients, while curve 1002 represents the ideal duobinary impulse response.

Further, curve 1003 represents the BER versus SNR curve for the direct detection scheme without any processing. Curve 1004 represents the BER versus SNR curve using a feed forward equalizer FFE scheme with an 8-tap finite impulse response FIR filter based on the least mean square LMS algorithm. Curve 1005 shows the BER versus SNR for the MLSE scheme using default duobinary channel coefficients. Curve 1006 represents the BER versus SNR curve for the MLSE scheme using learned channel coefficients according to an embodiment of the present disclosure.

As is clear from fig. 10, the scheme according to the embodiment of the present disclosure can achieve a lower bit error rate under the same SNR condition, especially when the SNR is high. Thus, the scheme according to embodiments of the present disclosure may achieve superior performance relative to other schemes of simulation.

FIG. 11 shows another performance comparison graph of a simulation according to an embodiment of the present disclosure and a conventional scheme. As shown in fig. 11, the abscissa represents the received signal power in dBm, and the ordinate represents the bit error rate BER.

In fig. 11,

curves

1101 and 1102 represent the BER versus received signal power for a direct detection scheme without any processing over 20km of fiber transmission and without fiber transmission, respectively.

Curves

1103 and 1104 represent the BER versus received signal power for 20km fiber transmission and no fiber transmission, respectively, using a feed forward equalizer FFE scheme (with an 8-tap finite impulse response FIR filter based on the least mean square LMS algorithm).

Curves

1105 and 1106 represent the BER versus received signal power for 20km fiber transmission and no fiber transmission, respectively, using the MLSE scheme with default duobinary channel coefficients.

Curves

1107 and 1108 represent the BER versus received signal power for 20km fiber transmission and no fiber transmission, respectively, using the MLSE scheme with learned channel coefficients according to an embodiment of the present disclosure.

As is clear from fig. 11, due to the strong bandwidth filtering, the signal is strongly distorted and difficult to resolve even with direct MLSE. Significant improvements can be achieved with an MLSE that performs appropriate channel compensation and improvements in accordance with embodiments of the present disclosure. That is, the scheme according to the embodiments of the present disclosure can achieve a lower bit error rate with the same received signal power, especially when the received signal power is high.

Fig. 12 shows a graph of an experimental test according to an embodiment of the present disclosure. As shown in fig. 12, the abscissa represents frequency in GHz and the ordinate represents normalized magnitude in decibels dB. Reference numeral 1201 shows the directly received eye pattern, reference numeral 1202 shows the eye pattern after channel learning, and reference numeral 1203 shows the learned channel impulse response.

In this experimental test, 10G-B/s PAM4 (i.e., 20Gb/s bit rate) was launched into a directly modulated laser DML having a 2.7GHz @3dB bandwidth. As illustrated by reference numeral 1201, the directly received duobinary PAM4 eye diagram shows poor performance. However, as illustrated by reference numeral 1202, the eye diagram after channel learning may be improved. Specifically, the BER is from 1.2E-1 (i.e., 1.2X 10)^-3) The improvement is 5.2E-3. After this, MLSE is implemented and the BER is further reduced below 3E-3 to meet specific forward error correction FEC threshold requirements. This shows that a signal rate of 20Gb/s can be achieved using only 2.5G class transmitters and receivers, achieving 8 times the spectral efficiency.

Fig. 13 shows a graph of another experimental test according to an embodiment of the present disclosure. As shown in fig. 13, the abscissa represents frequency in GHz and the ordinate represents normalized magnitude in decibel dB. Reference numeral 1301 shows the eye pattern directly received, reference numeral 1302 shows the eye pattern after channel learning, and reference numeral 1303 shows the channel impulse response obtained by learning.

In this experimental test, 25G-B/s PAM4 (i.e., 50Gb/s bit rate) was squeezed to have an end-to-end 4.6GHz @3dB bandwidth limited channel. As illustrated by reference numeral 1301, the directly received duobinary PAM4 eye shows poor performance. However, as illustrated by reference numeral 1302, the eye diagram after channel learning may be improved. Specifically, the BER improves from 1E-1 to 2E-3. After this, MLSE is implemented and the BER is further reduced below 6E-4. In contrast, conventional MLSE using default channel coefficients only reduces the BER to 5E-2.

The simulations and experiments described above show that the embodiments of the present disclosure can significantly improve the estimation accuracy of a bipolar transformed signal sequence by implementing adaptive channel learning prior to signal sequence estimation. By the embodiment of the disclosure, the data transmission of 20Gb/s at the peak can be realized under the condition that the BER is 3.8E-3 by using a DML laser with the bandwidth of 2.7G-3dB for a bipolar PAM4 signal, and the data transmission of 50Gb/s can be realized under the bandwidth of 4.6G-3dB by using a conventional device with the bandwidth of 10 GHz. Furthermore, embodiments of the present disclosure are equally effective in improving error correction under varying channel coefficients. That is, embodiments of the present disclosure are adaptive and intelligent for different communication devices (e.g., ONUs in a PON system), particularly in the uplink direction.

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.

It should be noted that the embodiments of the present disclosure can be realized by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided, for example, in programmable memory or on a data carrier such as an optical or electronic signal carrier.

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.

Claims

1. A method implemented at a receiver configured to receive a sequence of signals over a bandwidth-limited channel, the method comprising:

performing a bipolar transformation on a training sequence known to the receiver to obtain a reference sequence;

obtaining a distorted training sequence of the training sequence after the transmission and the bipolar transformation of the channel;

determining a sequence of channel coefficients for the channel based on the reference sequence and the distorted training sequence;

obtaining a distorted signal sequence after the signal sequence is transmitted through the channel and subjected to bipolar transformation; and

determining an estimate of the signal sequence based on the channel coefficient sequence and the distorted signal sequence.

2. The method of claim 1, wherein performing a bipolar transform on a training sequence to obtain a reference sequence comprises:

transforming the binary sequence into a duobinary sequence; or

The four-order pulse amplitude modulation (PAM4) sequence was transformed into a bipolar PAM4 sequence.

3. The method of claim 1, wherein determining a sequence of channel coefficients for the channel comprises performing the following at least once:

obtaining an intermediate training sequence based on a convolution of the distorted training sequence and a current channel coefficient sequence of the channel;

comparing the difference between the intermediate training sequence and the reference sequence to a threshold; and

adjusting the current channel coefficient sequence in response to determining that the difference is greater than the threshold.

4. The method of claim 1, wherein determining the estimate of the signal sequence comprises:

an estimate of the signal sequence is determined using a maximum likelihood sequence estimation algorithm.

5. The method of claim 4, wherein determining the estimate of the signal sequence using a maximum likelihood sequence estimation algorithm comprises:

determining an estimate of the signal sequence using the sequence of channel coefficients and the distorted signal sequence as inputs to the maximum likelihood sequence estimation algorithm.

6. The method of claim 4, wherein determining the estimate of the signal sequence using a maximum likelihood sequence estimation algorithm comprises:

performing distortion compensation on the distorted signal sequence based on the channel coefficient sequence;

determining an undistorted channel coefficient sequence of an undistorted bipolar channel; and

determining an estimate of the signal sequence using the distorted signal sequence and the undistorted channel coefficient sequence that are distortion compensated as inputs to the maximum likelihood sequence estimation algorithm.

7. A communication device including a receiver configured to receive a sequence of signals over a bandwidth-limited channel, the communication device comprising:

at least one processor; and

at least one memory including computer-executable instructions, the at least one memory and the computer-executable instructions configured to, with the at least one processor, cause the communication device to:

8. The communication device of claim 7, wherein the at least one memory and the computer-executable instructions are further configured to, with the at least one processor, cause the communication device to:

transforming the binary sequence into a duobinary sequence; or

9. The communication device of claim 7, wherein the at least one memory and the computer-executable instructions are further configured to, with the at least one processor, cause the communication device to perform at least one of:

10. The communication device of claim 7, wherein the at least one memory and the computer-executable instructions are further configured to, with the at least one processor, cause the communication device to:

11. The communication device of claim 10, wherein the at least one memory and the computer-executable instructions are further configured to, with the processor, cause the communication device to:

12. The communication device of claim 10, wherein the at least one memory and the computer-executable instructions are further configured to, with the at least one processor, cause the communication device to:

estimating the signal sequence using the distorted signal sequence and the undistorted channel coefficient sequence after distortion compensation as inputs to the maximum likelihood sequence estimation algorithm.

13. A computer program product tangibly stored on a non-volatile computer-executable medium and comprising machine-executable instructions that, when executed, cause a machine to perform the steps of the method of any of claims 1 to 6.

14. A receiver for use in a communication device, comprising:

a bipolar transformation module configured to perform a bipolar transformation on a training sequence known to the receiver to obtain a reference sequence;

a channel estimation module configured to

Obtaining a distorted training sequence of the training sequence after the transmission and the bipolar transformation of the channel; and is

Determining a sequence of channel coefficients for the channel based on the reference sequence and the distorted training sequence; and

a signal estimation module configured to

Obtaining a distortion signal sequence of the signal sequence after the transmission and the bipolar transformation of the channel; and is

15. The receiver of claim 14, wherein the bipolar transform module is further configured to:

transforming the binary sequence into a duobinary sequence; or

16. The receiver of claim 14, wherein the channel estimation module is configured to perform the following at least once:

17. The receiver of claim 14, wherein the signal estimation module is further configured to:

18. The receiver of claim 17, wherein the signal estimation module is further configured to:

19. The receiver of claim 17, wherein the signal estimation module is further configured to: