CN102136852A - Crosstalk counteracting method and system - Google Patents

Abstract

Embodiments of the present invention provide a centralized crosstalk cancellation method and system to reduce system complexity and system cost. The method includes: compressing the signal data of each line after pre-processing and transmitting to the processing card; the processing card calculates the precoding compensation value or crosstalk value of a current line according to the compressed signal data and calculates Transmitting the precoding compensation value or crosstalk value to a line card; the line card cancels the crosstalk received by the current certain line according to the precoding compensation value or crosstalk value and the signal data of the current certain line. Compared with the prior art, the present invention reduces the bandwidth occupied when data is transmitted between the processing card and the line card, greatly reduces the complexity of system design, thereby reducing the cost of the system.

Description

A method and system for canceling crosstalk

technical field

The invention relates to the field of data communication, in particular to a crosstalk cancellation method and system.

Background technique

xDSL is the general term for all kinds of digital subscriber lines (DSL, Digital Subscriber Line). The xDSL of passband transmission uses frequency division multiplexing technology to make xDSL and traditional telephone service (POTS,) coexist on the same pair of twisted pairs. Among them, xDSL The signal occupies the high frequency band, POTS occupies the baseband part below 4KHz, and the POTS signal is separated from the xDSL signal through a splitter. The xDSL for passband transmission adopts Discrete Multi-Tone Modulation (DMT, Discrete Multi-Tone), and the system that provides multiple xDSL access is called Digital Subscriber Line Access Multiplexer (DSLAM, Digital Subscriber Line Access Multiplexer).

Due to the principle of electromagnetic induction, the multiple signals connected to the DSLAM generate crosstalk (Crosstalk) with each other. The crosstalk can be divided into two types: near-end crosstalk (NEXT, Near End Crosstalk) and far-end crosstalk (FEXT, Far End Crosstalk). The crosstalk energy will increase as the frequency band increases. The xDSL uplink and downlink channels adopt frequency division multiplexing, and NEXT does not cause too much harm to system performance. Since the frequency band used by xDSL is getting wider and wider, FEXT will seriously affect the transmission performance of the line. This is because, in xDSL transmission, the crosstalk is reflected as part of the noise. According to Shannon's formula, serious FEXT reduces the channel rate significantly. When multiple users in a bundle of cables require to activate xDSL services, some lines will have low rate, unstable performance, or even fail to activate due to FEXT, which will eventually lead to a relatively low DSLAM outgoing line rate.

In response to the harm of FEXT, the industry has proposed Vectored-DSL technology, which mainly utilizes the possibility of joint transmission and reception at the DSLAM end, uses signal processing methods to offset the interference of FEXT, and finally eliminates the FEXT in each signal interference. Figure 1 and Figure 2 respectively show the working conditions of synchronous transmission and synchronous reception at the DSLAM side.

The shared channel H shown in Figure 1 and Figure 2 can use the matrix on the kth frequency interval (tone) in the frequency domain:

(Formula 1.1)

Indicates that h _ij is the transmission equation from pair j to pair i. In actual situations, j and i are equal and equal to the number of channels that have crosstalk relationship with each other in the shared channel, which is set to M here, and H is a channel transmission matrix of order M×M. Suppose x is an M×1 channel input vector, y is an M×1 channel output vector, and n is an M×1 noise vector, then the channel transmission equation is expressed as follows:

y=Hx+n (Formula 1.2)

For the uplink, the signal is jointly received and processed at the central office (CO) end, that is, a crosstalk canceller w is introduced at the receiving end, and the signal received at the uplink receiving end is:

$\widetilde{\mathrm{the\; y}}=\mathrm{Wy}=\mathrm{wxya}+W$ (Formula 1.3)

When WH is a diagonal matrix, each component of the received signal is equal to the corresponding component of the transmitted signal multiplied by a coefficient, plus noise, which eliminates the influence of other components of the transmitted signal, thereby eliminating crosstalk.

For the downlink, the joint transmission processing is performed on the signal at the CO end, that is, a vector precoder P is introduced at the CO end, then the transmitted signal is:

$\tilde{x}=\mathrm{Px}$ (Formula 1.4)

The signal received by the downlink receiver is

$\widetilde{\mathrm{the\; y}}=h\tilde{x}+\mathrm{no}=\mathrm{HPx}+\mathrm{no}$ (Formula 1.5)

Likewise, crosstalk is eliminated when HP is a diagonal array.

Through this joint processing of uplink and downlink, Vectored-DSL can achieve the effect of eliminating far-end crosstalk.

In the Vectored-DSL system, the DSLAM usually uses a board to access the subscriber line, and each board is called a line card (LC, Line Card). Currently, each LC can support about dozens of users; on the other hand, the number of users connected to the same DSLAM is usually about 200. In this case, multiple LCs must be used in one DSLAM to connect all users.

Because the Vectored-DSL system uses joint uplink and downlink processing to offset far-end crosstalk, when the system cancels the far-end crosstalk of a line connected to a DSLAM, except the LC connected to the line participates in joint processing, all other LCs connected to the same DSLAM Lines also need to participate in joint processing. Therefore, how to make the signal data of the connected lines on other LCs participate in the joint processing of one line on this LC becomes a problem to be solved.

For the above-mentioned technical problems, the industry usually adopts a centralized architecture to solve the problem, that is, the functional modules for joint processing of uplink and downlink to offset far-end crosstalk are configured on a dedicated processing card, and each LC transmits the line signal to the processing card, and the processing card performs joint processing. The processing cancels the far-end crosstalk of all lines, and then transmits the data back to each LC. The architecture diagram is shown in Figure 3. For the convenience of description, all the signal processing processes before the far-end crosstalk cancellation of the signal in the LC are called "pre-processing", and correspondingly, all the signal processing of the signal data after the far-end crosstalk cancellation from the processing card is received in the LC The process is called "post-processing".

In the prior art shown in Figure 3, for the uplink transmission direction (that is, the direction in which the signal is transmitted from the user terminal to the DSLAM), the original time-domain received signal data is input to the LC from the left side of the LC, and after a series of "filtering, analog-to-digital conversion, etc. After pre-processing", the received signal in the frequency domain is obtained, that is, y in the previous formula 1.2 or formula 1.3. The received signal is used as an input, and is transmitted to the processing card, and the processing card calculates the far-end crosstalk of each line, and cancels the calculated crosstalk. After the crosstalk is cancelled, the data of each line is transmitted back to the LC, and the LC continues to perform a series of "post-processing" such as Fast Fourier Transform (FFT, Fast Fourier Transform).

For the downlink transmission direction (that is, the direction in which the signal is transmitted from the DSLAM to the user terminal), the original time domain signal data is input to the LC from the left side of the LC, and after encoding, constellation point mapping, and Inverse Fast Fourier Transform (IFFT, Inverse Fast Fourier Transform ) and a series of "pre-processing", the original signal to be sent in the frequency domain is obtained, that is, x in formula 1.2 or formula 1.4. The transmitted signal is transmitted to the processing card, which jointly precodes each line, and transmits the precoded signal back to the LC, and the LC continues to perform a series of "post-processing" such as digital-to-analog conversion and filtering.

The inventor found through research on the above-mentioned prior art: in the centralized architecture solution shown in Figure 3, in order to ensure the correctness of signal transmission and reception, it is necessary to ensure the accuracy of the signal transmission between the LC and the processing card, for example, the signal needs It is expressed by 32 or higher digits and then transmitted, which significantly increases the data flow transmitted from the LC to the processing card, and the increase in data flow means that the bandwidth between the LC and the processing card increases. In order to meet the high bandwidth requirements between the LC and the processing card, the system will need to design more components, deploy more data lines, etc. These requirements greatly increase the complexity of the system and eventually lead to a surge in cost.

Contents of the invention

Embodiments of the present invention provide a crosstalk cancellation method and system to reduce system complexity and system cost.

An embodiment of the present invention provides a centralized crosstalk cancellation method, including: compressing the signal data of each line after pre-processing and transmitting it to a processing card; The precoding compensation value or crosstalk value of the line and transmit the precoding compensation value or crosstalk value to the line card; the line card cancels according to the precoding compensation value or crosstalk value and the signal data of the current certain line The crosstalk received by the current line.

An embodiment of the present invention provides a centralized crosstalk cancellation method, including: the processing card calculates the precoding compensation value or crosstalk value of a current line according to the pre-processed signal data of each line; The crosstalk value is compressed and then transmitted to the line card; the line card converts the compressed precoding compensation value or crosstalk value into a processable format; The converted precoding compensation value or crosstalk value cancels the crosstalk received by the current line.

An embodiment of the present invention provides a centralized crosstalk cancellation system, the system includes a processing card and a line card, the line card includes a first compression module and a first crosstalk cancellation module, and the processing card includes a first computing module; The first compression module is used to compress the signal data of each line after pre-processing and transmit it to the first computing module; the first computing module is used to Signal data calculates the precoding compensation value or crosstalk value of a current line and transmits the precoding compensation value or crosstalk value to the first crosstalk cancellation module; the first crosstalk cancellation module is configured to The signal data of a certain line and the precoding compensation value or crosstalk value calculated by the first calculation module cancel the crosstalk received by the current certain line.

An embodiment of the present invention provides a centralized crosstalk cancellation system, the system includes a processing card and a line card, the processing card includes a third calculation module and a third compression module, and the line card includes a third conversion module and a third Crosstalk cancellation module; the third calculation module is used to calculate the precoding compensation value or crosstalk value of a current line according to the signal data of each line after pre-processing; the third compression module is used to convert the The precoding compensation value or crosstalk value calculated by the third calculation module is compressed and then transmitted to the third conversion module; the third conversion module is used to compress the precoding compensation value or crosstalk value compressed by the third compression module value into a processable format; the third crosstalk cancellation module is used to cancel the current signal data of a certain line and the precoding compensation value or crosstalk value converted by the third conversion module The crosstalk experienced by a line.

In the embodiment provided by the present invention, since the precoding compensation value or crosstalk value of a certain current line is calculated in the processing card, the crosstalk received by the current certain line is counteracted in the line card, that is, the same processing card or The functions processed in the line card are processed separately, and the pre-processed signal data of each line is compressed before being transmitted to the processing card. Therefore, compared with the prior art, the data transmission between the processing card and the line card is reduced. The occupied bandwidth greatly reduces the complexity of system design, thereby reducing the cost of the system.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the prior art or the accompanying drawings that need to be used in the description of the embodiments. Obviously, the accompanying drawings in the following description are only some of the present invention. Embodiments, for those of ordinary skill in the art, other drawings can also be obtained like these drawings without paying creative labor.

Fig. 1 is a schematic diagram of synchronous transmission at the digital subscriber line access multiplexer end;

Fig. 2 is a schematic diagram of synchronous reception at the digital subscriber line access multiplexer end;

FIG. 3 is a schematic diagram of a centralized Vectored-DSL system architecture provided by the prior art;

FIG. 4 is a schematic flowchart of a basic crosstalk cancellation method provided by an embodiment of the present invention;

FIG. 5 is a schematic flow chart of a crosstalk cancellation method provided by another embodiment of the present invention;

Fig. 6 is a flow chart of calculating the real part when the complex number is expressed in the "MME" format provided by the embodiment of the present invention;

Fig. 7 is a flow chart of calculating the imaginary part when complex numbers are represented in the "MME" format provided by the embodiment of the present invention;

Fig. 8 is a flow chart of index adjustment for the real part and the imaginary part when the complex number is expressed in the "MME" format provided by the embodiment of the present invention;

Fig. 9 is a flow chart of calculating the real part when multiplying complex numbers in the "MME" format provided by the embodiment of the present invention;

Fig. 10 is a flow chart of calculating the imaginary part when multiplying complex numbers in the "MME" format provided by an embodiment of the present invention;

FIG. 11 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by an embodiment of the present invention;

Fig. 12 is a schematic diagram of a basic logic structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 13 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 14 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 15 is a schematic diagram of a basic logical structure of a centralized crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 16 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 17 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 18 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

FIG. 19 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

FIG. 20 is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention;

Fig. 21 is a schematic diagram of a basic logic structure of a crosstalk cancellation system provided by another embodiment of the present invention.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

Please refer to accompanying drawing 4, which is a schematic flowchart of a centralized crosstalk cancellation method provided by an embodiment of the present invention, which mainly includes steps:

S401. Compress the pre-processed line signal data and transmit it to the processing card.

In this embodiment, the concept of "pre-processing" is the same as the concept of pre-processing described in the background art, that is, it refers to all signal processing processes in the line card before the signal far-end crosstalk is canceled.

In the line card, after the signal data of each line is pre-processed, the frequency domain signal of these lines in each frequency interval (tone) is a complex number. For example, in a 16-bit fixed-point system, the real part of the complex number and The imaginary part is represented by a 16-bit integer, and the entire complex number is represented by a 32-bit integer. By compressing these complex numbers, they can be compressed into a format that only requires 16-bit integers to represent the entire complex number. For specific compression methods, please refer to the following embodiments.

S402. The processing card calculates a precoding compensation value or crosstalk value of a current line according to the compressed signal data, and transmits the precoding compensation value or crosstalk value to the line card.

The crosstalk value of a current line is the sum of the products obtained by multiplying the crosstalk coefficients of other lines to the current line except the current line and the frequency domain signals sent by the other lines, The precoding compensation value of a current line is numerically equal to the sum of the products obtained by multiplying the precompensation coefficients of other lines except the current line with the signal data on the other lines.

S403. The line card cancels the crosstalk received by the current certain line according to the precoding compensation value or the crosstalk value obtained in S402 and the signal data of the current certain line.

Different from the prior art, in the embodiment of the present invention, the crosstalk value and precoding compensation value required to cancel the crosstalk of a certain line are concentrated on the processing card for processing, and the current crosstalk received by a certain line is determined by the Line card offset, this processing method provides the first premise for compressing the signal data of each line after pre-processing, and before transmitting from the line card to the processing card, the signal data of each line is compressed before transmission to the processing card, reducing the bandwidth occupied by the signal transmission from the line card to the processing card, which will be described in detail below.

In the uplink transmission direction, the crosstalk received by a single line is related to the signal data received by other lines of all line cards that are in working state (that is, users connected to the lines are using the lines for data transmission). Assuming that the current system is connected with M lines in working state, the crosstalk canceller in Figure 2 can be expressed as:

(公式1.6)

(Formula 1.6)

Then, the crosstalk value of the signal on the kth line can be expressed as:

${\mathrm{UP}-\mathrm{STREAM}-\mathrm{FEXT}}_k=\underset{i\≠k}{\mathrm{\Σ}}{w}_{\mathrm{the\; ki}}{\mathrm{the\; y}}_i$ (Formula 1.7)

Among them, y _i represents the signal received by the i-th line, and the crosstalk value of the signal on the k-th line is the crosstalk cancellation coefficient of each line except the k-th line to the k-th line (that is, the crosstalk cancellation matrix W The sum of the products obtained by multiplying the non-diagonal elements in the middle and the signal data received on the other lines respectively. In the embodiment of the present invention, the implementation process of Formula 1.7 is concentrated in the processing card.

In the downlink transmission direction, for a signal on a single line, before it leaves the processing card, a precoding compensation value of the signal on the single line can be calculated. The transmitted signal data will cause far-end crosstalk to the single line when it is transmitted through the line. Assuming that there are M lines in working state connected to the current system, the vector precoder in Figure 1 can be expressed as a precoding matrix P:

(Formula 1.8)

Then, the precoding compensation value for the signal on the kth line can be expressed as:

${\mathrm{DOWN}-\mathrm{STREAM}-\mathrm{PRECOMPENSATION}}_k=\underset{i\≠k}{\mathrm{\Σ}}{p}_{\mathrm{the\; ki}}{x}_i$ (Formula 1.9)

Among them, x _i represents the signal sent by the i-th line, and the precoding compensation value of the signal on the k-th line is numerically equal to the pre-compensation coefficients of other lines except the k-th line (that is, the precoding matrix P The sum of the products obtained by multiplying the non-diagonal elements in the middle and the signal data on the other lines respectively. In the embodiment of the present invention, the implementation process of Formula 1.8 is concentrated in the processing card.

For the uplink transmission direction, the crosstalk value (Equation 1.7) of the signal data on the kth line can be offset by the following method, namely:

${\widetilde{\mathrm{the\; y}}}_k=w_{\mathrm{kk}}{\mathrm{the\; y}}_k-\mathrm{UP}-{\mathrm{STREAM}-\mathrm{FEXT}}_k$ (Formula 1.10)

为第k条线路进行串扰抵消后得到的信号。在本发明实施例中，公式1.10的实现过程在线卡中进行。Among them, w _kk is the compensation value of frequency domain fading of the kth line,

The signal obtained after performing crosstalk cancellation for the kth line. In the embodiment of the present invention, the implementation process of Formula 1.10 is performed in the line card.

For the downlink transmission direction, the precoding compensation value (Equation 1.9) of the signal on the kth line will be offset by the far-end crosstalk generated by other lines on the kth line after transmission, so that the user side can receive The signal of can be offset by the following method, namely:

${\tilde{x}}_k=p_{\mathrm{kk}}{x}_k-{\mathrm{DOWN}-\mathrm{STREAM}-\mathrm{PRECOMPENSATION}}_k$ (Formula 1.11)

为第k条线路进行串扰抵消后得到的信号。在本发明实施例中，公式1.11的实现过程在线卡中进行。Among them, p _kk is the precompensation value of frequency domain fading of the kth line,

The signal obtained after performing crosstalk cancellation for the kth line. In the embodiment of the present invention, the implementation process of Formula 1.11 is performed in the line card.

The above is the analysis of the first premise provided by the compression of the signal data of each line after the pre-processing, and the analysis of the second premise will continue below.

In the Vectored-DSL system, the crosstalk generated by the i-th line to the j-th line is reflected in the line transmission as the disturbance of the receiving side voltage, and in the frequency domain as the crosstalk coefficient of the i-th line to the j-th line Multiplied by the frequency domain signal sent by the i-th line. Regardless of the uplink transmission direction or the downlink transmission direction, the result proved by theory and verified by experimental measurement data is that the absolute value of the crosstalk cancellation coefficient is very small. Corresponding to the channel transmission matrix H (Formula 1.1), it is reflected that the absolute value of the off-diagonal elements of each row is much smaller than the absolute value of the diagonal elements, that is

h _ij << h _ii , j≠i (Formula 1.12)

This property of the transmission matrix causes the crosstalk cancellation matrix W to have the same property as the precoding matrix P, namely

w _ij << w _ii , j≠i

p _ij ＜＜p _ii , j≠i (Formula 1.13)

It can be seen from formula 1.13 that since the absolute values of the crosstalk cancellation coefficient w _ij and the pre-compensation coefficient p _ij are much smaller than the absolute values of the diagonal elements w _ii and p _ii respectively, when calculating the crosstalk value (formula 1.7) and the pre-compensation coefficient of the signal on the kth line When encoding the compensation value (formula 1.9) and canceling the crosstalk of the signal on the k-th line (formula 1.10 and formula 1.11), for the signal y _i received by the i-th line and the signal x _i sent by the i-th line, in terms of accuracy It is not necessary at all to be the same as x _k and y _k in Equation 1.10 and Equation 1.11. For example, assuming that x _k and y _k are represented by 32 bits, then signal y _i and signal _xi can be represented by 16 bits or even lower. Compression of line signal data provides the second premise.

Since in the embodiment of the present invention, the implementation process of Formula 1.7 and Formula 1.8 is concentrated in the processing card, therefore, for each line signal data after pre-processing, before being transmitted from the line card to the processing card, each The signal data of each line is compressed and then transmitted to the processing card, thus reducing the bandwidth occupied by the signal transmission from the line card to the processing card.

In the embodiment shown in FIG. 4 , the compressed line signal data transmitted from the line card may not be directly processed by the processing card. For example, the processing card cannot directly process the compressed data due to different formats. Therefore, as another embodiment of the present invention, before the processing card calculates the precoding compensation value or crosstalk value of a current line according to the compressed signal data, the processing card further converts the compressed signal data into format.

In order to reduce the occupancy of the bandwidth during the transmission of the precoding compensation value or crosstalk value of a certain line from the processing card to the line card, as another embodiment of the present invention, the processing card can calculate the precoding value of the current certain line according to the calculation. After encoding the compensation value or the crosstalk value and before transmitting the precoded compensation value or the crosstalk value to the line card, compress the precoded compensation value or the crosstalk value. This processing method makes the sum of the bandwidth used when transmitting the precoding compensation value or crosstalk value to the line card after processing is smaller than the sum of the actual bandwidth occupied by the data between the "pre-processing" and the "crosstalk cancellation module" inside the line card in the prior art.

Considering that the compressed precoding compensation value or crosstalk value transmitted from the processing card may not be directly processed by the line card, for example, due to the different formats, the line card cannot directly use the precoding compensation value or crosstalk value to offset a current line received crosstalk. Therefore, as another embodiment of the present invention, before the line card cancels the crosstalk received by a certain current line according to the precoding compensation value or crosstalk value and the signal data of the current certain line, the line card further uses the compressed precoding compensation value or crosstalk values into a format it can handle.

If it is not considered to reduce the occupation of bandwidth during the process of transmitting the precoding compensation value or crosstalk value from the processing card to the line card, and the conversion of the precoding compensation value or crosstalk value is not performed in the line card, then as another implementation of the present invention For example, the processing card can convert the calculated precoding compensation value or crosstalk value into a format that can be processed by the line card and then transmit it to the line card, and then the line card can The signal data of the current line cancels the crosstalk received by the current line.

In the above embodiment, the line card cancels the crosstalk received by the current line according to the precoding compensation value or crosstalk value and the signal data of the current line, including the steps of: combining the signal data y _k received in the upstream direction of the current line Multiply the compensation value w _kk of the frequency domain fading of the current certain line in the cancellation matrix; subtract the crosstalk value from the product of y _k and w _kk . For the specific calculation formula, please refer to formula 1.10 above.

As another embodiment of the present invention, the line card may offset the crosstalk received by the current line according to the precoding compensation value or crosstalk value and the signal data of the current line: the signal data x sent in the downlink direction of the current line _k is multiplied by the pre-compensation value p _kk of frequency domain fading of a certain current line in the pre-coding matrix; the product of x _k and p _kk is subtracted from the pre-coding compensation value. For the specific calculation formula, please refer to formula 1.11 above.

It should be noted that, in the above-mentioned embodiments, whether it is the "conversion" action of the processing card or the line card, if there is an action of compressing data before, the "conversion" action performed actually includes the compression of the compressed data. decompression action.

Please refer to FIG. 5 , which is a schematic flow chart of a basic crosstalk cancellation method provided by another embodiment of the present invention, which mainly includes steps:

S501. The processing card calculates a precoding compensation value or a crosstalk value of a current line according to the line signal data after pre-processing.

In this embodiment, the line card may not compress the signal data of each line after pre-processing, but directly transmit it to the processing card, and the processing card calculates the current certain The calculation method of the precoding compensation value or crosstalk value of each line is the same as that of the embodiment shown in FIG.

S502. The processing card compresses the precoding compensation value or the crosstalk value and then transmits it to the line card.

S503. The line card converts the compressed precoding compensation value or crosstalk value into a processable format.

Similar to the foregoing embodiments, the "conversion" action performed by the line card actually includes decompression of the compressed precoding compensation value or crosstalk value.

S504. The line card cancels the crosstalk received by the current line according to the signal data of the current line and the converted precoding compensation value or crosstalk value.

In this embodiment, the line card cancels the crosstalk received by a certain current line is similar to the foregoing embodiments. Specifically, for the uplink direction, it includes the steps of: adding the signal data y _k received in the uplink direction of the current certain line to the crosstalk cancellation matrix Multiply the compensation value w _kk of frequency domain fading of the current certain line; subtract the crosstalk value from the product of y _k and w _kk . For the specific calculation formula, please refer to formula 1.10 above.

For the downlink direction, the line card counteracts the crosstalk received by the current certain line including the steps of: combining the signal data x _k sent in the downlink direction of the current certain line with the precompensation value p _kk of the frequency domain fading of the current certain line described in the precoding matrix Multiply; subtract the precoding compensation value from the product of x _k and p _kk . For the specific calculation formula, please refer to formula 1.11 above.

In the embodiment shown in Fig. 5, in order to reduce the occupation of the bandwidth during the transmission process of the signal data of each line after the pre-processing from the line card to the processing card, as another embodiment of the present invention, the Before the processing card transmits the pre-processed signal data of each line, the line card may compress the pre-processed signal data of each line and then transmit it to the processing card. This processing method makes the sum of the bandwidth used by the signal data of each line transmitted to the processing card smaller than the sum of the actual bandwidth occupied by the data between the "pre-processing" and the "crosstalk cancellation module" inside the line card in the prior art.

Considering that the compressed line signal data transmitted from the line card may not be directly processed by the processing card, for example, due to the different formats, the processing card cannot directly use the line signal data to calculate the precoding compensation value or crosstalk of a current line value. Therefore, as another embodiment of the present invention, before the processing card calculates the precoding compensation value or crosstalk value of a current line based on the pre-processed signal data of each line, the processing card converts these compressed line signals to Data is converted into a format it can handle.

In the above-mentioned embodiments of the present invention, whether the line card transmits the pre-processed signal data of each line to the processing card, or the processing card transmits the calculated precoding compensation value or crosstalk value of a current line to the line card, or using both solutions at the same time, reduces the bandwidth occupied by the line card and the processing card when transmitting data, thus greatly reducing the complexity of the system and ultimately reducing the cost of the system.

The method adopted for compressing the pre-processed signal data of each line and the precoding compensation value or crosstalk value in the above embodiment will be described in detail below.

In the line card, after the signal data of each line is "pre-processed", the frequency domain signal of each frequency interval (tone) of each line is a complex number. In a 16-bit fixed-point system, the real and imaginary parts of the complex number are represented by 16-bit integers, namely

x=a+bj (Formula 1.14)

Among them, a and b represent 16-bit signed integers respectively. When transmitting between boards (that is, between a line card and a processing card), the frequency domain data in this frequency range occupies 32 bits. The following uses an efficient compression format to reduce the accuracy of the frequency domain signal to the following form:

$x\&ap;(a_m+b_{m}j)\&times;2^{q_e}$ (Formula 1.15)

Among them, a _m and b _m are m-bit signed integers, q _e is e-bit unsigned integers, and the representation format of this complex number is called "MME" format. When m is set as 6 and e is set as 4 in the system, the entire complex number uses 16 (=6+6+4) bits, and the number of bits occupied by signal data in each frequency interval is reduced by half.

Under the given values of m and e, record M ₊ =2 ^m-1 -1, M _- =-2 ^m-1 , E=2 ^e , to express a complex number x into "MM" format can be done through the following three steps accomplish:

S61, calculate the real part, the specific flow chart is as shown in Figure 6, wherein, the function round (r/2) represents the processing of rounding after taking the modulus of 2 for r;

S71, calculate the imaginary part, the specific flow chart is as shown in Figure 7, wherein, the function round (i/2) represents the processing of rounding after taking the modulus of 2 for i;

S81, adjusting the indices of the real part and the imaginary part calculated in S61 and S71 to be equal, the specific flow chart is as shown in Figure 8, wherein, the function round (r/2) represents the processing of rounding after taking the modulus of 2 for r , the function round(i/2) means rounding after taking the modulus of 2 for i.

Through the above three steps, the real part and the imaginary part of each 16-bit complex number (that is, the signal data of each line from the line card to the processing card in the previous embodiment or the calculated precoding compensation value or crosstalk value) can be compressed into 16 bit "MME" data format.

In the embodiment of the present invention, if the coefficients in the processing card use the same "MME" format as the data sent by the line card (the real part and the imaginary part use m ¹ bit respectively, and the exponent uses e ¹ bit), complex numbers can be directly compared Multiply and add, and can be directly converted to another "MME" format: the real part and the imaginary part use m ² bits respectively, and the exponent uses e ² bits.

则采用以下三个步骤实现：For example, when multiplying complex numbers, suppose the input numbers are

This is accomplished in three steps:

S91, calculate the real part, the specific flow chart is as shown in Figure 9, wherein, the function round (r/2) represents the processing of rounding after taking the modulus of 2 for r;

S101, calculate the imaginary part, the specific flow chart is as shown in Figure 10, wherein, the function round (i/2) represents the processing of rounding after taking the modulus of 2 for i;

S111. Adjusting the indices of the real part and the imaginary part calculated in S91 and S101 to be equal. The specific flow chart is similar to that in FIG. 8 and will not be repeated here.

则采用以下两个步骤即可实现：The addition operation is relatively simple, for example, suppose the input numbers are

This can be achieved by following two steps:

S121, adjusting the indices of _x1 and _x2 to be equal, the specific flow chart is similar to that in Fig. 8, and will not be described in detail;

S131. Add the real part and the imaginary part of the complex number obtained after the exponent adjustment, respectively.

Through the above compression method, the bandwidth when transmitting data between boards (that is, between the line card and the processing card) can be reduced by half. The specific analysis is as follows:

Before compression, assume that a single line card is connected to M users, the uplink and downlink signals of each line use a total of T frequency intervals (tone), and the system sends and receives one symbol (symbol) within t seconds, that is The symbol rate (symbol rate) of the system is 1/t; at this time, the system needs to perform crosstalk cancellation on all M×T tones within t seconds. Assume that the signal data of each tone occupies b bits (bit), so the bandwidth between boards must be at least

For example, each line card is connected to 32 users; the uplink and downlink signals of each user occupy a total of 4096 tones; the signal of each tone is a complex number, and in the fixed-point system, it is assumed that the real part and the imaginary part respectively use an int16 fixed-point Points indicate that the signal data of each tone occupies 32 bits at this time; the symbol rate used by the system is 4000, that is, one symbol is sent and received every 1/4000 second. At this time, the inter-board bandwidth is at least:

$\frac{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; T</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; b</span>}}{\mathrm{<span\; class="notranslate">\; t</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 32</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 4096</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 32</span>}<span\; class="notranslate">\; \&times;</span>\mathrm{<span\; class="notranslate">\; 4000</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 15.625</span>}\mathrm{<span\; class="notranslate">\; Gbits</span>}<span\; class="notranslate">\; /</span>\mathrm{<span\; class="notranslate">\; the\; s</span>}$

After compression, since the signal data of each tone only occupies 16 bits, the inter-board bandwidth is reduced to 15.625/2Gbits/s=7.813Gbits/s, that is, the bandwidth is reduced by half.

Please refer to FIG. 11 , which is a schematic diagram of a basic logical structure of a centralized crosstalk cancellation system provided by an embodiment of the present invention. For ease of description, only parts related to the embodiments of the present invention are shown. The system includes a line card 111 and a processing card 112. The line card 111 further includes a first compression module 1111 and a first crosstalk cancellation module 1112. The processing card 112 includes a first calculation module 1121, wherein:

The first compression module 1111 is configured to compress the pre-processed line signal data and transmit it to the first calculation module 1121;

The first calculation module 1121 is configured to calculate the precoding compensation value or crosstalk value of a current line according to the signal data compressed by the first compression module 1111 and transmit the precoding compensation value or crosstalk value to the first crosstalk cancellation module 1112 ;

The first crosstalk cancellation module 1112 is configured to cancel the crosstalk received by the current line according to the signal data of the current line and the precoding compensation value or crosstalk value calculated by the first calculation module 1121 .

The processing card 112 illustrated in FIG. 11 further includes a first conversion module 1122 , which is a crosstalk cancellation system provided by another embodiment of the present invention as shown in FIG. 12 . The first conversion module 1122 is used for converting the signal data compressed by the first compression module 1111 into a format that the first calculation module 1121 can calculate.

The processing card 112 illustrated in FIG. 12 further includes a second compression module 1123 , which is a crosstalk cancellation system provided by another embodiment of the present invention as shown in FIG. 13 . The second compression module 1123 is configured to compress the precoding compensation value or the crosstalk value calculated by the first calculation module 1121 .

The line card 111 illustrated in FIG. 11 further includes a second conversion module 1113 , which is a crosstalk cancellation system provided by another embodiment of the present invention as shown in FIG. 14 . The second conversion module 1113 is configured to convert the precoding compensation value or crosstalk value compressed by the second compression module 1123 into a format that can be processed by the first crosstalk cancellation module 1112 .

The first crosstalk cancellation module 1112 in any one of the examples in FIGS. 11 to 14 may further include a first multiplication unit 1501 and a first subtraction unit 1502, as shown in FIG. 15 , a crosstalk cancellation system provided by another embodiment of the present invention ,in:

The first multiplication unit 1501 is configured to multiply the signal data y _k received in the uplink direction of a certain current line by the compensation value w _kk of frequency domain fading of the current certain line in the cancellation matrix;

The first subtraction unit 1502 is configured to subtract the crosstalk value from the product of y _k and w _kk calculated by the first multiplication unit 1501 .

The first crosstalk cancellation module 1112 in any one of the examples in FIG. 11 to FIG. 14 may further include a second multiplication unit 1603 and a second subtraction unit 1604, as shown in FIG. 16 , a crosstalk cancellation system provided by another embodiment of the present invention ,in:

The second multiplying unit 1603 is configured to multiply the signal data x _k sent in the downlink direction of a certain current line by the precompensation value p _kk of the frequency domain fading of the current certain line in the precoding matrix;

The second subtraction unit 1604 is configured to subtract the precoding compensation value from the product of x _k and p _kk calculated by the second multiplication unit 1603 .

Please refer to FIG. 17 , which is a schematic diagram of a basic logical structure of a crosstalk cancellation system provided by another embodiment of the present invention. For ease of description, only parts related to the embodiments of the present invention are shown. The system includes a line card 171 and a processing card 172. The line card 171 further includes a third conversion module 1711 and a third crosstalk cancellation module 1712. The processing card 172 includes a third calculation module 1721 and a third compression module 1722, wherein:

The third calculation module 1721 is used to calculate the precoding compensation value or crosstalk value of a certain current line according to the line signal data after pre-processing;

The third compression module 1722 is configured to compress the precoding compensation value or crosstalk value calculated by the third calculation module 1721 and transmit it to the third conversion module 1711;

The third conversion module 1711 is configured to convert the precoding compensation value or crosstalk value compressed by the third compression module 1722 into a processable format;

The third crosstalk cancellation module 1712 is configured to cancel the crosstalk received by a current line according to the signal data of the current line and the precoding compensation value or crosstalk value converted by the third conversion module 1711 .

The line card 171 illustrated in FIG. 17 further includes a fourth compression module 1811 , which is a crosstalk cancellation system provided by another embodiment of the present invention as shown in FIG. 18 . The fourth compression module 1811 is configured to compress the signal data of each line after pre-processing.

The processing card 172 illustrated in FIG. 18 further includes a fourth conversion module 1911 , which is a crosstalk cancellation system provided by another embodiment of the present invention as shown in FIG. 19 . The fourth conversion module 1911 is used for converting the line signal data compressed by the fourth compression module 1811 into a format that the third calculation module 1721 can process.

The third crosstalk cancellation module 1712 in any one of the examples in FIGS. 17 to 19 may further include a third multiplication unit 2001 and a third subtraction unit 2002, as shown in FIG. 20 , a crosstalk cancellation system provided by another embodiment of the present invention ,in:

The third multiplication unit 2001 is used to multiply the signal data y _k received in the uplink direction of a certain current line by the compensation value w _kk of the frequency domain fading of the current certain line in the cancellation matrix;

The third subtraction unit 2002 is configured to subtract the crosstalk value from the product of y _k and w _kk calculated by the third multiplication unit 2001 .

The third crosstalk cancellation module 1712 in any one of the examples in FIGS. 17 to 19 may further include a fourth multiplication unit 2101 and a fourth subtraction unit 2102, as shown in FIG. 20 , a crosstalk cancellation system provided by another embodiment of the present invention ,in:

The fourth multiplication unit 2101 is configured to multiply the signal data x _k sent in the downlink direction of a current line by the precompensation value p _kk of the frequency domain fading of the current line in the precoding matrix;

The fourth subtraction unit 2102 is configured to subtract the precoding compensation value from the product of x _k and p _kk calculated by the fourth multiplication unit 2101 .

It should be noted that the information interaction and execution process among the modules/units of the above-mentioned system are based on the same idea as the method embodiment of the present invention, and the technical effect it brings is the same as that of the method embodiment of the present invention, and the specific content can be Refer to the descriptions in the method embodiments of the present invention, and details are not repeated here.

A crosstalk cancellation method and system provided by the embodiments of the present invention have been described above in detail. In this paper, specific examples are used to illustrate the principle and implementation of the present invention. The descriptions of the above embodiments are only used to help understand the present invention. method and its core idea; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. Invention Limitations.

Claims (15)

1. A crosstalk cancellation method, characterized in that, comprising: Compress the pre-processed line signal data and transmit it to the processing card; The processing card calculates the precoding compensation value or crosstalk value of a current line according to the compressed signal data, and transmits the precoding compensation value or crosstalk value to the line card; The line card cancels the crosstalk received by the current certain line according to the precoding compensation value or the crosstalk value and the signal data of the current certain line. 2. The method according to claim 1, wherein, before the processing card transmits the precoding compensation value or the crosstalk value to the line card, it further comprises: Compressing the precoding compensation value or crosstalk value. 3. The method according to claim 1 or 2, wherein the line card cancels the signal data received by the current certain line according to the precoding compensation value or crosstalk value and the signal data of the current certain line. Crosstalk is specifically: Multiplying the signal data y _k received in the uplink direction of the current certain line by the compensation value w _kk of the frequency domain fading of the current certain line in the crosstalk cancellation matrix; Subtract the crosstalk value from the product of y _k and w _kk . 4. The method according to claim 1 or 2, wherein the line card offsets the signal data received by the current certain line according to the precoding compensation value or crosstalk value and the signal data of the current certain line. Crosstalk is specifically: Multiplying the signal data x _k sent in the downlink direction of the current certain line by the precompensation value p _kk of the frequency domain fading of the current certain line in the precoding matrix; Subtract the precoding compensation value from the product of x _k and p _kk . 5. A crosstalk cancellation method, comprising: The processing card calculates the precoding compensation value or crosstalk value of a current line according to the line signal data after pre-processing; compressing the precoding compensation value or the crosstalk value and then transmitting it to the line card; The line card converts the compressed precoding compensation value or crosstalk value into a processable format; The line card cancels the crosstalk received by the current certain line according to the signal data of the current certain line and the converted precoding compensation value or crosstalk value. 6. The method according to claim 5, wherein the processing card further comprises: The line card compresses the pre-processed line signal data and transmits it to the processing card. 7. The method according to claim 5 or 6, wherein the line card cancels the current certain line according to the signal data of the current certain line and the converted precoding compensation value or crosstalk value The crosstalk received is specifically: multiplying the signal data y _k received in the uplink direction of the current certain line by the compensation value w _kk of the frequency domain fading of the current certain line in the cancellation matrix; Subtract the crosstalk value from the product of y _k and w _kk . 8. The method according to claim 5 or 6, wherein the line card cancels the current certain line according to the signal data of the current certain line and the converted precoding compensation value or crosstalk value The crosstalk received is specifically: Multiplying the signal data x _k sent in the downlink direction of the current certain line by the precompensation value p _kk of the frequency domain fading of the current certain line in the precoding matrix; Subtract the precoding compensation value from the product of x _k and p _kk . 9. A crosstalk cancellation system, characterized in that the system includes a processing card and a line card, the line card includes a first compression module and a first crosstalk cancellation module, and the processing card includes a first calculation module; The first compression module is configured to compress the pre-processed line signal data and transmit it to the first computing module; The first calculation module is used to calculate the precoding compensation value or crosstalk value of a current line according to the signal data compressed by the first compression module and transmit the precoding compensation value or crosstalk value to the second A crosstalk cancellation module; The first crosstalk canceling module is configured to cancel the crosstalk suffered by the current certain line according to the signal data of the current certain line and the precoding compensation value or crosstalk value calculated by the first calculation module. 10. The system according to claim 9, wherein the processing card further comprises: The second compression module is configured to compress the precoding compensation value or the crosstalk value calculated by the first calculation module. 11. The system according to claim 9 or 10, wherein the first crosstalk cancellation module further comprises: The first multiplication unit is used to multiply the signal data y _k received in the uplink direction of the current certain line by the compensation value w _kk of the frequency domain fading of the current certain line in the cancellation matrix; The first subtraction unit is configured to subtract the crosstalk value from the product of y _k and w _kk calculated by the first multiplication unit. 12. The system according to claim 9 or 10, wherein the first crosstalk cancellation module further comprises: The second multiplication unit is used to multiply the signal data x _k sent in the downlink direction of the current certain line by the precompensation value p _kk of the frequency domain fading of the current certain line in the precoding matrix; The second subtraction unit is configured to subtract the precoding compensation value from the product of x _k and p _kk calculated by the second multiplication unit. 13. A crosstalk cancellation system, characterized in that the system includes a processing card and a line card, the processing card includes a third calculation module and a third compression module, and the line card includes a third conversion module and a third crosstalk module Offset module; The third calculation module is used to calculate the precoding compensation value or crosstalk value of a certain current line according to the line signal data after pre-processing; The third compression module is configured to compress the precoding compensation value or crosstalk value calculated by the third calculation module and transmit it to the third conversion module; The third conversion module is configured to convert the precoding compensation value or crosstalk value compressed by the third compression module into a processable format; The third crosstalk canceling module is configured to cancel the crosstalk received by the current certain line according to the signal data of the current certain line and the precoding compensation value or crosstalk value converted by the third conversion module. 14. The system according to claim 13, wherein the line card further comprises: The fourth compression module is configured to compress the pre-processed signal data of each line. 15. The system according to claim 13 or 14, wherein the third crosstalk cancellation module further comprises a third multiplication unit and a third subtraction unit; The third multiplication unit is configured to multiply the signal data y _k received in the uplink direction of the current certain line by the compensation value w _kk of the frequency domain fading of the current certain line in the cancellation matrix; The third subtraction unit is configured to subtract the crosstalk value from the product of y _k and w _kk calculated by the third multiplication unit; or The third crosstalk cancellation module further includes a fourth multiplication unit and a fourth subtraction unit; The fourth multiplication unit is configured to multiply the signal data x _k sent in the downlink direction of the current certain line by the precompensation value p _kk of the frequency domain fading of the current certain line in the precoding matrix; The fourth subtraction unit is configured to subtract the precoding compensation value from the product of x _k and p _kk calculated by the fourth multiplication unit.

Images