CN116781462A - An equalizer tap adjustment method, equipment, storage medium and system - Google Patents

Abstract

Embodiments of the present application provide an equalizer tap adjustment method, device, storage medium, and system. The method is applied to a receiving device. The receiving device includes a decision feedback equalizer, and the decision feedback equalizer includes at least one Floating tap; the method includes periodically acquiring an input signal, performing equalization processing on the input signal to obtain an equalized signal; determining a decision signal from at least one preset transmission signal according to the equalized signal, and determining the decision signal according to the equalized signal. Determine the signal and the equalization signal to obtain an error signal; update the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal cached in the at least one floating tap; according to the at least one floating tap The accumulated error value determines at least one target tap among the at least one floating tap. This application is used to eliminate reflected ISI with low power consumption.

Description

An equalizer tap adjustment method, equipment, storage medium and system

Technical field

The present application relates to the field of communication technology, and specifically to an equalizer tap adjustment method, equipment, storage medium and system.

Background technique

With the rapid development of digital signal technology towards high speed and large capacity, the demand for high-speed signal processing technology is becoming more and more urgent. Inter Symbol Interference (ISI) generated during the signal transmission process is a key factor restricting the increase in signal rate. ISI will cause pulse broadening, cause the signal voltage amplitude to be unstable, cause signal data edge jitter, and lead to channel errors. The bit rate (Bit Error Ratio, BER) increases. In the existing technology, ISI in signals can be eliminated through channel equalization. Decision Feedback Equalizer (DFE), as a commonly used equalization method, can be used to eliminate the effects of ISI and part of the noise and improve the signal-to-noise ratio.

Take the third-order DFE as an example to illustrate. As shown in Figure 1, the third-order DFE includes three taps, an adder and a decider. Each tap includes an adder, a multiplier and a register. For each received signal y[n], the registers in the taps of the third-order DFE store delayed signals respectively. Part of the ISI in the received signal y[n] can be eliminated through this delayed signal. Usually, the impulse response of the channel decreases rapidly as time increases. Therefore, only a small number of taps can be considered to eliminate most of the ISI in the signal and achieve better equalization effects. Since the impedance in the channel is discontinuous, signal reflection occurs in the channel. For long-distance channels, the ISI caused by signal reflection is large and cannot be ignored. Figure 2 shows a channel impulse response with reflections. It can be seen from Figure 2 that, except for the larger coefficient values of the first few taps, the coefficient values of the taps at n21 and n=25, that is, |h[21]| and |h[25]| are significantly larger than those of other taps. coefficient value. Therefore, targeted elimination of the ISI corresponding to these taps with larger coefficient values can effectively improve SNR and reduce the bit error rate.

In order to eliminate reflected ISI, it is necessary to first determine the position of reflected ISI, and then set the corresponding tap switch in the DFE according to the position of reflected ISI to eliminate reflected ISI. In this method, the ISI generated at the fixed position can be eliminated by turning on the tap at the fixed position, and after the reflected ISI position is determined, the tap at the reflected ISI position can be turned on to eliminate the reflected ISI at the corresponding position. The above method of eliminating reflected ISI requires first determining the location of the reflected ISI. In the prior art, when L floating taps are included, L times of N+1 adaptive DFE can be used to determine M target taps. At this time, N fixed taps and 1 candidate floating tap can be turned on each time. After convergence, the position of the floating tap and the corresponding tap coefficient are recorded. When all L times of adaptive DFE are completed, M floating taps with the largest absolute values of tap coefficients are selected from the L candidate floating taps as target taps. Reflected ISI is eliminated through M target taps.

However, the above method requires L times of N+1 adaptive DFE, waiting for each convergence to record the coefficients of the candidate taps, which consumes a lot of power and has a slow convergence speed.

Contents of the invention

In view of this, the present application provides an equalizer tap adjustment method, device, storage medium, and system to help solve the problems of high power consumption and long time when adjusting the equalizer tap in the prior art.

In a first aspect, embodiments of the present application provide a tap adjustment method of an equalizer, which is applied to a receiving device. The receiving device includes: a decision feedback equalizer, the decision feedback equalizer includes at least one floating tap; Historical decision signals are cached in at least one floating tap, and the historical decision signals cached in different floating taps have different signal acquisition times; the method includes:

Periodically acquire input signals, perform equalization processing on the input signals, and obtain equalized signals;

According to the equalized signal, a decision signal is determined from at least one preset transmission signal, and an error signal is obtained based on the decision signal and the equalized signal;

Update the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal cached in the at least one floating tap;

At least one target tap is determined among the at least one floating tap according to the accumulated error value of the at least one floating tap.

In an implementation manner of the first aspect, it also includes:

According to the decision signal, the historical decision signal buffered by each floating tap in the at least one floating tap is updated.

In an implementation of the first aspect, updating the error cumulative value of the at least one floating tap based on the error signal and the historical decision signal cached in the at least one floating tap includes:

For each floating tap in the at least one floating tap, according to the error signal and the historical decision signal buffered in the floating tap, the error signal and the historical decision signal buffered in the floating tap are both positive signals, or both are positive signals. When it is a negative signal, increase the error accumulation value of the floating tap by the preset threshold;

Alternatively, when one of the error signal and the historical decision signal buffered in the floating tap is a positive signal and the other is a negative signal, the accumulated error value of the floating tap is reduced by a preset threshold.

In an implementation manner of the first aspect, before determining at least one target tap among the at least one floating tap based on the accumulated error value of the at least one floating tap, the method further includes:

Detect whether the accumulated number of errors of the at least one floating tap reaches a preset number threshold;

Determining at least one target tap among the at least one floating tap according to the error cumulative value of the at least one floating tap includes:

When the error accumulation number of the at least one floating tap reaches the preset number threshold, at least one floating tap is determined among the at least one floating tap according to the error accumulation value of the at least one floating tap.

In an implementation manner of the first aspect, it also includes:

When the cumulative number of errors of the at least one floating tap does not reach the preset threshold, continue to perform the step of periodically acquiring the input signal, performing equalization processing on the input signal to obtain an equalized signal, and then proceed to the step of obtaining an equalized signal according to the error signal and The cached historical decision signal in the at least one floating tap updates the error accumulation value of the at least one floating tap until the error accumulation number of the at least one floating tap reaches a preset number threshold.

In an implementation of the first aspect, the method further includes:

The preset tap coefficient of the at least one target tap is updated according to the error signal and the historical decision signal cached in the at least one target tap.

In an implementation manner of the first aspect, determining at least one target tap in the at least one floating tap according to the error accumulation value of the at least one floating tap includes:

According to the error cumulative value of the at least one floating tap, a preset number of target taps are determined in the at least one floating tap in order of the absolute value of the error cumulative value from large to small.

In an implementation manner of the first aspect, determining at least one target tap in the at least one floating tap according to the error accumulation value of the at least one floating tap includes:

According to a preset order, the at least one floating tap is divided into at least one floating tap group; each floating tap group contains at least one floating tap;

Calculate the error cumulative value corresponding to each floating tap group in the at least one floating tap group according to the error cumulative value of the at least one floating tap group;

According to the accumulated error value corresponding to each floating tap group, the target tap is determined in the at least one floating tap group.

In a second aspect, embodiments of the present application provide a receiving device, including: a receiver, a decision feedback equalizer and a processor; wherein the decision feedback equalizer includes at least one floating tap; the at least one floating tap There are historical decision signals in the buffer, and the signal acquisition times of the historical decision signals in different floating tap buffers are different;

The receiver is used to periodically acquire input signals;

The decision feedback equalizer is used to equalize the input signal to obtain an equalized signal;

The decision feedback equalizer is further configured to determine a decision signal from at least one preset transmission signal according to the equalized signal;

The processor is used to obtain an error signal according to the decision signal and the equalization signal;

The processor is further configured to update the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal cached in the at least one floating tap;

The processor is further configured to determine at least one target tap among the at least one floating tap according to the accumulated error value of the at least one floating tap.

In an implementation manner of the second aspect, the decision feedback equalizer is further configured to update the historical decision signal buffered by each floating tap in the at least one floating tap according to the decision signal.

In an implementation manner of the second aspect, the processor is specifically configured to, for each floating tap in the at least one floating tap, based on the error signal and the historical decision signal cached in the floating tap, in the When the error signal and the historical decision signal cached in the floating tap are both positive signals, or both are negative signals, the accumulated error value of the floating tap is increased by the preset threshold;

Alternatively, when one of the error signal and the historical decision signal buffered in the floating tap is a positive signal and the other is a negative signal, the accumulated error value of the floating tap is reduced by a preset threshold.

In an implementation manner of the second aspect, the processor is further configured to detect whether the accumulated number of errors of the at least one floating tap reaches a preset number threshold;

The processor is specifically configured to determine at least one of the at least one floating taps based on the accumulated error value of the at least one floating tap when the error cumulative number of the at least one floating tap reaches a preset threshold. A floating tap.

In an implementation manner of the second aspect, the processor is further configured to trigger the receiver to perform the step of periodically acquiring input when the accumulated number of errors of the at least one floating tap does not reach a preset number threshold. signal, triggering the decision feedback equalizer to execute the step of equalizing the input signal to obtain an equalized signal to the step of determining a decision signal from at least one preset transmission signal according to the equalized signal, and re-executing the step according to The decision signal and the equalization signal, and the step of obtaining the error signal, update the error cumulative value of the at least one floating tap according to the error signal and the cached historical decision signal in the at least one floating tap until the The accumulated number of errors of at least one floating tap reaches the preset number threshold.

In an implementation manner of the second aspect, the processor is further configured to update the tap coefficient in the at least one target tap according to the error signal and the historical decision signal cached in the at least one target tap.

In an implementation manner of the second aspect, the processor is specifically configured to, according to the error cumulative value of the at least one floating tap, in the order of the absolute value of the error cumulative value from large to small, in the at least one A preset number of target taps are determined among the floating taps.

In an implementation manner of the second aspect, the processor is specifically configured to divide the at least one floating tap into at least one floating tap group according to a preset order; each floating tap group includes at least one floating tap

Calculate the error cumulative value corresponding to each floating tap group in the at least one floating tap group according to the error cumulative value of the at least one floating tap group;

According to the accumulated error value corresponding to each floating tap group, the target tap is determined in the at least one floating tap group.

In a third aspect, embodiments of the present application provide a computer-readable storage medium. The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned The method described in any one of the first aspects.

In a fourth aspect, embodiments of the present application provide a communication system, including: a sending device and a receiving device according to any one of the above second aspects.

Using the solution provided by the embodiments of the present application, the receiving device periodically obtains the input signal, and after equalizing the input signal with at least one fixed tap, obtains an equalized signal. Determine the decision signal based on the equalization signal, and obtain the error signal based on the decision signal and the equalization signal; update the error cumulative value of at least one floating tap based on the error signal and the historical decision signal cached in at least one floating tap; The accumulated error value of the tap determines at least one target tap among at least one floating tap. In this way, in this application, the error cumulative value of each floating tap can be updated according to the historical decision signal and error signal cached in each floating tap in each cycle, so that according to the The accumulated error value determines the target tap. In this application, only the error accumulation corresponding to each floating tap is calculated, the power consumption and implementation complexity are low, the target tap can be determined quickly, the delay is low, and the implementation speed is fast.

Description of drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without exerting any creative effort.

Figure 1 is an example diagram of a third-order decision feedback equalizer provided by an embodiment of the present application;

Figure 2 is an example diagram of a channel impact response scenario with reflected ISI provided by an embodiment of the present application;

Figure 3 is a schematic diagram of a scenario of a floating decision feedback equalizer provided by an embodiment of the present application;

Figure 4 is a schematic structural diagram of a receiving device provided by an embodiment of the present application;

Figure 5 is a schematic flow chart of tap adjustment of an equalizer provided by an embodiment of the present application;

Figure 6 is a schematic flowchart of another equalizer tap adjustment method provided by an embodiment of the present application;

Figure 7 is a schematic structural diagram of another receiving device provided by an embodiment of the present application;

Figure 8 is a schematic structural diagram of a decision feedback equalizer provided by an embodiment of the present application;

Figure 9 is a schematic structural diagram of another receiving device provided by an embodiment of the present application;

Figure 10 is a schematic structural diagram of a communication system provided by an embodiment of the present application;

Figure 11 is a schematic diagram of a communication system provided by an embodiment of the present application.

Detailed ways

In order to better understand the technical solution of the present application, the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

It should be clear that the described embodiments are only some of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of this application.

The terminology used in the embodiments of the present application is only for the purpose of describing specific embodiments and is not intended to limit the present application. As used in the embodiments and the appended claims, the singular forms "a," "the" and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise.

It should be understood that the term "and/or" used in this article is only an association relationship describing related objects, indicating that there can be three relationships, for example, A and/or B, which can mean: A alone exists, and A and A exist simultaneously. B, there are three situations of B alone. In addition, the character "/" in this article generally indicates that the related objects are an "or" relationship.

Before giving a detailed introduction to the embodiments of the present application, terminology that is or may be applied in the embodiments of the present application is first explained.

UI: English is unit interval, literally translated as unit interval or single signal period, which represents the signal period after modulation. At the same time, UI is often used to refer to the modulated signal itself. For example, expressions such as "current UI" and "previous UI" actually refer to the current signal and the previous signal.

Tap: Tap in English, which represents a feedback coefficient of a multi-tap decision feedback equalizer (multi-tap Decision FeedbackEqualizer, multi-tap DFE) (sometimes called a multi-order decision feedback equalizer).

In mobile communications and high-speed wireless data communications, multipath effects, limited channel bandwidth, and imperfections in channel characteristics lead to inevitable inter-code interference during data transmission, which has become the main factor affecting communication quality, and the balance of the channel Technology can eliminate inter-symbol interference and noise and reduce bit error rates. Among them, the decision feedback equalizer (DFE) is a very effective and widely used measure to deal with multipath interference. Since the impedance in the channel is discontinuous, signal reflection occurs in the channel. For long-distance channels, the ISI caused by signal reflection is large and cannot be ignored. Figure 2 shows a channel impulse response with reflections. It can be seen from Figure 2 that, in addition to the larger coefficient values of the first few taps, the coefficient values of the taps at n=21 and n=25, that is, |h[21]| and |h[25]| are significantly larger than others. The coefficient value of the tap. Therefore, targeted elimination of the ISI corresponding to these taps with larger coefficient values can effectively improve SNR and reduce the bit error rate.

In order to eliminate reflected ISI, it is necessary to first determine the position of reflected ISI, and then set the corresponding tap switch in the DFE according to the position of reflected ISI to eliminate reflected ISI. In the prior art, reflected ISI is eliminated through a floating tap decision feedback equalizer. Among them, the floating tap decision feedback equalizer includes a fixed tap to eliminate ISI at the first position and multiple floating taps. By activating at least one floating tap among the multiple floating taps, the reflected ISI at other positions can be eliminated, such as As shown in Figure 3. In order to reduce power consumption, the ISI at a position with a large channel impulse response coefficient can be selected for equalization elimination. In this case, the target tap needs to be determined among multiple floating taps. When L floating taps are included, L times of N+1 adaptive DFE can be used to determine M target taps. At this time, N fixed taps and 1 floating tap can be turned on each time. After convergence, the position of the floating tap and the corresponding tap coefficient are recorded. When all L times of adaptive DFE are completed, M floating taps with the largest absolute values of tap coefficients are selected from the L floating taps as target taps. Eliminate reflective ISI through M target floating DFEs.

This method requires L times of adaptive DFE and waits for each convergence to record the tap coefficient of the floating tap. It consumes large power and has a slow convergence speed.

To address the above problems, embodiments of the present application provide a tap adjustment method for an equalizer. The receiving device periodically acquires input signals, and equalizes the input signals with at least one fixed tap to obtain an equalized signal. Determine the decision signal based on the equalization signal, and obtain the error signal based on the decision signal and the equalization signal; update the error cumulative value of at least one floating tap based on the error signal and the historical decision signal cached in at least one floating tap; The accumulated error value of the tap determines at least one target tap among at least one floating tap. In this way, in this application, the error cumulative value of each floating tap can be updated according to the historical decision signal and error signal cached in each floating tap in each cycle, so that according to the The accumulated error value determines the target tap. In this application, only the error accumulation corresponding to each floating tap is calculated, the power consumption and implementation complexity are low, the target tap can be determined quickly, the delay is low, and the implementation speed is fast. The details are explained below.

Refer to Figure 4, which is a schematic structural diagram of a receiving device provided by an embodiment of the present application. As shown in Figure 4, the receiving device includes: a decision feedback equalizer 40 and a processor 47. The decision feedback equalizer 40 includes at least one floating tap 42 . Historical decision signals are cached in at least one floating tap 42 , and the historical decision signals cached in different floating taps 42 have different signal acquisition times.

As a possible implementation manner, as shown in FIG. 4 , the above-mentioned decision feedback equalizer 40 also includes at least one fixed tap (Fixed tap) 41 , a slicer (slicer) 43 and an adder 44 . Among them, at least one fixed tap 41 has cached historical decision signals, and the signal acquisition time of the historical decision signal cached in at least one fixed tap 41 is different from the signal acquisition time of the historical decision signal cached in at least one floating tap 42, and they are different. The signal acquisition time of the historical decision signal buffered in the fixed tap 41 is also different.

At least one fixed tap 41 and at least the first floating tap 42 are connected to the first input end of the adder 44, and the second input end of the adder 44 is used to receive the input signal. The output terminal of the adder 44 is connected to the input terminal of the decider 43 . The output of the decider 43 is connected to the processor 47 and the register 45 in at least one fixed tap 41 .

It should be noted that the fixed tap 41 refers to a tap in the decision feedback equalizer that is in working state. The floating tap 42 is a tap in the decision feedback equalizer that is in an inoperative state.

It should be noted that each tap (including the fixed tap 41 and the floating tap 42) included in the decision feedback equalizer 40 includes a register 45 and a multiplier 46. Among them, register 45 is used to cache historical decision signals. The multiplier 46 is used to equalize the input signal.

It should be noted that the signal acquisition time of the historical judgment signal refers to the time when the historical judgment signal is acquired.

Refer to FIG. 5 , which is a schematic flow chart of a tap adjustment method of an equalizer provided by an embodiment of the present application. This method is applied to the receiving device as shown in Figure 4. The methods include:

Step S501: Periodically acquire the input signal, perform equalization processing on the input signal, and obtain an equalized signal.

In the embodiment of the present application, since ISI will be generated during signal transmission, the receiving device needs to perform ISI elimination processing on the received input signal. At this time, the receiving device can periodically acquire the input signal. For each input signal obtained, the input signal is equalized to eliminate part of the ISI and obtain an equalized signal.

As a possible implementation, the receiving device can transmit the received input signal to at least one fixed tap in the decision feedback equalizer, and through the equalization processing of the at least one fixed tap, part of the ISI in the input signal is eliminated to obtain an equalized signal. .

As a possible implementation, the receiving device receives a signal (UI) every cycle, which is the input signal. After receiving the input signal, the receiving device needs to equalize the input signal. At this time, at least one fixed tap in working state in the decision feedback equalizer of the receiving device can perform equalization processing on the input signal. At this time, each fixed tap in at least one fixed tap reads the stored historical decision signal from its internal register, and compares the read historical decision signal cached in the fixed tap internal register with the fixed tap internal multiplier The corresponding tap coefficients are multiplied, and the product is inverted and transmitted to the adder. The adder adds the signal to the input signal to obtain an equalized signal. An equalized signal is a signal obtained by eliminating part of the ISI in the input signal through at least one fixed tap. Among them, at least one fixed tap is used to eliminate which part of the ISI in the input signal can be set in advance. At this time, the signal cached in the register in at least one fixed tap is the historical decision signal corresponding to the ISI that needs to be eliminated. For example, the input signal acquired in the current period contains ISI. This ISI is formed due to the input signal of the previous period and the input signals of other earlier periods. In order to eliminate ISI, the input signal of the previous period and the input signals of other earlier periods contained in the input signal acquired in the current period need to be eliminated, then message ISI can be achieved. Therefore, it is possible to set in advance the input signals at which times included in the input signal acquired in the current period that need to be eliminated, for example, the input signal of the previous period included in the input signal of the current period needs to be eliminated. At this time, the tap of the historical decision signal cached in the register is the decision signal acquired in the previous cycle and is determined as a fixed tap, so that the fixed tap can be used to eliminate the input signal of the previous cycle contained in the input signal acquired in the current cycle.

As a possible implementation method, the following formula can be used based on the input signal and the historical decision signal cached in each register in at least one fixed tap. Calculate the balanced signal. Among them, w[n] represents the balanced signal in the nth period, y[n] represents the input signal in the nth period, j represents the jth fixed tap, N represents the number of at least one fixed tap, and c _j represents The tap coefficient corresponding to the multiplier in the j-th fixed tap, data_reg[n,j] represents the historical decision signal cached in the register in the j-th fixed tap in the n-th cycle, n is an integer greater than 0.

Step S502: Determine a decision signal from at least one preset transmission signal based on the equalized signal, and obtain an error signal based on the decision signal and the equalized signal.

Wherein, at least one transmission signal is at least one signal that the sending device may send to the receiving device.

In the embodiment of the application, after the receiving device obtains the equalized signal, it can determine the signal closest to the equalized signal among the multiple preset transmission signals based on the equalized signal, and determine it as the decision signal. After the decision signal is determined, the error signal can be calculated based on the decision signal and the equalization signal.

As a possible implementation method, the error signal can be calculated through the formula error[n]=x[n]-w[n] based on the decision signal and the equalization signal. Among them, error[n] represents the error signal in the nth cycle, and x[n] represents the decision signal in the nth cycle.

It should be noted that in the embodiment of the present application, various signals that may be transmitted by the sending device are preset in the receiving device. After the receiving device obtains the balanced signal, it can determine the signal closest to the balanced signal among the signals that may be transmitted by the preset sending device based on the equalized signal, and determine it as the sending device received in the current cycle. The emitted signal is determined to be the decision signal.

After the decision signal is determined, ideally, the equalization signal and the decision signal should be the same. However, since the equalized signal also contains noise such as reflected ISI, the equalized signal is not exactly the same as the decision signal. At this time, the error signal can be calculated based on the decision signal and the equalized signal, that is, the interference signals such as noise and reflected ISI contained in the equalized signal are calculated.

Step S503: Update the error cumulative value of at least one floating tap according to the error signal and the historical decision signal buffered in at least one floating tap.

In this embodiment of the present application, the error accumulation value of the floating tap is used to identify the size of the ISI of the historical decision signal buffered in the floating tap internal register to the input signal of the current cycle. The larger the accumulated error value of the floating tap is, the greater the interference of the historical decision signal cached in the register in the floating tap to the input signal of the current period. After acquiring the error signal, the receiving device can update the accumulated error value of each floating tap based on the error signal and the historical decision signal buffered in each floating tap.

As a possible implementation, updating the error cumulative value of at least one floating tap based on the error signal and the historical decision signal cached in at least one floating tap includes:

For each floating tap in at least one floating tap, according to the error signal and the historical decision signal buffered in the floating tap, when the error signal and the historical decision signal buffered in the floating tap are both positive signals, or both are negative signals , increase the accumulated error value of the floating tap by the preset threshold;

Or, when one of the error signal and the historical decision signal buffered in the floating tap is a positive signal and the other is a negative signal, the accumulated error value of the floating tap is reduced by a preset threshold.

Specifically, after acquiring the error signal of the current period, the receiving device can update the accumulated error value of each floating tap based on the error signal of the current period. Since the historical decision signal buffered in the floating tap is the historical decision signal buffered in the register in the floating tap and the error signal of the current period relative to the preset reference signal, such as a 0-level signal, the sign may be the same. The sign may be different in sign. When the historical decision signal cached in the floating tap and the error signal of the current period are of the same sign relative to the preset reference signal, it means that the error signal contains the same signal as the historical decision signal cached in the floating tap. To enhance the error signal, at this time, it is necessary to add a preset threshold to the accumulated error value corresponding to the floating tap. When the historical decision signal cached in the floating tap and the error signal of the current period are not the same relative to the preset reference signal, it means that the error signal contains the same signal as the historical decision signal cached in the floating tap. , which weakens the error signal. At this time, the error accumulation value corresponding to the floating tap needs to be reduced by the preset threshold.

Based on this, after acquiring the error signal of the current period, the receiving device reads the historical decision signal cached in the floating tap for each floating tap of at least one floating tap, and detects the historical decision signal and the error signal of the current period. Are they all positive signals, or are they all negative signals. When the historical decision signal and the error signal of the current period are both positive signals, or both are negative signals, it means that the historical decision signal and the error signal of the current period are signals with the same sign. At this time, the history cached in the floating tap The decision signal has an enhancing effect on the error signal of the current period. Therefore, the error accumulation value of the floating tap increases by the preset threshold. For example, the error accumulation value of a floating tap is increased by 1. When one of the historical decision signal and the error signal of the current period is a positive signal and the other is a negative signal, for example, the historical decision signal is a positive signal and the error signal of the current period is a negative signal, or the historical decision signal is a negative signal and the error signal of the current period is a negative signal. When the error signal is a positive signal, the historical decision signal and the error signal of the current period are signals with different signs. At this time, the historical decision signal buffered in the floating tap has a weakening effect on the error signal of the current period. Therefore, the floating tap The accumulated error value is reduced by the preset threshold. For example, the error accumulation value of a floating tap is reduced by 1. Through the above method, the error accumulation value of each floating tap in the at least one floating tap can be updated.

It should be noted that a positive signal refers to a signal that is greater than 0 relative to the preset reference signal, and a negative signal refers to a signal that is less than 0 relative to the preset reference signal.

It should be noted that the preset threshold is set in advance according to actual needs, and the amount of change in the accumulated error value is changed each time. It can be a value of 1, or it can be a value of 2 or other other values, which is not limited in this application.

As a possible implementation, for each floating tap in at least one floating tap, the receiving device uses the formula Error_Accum[n, i]=Error_Accum[n-1 based on the error signal and the historical judgment signal buffered in the floating tap. ,i]+sign(error[n])×sign(data_reg[n,i]), update the error cumulative value of the floating tap. Among them, Error_Accum[n, i] represents the accumulated error value of the i-th floating tap in the n-th period; Error_Accum[n-1, i] represents the error of the i-th floating tap in the n-1 period. Cumulative error value, sign(error[n]) indicates the positive or negative of the error signal relative to the preset reference signal in the nth period; sign(data_reg[n,i]) indicates the i-th value in the nth period The positive and negative nature of the historical decision signal of the floating tap buffer relative to the preset reference signal.

It should be noted that the preset reference signal is set in advance and can be a 0-level signal.

Step S504: Determine at least one target tap among the at least one floating tap based on the accumulated error value of the at least one floating tap.

In the embodiment of the present application, the receiving device can update the error cumulative value of each floating tap in each cycle through the above step S503. After reaching the cumulative number of times, the receiving device can update the error cumulative value of each floating tap according to the cumulative number of errors. On at least one floating tap, at least one target tap is selected. For example, at least one target tap can be selected in descending order of the cumulative error value of each floating tap.

As a possible implementation, determining at least one target tap in at least one floating tap according to the error cumulative value of at least one floating tap includes:

According to the error cumulative value of at least one floating tap, a preset number of target taps are determined in at least one floating tap in order of the absolute value of the error cumulative value from large to small.

Specifically, when the receiving device needs to determine the target tap, it can determine the target tap in at least one floating tap according to the error cumulative value of each floating tap and in order from the largest to the smallest absolute value of the error cumulative value of each floating tap. Outputs a preset number of target taps. That is, a preset number of floating taps with the largest absolute value of the error cumulative value among at least one floating tap are determined as target taps.

As a possible implementation, determining at least one target tap in at least one floating tap according to the error cumulative value of at least one floating tap includes:

Divide at least one floating tap into at least one group of floating taps according to a preset order; calculate the error cumulative value corresponding to each group of floating taps in the at least one group of floating taps based on the error cumulative value of the at least one floating tap; based on each group of floating taps The corresponding accumulated error value determines the target tap in at least one group of floating taps.

In the embodiment of the present application, in order to meet user requirements, at least one floating tap can also be divided into at least one group of floating taps. Each group of floating taps contains at least one floating tap. In this way, the cumulative error value corresponding to each group can be calculated based on the cumulative error value of at least one floating tap included in each group of floating taps. For example, the cumulative error value of at least one floating tap included in each group of floating taps The absolute values are accumulated to calculate the accumulated error value corresponding to each group of floating taps. In this way, at least one target tap group can be determined in at least one group of floating taps based on the accumulated error value corresponding to each group, so that the floating taps included in at least one group of floating tap groups can be determined as the target taps.

Refer to FIG. 6 , which is a schematic flow chart of another equalizer tap adjustment method provided by an embodiment of the present application. This method is applied to the receiving device as shown in Figure 4. Compared with the embodiment described in Figure 5, the embodiment of the present application adds the step of updating the historical decision signal cached in the floating tap. The method includes:

Step S601: Periodically acquire the input signal, perform equalization processing on the input signal, and obtain an equalized signal.

For details, please refer to step S501, which will not be described again here.

Step S602: Determine a decision signal from at least one preset transmission signal according to the equalized signal, and obtain an error signal based on the decision signal and the equalized signal.

For details, please refer to step S502, which will not be described again here.

Step S603: Update the error cumulative value of at least one floating tap according to the error signal and the historical decision signal buffered in at least one floating tap.

For details, please refer to step S503, which will not be described again here.

Step S604: Update the historical decision signal buffered by each floating tap in at least one floating tap according to the decision signal.

In the embodiment of the present application, since the historical decision signal buffered in at least one floating tap is not static, but after the decision signal is determined in each cycle, the currently stored historical decision signal will be updated. At this time, for each floating tap in at least one floating tap, the historical decision signal cached in the floating tap is transmitted to the floating tap after the floating tap in a preset order, and the floating tap receives the previous floating tap of the floating tap. The floating tap transmits the historical decision signal and caches the received historical decision signal. According to the preset sequence, when there is no other tap before the first floating tap, the decision signal is sent to the first floating tap, and the first floating tap caches the received decision signal as a historical decision signal.

As a first possible implementation, the decision feedback equalizer contains at least one fixed tap, and according to the preset sequence, when at least one fixed tap is before the first floating tap, the first floating tap needs to receive its The historical decision signal cached in the previous fixed tap is stored. The decision signal determined in the current cycle can be cached in the first fixed tap.

As a possible implementation, the decision feedback equalizer includes at least one fixed tap and at least one floating tap. Historical decision signals are cached in at least one fixed tap and at least one floating tap, and the histories cached in different taps are The signal acquisition time of the decision signal is different. At this time, after the decision signal is determined, the historical decision signals cached in all taps in the decision feedback equalizer need to be updated. At this time, for each tap of at least one fixed tap and at least one floating tap in the decision feedback equalizer, according to the decision signal, the historical decision signal saved in the tap is transmitted to the subsequent tap in a preset order, and Receive the historical decision signal sent by the previous tap and store the historical decision signal.

In the embodiment of this application, since the input signal needs to be acquired periodically, after the error cumulative value of each floating tap is updated in the current cycle, the historical decision stored in each tap needs to be updated based on the decision signal determined in the current cycle. Signal. At this time, for each tap of at least one fixed tap and at least one floating tap, the historical decision signal saved in the tap is transmitted to the next tap of the tap in a preset order, and the tap receives the previous tap of the tap. The historical decision signal transmitted by the tap, and the historical decision signal received by the tap are cached. Through the above process, the update of the historical decision signal cached in each tap can be completed. According to the preset sequence, the decision signal determined in the current cycle is used as the first tap and the stored historical decision signal needs to be updated. At this time, the historical decision signal stored in the first tap needs to be transmitted to the second tap first. Store the decision signal determined in the current cycle into the first tap. Moreover, according to the preset sequence, the last tap does not have a subsequent tap, and there is no need to transmit its saved historical decision signal to the subsequent tap. At this time, the last tap can receive the historical decision signal transmitted by its previous tap and cache the historical decision signal received by the last tap.

It should be noted that the preset sequence is a preset arrangement sequence between at least one fixed tap and at least one floating tap.

It should be noted that in this embodiment of the present application, the signal acquisition time of the historical decision signal buffered in each tap is different. However, the signal acquisition time of the historical judgment signal that needs to be stored in each tap is fixed. For example, relative to the current cycle, the signal stored in tap a is the determined decision signal of the previous cycle, and the signal stored in tap b is the determined decision signal of the previous cycle. At this time, each time the time changes, the signals stored in tap a and tap b also change, but the stored signals remain unchanged relative to the signal acquisition time of the current cycle. That is, relative to the current cycle, the signal stored in tap a is always the decision signal determined in the previous cycle, and the signal stored in tap b is the decision signal determined in the previous cycle.

As a possible implementation manner, in a preset order, at least one fixed tap is located before at least one floating tap.

Step S605: Update the tap coefficient of at least one fixed tap according to the error signal and the historical decision signal buffered in at least one fixed tap.

In the embodiment of the present application, in order to set the tap coefficients more accurately so as to perform more accurate ISI elimination on the input signal, a preset adaptive algorithm can be used to calculate the error signal based on the error signal and the historical judgment signal cached in each fixed tap. The tap coefficients of each fixed tap are updated. At this time, the adaptive algorithm is set in advance. After acquiring the error signal, the receiving device can adjust the tap coefficient in each fixed tap, that is, the multiplication coefficient of each multiplier, based on the error signal. That is to say, after the receiving device obtains the error signal, it can use the preset adaptive algorithm to calculate the adjustment value of the tap coefficient based on the historical judgment signal and error signal cached in each fixed tap. After the adjustment value of the tap coefficient is obtained, the tap coefficient of each fixed tap can be updated according to the adjustment value of the tap coefficient.

It should be noted that when adjusting the tap coefficient of each fixed tap, the error signal used for each fixed tap is the error signal obtained in the above step S602. Since the historical decision signals cached in each fixed tap are different, the calculated adjustment values of the tap coefficients corresponding to each fixed tap are not exactly the same.

It should be noted that the preset adaptive algorithm can be the minimum average variance algorithm, and of course it can also be other algorithms, which is not limited in this application.

It should be noted that the tap coefficients of at least one fixed tap and at least one floating tap in the decision feedback equalizer can be set in advance.

Step S606: Detect whether the accumulated number of errors of at least one floating tap reaches a preset number threshold.

In the embodiment of the present application, in order to improve the accuracy of the error accumulation value of the floating tap, the errors of the floating taps in multiple periods can be accumulated. At this time, before determining the target tap based on the error accumulation value of at least one floating tap, it is first detected whether the error accumulation has been completed a preset number of times. At this time, after updating the error accumulation value of at least one floating tap in the current cycle, the receiving device can detect whether the error accumulation number of each floating tap in the at least one floating tap reaches the preset number threshold. For example, the preset number threshold is 1000 times. At this time, after the receiving device updates the error cumulative value of at least one floating tap in the current cycle, it can accumulate the error cumulative number of each floating tap by 1, and update the error cumulative value of each floating tap. Accumulated number of errors. Check whether the updated cumulative error count of each floating tap reaches 1000 times. If it does not reach 1000 times, you need to continue to accumulate the cumulative error value of each floating tap. At this time, the following step S607a can be performed. If it reaches 1000 times, it means that the accumulated error value of each floating tap can be used to determine the target tap, and the following step S607b can be executed.

Step S607a, when the cumulative number of errors of at least one floating tap does not reach the preset threshold, continue to perform the step of periodically acquiring the input signal, equalizing the input signal to obtain an equalized signal, and then proceed to step S607a according to the error signal and at least one floating tap. The cached historical decision signal in the tap updates the error accumulation value of at least one floating tap until the error accumulation number of at least one floating tap reaches the preset number threshold.

In this embodiment of the present application, when the receiving device determines that the cumulative number of errors of at least one floating tap has not reached the preset threshold, it needs to continue to accumulate the cumulative error value of at least one floating tap. At this time, the above steps can be continued. From S601 to step S606, the input signal is continuously acquired periodically and the input signal is equalized to obtain an equalized signal. The decision signal is determined based on the equalized signal, and the error signal is determined based on the decision signal and the equalized signal. The error accumulation value of each floating tap is updated according to the error signal and the historical decision signal cached in each floating tap. And detect whether the cumulative number of errors of at least one floating tap reaches the preset threshold. If not, steps S601 to S606 are re-executed until the cumulative number of errors of at least one floating tap reaches the preset threshold.

Step S607b: When the cumulative number of errors of at least one floating tap reaches the preset threshold, at least one target tap is determined in the at least one floating tap based on the cumulative error value of the at least one floating tap.

In this embodiment of the present application, when the receiving device determines that the cumulative number of errors of at least one floating tap reaches the preset threshold, it means that the cumulative error value of at least one floating tap has been accumulated for the preset threshold. At this time, it can The target tap is judged based on the accumulated error value of at least one floating tap. The receiving device may determine at least one target tap among the at least one floating tap based on the accumulated error value of the at least one floating tap. For details, reference may be made to step S504, which will not be described again here.

In this application, in each cycle, the error cumulative value of each floating tap can be updated based on the historical decision signal and error signal cached in each floating tap, so that according to the error cumulative value of each floating tap, Determine the target tap. In this application, only the error accumulation corresponding to each floating tap is calculated, the power consumption and implementation complexity are low, the target tap can be determined quickly, the delay is low, and the implementation speed is fast.

Step S608: The input signal is periodically acquired, and the input signal is equalized to obtain an equalized signal.

In the embodiment of the present application, after determining the target decision feedback equalizer, the target tap can be turned on. After acquiring the input signal, the input signal can be passed through the fixed tap in the decision feedback equalizer and the equalization adopted by at least one target. After processing, an equalized signal with part of the ISI eliminated is obtained.

As a possible implementation, the receiving device can use the formula according to the input signal Calculate the balanced signal. Among them, w[n] represents the balanced signal of the nth period, y[n] represents the input signal of the nth period, j represents the jth fixed tap, N represents the number of fixed taps, and c _j represents the jth The tap coefficient of the multiplier in the fixed tap, data_reg[n,j] represents the historical decision signal cached in the j-th fixed tap in the n-th period. data_reg[n,i] represents the historical decision signal cached in the i-th target tap in the n-th period, i represents the i-th target tap, and M represents the number of target taps.

Step S609: Determine the decision signal based on the equalized signal, and obtain the error signal based on the decision signal and the equalized signal.

Specifically, step S602 can be parameterized, which will not be described again here.

Step S610: Update the tap coefficient in at least one tap in the working state based on the error signal and the historical decision signal buffered in the at least one tap in the working state.

Wherein, at least one tap in working state includes at least one fixed tap and at least one target tap.

For details, reference may be made to step S605, which will not be described again here.

In this application, in each cycle, the error cumulative value of each floating tap can be updated based on the historical decision signal and error signal cached in each floating tap, so that according to the error cumulative value of each floating tap, Determine the target tap. In this application, only the error accumulation corresponding to each floating tap is calculated, the power consumption and implementation complexity are low, the target tap can be determined quickly, the delay is low, and the implementation speed is fast.

Refer to Figure 7, which is a schematic structural diagram of a receiving device provided by an embodiment of the present application. As shown in Figure 7, the receiving device includes: a receiver 701, a decision feedback equalizer 702 and a processor 703. As shown in FIG. 8 , the decision feedback equalizer 702 includes at least one floating tap 7021 . Historical decision signals are cached in at least one floating tap 7021, and the signal acquisition times of the historical decision signals cached in different floating taps are different.

Receiver 701 is used to periodically acquire input signals.

The decision feedback equalizer 702 is used to equalize the input signal to obtain an equalized signal.

The decision feedback equalizer 702 is also used to determine a decision signal from at least one preset transmission signal according to the equalized signal.

The processor 703 is configured to obtain an error signal according to the decision signal and the equalization signal.

The processor 703 is also configured to update the accumulated error value of at least one floating tap 7021 based on the error signal and the historical decision signal cached in the at least one floating tap 7021 .

In a possible implementation, the processor 703 is specifically configured to, for each floating tap 7021 in the at least one floating tap 7021, calculate the difference between the error signal and the floating tap according to the error signal and the historical decision signal cached in the floating tap. When the historical decision signals cached in the tap 7021 are all positive signals, or all are negative signals, the error accumulation value of the floating tap 7021 is increased by the preset threshold;

Alternatively, when one of the error signal and the historical decision signal buffered in the floating tap 7021 is a positive signal and the other is a negative signal, the accumulated error value of the floating tap 7021 is reduced by a preset threshold.

The processor 703 is also configured to determine at least one target tap in the at least one floating tap 7021 according to the error accumulation value of the at least one floating tap 7021.

As a possible implementation manner, the processor 703 is specifically configured to determine the predetermined value in at least one floating tap 7021 according to the error accumulated value of at least one floating tap 7021 in order of the absolute value of the error accumulated value from large to small. Set a number of target taps.

Or, as a possible implementation, the processor 703 is specifically configured to divide at least one floating tap 7021 into at least one floating tap group according to a preset order; calculate at least The error cumulative value corresponding to each floating tap group in a group of floating tap groups; according to the error cumulative value corresponding to each floating tap group, the target tap is determined in at least one floating tap group.

Each floating tap group contains at least one floating tap 7021.

As a possible implementation manner, the above-mentioned decision feedback equalizer 702, as shown in FIG. 8, also includes: at least one fixed tap 7022, an adder 7023 and a decider 7024. Among them, at least one fixed tap 7022 and at least one floating tap 7023 are both connected to one input end of the adder 7023, and the other input end of the adder 7023 is connected to the receiver 701. The output terminal of the adder 7023 is connected to the input terminal of the determiner 7024, and the output terminal of the determiner 7024 is connected to the first tap in a preset sequence among at least one fixed tap 7022 and at least one floating tap 7023. Moreover, according to a preset order, at least one fixed tap 7022 and at least one floating tap 7023 are connected in sequence. Historical decision signals are buffered in at least one fixed tap 7022, and the signal acquisition times of the historical decision signals buffered in at least one fixed tap 7022 are different, and the signal acquisition times of the historical decision signals buffered in at least one floating tap 7022 are different. .

As a possible implementation manner, each of the at least one fixed tap 7022 and the at least one floating tap includes a register 7025 and a multiplier 7026 connected to the register 7025. At this time, according to the preset sequence, at least one fixed tap 7022 and at least one floating tap 7023 are connected in sequence, including: according to the preset sequence, the register 7025 in at least one fixed tap 7022 and the register 7025 of at least one floating tap 7023 are connected in sequence, such as As shown in Figure 8.

As a possible implementation manner, each of the at least one fixed tap 7022 and the at least one floating tap further includes an equalization adder 7027. In this case, one input of the equalization adder 7027 is connected to the output of the multiplier 7026 within this tap. According to the preset sequence, when the tap is not the first tap in the decision feedback equalizer, the other input end of the intra-tap equalization adder 7027 is connected to the output end of the intra-tap equalization adder 7027 after the tap. , and the output end of the equalization adder 7027 of this tap is connected to an input end of the previous equalization adder 7027 of this tap. According to the preset sequence, when the tap is the first tap in the decision feedback equalizer, the output end of the equalization adder 7027 of the tap is connected to the adder 7023.

At this time, the above-mentioned decision feedback equalizer 702 performs equalization processing on the input signal, and the equalized signal obtained includes: at least one fixed tap 7022 outputs a signal that eliminates part of the ISI to the adder 7023, and the adder 7023 outputs according to the input signal and at least one fixed tap 7022 The signal that eliminates part of the ISI is subjected to equalization processing to obtain an equalized signal.

The above decision feedback equalizer 702 determines a decision signal in at least one preset transmission signal according to the equalization signal, including: the decision unit 7024 determines a decision signal in at least one preset transmission signal according to the equalization signal.

As a possible implementation manner, the decision feedback equalizer 702 is also configured to update the historical decision signal buffered by each floating tap in at least one floating tap according to the decision signal.

That is, each of the at least one fixed tap 7022 and the at least one floating tap 7023 included in the decision feedback equalizer 702, in a preset order, when the tap is not the first tap in the decision feedback equalizer 702, Transmit the historical decision signal cached in the register 7025 of this tap to the register 7025 of the next tap, receive the historical decision signal transmitted by the register 7025 of the previous tap of this tap, and store the received historical decision signal to the register of this tap 7025 in. According to the preset sequence, when the tap is the first tap in the decision feedback equalizer 702, the historical decision signal buffered in the register 7025 in the tap is transferred to the register 7025 in the subsequent tap. And receive the decision signal output by the decision device 7024, and cache the decision signal into the register 7025 in the tap.

As a possible implementation, the processor 706 is also configured to detect whether the accumulated number of errors of at least one floating tap 703 reaches a preset number threshold.

The processor 706 is specifically configured to determine at least one target tap in the at least one floating tap 7021 based on the accumulated error value of the at least one floating tap 7021 when the error accumulation number of the at least one floating tap 703 reaches a preset number threshold.

The processor 706 is also configured to trigger the receiver 701 to perform steps to periodically obtain the input signal, and trigger the decision feedback equalizer 702 to perform equalization processing when the cumulative number of errors of at least one floating tap 7021 does not reach the preset number threshold. Obtain the equalized signal to the step of determining a decision signal from at least one preset transmission signal according to the equalized signal, and re-execute the step of obtaining the error signal based on the decision signal and the equalized signal according to the error signal and at least one floating tap 7021 The cached historical decision signal updates the accumulated error value of at least one floating tap 7021 until the accumulated number of errors of at least one floating tap 7021 reaches the preset threshold.

As a possible implementation manner, the processor 703 is also configured to update the preset tap coefficient of at least one target tap based on the error signal and the historical decision signal cached in the at least one target tap.

As a possible implementation, the processor 703 is also configured to update the preset tap coefficient of at least one fixed tap based on the error signal and the historical decision signal cached in at least one fixed tap.

As a possible implementation manner, referring to FIG. 9 , the above-mentioned processor 703 includes: an error adder 7031 , an adaptive algorithm unit 7032 and an error accumulator 7033 . Among them, the error adder 7031 is connected to the decision unit 7024 and the adder 7023, and is used to obtain the error signal according to the decision signal and the equalization signal.

The error accumulator 7033 is connected to the register 7025 and the error adder 7031 in at least one floating tap 7021, and is used to update the error cumulative value of at least one floating tap based on the error signal and the historical decision signal buffered in the at least one floating tap.

The adaptive algorithm unit 7032 has a register 7025 in at least one fixed tap 7022, a register 7025 in at least one floating tap 7021, an error adder 7031, a multiplier 7026 in at least one fixed tap 7022, and a multiplication in at least one floating tap 7021. The processor 7026 is connected and used to update the tap coefficients in the tap in the working state according to the error signal and the historical decision signal buffered in at least one tap in the working state.

Corresponding to the above embodiment, this application also provides a communication system. As shown in Figure 10, the above communication system includes: a sending device 1001 and the receiving device 1002 described in the above embodiment. Among them, the sending device 1001 and the receiving device 1002 are connected through a communication channel.

As a possible implementation method, the above communication system can be applied to the field of ICT (Information and Communications Technology). At this time, the receiving device 1002 may be a high-speed serial interface, including a network device or a computer high-speed interface, such as a backplane interface, a panel interface, and an intra-board interface, as shown in FIG. 11 .

In specific implementation, the present invention also provides a computer storage medium, wherein the computer storage medium can store a program, and when executed, the program can include some or all of the embodiments of the equalizer tap adjustment method provided by the present invention. step. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus the necessary general hardware platform. Based on this understanding, the technical solutions in the embodiments of the present invention can be embodied in the form of software products in essence or those that contribute to the existing technology. The computer software products can be stored in storage media, such as ROM/RAM. , magnetic disk, optical disk, etc., including a number of instructions to cause a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

The same and similar parts among the various embodiments in this specification can be referred to each other. In particular, for the device embodiment and the terminal embodiment, since they are basically similar to the method embodiment, the description is relatively simple. For relevant details, please refer to the description in the method embodiment.

Claims (18)

1. A tap adjustment method of an equalizer, characterized in that it is applied to a receiving device, the receiving device includes: a decision feedback equalizer, the decision feedback equalizer includes at least one floating tap; the at least one floating tap There are historical decision signals cached in the taps, and the signal acquisition times of the historical decision signals cached in different floating taps are different; the method includes: Periodically acquire input signals, perform equalization processing on the input signals, and obtain equalized signals; According to the equalized signal, a decision signal is determined from at least one preset transmission signal, and an error signal is obtained based on the decision signal and the equalized signal; Update the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal cached in the at least one floating tap; At least one target tap is determined among the at least one floating tap according to the accumulated error value of the at least one floating tap. 2. The method of claim 1, further comprising: According to the decision signal, the historical decision signal buffered by each floating tap in the at least one floating tap is updated. 3. The method according to claim 1, wherein updating the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal buffered in the at least one floating tap includes: For each floating tap in the at least one floating tap, according to the error signal and the historical decision signal buffered in the floating tap, the error signal and the historical decision signal buffered in the floating tap are both positive signals, or both are positive signals. When it is a negative signal, increase the error accumulation value of the floating tap by the preset threshold; Alternatively, when one of the error signal and the historical decision signal buffered in the floating tap is a positive signal and the other is a negative signal, the accumulated error value of the floating tap is reduced by a preset threshold. 4. The method according to claim 1, characterized in that, before determining at least one target tap in the at least one floating tap according to the error cumulative value of the at least one floating tap, it further includes: Detect whether the accumulated number of errors of the at least one floating tap reaches a preset number threshold; Determining at least one target tap among the at least one floating tap according to the error cumulative value of the at least one floating tap includes: When the error accumulation number of the at least one floating tap reaches the preset number threshold, at least one floating tap is determined among the at least one floating tap according to the error accumulation value of the at least one floating tap. 5. The method of claim 4, further comprising: When the cumulative number of errors of the at least one floating tap does not reach the preset threshold, continue to perform the step of periodically acquiring the input signal, performing equalization processing on the input signal to obtain an equalized signal, and then proceed to the step of obtaining an equalized signal according to the error signal and The cached historical decision signal in the at least one floating tap updates the error accumulation value of the at least one floating tap until the error accumulation number of the at least one floating tap reaches a preset number threshold. 6. The method according to claim 1, characterized in that, the method further comprises: The tap coefficient of the at least one target tap is updated according to the error signal and the historical decision signal cached in the at least one target tap. 7. The method according to claim 1, wherein determining at least one target tap among the at least one floating tap according to the accumulated error value of the at least one floating tap includes: According to the error cumulative value of the at least one floating tap, a preset number of target taps are determined in the at least one floating tap in order of the absolute value of the error cumulative value from large to small. 8. The method according to any one of claims 1 to 7, wherein determining at least one target tap in the at least one floating tap according to the accumulated error value of the at least one floating tap includes: According to a preset order, the at least one floating tap is divided into at least one floating tap group; each floating tap group contains at least one floating tap; Calculate the error cumulative value corresponding to each floating tap group in the at least one floating tap group according to the error cumulative value of the at least one floating tap group; According to the accumulated error value corresponding to each floating tap group, the target tap is determined in the at least one floating tap group. 9. A receiving device, characterized in that it includes: a receiver, a decision feedback equalizer and a processor; wherein, the decision feedback equalizer contains at least one floating tap; and the at least one floating tap has history cached in it. Decision signal, and the signal acquisition time of the historical decision signal of different floating tap buffers is different; The receiver is used to periodically acquire input signals; The decision feedback equalizer is used to equalize the input signal to obtain an equalized signal; The decision feedback equalizer is further configured to determine a decision signal from at least one preset transmission signal according to the equalized signal; The processor is used to obtain an error signal according to the decision signal and the equalization signal; The processor is further configured to update the error cumulative value of the at least one floating tap according to the error signal and the historical decision signal cached in the at least one floating tap; The processor is further configured to determine at least one target tap among the at least one floating tap according to the accumulated error value of the at least one floating tap. 10. The receiving device according to claim 9, characterized in that, The decision feedback equalizer is also configured to update the historical decision signal buffered by each floating tap in the at least one floating tap according to the decision signal. 11. The receiving device according to claim 9, characterized in that, The processor is specifically configured to, for each floating tap in at least one floating tap, calculate the difference between the error signal and the historical decision signal cached in the floating tap according to the error signal and the historical decision signal cached in the floating tap. When the signals are all positive signals or all negative signals, the error accumulation value of the floating tap is increased by the preset threshold; Alternatively, when one of the error signal and the historical decision signal buffered in the floating tap is a positive signal and the other is a negative signal, the accumulated error value of the floating tap is reduced by a preset threshold. 12. The receiving device according to claim 9, characterized in that, The processor is also configured to detect whether the accumulated number of errors of the at least one floating tap reaches a preset number threshold; The processor is specifically configured to determine at least one of the at least one floating taps based on the accumulated error value of the at least one floating tap when the error cumulative number of the at least one floating tap reaches a preset threshold. A floating tap. 13. The receiving device according to claim 12, characterized in that, The processor is also configured to trigger the receiver execution step to periodically obtain the input signal and trigger the decision feedback equalizer execution step when the cumulative number of errors of the at least one floating tap does not reach a preset threshold. Perform equalization processing on the input signal to obtain an equalized signal. According to the equalized signal, determine a decision signal from at least one preset transmission signal, and re-execute the step of obtaining an equalized signal based on the decision signal and the equalized signal. The error signal to step is to update the error cumulative value of the at least one floating tap according to the error signal and the buffered historical decision signal in the at least one floating tap until the error cumulative number of the at least one floating tap reaches a preset value. times threshold. 14. The receiving device according to claim 13, characterized in that, The processor is further configured to update the tap coefficient in the at least one target tap according to the error signal and the historical decision signal cached in the at least one target tap. 15. The receiving device according to claim 9, characterized in that, The processor is specifically configured to determine a preset number of target taps in the at least one floating tap according to the error cumulative value of the at least one floating tap in order of the absolute value of the error cumulative value from large to small. . 16. The receiving device according to any one of claims 9-15, characterized in that, The processor is specifically configured to divide the at least one floating tap into at least one floating tap group according to a preset order; each floating tap group contains at least one floating tap. Calculate the error cumulative value corresponding to each floating tap group in the at least one floating tap group according to the error cumulative value of the at least one floating tap group; According to the accumulated error value corresponding to each floating tap group, the target tap is determined in the at least one floating tap group. 17. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to execute claims 1-8 any of the methods described. 18. A communication system, characterized by comprising: a sending device and a receiving device according to any one of claims 9-16.