Equalizer adjustment method, adjustment device and receiver

Abstract

The embodiment of the present application discloses an adjustment method, an adjustment device and a receiver of an equalizer, which belongs to the field of communication technology. The method includes: obtaining a first signal after equalization processing by an analog equalizer. Performing an inverse transformation on the first signal to obtain a second signal. According to the first signal and the second signal, adjusting the tap coefficient of the analog equalizer. There is no need to separate the received signal, and there is no need for a large input optical power, so the requirements on the transmitting end are also reduced, which can reduce the cost of the transmitting end to a certain extent.

Description

Technical Field

[0001] The present application relates to the field of communication technology, and in particular to an adjustment method, an adjustment device and a receiver of an equalizer. Background Art

[0002] In the field of optical communications, the signal of the signal transmission system is sent out by the transmitter and transmitted to the receiver through the transmission channel. In the process of signal transmission in the transmission channel, inter-symbol interference (ISI) will occur. In order to deal with ISI, an equalizer is usually set at the receiving end of the signal. After the receiving end receives the signal, the signal is first input into the equalizer, and the equalizer equalizes the received signal before outputting it, which can compensate for the impact of ISI on the signal to a certain extent. The equalization process can be specifically as follows: the equalizer convolves the signal with the tap coefficient of the equalizer to adjust the signal amplitude of each frequency component in the signal. However, the degree of ISI may change dynamically due to the influence of medium bending, vibration, etc. Therefore, it is necessary to periodically update and adjust the tap coefficients in the equalizer.

[0003] In the related art, the signal received by the receiving end for adjusting the tap coefficient needs to be separated into two identical parts, one of which is input into the equalizer for equalization and then output, and the error between the signal output by the equalizer and the corresponding target signal is calculated, and the target signal is pre-stored. Then, the other separated signal is converted into digital form and multiplied with the error, and the tap coefficient is adjusted according to the result of the multiplication.

[0004] Because the input signal needs to be separated into two parts, a larger input optical power is required, which places higher requirements on the signal transmitter. Summary of the invention

[0005] The embodiments of the present application provide an equalizer adjustment method, an adjustment device and a receiver, which can solve the problem in the related art that a large optical power is required to separate the input optical signal. The technical solution is as follows:

[0006] In a first aspect, a method for adjusting an analog equalizer is provided, the method comprising: obtaining a first signal equalized by the analog equalizer, performing an inverse transformation on the first signal to obtain a second signal, and adjusting a tap coefficient of the analog equalizer according to the first signal and the second signal.

[0007] In the scheme shown in the embodiment of the present application, by inverse transforming the equalized signal, a signal approximately equal to the signal received by the receiver and not subjected to equalization processing is obtained, and the obtained signal is used to adjust the tap coefficients of the analog equalizer. In this way, there is no need to separate the received signal, and there is no need to input a large optical power, thereby reducing the requirements on the transmitting end and reducing the cost of the transmitting end to a certain extent.

[0008] In a possible implementation, the method for adjusting the tap coefficients of the analog equalizer may be as follows: calculating an error value between a first signal and a target signal corresponding to the first signal, and then adjusting the tap coefficients of the analog equalizer according to the error value, the second signal, and a preset update step size.

[0009] In the scheme shown in the embodiment of the present application, the target signal is a digital signal corresponding to the training signal sent by the transmitter without being transmitted through the signal transmission channel. When calculating the error value between the first signal and the target signal, a data can be selected from the first signal, and the data at the corresponding position in the target signal can be obtained, and the difference between the two data can be calculated as the error value between the first signal and the target signal. Among them, the data at the corresponding position in the target signal can be pre-stored, and when it is needed, it can be directly read from the specified storage location.

[0010] In addition, the continuous N data in the second signal are used as the gradient corresponding to the above error value, and then the error value, the gradient and the preset update step are multiplied to obtain a vector of length N. N is the number of taps of the analog equalizer. Finally, the obtained vector of length N is added to the vector composed of the current N tap coefficients to obtain the vector corresponding to the adjusted tap coefficients.

[0011] In one possible implementation, the inverse transformation method may be a pseudo-inverse matrix method. Accordingly, the inverse transformation process of the first signal may be: obtaining the pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer, multiplying the pseudo-inverse matrix to be used and the first signal to obtain the second signal.

[0012] In the solution shown in the embodiment of the present application, the calculation method of the pseudo-inverse matrix can be Moore-Penrose method, singular value decomposition (SVD) method, etc.

[0013] In a possible implementation, when the inverse transformation adopts the pseudo-inverse matrix method, because the signal transmission channel characteristics will not change very quickly, in order to save computing resources, it is not necessary to recalculate the pseudo-inverse matrix to be used every time the tap coefficients are updated. Accordingly, a pseudo-inverse matrix update period can be set. If the pseudo-inverse matrix update period is currently reached, the convolution matrix composed of the tap coefficients of the analog equalizer is pseudo-inversely calculated to obtain the pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer. If the pseudo-inverse matrix update period is currently reached, the pseudo-inverse matrix obtained by the latest pseudo-inverse calculation is obtained as the pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

[0014] In a possible implementation, the inverse transformation method may also be a convolution calculation method, and accordingly, the inverse transformation process of the first signal may be: obtaining the time domain data to be used corresponding to the tap coefficients of the analog equalizer, and convolving the first signal with the time domain data to be used to obtain the second signal.

[0015] In a possible implementation, when the inverse transform adopts the convolution calculation method, because the characteristics of the signal transmission channel will not change very quickly, in order to save computing resources, it is not necessary to recalculate the time domain data to be used every time the tap coefficient is updated. Accordingly, the time domain data update cycle can be set. If the time domain data update cycle is currently reached, the tap coefficients of the analog equalizer are subjected to a fast Fourier transform FFT to obtain the frequency domain data corresponding to the tap coefficients of the analog equalizer. The reciprocal of each element in the frequency domain data is calculated to obtain the spectrum data corresponding to the tap coefficients of the analog equalizer. The spectrum data is subjected to an inverse fast Fourier transform IFFT to obtain the time domain data to be used corresponding to the tap coefficients of the analog equalizer. If the time domain data update cycle is not currently reached, the time domain data corresponding to the tap coefficients of the analog equalizer obtained by the latest calculation are obtained as the time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0016] In a second aspect, a receiver is provided, the receiver comprising an analog equalizer, an analog-to-digital converter and a digital signal processor, wherein:

[0017] An analog equalizer, used for performing equalization processing on the received first input signal;

[0018] an analog-to-digital converter, configured to perform analog-to-digital conversion on the equalized first input signal to obtain a first signal, wherein the first signal is a digital signal;

[0019] A digital signal processor is used to perform an inverse transformation on the first signal to obtain a second signal, and to adjust the tap coefficients of the analog equalizer according to the first signal and the second signal.

[0020] In a possible implementation, the digital signal processor is used to:

[0021] Calculating an error value between the first signal and a target signal corresponding to the first signal;

[0022] The tap coefficients of the analog equalizer are adjusted according to the error value, the second signal and a preset update step size.

[0023] In a possible implementation, the digital signal processor is used to:

[0024] Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer;

[0025] The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

[0026] In a possible implementation, the digital signal processor is used to:

[0027] A pseudo-inverse calculation is performed on the convolution matrix composed of the tap coefficients of the analog equalizer to obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

[0028] In a possible implementation, the digital signal processor is used to:

[0029] Acquire time domain data to be used corresponding to the tap coefficients of the analog equalizer;

[0030] The first signal and the time domain data to be used are convolved to obtain a second signal.

[0031] In a possible implementation, the digital signal processor is further configured to output the adjusted tap coefficient to the analog equalizer;

[0032] The analog equalizer is further used to perform equalization processing on the received second input signal according to the adjusted tap coefficients.

[0033] In a possible implementation manner, the first input signal is an optical signal, and the receiver further includes:

[0034] A photoelectric converter, used for performing photoelectric conversion on the equalized first input signal to obtain a third signal, wherein the third signal is an electrical signal;

[0035] The analog-to-digital converter is used to perform analog-to-digital conversion on the third signal to obtain a first signal.

[0036] In a third aspect, a device for adjusting an equalizer is provided, the device comprising:

[0037] An acquisition module, used for acquiring the first signal after being equalized by the analog equalizer;

[0038] an inverse transformation module, configured to perform an inverse transformation on the first signal to obtain a second signal;

[0039] An adjustment module is used to adjust the tap coefficients of the analog equalizer according to the first signal and the second signal.

[0040] In a possible implementation, the adjustment module is used to:

[0041] Calculating an error value between the first signal and a target signal corresponding to the first signal;

[0042] The tap coefficients of the analog equalizer are adjusted according to the error value, the second signal and a preset update step size.

[0043] In a possible implementation, the inverse transformation module is used to:

[0044] Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer;

[0045] The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

[0046] In a possible implementation, the inverse transformation module is used to:

[0047] A pseudo-inverse calculation is performed on the convolution matrix composed of the tap coefficients of the analog equalizer to obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

[0048] In a possible implementation, the inverse transformation module is used to:

[0049] Acquire time domain data to be used corresponding to the tap coefficients of the analog equalizer;

[0050] The first signal and the time domain data to be used are convolved to obtain a second signal.

[0051] In a possible implementation, the inverse transformation module is used to:

[0052] Performing a fast Fourier transform (FFT) on the tap coefficients of the analog equalizer to obtain frequency domain data corresponding to the tap coefficients of the analog equalizer;

[0053] Calculating the reciprocal of each element in the frequency domain data to obtain frequency spectrum data corresponding to the tap coefficients of the analog equalizer;

[0054] Performing an inverse fast Fourier transform (IFFT) on the frequency spectrum data to obtain time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0055] In a fourth aspect, a signal transmission system is provided, the signal transmission system comprising a transmitter and a receiver as described in the second aspect above.

[0056] In a fifth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes instructions, and when the computer-readable storage medium is run in a receiver, the receiver executes the equalizer adjustment method as described in the first aspect above.

[0057] In a sixth aspect, a computer program product is provided. When the computer program product is run in a receiver, the receiver executes the equalizer adjustment method as described in the first aspect above. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a schematic diagram of the architecture of a signal transmission system provided in an embodiment of the present application;

[0059] Figure 2 It is a schematic diagram of the architecture of a signal transmission system provided in an embodiment of the present application;

[0060] Figure 3 is a structural diagram of an equalizer provided in an embodiment of the present application;

[0061] Figure 4A and 4B It is a schematic diagram of the architecture of a signal transmission system provided in an embodiment of the present application;

[0062] Figure 5 is a flow chart of an equalizer adjustment method provided in an embodiment of the present application;

[0063] Figure 6 is a flow chart of an equalizer adjustment method provided in an embodiment of the present application;

[0064] Figure 7 is a schematic diagram of a convolution matrix provided in an embodiment of the present application;

[0065] Figure 8 is a flow chart of an adjustment method of an equalizer provided in an embodiment of the present application;

[0066] Fig. 9 is a flow chart of an equalizer adjustment method provided in an embodiment of the present application;

[0067] Fig.10 It is a schematic diagram of a convolution process provided in an embodiment of the present application;

[0068] Fig.11 It is a structural schematic diagram of an adjustment device of an equalizer provided in an embodiment of the present application. DETAILED DESCRIPTION

[0069] An embodiment of the present application provides a method for adjusting an equalizer, which can be applied to a signal transmission system.

[0070] The architecture of the signal transmission system is described below.

[0071] like Figure 1 As shown, the signal transmission system may include a transmitting end 110, a transmission channel 120, and a receiving end 130. The transmitting end 110 may be an optical transmitting end (also referred to as an optical transmitter), and accordingly, the transmission channel may be an optical fiber, and the receiving end 130 may be an optical receiving end. Alternatively, the transmitting end 110 may be an electrical transmitting end (also referred to as an electrical transmitter), and accordingly, the transmission channel 120 may be a wireless link, a cable, a network cable, etc., and the receiving end 130 may be an electrical receiving end (also referred to as an electrical receiver).

[0072] The signal transmission system 100 is described below by taking the transmitting end 110 as an optical transmitting end, the transmission channel 120 as an optical fiber, and the receiving end 130 as an optical receiving end as an example.

[0073] like Figure 2 As shown, the transmitting end 110 may include a signal generator 111 and a modulator 112. The signal generator 111 is used to generate an electrical signal. Specifically, the electrical signal may be generated locally or received from the outside. In addition, the signal generator may also perform pulse shaping on the generated electrical signal and send it to the modulator 112. The modulator 112 is used to modulate the electrical signal onto an optical carrier so that it is sent out in the form of an optical signal.

[0074] The transmission channel 120 is used to transmit the optical signal to the receiving end 130 .

[0075] The receiving end 130 may include an analog equalizer 131, an analog-to-digital converter (ADC) 132, a digital signal processor (DSP) 133 and an optical-electrical converter 134. Among them, the analog equalizer 131 may be an analog equalizer. The analog equalizer 131 is used to equalize the received signal (for example, the first input signal) to offset the influence of inter-symbol interference (ISI) received by the signal as much as possible. The analog-to-digital converter 132 is used to convert the analog signal output by the analog equalizer 131 into a digital signal, and input it into the digital signal processor 133 for processing. The digital signal processor 133 is used to adjust the tap coefficient in the analog equalizer 131 according to the input signal. The optical-electrical converter 134 is used to convert the optical signal into an electrical signal. The optical-electrical converter 134 may be arranged before the analog equalizer 131, or between the analog equalizer 131 and the digital signal processor. Figure 2 The position of the photoelectric converter 134 is only an example. In addition, it should be noted that, when the signal transmission system is an electrical signal transmission system, the receiving end 130 does not include the photoelectric converter 134.

[0076] Next, the equalization processing of the signal by the analog equalizer 131 is described.

[0077] The essence of the analog equalizer 131 is a filter. Specifically, it can be a transversal filter, also known as a tapped-delay line filter or a finite impulse response filter (FIR filter).

[0078] like Figure 3 As shown, the number of taps of the analog equalizer 131 is N, that is, the analog equalizer is provided with N tap coefficients, which are W0, W1, W2...W N-1 Accordingly, the analog equalizer 131 includes N-1 unit delay units (Z -1 ), N multipliers and N-1 accumulators.

[0079] The unit delay unit is used to obtain and output the signal of the previous moment of the input signal of the current unit delay unit. The multiplier is used to multiply the input signal and the specified tap coefficient and output it, wherein each multiplier corresponds to a specified tap coefficient. The accumulator is used to accumulate the input signal and output it.

[0080] After the signal r(n) is input into the analog equalizer, the result obtained after passing through the first unit delay unit is r(n-1), and the result obtained after r(n-1) passes through the second unit delay unit is r(n-2), and so on. Finally, the result obtained after r(n-N+2) passes through the last unit delay unit is r(n-N+1). The final output of the entire analog equalizer is: z(n) = W0r(n) + W1r(n-1) + ... W N-1 r(n-N+1).

[0081] In order to facilitate the understanding of this application, Figure 4A and Figure 4B The adjustment method of the equalizer provided in the embodiment of the present application is briefly described.

[0082] exist Figure 4A In the embodiment, system 1 is a transmission channel, system 2 includes an analog-to-digital converter, and when the signal transmission system is an optical signal transmission system and no photoelectric converter is provided before the analog equalizer, system 2 may also include a photoelectric converter.

[0083] The signal r(n) output by system 1 is input to the analog equalizer, and the signal z(n) is output after equalization processing by the analog equalizer. The signal z(n) is converted into a digital signal y(n) by system 2. The signal y(n) is input to the digital signal processor. The digital signal processor performs an inverse transformation on y(n) to obtain the signal x′(n). The signal x′(n) is an estimated signal of the signal x(n), and the signal x′(n) is approximately the signal x(n). See Figure 4B , signal x(n) is the signal r(n) output by system 1 that is directly obtained by passing through system 2 without passing through the analog equalizer. Finally, the digital signal processor adjusts the tap coefficients according to the signals y(n) and x′(n), and outputs the adjusted tap coefficients to the analog equalizer.

[0084] It can be seen that in the method provided in the embodiment of the present application, a signal approximate to x(n) is obtained by inverse transforming the equalized signal to replace x(n). In this way, there is no need to separate the signal output by system 1, and there is no need to input a large optical power, thereby reducing the requirements on the transmitting end and reducing the cost of the transmitting end to a certain extent.

[0085] The above-mentioned inverse transformation can be achieved through a pseudo-inverse matrix, convolution, etc. The following takes the pseudo-inverse matrix to achieve the inverse transformation and the convolution to achieve the inverse transformation as examples to illustrate the adjustment method of the equalizer provided in the embodiment of the present application.

[0086] like Figure 5 As shown, in the case where the pseudo inverse matrix realizes the inverse transformation, the equalizer adjustment method provided in the embodiment of the present application may include the following processing steps:

[0087] Step 501: Obtain a first signal after being equalized by an analog equalizer.

[0088] In implementation, the transmitting end of the signal transmission system may send a training signal to the receiving end according to a preset period, and the training signal may also be referred to as a training sequence. Accordingly, the receiving end may receive the training signal sent by the transmitting end every preset period. Specifically, the receiving end may be provided with a timing module, and an update signal is triggered whenever the preset period is reached. When the digital signal processor receives the trigger update signal, it executes to obtain the first signal equalized by the analog equalizer.

[0089] The above preset period can be set by the technician according to the actual situation of the signal transmission system. For example, when setting the preset period, the technician can consider the signal transmission speed of the signal transmission system. Specifically, for a signal transmission system with a faster signal transmission speed, the preset period can be set to be relatively small. For a signal transmission system with a slower signal transmission speed, the preset period can be set to be relatively large. For example, if the signal transmission system is an optical transmission system with a very fast signal transmission speed, then the preset period can be set to between 0.1 microseconds (μs) and 1.2μs.

[0090] After the receiving end receives the training signal, if an optoelectronic converter is provided before the analog equalizer, the training signal is first converted into an electrical signal by the optoelectronic converter and then input into the analog equalizer. The analog equalizer equalizes the input training signal and outputs it. For specific equalization processing, please refer to the above Figure 3 , I will not go into details here.

[0091] Because the signal output by the analog equalizer is an analog signal, and a digital signal is required in subsequent processing, the training signal output by the analog equalizer must first be input to an analog-to-digital converter, converted by the analog-to-digital converter into a digital signal and output.

[0092] The training signal output by the analog-to-digital converter is input to the digital signal processor. The digital signal processor can select a preset number of data that are continuous in time. For the convenience of description, the preset number of data selected here is referred to as the first signal below.

[0093] Specifically, the digital signal processor may continuously acquire a preset number of data starting from the first data corresponding to the input training signal as the first signal.

[0094] The preset number may be 2M+1, where M may be close to the value of the number of taps of the analog equalizer, for example, when the number of taps is 5, M may be 4. M may be referred to as the number of single-sided taps.

[0095] The first signal may be represented in the form of a vector. Specifically, the first signal may be as follows:

[0096] Y=[y(1),y(2),y(3),...,y(2M+1)]

[0097] Step 502: Obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

[0098] In implementation, because the characteristics of the signal transmission channel do not change very quickly, in order to save computing resources, it is not necessary to recalculate the pseudo-inverse matrix to be used every time the tap coefficients are updated.

[0099] Based on this, a pseudo-inverse matrix update period can be set. When obtaining the pseudo-inverse matrix to be used, how to obtain the pseudo-inverse matrix to be used can be determined based on whether the pseudo-inverse matrix update period is currently reached.

[0100] Specifically, Figure 6 As shown, step 502 can be replaced by the following steps 5021 to 5023.

[0101] Step 5021: Determine whether the pseudo-inverse matrix update period has been reached.

[0102] The tap coefficient updates may be set K times as a pseudo-inverse matrix update cycle, and the value of K may be set by a technician according to actual conditions. For example, the value of K may be between 10 and 500.

[0103] In addition, a variable i can be set, and the initial value of i in a pseudo-inverse matrix update cycle is 1. In a pseudo-inverse matrix update cycle, a tap coefficient update is completed. If i is not equal to K, the value of i is increased by 1. If i is equal to K, the value of i is initialized to 1. Correspondingly, the method for determining whether the pseudo-inverse matrix update cycle is reached can be as follows:

[0104] If the current i is equal to 1, it is determined that the current pseudo-inverse matrix update period has been reached; if the current i is not equal to 1, it is determined that the current pseudo-inverse matrix update period has not been reached.

[0105] For example, K=3. Before obtaining the pseudo-inverse matrix to be used, the process of judging whether the pseudo-inverse matrix update period has been reached can be as follows: if the current i is equal to 1, it is determined that the pseudo-inverse matrix update period has been reached, and after the tap coefficient update is subsequently completed, i is updated to i+1=2; if the current i is equal to 2, it is determined that the pseudo-inverse matrix update period has not been reached, and after the tap coefficient update is subsequently completed, i is updated to i+1=3; if the current i is equal to 3, it is determined that the pseudo-inverse matrix update period has not been reached, and after the tap coefficient update is subsequently completed, i is initialized to 1.

[0106] It should be noted that the timing of updating i mentioned above is only an example. i can be updated after determining whether the current pseudo-inverse matrix update cycle has been reached and at any time before the next determination of whether the current pseudo-inverse matrix update cycle has been reached. The embodiments of the present application do not limit this.

[0107] Step 5022: If the pseudo-inverse matrix update period is currently reached, a pseudo-inverse calculation is performed on the convolution matrix composed of the tap coefficients of the analog equalizer to obtain the pseudo-inverse matrix to be used.

[0108] First, a convolution matrix composed of tap coefficients of an analog equalizer is obtained. The size of the convolution matrix and the position of the tap coefficients in the convolution matrix are preset. When obtaining the convolution matrix, the convolution matrix can be generated according to the preset tap coefficients in the convolution matrix and the preset size of the convolution matrix.

[0109] In the case where the first signal includes 2M+1 data, the size of the convolution matrix composed of tap coefficients may be (2M+1)×(2M+N), that is, the convolution matrix has 2M+1 rows and 2M+N columns. Figure 7 , is an example of the convolution matrix, taking M=4, N=5 as an example, where W0, W1, W2, W3 and W4 are 5 tap coefficients and the blank spaces are 0.

[0110] After obtaining the convolution matrix corresponding to the tap coefficients, the convolution matrix is ​​pseudo-inversely calculated to obtain the corresponding pseudo-inverse matrix. The size of the pseudo-inverse matrix is ​​(2M+N)×(2M+1), that is, the pseudo-inverse matrix has 2M+N rows and 2M+1 columns.

[0111] Specifically, the pseudo-inverse calculation method may be a Moore-Penrose generalized inverse (Moore-Penrose) method, a singular value decomposition (SVD) method, or the like.

[0112] Step 5023: If the pseudo-inverse matrix update period has not been reached, the latest calculated pseudo-inverse matrix is ​​obtained as the pseudo-inverse matrix to be used.

[0113] It should be noted that, in step 502, the pseudo-inverse matrix to be used may be recalculated according to step 5022 each time the tap coefficients are updated, so that the updated tap coefficients may be more accurate.

[0114] Step 503: multiply the pseudo-inverse matrix to be used and the first signal to obtain a second signal.

[0115] In implementation, after obtaining the pseudo-inverse matrix to be used, the pseudo-inverse matrix to be used is multiplied by the first signal to obtain the second signal. Specifically, the calculation formula may be as follows:

[0116] X′V·Y T

[0117] Among them, X′ is the second signal, V is the pseudo-inverse matrix, and Y T is the transpose of the first signal Y.

[0118] When the length of the first signal is 2M+1 and the size of the pseudo-inverse matrix to be used is (2M+N)×(2M+1), the second signal is a vector with a length of 2M+N.

[0119] Step 504: Adjust the tap coefficients of the analog equalizer according to the first signal and the second signal.

[0120] In implementation, the digital signal processor calculates an error value between the first signal and a target signal corresponding to the first signal, and adjusts the tap coefficients of the analog equalizer according to the error value, the second signal and a preset update step size.

[0121] When calculating the error value between the first signal and the target signal corresponding to the first signal, the Pth data in the first signal can be obtained. For the convenience of description, the Pth data in the first signal is referred to as the first data below. In addition, the second data in the target signal corresponding to the first signal at the same position as the first data is also obtained. The target signal is a digital signal corresponding to the training signal sent by the transmitter without being transmitted through the signal transmission channel.

[0122] In the case where the first signal is selected by the digital signal processor starting from the first data of the received training signal, the second signal is the Pth data in the target signal.

[0123] After the first data and the second data are acquired, the difference between the first data and the second data is calculated as the error value between the first signal and the corresponding target signal, which is recorded as e(n).

[0124] In addition, because the training signal sent by the transmitting end each time is the same, the second data can be pre-stored at the receiving end, so that when calculating the error value, the stored second data can be directly obtained.

[0125] In addition to calculating the error value, it is also necessary to select N consecutive data in the second signal as the gradient Δ of the error value.

[0126] The selection rule of N consecutive data in the second signal may be:

[0127] Starting from the Pth data in the second signal, N data are selected continuously. Based on this, when setting the value of P, it is necessary to satisfy that the Pth data to the last data in the second signal includes greater than or equal to N data.

[0128] In addition, because the second signal is calculated using a pseudo-inverse matrix, the data in the middle of the second signal has a higher accuracy, so the selected N data can be the N data in the middle of the second signal. When the length of the second signal is 2M+N, the N data in the middle of the second signal are the N data between the M+1th and the M+Nth. That is, P=M+1. When P=M+1 and the length of the first signal is 2M+1, the first data is also the data in the middle of the first signal.

[0129] The preset update step size μ can be set according to actual needs, for example, it can be set between 0.01 and 0.1.

[0130] Based on the above error value e(n), gradient Δ and preset update step size μ, the method for updating the tap coefficients can be shown in the following formula:

[0131] W′=W+μe(n)Δ

[0132] Where W is the vector corresponding to the tap coefficients before the update, which is composed of N tap coefficients before the update. W′ is the vector corresponding to the tap coefficients after the update, which is composed of N tap coefficients after the update. μe(n)Δ can obtain a vector of length N, and each element in the vector is added to each element in W to obtain W′.

[0133] After the tap coefficients are properly updated, the digital signal processor outputs the updated tap coefficients to the digital-to-analog converter, which converts the updated tap coefficients into analog signals and outputs them to the analog equalizer. In this way, the analog equalizer can use the updated tap coefficients for equalization when equalizing the subsequently received signal (e.g., the second input signal).

[0134] In an embodiment of the present application, an inverse transform is performed on the equalized signal to obtain a signal that is similar to a signal that has not been equalized after reception, and this signal is used to replace the signal that has not been equalized after reception to adjust the tap parameters of the analog equalizer. In this way, there is no need to separate the received signal, and there is no need to input a large optical power, so the requirements for the transmitting end are also reduced, which can reduce the cost of the transmitting end to a certain extent.

[0135] like Figure 8 As shown, in the case where the inverse transform is implemented by convolution, the equalizer adjustment method provided in the embodiment of the present application may include the following processing steps:

[0136] Step 801: Obtain a first signal after being equalized by an analog equalizer.

[0137] In implementation, the specific processing of step 801 and Figure 5The specific processing of step 501 is similar and will not be repeated here.

[0138] It should be noted that the difference between the specific processing of step 801 and step 501 is that the number of data included in the first signal obtained in step 801 is different from that in step 501. Specifically, the number of data included in the first signal in step 801 may be 2N+A+B-1. Among them, A and B may be integers close to N / 2, and A and B may be the same or different. A and B are introduced in step 802.

[0139] Step 802: Obtain the time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0140] In implementation, because the characteristics of the signal transmission channel do not change very quickly, in order to save computing resources, it is not necessary to recalculate the time domain data to be used every time the tap coefficients are updated.

[0141] Based on this, a time domain data update cycle can be set. When acquiring the time domain data to be used, how to acquire the time domain data to be used can be determined based on whether the time domain data update cycle is currently reached.

[0142] Specifically, Fig. 9 As shown, step 802 can be replaced by the following steps 8021 to 8023.

[0143] Step 8021: Determine whether the time domain data update cycle has been reached.

[0144] The time domain data update cycle in step 8021 and Figure 6 The pseudo inverse matrix update period in step 5021 shown in FIG. 5021 may be the same as that in step 8021, and the method for determining whether the time domain data update period has been reached is the same as that in step 8021. Figure 6 The method for determining whether the pseudo-inverse matrix update period has been reached in step 5021 is the same as that shown, and will not be described in detail here.

[0145] Step 8022: If the time domain data update period is reached, the tap coefficients of the analog equalizer are subjected to fast Fourier transform (FFT) to obtain frequency domain data corresponding to the tap coefficients of the analog equalizer. Then, the reciprocal of each element in the frequency domain data is calculated to obtain the spectrum data corresponding to the tap coefficients of the analog equalizer. Finally, the spectrum data is subjected to inverse Fourier transform (IFFT) to obtain the time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0146] Specifically, when performing FFT on the tap coefficients, in order to obtain a better spectrum resolution corresponding to the tap coefficients, A zeros may be added before the first element in the vector corresponding to the tap coefficients, and B zeros may be added after the last element. Then, FFT is performed on the adjusted vector to obtain the frequency domain result F.

[0147] F=[F1,F2,F3,...,F N+A+B ]

[0148] Then calculate the reciprocal of each element in F to obtain the corresponding spectrum G.

[0149] G=[1 / F1, 1 / F2, 1 / F3,..., 1 / F N+A+B ]

[0150] Finally, perform IFFT on G to obtain the time domain data V to be used corresponding to the tap coefficients.

[0151] V = [v1, v2, v3, ..., v N+A+B ]

[0152] Step 8023: If the time domain data update period has not been reached, the time domain data corresponding to the tap coefficients of the analog equalizer calculated most recently are obtained as the time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0153] It should be noted that, in step 802, the time domain data to be used may be recalculated according to step 8022 each time the tap coefficients are updated, so that the updated tap coefficients may be more accurate.

[0154] Step 803: Convolve the first signal and the time domain data to be used to obtain a second signal.

[0155] In implementation, the first signal and the time domain data to be used are convolved without zero padding to obtain the second signal. Specifically, during the convolution, the time domain data to be used is slid with a step length of 1 to obtain the second signal with a length of N.

[0156] See also Fig.10 , taking N=5, A=B=2 as an example, the calculation process of the first signal and the time domain data V to be used without zero padding convolution is shown. When N=5, A+B=2, the length of the first signal is 2N+A+B-1=13, and the length of the time domain data V to be used is N+A+B=9. Fig.10 The convolution process shown in the figure, the five elements x1′, x2′, x3′, x4′, x5′ included in the obtained second signal are as follows:

[0157] x1′=y(1)v1+y(2)v2+y(3)v3+y(4)v4+y(5)v5+y(6)v6+y(7)v7+y(8)v8+y(9)v9

[0158] x2′=y(2)v1+y(3)v2+y(4)v3+y(5)v4+y(6)v5+y(7)v6+y(8)v7+y(9)v8+y(10)v9

[0159] x3′=y(3)v1+y(4)v2+y(5)v3+y(6)v4+y(7)v5+y(8)v6+y(9)v7+y(10)v8+y(11)v9

[0160] x4′=y(4)v1+y(5)v2+y(6)v3+y(7)v4+y(8)v5+y(9)v6+y(10)v7+y(11)v8+y(12)v9

[0161] x5′=y(5)v1+y(6)v2+y(7)v3+y(8)v4+y(9)v5+y(10)v6+y(11)v7+y(12)v8+y(13)v9

[0162] Step 804: Adjust the tap coefficients of the analog equalizer according to the first signal and the second signal.

[0163] The specific processing of step 804 is similar to step 504, which is not described here. The difference from step 504 is that when selecting the first data in the first signal in step 804, a data in the middle position of the first signal can be selected as the first data. In addition, the gradient Δ corresponding to the error value in step 804 is the second signal itself.

[0164] In an embodiment of the present application, an inverse transform is performed on the equalized signal to obtain a signal that is similar to a signal that has not been equalized after reception, and this signal is used to replace the signal that has not been equalized after reception to adjust the tap parameters of the analog equalizer. In this way, there is no need to separate the received signal, and there is no need to input a large optical power, so the requirements for the transmitting end are also reduced, which can reduce the cost of the transmitting end to a certain extent.

[0165] Based on the same technical concept, the embodiment of the present application also provides an adjustment device for an equalizer, such as Fig.11 As shown, the device comprises:

[0166] The acquisition module 1110 is used to acquire the first signal after being equalized by the analog equalizer. Figure 5 and Figure 6 A detailed description of step 501 in the embodiment shown, or the above Figure 8 and Fig. 9 The detailed description of step 801 in the illustrated embodiment will not be repeated here.

[0167] The inverse transformation module 1120 is used to perform an inverse transformation on the first signal to obtain a second signal. Figure 5 Detailed description of step 502 and step 503, or the above Figure 6 Detailed description of steps 5021 to 5023 in the embodiment shown, or the above Figure 8 Detailed description of step 802, or the above Fig. 9 The detailed description of steps 8021 to 8023 will not be repeated here.

[0168] The adjustment module 1130 is used to adjust the tap coefficients of the analog equalizer according to the first signal and the second signal. Figure 5 and Figure 6 Detailed description of step 504, or the above Figure 8 and Fig. 9 The detailed description of step 804 is omitted here.

[0169] In a possible implementation, the adjustment module 1130 is configured to:

[0170] Calculating an error value between the first signal and a target signal corresponding to the first signal;

[0171] The tap coefficients of the analog equalizer are adjusted according to the error value, the second signal and a preset update step size.

[0172] In a possible implementation, the inverse transform module 1120 is used to:

[0173] Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer;

[0174] The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

[0175] In a possible implementation, the inverse transform module 1120 is used to:

[0176] A pseudo-inverse calculation is performed on the convolution matrix composed of the tap coefficients of the analog equalizer to obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

[0177] In a possible implementation, the inverse transform module 1120 is used to:

[0178] Acquire time domain data to be used corresponding to the tap coefficients of the analog equalizer;

[0179] The first signal and the time domain data to be used are convolved to obtain a second signal.

[0180] In a possible implementation, the inverse transform module 1120 is used to:

[0181] Performing a fast Fourier transform (FFT) on the tap coefficients of the analog equalizer to obtain frequency domain data corresponding to the tap coefficients of the analog equalizer;

[0182] Calculating the reciprocal of each element in the frequency domain data to obtain frequency spectrum data corresponding to the tap coefficients of the analog equalizer;

[0183] Performing an inverse fast Fourier transform (IFFT) on the frequency spectrum data to obtain time domain data to be used corresponding to the tap coefficients of the analog equalizer.

[0184] In an embodiment of the present application, an inverse transform is performed on the equalized signal to obtain a signal that is similar to a signal that has not been equalized after reception, and this signal is used to replace the signal that has not been equalized after reception to adjust the tap parameters of the analog equalizer. In this way, there is no need to separate the received signal, and there is no need to input a large optical power, so the requirements for the transmitting end are also reduced, which can reduce the cost of the transmitting end to a certain extent.

[0185] It should be noted that: the equalizer adjustment device provided in the above embodiment only uses the division of the above functional modules as an example when performing equalizer adjustment. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the digital signal processor is divided into different functional modules to complete all or part of the functions described above. Figure 5 , Figure 6 , Fig. 9 and Fig.10 The provided equalizer adjustment method embodiments belong to the same concept, and the specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0186] An embodiment of the present application provides a computer program product. When the computer program product is run on a receiver, the receiver executes the equalizer adjustment method as described in the first aspect above.

[0187] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the device, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by the device or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape, etc.), an optical medium (e.g., a digital video disk (DVD), etc.), or a semiconductor medium (e.g., a solid-state hard disk, etc.).

[0188] A person skilled in the art will understand that all or part of the steps to implement the above embodiments may be accomplished by hardware or by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a disk or an optical disk, etc.

[0189] The above description is only one embodiment of the present invention and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for adjusting an equalizer, characterized in that: The method comprises: Acquire a first signal obtained by equalizing a signal output from a transmission channel by an analog equalizer; Performing an inverse transformation on the first signal to obtain a second signal, wherein the second signal is an estimated signal of a signal output by the transmission channel; Calculating an error value between the first signal and a target signal corresponding to the first signal; The tap coefficients of the analog equalizer are adjusted according to the error value, the second signal and a preset update step size.

2. The method according to claim 1, characterized in that The inverse transformation of the first signal to obtain a second signal includes: Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer; The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

3. The method according to claim 2, characterized in that The step of obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer includes: A pseudo-inverse calculation is performed on a convolution matrix composed of tap coefficients of the simulated equalizer to obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the simulated equalizer.

4. The method according to claim 1, characterized in that: The inverse transformation of the first signal to obtain a second signal includes: Acquire time domain data to be used corresponding to the tap coefficients of the analog equalizer; The first signal and the time domain data to be used are convolved to obtain a second signal.

5. A receiver, characterized in that: include: An analog equalizer, used for performing equalization processing on the received first input signal; an analog-to-digital converter, configured to perform analog-to-digital conversion on the equalized first input signal to obtain a first signal, wherein the first signal is a digital signal; A digital signal processor is used to perform an inverse transformation on the first signal to obtain a second signal, wherein the second signal is an estimated signal of the signal obtained after analog-to-digital conversion of the received first input signal; calculate the error value between the first signal and a target signal corresponding to the first signal; and adjust the tap coefficient of the analog equalizer according to the error value, the second signal and a preset update step size.

6. The receiver according to claim 5, characterized in that The digital signal processor is used for: Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer; The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

7. The receiver according to claim 6, characterized in that The digital signal processor is used for: A pseudo-inverse calculation is performed on the convolution matrix composed of the tap coefficients of the analog equalizer to obtain a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer.

8. The receiver according to claim 5, characterized in that The digital signal processor is used for: Acquire time domain data to be used corresponding to the tap coefficients of the analog equalizer; The first signal and the time domain data to be used are convolved to obtain a second signal.

9. The receiver according to any one of claims 5 to 8, characterized in that: The digital signal processor is further used to output the adjusted tap coefficients to the analog equalizer; The analog equalizer is further used to perform equalization processing on the received second input signal according to the adjusted tap coefficients.

10. The receiver according to any one of claims 5 to 8, characterized in that: The first input signal is an optical signal, and the receiver further includes: A photoelectric converter, used for performing photoelectric conversion on the equalized first input signal to obtain a third signal, wherein the third signal is an electrical signal; The analog-to-digital converter is used to perform analog-to-digital conversion on the third signal to obtain a first signal.

11. An adjustment device for an equalizer, characterized in that: The device comprises: An acquisition module, used for acquiring a first signal obtained after equalization processing of a signal output by a transmission channel by an analog equalizer; an inverse transform module, configured to perform an inverse transform on the first signal to obtain a second signal, wherein the second signal is an estimated signal of a signal output by the transmission channel; An adjustment module is used to calculate an error value between the first signal and a target signal corresponding to the first signal; and adjust the tap coefficient of the analog equalizer according to the error value, the second signal and a preset update step size.

12. The device according to claim 11, characterized in that The inverse transformation module is used for: Obtaining a pseudo-inverse matrix to be used corresponding to the tap coefficients of the analog equalizer; The first signal is processed according to the pseudo-inverse matrix to be used to obtain a second signal.

13. A signal transmission system, characterized in that: The signal transmission system comprises a transmitter and a receiver as claimed in any one of claims 5-10.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium comprises instructions, and when the computer-readable storage medium is run in a receiver, the receiver executes the equalizer adjustment method according to any one of claims 1 to 4.

Images