KR102613262B1 - Apparatus and method of adaptive equalization - Google Patents

Abstract

We present an adaptive equalization device and method that optimizes digital algorithms, is resistant to dynamic environmental changes, and can adjust the monitoring range according to signal size. The presented device includes a first equalization filter that compensates for and outputs components in the high-frequency band of the received signal, a second equalization filter installed in parallel with the first equalization filter and monitoring the received signal, and a second equalization filter for each cycle of the asynchronous clock signal. A size comparison unit that samples the size of the monitoring signal from the filter, and the equalizer monitoring code to be provided to the second equalization filter, and the reference signal of the size comparison unit are changed to collect comparison data from the size comparison unit and collect the collected comparison data. It includes a digital control unit that finds the optimal equalizer control code and provides it to the first equalization filter.

Description

Adaptive equalization device and method {Apparatus and method of adaptive equalization}

The present invention relates to an adaptive equalization device and method, and more specifically, to an adaptive equalization device and method that detects a level of distortion and outputs an optimal signal.

An equalizer compensates for signal attenuation or distortion that occurs in the process of transmitting a signal using a predetermined channel.

In general, equalizer technologies used in high-speed adaptive equalizers are designed with various structures, such as decision feedback equalizers and tap-delay line filters.

Meanwhile, in order to effectively compensate for changes in channel characteristics due to changes in process or temperature, various techniques are applied to high-speed adaptive equalizers. For example, in a conventional high-speed adaptive equalizer, the power of the high-frequency component of the equalization filter output signal is compared with the power of the high-frequency component of the reference signal and the high-frequency voltage gain of the equalization filter is adjusted to minimize the power difference.

Accordingly, an adaptive equalization device (Korean Patent No. 10-1074454) that reduces power consumption by not requiring a high-speed clock generator has been applied for and registered as a patent.

The adaptive equalization device of the above-mentioned Korean Patent No. 10-1074454 extracts a histogram from the received signal, calculates the Q-factor, and then evaluates the transmission quality of the signal. Since the size of the maximum value on the histogram appears the largest in the optimal equalization state, the Q-factor at this time is determined to be the best and the current state is set as the optimal state and applied to the equalizer. By sampling with a non-synchronized clock, the voltage magnitude at a random point can be read.

In addition, the adaptive equalization device of the above-described Korean Patent No. 10-1074454 randomly samples the input signal and records a smaller number of times than the comparison voltage. The recorded results appear in the form of a cumulative density function (CDF).

However, the adaptive equalization device of the above-mentioned Korean Patent No. 10-1074454 stores all data about the cumulative density function (CDF) until adaptation is completed, so there is a problem of using a lot of registers and increasing the digital circuit area. there is.

In addition, the adaptive equalization device of the above-mentioned Korean Patent No. 10-1074454 has a problem in that it distorts the signal by monitoring the signal. In other words, since all equalizer control codes must be monitored to find the optimal signal, changes in the equalizer control code are immediately reflected in the output signal. As a result, the areas in which it can be applied are limited. Because it is fixed in the initial calibration state, it cannot keep up with dynamic changes in the environment.

In addition, the adaptive equalization device of the above-described Korean Patent No. 10-1074454 has a problem in that signal size affects accuracy. In other words, if the input signal is too small, the discrimination power of the histogram decreases, so the size of the minimum resolution must be reduced. If the input signal is too large, the histogram cannot be observed properly, so the maximum observation range must be increased. To satisfy these two conditions, the signal must be observed in detail and over a wide range, so it takes a long time to complete adaptation and the size of the monitoring circuit increases.

Prior art 1: Republic of Korea Patent No. 10-1074454 (adaptive equalization device and equalization method)

The present invention was proposed to solve the above-described conventional problems, and its purpose is to provide an adaptive equalization device and method that optimizes digital algorithms, is resistant to dynamic environmental changes, and can adjust the monitoring range according to signal size. there is.

In order to achieve the above object, an adaptive equalization device according to a preferred embodiment of the present invention includes a first equalization filter that compensates for and outputs components of a high frequency band of a received signal; a second equalization filter installed in parallel with the first equalization filter and monitoring the received signal; a magnitude comparison unit that samples the magnitude of the monitoring signal from the second equalization filter every cycle of the asynchronous clock signal; and collecting comparison data from the size comparison unit by changing the equalizer monitoring code to be provided to the second equalization filter and the reference signal of the size comparison unit, and finding the optimal equalizer control code based on the collected comparison data. a digital control unit providing the first equalization filter; and the second equalization filter according to the equalizer monitoring code of the digital control unit to enable the digital control unit to find an optimal equalizer control code. The received signal is compensated and output.

The size comparison unit includes a reference signal generator that generates a reference signal of an analog component corresponding to a reference signal control code from the digital control unit; an analog comparator that obtains the difference between the monitoring signal from the second equalization filter and the reference signal from the reference signal generator and outputs the difference as an analog signal; and a sampling circuit unit that samples and digitizes the output from the analog comparison unit at every cycle of the input asynchronous clock signal.

The reference signal control code is a code that determines the level of the reference signal, and may be any one of N codes having different levels.

The digital control unit adjusts the size of the input signal so that the size comparison unit falls within a specific reference signal range, and after completion of adjustment, changes the equalizer monitoring code of the second equalization filter or the reference signal of the size comparison unit to create a histogram. The peak value of can be calculated, and the optimal equalizer control code found based on the calculated peak value can be applied to the first equalization filter.

Meanwhile, the adaptive equalization method according to a preferred embodiment of the present invention includes a first equalization filter that compensates for and outputs components in the high frequency band of the received signal, installed in parallel with the first equalization filter, and monitoring the received signal. A second equalization filter, a magnitude comparison unit that samples the magnitude of the monitoring signal from the second equalization filter every cycle of the asynchronous clock signal, and an equalizer monitoring code to be provided to the second equalization filter and a standard for the magnitude comparison unit. Adaptation of an adaptive equalization device including a digital control unit that changes signals, collects comparison data from the magnitude comparison unit, finds the optimal equalizer control code based on the collected comparison data, and provides it to the first equalization filter. As a type equalization method,

adjusting the size of the input signal received by the size comparison unit so that the input signal falls within a specific reference signal range; and calculating the peak value of the histogram by changing the equalizer monitoring code of the second equalization filter or the reference signal of the magnitude comparison unit, and calculating the optimal equalizer control code based on the calculated peak value. 1. It includes the step of applying to the equalization filter.

It may further include moving to standby mode when the input signal to the adaptive equalization device is too small, and moving to standby mode when the applying step is completed.

The adjusting step includes setting the amplification gain of the second equalization filter to a minimum; setting the reference signal to the middle of the reference signal and then collecting N times sampling results from the size comparison unit; If all of the sampling results are high, setting the reference signal to the middle of the + reference signal and then collecting N times sampling results from the size comparison unit; After setting the reference signal to the middle of the + reference signal, if all N times sampling results in the magnitude comparison unit are low, determining whether the current amplification gain is maximum; And if the current amplification gain is not the maximum, increasing the current amplification gain, setting the reference signal to the middle of the reference signal, and then returning to the step of collecting N times sampling results from the magnitude comparison unit. It can be included.

After setting the reference signal to the middle of the -reference signal, all N-time sampling results in the magnitude comparison unit are not high, or after setting the reference signal to the middle of the +reference signal, the magnitude comparison unit If all N-time sampling results are not low, the process can move to the application step and perform the operation of the corresponding step.

If the current amplification gain is maximum, the standby mode can be entered.

The adjusting step includes setting the reference signal range in the size comparison unit to the maximum; setting the reference signal to the middle of the reference signal and then collecting N times sampling results from the size comparison unit; If all of the sampling results are high, setting the reference signal to the middle of the + reference signal and then collecting N times sampling results from the size comparison unit; After setting the reference signal to the middle of +reference signal, if all N-time sampling results in the size comparison unit are low, determining whether the range of the current reference signal is minimum; And if the range of the current reference signal is not the minimum, reducing the current reference signal range, setting the reference signal to the middle of the reference signal, and then returning to the step of collecting N times sampling results from the size comparison unit. It can be included.

After setting the reference signal to the middle of the -reference signal, all N-time sampling results in the magnitude comparison unit are not high, or after setting the reference signal to the middle of the +reference signal, the magnitude comparison unit If all N-time sampling results are not low, the process can move to the application step and perform the operation of the corresponding step.

If the range of the current reference signal is minimum, the standby mode can be entered.

The applying step includes setting K, the equalizer monitoring code, to 0 (zero), S, the reference signal control code, to 0 (zero), the maximum peak to 0 (zero), and the optimal code to 0 (zero); Inputting a Kth equalizer monitoring code into the second equalization filter; Generating a reference signal corresponding to the S reference signal control code, collecting X times sampling results from the size comparison unit, and counting the high number of the collected sampling results; determining whether the absolute value of the difference between the current counted value and the previous counted value is greater than the maximum peak; If the absolute value of the difference between the current counting value and the previous counting value is greater than the maximum peak, the maximum peak is replaced with the absolute value of the difference between the current counting value and the previous counting value, and the optimal code is monitored by the current equalizer. Replacing the value of the code and replacing the previous counting value with the current counting value; determining whether the value of the reference signal control code is the maximum value of the preset reference signal control code; If the value of the reference signal control code is the maximum value of the reference signal control code, determining whether the value of the equalizer monitoring code is the maximum value of the preset equalizer monitoring code; and, if the value of the equalizer monitoring code is the maximum value of preset equalizer monitoring codes, reflecting the optimal code as an optimal equalizer control code in the first equalization filter.

If the absolute value of the difference between the current counting value and the previous counting value is not greater than the maximum peak, the previous counting value can be replaced with the current counting value without replacing the maximum peak and the optimal code.

As a result of the determination in the step of determining whether the value of the reference signal control code is the maximum value of the preset reference signal control code, if the value of the reference signal control code is not the maximum value of the reference signal control code, the reference signal control code By setting the value to +1, you can move to the step of generating a reference signal corresponding to the S reference signal control code, collecting X times of sampling results from the size comparison unit, and counting the high number of the collected sampling results. there is.

As a result of the determination in the step of determining whether the value of the equalizer monitoring code is the maximum value of the preset equalizer monitoring code, if the value of the equalizer monitoring code is not the maximum value of the preset equalizer monitoring code, the equalizer is monitored. By increasing the value of the code by +1, you can move to the step of inputting the Kth equalizer monitoring code into the second equalization filter.

The standby mode may turn off the power of the first equalization filter if it is determined that there is no signal in the adjustment step.

The standby mode may turn off the power of the second equalization filter and the size comparison unit after the applying step.

According to the present invention with this configuration, the area of the digital control unit can be further reduced by optimizing the algorithm of the digital control unit.

By adding an equalization filter dedicated to signal monitoring, the digital control unit changes the algorithm to continuously adapt the equalizer, making it resistant to dynamic environmental changes.

The monitoring range can be adjusted according to the signal size, and a signal detector function can be performed by adding an input signal size determination function to the digital control unit.

1 is a configuration diagram of an adaptive equalization device according to an embodiment of the present invention.
FIG. 2 is an internal configuration diagram of the size comparison unit shown in FIG. 1.
FIGS. 3 and 4 and FIGS. 5A and 5B are diagrams that can be used to explain the size comparison unit shown in FIG. 1.
Figure 6 is a diagram for explaining an adaptive equalization method according to an embodiment of the present invention.
FIGS. 7 to 13 are diagrams for explaining the input signal size determination step shown in FIG. 6.
Figures 14 to 16 are diagrams for explaining the equalization filter characteristic observation step shown in Figure 6.

Since the present invention can make various changes and have various embodiments, specific embodiments will be illustrated in the drawings and described in detail.

However, this is not intended to limit the present invention to specific embodiments, and should be understood to include all changes, equivalents, and substitutes included in the spirit and technical scope of the present invention.

The terms used in this application are only used to describe specific embodiments and are not intended to limit the invention. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, terms such as “comprise” or “have” are intended to designate the presence of features, numbers, steps, operations, components, parts, or combinations thereof described in the specification, but are not intended to indicate the presence of one or more other features. It should be understood that this does not exclude in advance the possibility of the existence or addition of elements, numbers, steps, operations, components, parts, or combinations thereof.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the technical field to which the present invention pertains. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related technology, and should not be interpreted in an ideal or excessively formal sense unless explicitly defined in the present application. No.

Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In order to facilitate overall understanding when describing the present invention, the same reference numerals are used for the same components in the drawings, and duplicate descriptions for the same components are omitted.

1 is a configuration diagram of an adaptive equalization device according to an embodiment of the present invention.

The adaptive equalization device according to an embodiment of the present invention includes a first equalization filter 10, a second equalization filter 20, a size comparison unit 30, and a digital control unit 40.

The first equalization filter 10 can compensate and output components in the high frequency band of the actually received signal. The first equalization filter 10 receives an equalizer control code from the digital control unit 40. The first equalization filter 10 may select an equalization coefficient in response to an equalizer control code and perform equalization with an equalization gain corresponding to the selected equalization coefficient.

The second equalization filter 20 is installed in parallel with the first equalization filter 10 at the receiving signal end.

Like the first equalization filter 10, the second equalization filter 20 has the function of compensating for and outputting the components of the high frequency band of the received signal, so the second equalization filter 20 is similar to the first equalization filter 10. It can be said that they have the same operating characteristics.

Meanwhile, the second equalization filter 20 can be said to be an equalization filter exclusively for signal monitoring. The second equalization filter 20 performs monitoring to find the optimal equalizer control code that can optimally compensate the received signal according to the equalizer monitoring code of the digital control unit 40.

That is, the second equalization filter 20 can compensate and output the received signal according to the equalizer monitoring code of the digital control unit 40 in order to find the optimal equalizer control code in the digital control unit 40. . And, the signal output from the second equalization filter 20 can be called a monitoring signal.

The above-described first equalization filter 10 and second equalization filter 20 may include an input signal stage, an output signal stage, a low-frequency gain control stage, and a high-frequency gain control stage, respectively. Here, the low-frequency gain control stage receives a signal that controls the amount of amplification of the low-frequency band of the signal, and the high-frequency gain control stage receives a signal that controls the amount of amplification of the high-frequency band of the signal. Low-frequency gain control can control the size of the input signal and thus can control the amplification gain. High-frequency gain control can control the transition speed of an input signal and change the waveform of the signal itself.

The equalizer control code of the digital control unit 40 is input to the low frequency gain control stage or the high frequency gain control stage of the first equalization filter 10 depending on the situation. And, the equalizer monitoring code of the digital control unit 40 is input to the low-frequency gain control stage or the high-frequency gain control stage of the second equalization filter 20 depending on the situation. That is, in order to achieve an amplification gain for the input signal of the second equalization filter 20, a predetermined equalizer monitoring code will be input to the low-pi gain control stage. When it is desired to change the waveform of the input signal of the second equalization filter 20 (to control the transition speed of the signal), a predetermined equalizer monitoring code is input to the high frequency gain control stage.

The size comparison unit 30 receives a predetermined clock signal (eg, asynchronous clock), a monitoring signal from the second equalization filter 20, and a reference signal control code from the digital control unit 40.

Accordingly, the size comparison unit 30 measures (samples) the size of the input signal (monitoring signal) from the second equalization filter 20 every cycle of the input clock signal (asynchronous clock). In other words, the size comparison unit 30 can sample the input signal every cycle of the clock signal (asynchronous clock) and compare it with a digitally controlled reference voltage to digitally output high/low. A more detailed description of the size comparison unit 30 will be described later.

The digital control unit 40 provides an equalizer control code to the first equalization filter 10, an equalizer monitoring code (also known as a high-frequency band gain control code) to the second equalization filter 20, and controls the reference signal. The code can be provided to the size comparison unit 30.

The digital control unit 40 collects comparison data from the size comparison unit 30 by changing the equalizer monitoring code to be provided to the second equalization filter 20 and the reference signal of the size comparison unit 30, and compares the collected data. The activity of finding the optimal equalizer control code based on data is continuously performed. The optimal equalizer control code found is applied to the first equalization filter (10).

That is, the digital control unit 40 provides an equalizer monitoring code to be provided to the second equalization filter 20 (for example, it can be said to be a high-frequency band gain code input to the high-frequency gain control stage) and the magnitude comparison unit 30. By changing the reference signal control code to be provided, the number of highs in the comparison data is counted each time. Then, the digital control unit 40 calculates the difference between the current counting value and the previous counting value, determines the value with the largest peak in the probability density function (PDF) as optimal, and sends the corresponding equalizer control code to the first equalization filter (10). ) can be provided to.

In other words, the digital control unit 40 adjusts the size of the input signal received from the adaptive equalization device, and based on this, the code for the optimal high-frequency band gain found by observing the characteristics of the second equalization filter 20 ( That is, the equalizer control code) can be applied to the first equalization filter 10.

FIG. 2 is an internal configuration diagram of the size comparison unit 30 shown in FIG. 1, and FIGS. 3, 4, 5A and 5B are diagrams that can be used to explain the size comparison unit 30 shown in FIG. 1. .

The size comparison unit 30 may include a reference signal generation unit 31, an analog comparison unit 32, and a sampling circuit unit 33.

The reference signal generator 31 generates a reference signal (also referred to as a reference voltage) of an analog component corresponding to the reference signal control code from the digital control unit 40. Here, the reference signal control code is a code that determines the level of the reference voltage, and may be, for example, any one of about 16 codes having different levels.

Accordingly, the reference signal control code applied to the reference signal generator 31 may be any one of about 16 codes having different levels. Additionally, the reference signal output from the reference signal generator 31 varies depending on the reference signal control code.

The analog comparator 32 obtains the difference between the input signal (i.e., the monitoring signal from the second equalization filter 20) and the reference signal from the reference signal generator 31 and outputs it as an analog signal.

The sampling circuit unit 33 samples and digitizes the output from the analog comparator 32 at every cycle of the input clock signal (eg, asynchronous clock). Accordingly, the sampling circuit unit 33 outputs predetermined digital data (eg, high or low).

Here, the clock signal is a signal that provides a point in time to perform comparison, and generally may use a rising edge or falling edge.

In Figure 2, the analog comparison unit 32 and the sampling circuit unit 33 are configured separately, but they can be integrated if necessary.

The size comparison unit 30 configured as described above compares the size of the input signal with an internally generated reference signal, as shown in FIG. 3. Then, the size comparison unit 30 samples the clock signal every cycle, determines the size, and outputs digital data corresponding to it. The reference signal to be compared varies depending on the control.

Meanwhile, the size comparison unit 30 can also be implemented by creating a signal by subtracting the difference from the input signal by the amount of the reference signal and comparing it based on 0 (zero), as shown in FIG. 4.

Unlike the case where only the center of the data is sampled with the synchronized clock of FIG. 5A, the size comparison unit 30 described above can randomly sample the input signal when it receives an asynchronous clock as shown in FIG. 5B. In this case, the probability that the input signal is high/low is 50:50. Since each input signal bit is not sampled at a specific point in time, the sampling value at a random point in time can be obtained. Although there is no data at the time of sampling, data on the distribution of the sampled voltage (i.e., allowing counting the number of highs and lows in the comparison data) can be obtained.

If the clock is asynchronous, random sampling occurs, so a low-speed clock can be used. That is, compared to the case of sampling only the center of the data with the synchronous clock of Figure 5a, using a low-speed asynchronous clock as shown in Figure 5b reduces the burden of operating speed, so power consumption and circuit size can be reduced. .

Figure 6 is a diagram for explaining an adaptive equalization method according to an embodiment of the present invention.

The adaptive equalization method according to an embodiment of the present invention includes the step of determining the size of the input signal (S100), the step of observing the characteristics of the equalization filter (S200), and the step of waiting to reduce power consumption (S300). , has a circular structure. The adaptive equalization method according to an embodiment of the present invention is performed in the adaptive equalization device described with reference to FIGS. 1 to 5.

In the input signal size determination step (S100), the range of the input signal received from the adaptive equalization device (i.e., the signal (monitoring signal) input to the size comparison unit 30) and the range of the internally generated reference signal are determined. After comparison, adjust the size of the input signal so that it is approximately 1/2 of the reference signal range. After adjusting the size of the input signal, the process moves to the equalization filter characteristic observation step (S200). If the input signal is too small, it is considered as no signal and moves to the standby step (S300). Here, the input signal size determination step (S100) can be said to be an input signal size adjustment step.

The reason why the size of the input signal is adjusted so that the input signal is approximately 1/2 of the reference signal range (i.e., between the maximum and minimum values of the reference signal) in the input signal size determination step (S100) is because the input signal is too large. This is because if it is too small, the accuracy in the equalization filter characteristic observation step (S200) decreases. In other words, if the input signal is approximately 1/2 of the reference signal range, the characteristics of the second equalization filter 30, which is the monitoring target, can be better observed, and the optimal equalizer control code can be set to the first equalization filter 10. can be provided to.

In the equalization filter characteristic observation step (S200), the peak value of the histogram is calculated by changing the high-frequency band gain or reference signal of the second equalization filter 20, and the optimal value found using the second equalization filter 20 is calculated. A code for the high frequency band gain (i.e., equalizer control code) is applied to the first equalization filter 10. Afterwards, a flag indicating that adaptation of the adaptive equalizer has been completed is raised and the process moves to the waiting stage (S300). Here, the equalization filter characteristic observation step (S200) can be said to be an equalizer control code application step.

In the standby step (S300) (i.e., standby mode), the second equalization filter 20 and the size comparison unit 30 are turned off to reduce power consumption after the equalization filter characteristic observation step (S200). In addition, if it is determined that there is no signal in the input signal size determination step (S100), the power of the first equalization filter 10 is also turned off. After a certain period of time has elapsed, the process moves to the input signal size determination step (S100).

The above-described input signal size determination step (S100), equalization filter characteristic observation step (S200), and waiting step (S300) can be sufficiently performed under the supervision of the digital control unit 40.

FIGS. 7 to 13 are diagrams for explaining the input signal size determination step (S100) shown in FIG. 6.

If the input signal input to the adaptive equalization device according to an embodiment of the present invention is too large or too small, the accuracy in the equalization filter characteristic observation step (S200) is reduced. Accordingly, in the input signal size determination step (S100), it is desirable to adjust the range of the input signal or reference signal so that the size of the input signal is approximately ½ of the reference signal range. If the input signal is too small, it is determined that there is no signal and the standby mode (i.e., standby stage (S300)) can be entered.

In the input signal size determination step (S100), +/- reference signals of the same size are used based on 0 (zero) as shown in FIG. 7. As shown in Figure 7, when the input signal is compared with the - reference signal, the comparison values of the size comparison unit 30 are all high, and when the input signal is compared with the + reference signal, the size comparison unit 30 If the comparison values of are all low, the digital control unit 40 may determine that the range of the corresponding input signal is smaller than the range of the reference signal. Meanwhile, as shown in FIG. 8, when values opposite to the comparison values in the size comparison unit 30 begin to be mixed, the digital control unit 40 may determine that the range of the input signal is larger than the range of the reference signal.

In this way, in order to determine the size of the input signal by comparing the input signal and the reference signal, a method of gradually increasing the amplification gain of the second equalization filter 20 or a method of reducing the range of the reference signal can be adopted.

First, the case of using a method of gradually increasing the amplification gain of the second equalization filter 20 to find a gain that is greater than the reference signal will be described with reference to FIG. 9. In Figure 9, the amplification gain of the second equalization filter 20 is gradually increased to find the initial gain that becomes larger than the reference signal. In other words, if you increase the amplification gain of the second equalization filter 20 step by step, you can find the first code (i.e., the equalizer monitoring code input to the low-pi gain control stage) that becomes greater than the middle of the +/- reference signal. . When the first code that becomes larger than the +/- middle of the reference signal is found, select it and move to the next step (i.e., equalization filter characteristic observation step (S200)). If the amplification gain is made too large, the characteristics of the second equalization filter 20 cannot be observed. Meanwhile, even if the amplifier gain is increased, if the input signal is not greater than the middle of the reference signal, it is determined that there is no signal.

This time, the case of using a method of reducing the range of the reference signal without increasing the amplification gain of the second equalization filter 20 will be described with reference to FIG. 10. In Figure 10, the range of the reference signal is reduced without increasing the amplification gain of the second equalization filter 20, and a code (i.e., reference signal control code) whose input signal is larger than the middle of the reference signal range is searched. If you find a code (i.e., reference signal control code) whose input signal is larger than the middle of the reference signal range, select the corresponding code (i.e., reference signal control code) and proceed to the next step (i.e., equalization filter characteristic observation step (S200)). Go to This method is relatively more advantageous in terms of performance/implementation than the method of increasing the amplification gain. Even if the range of the reference signal is reduced to the minimum, if the input signal is smaller than the mid-range of the reference signal, it is determined that there is no signal.

Meanwhile, unlike the above-described FIGS. 9 and 10, it is also possible to narrow the reference signal range without increasing the amplification gain of the second equalization filter 20, as shown in FIG. 11, and then change only the code used.

Ultimately, in the input signal size determination step (S100), the size of the input signal can be compared while changing the size of the input signal or the range of the reference signal.

In the following, the case of comparing the size of the input signal by adjusting the amplification gain will be described again with reference to FIG. 12, and the case of comparing the size of the input signal by adjusting the range of the reference signal will be described again with reference to FIG. 13. Let me explain.

Referring to FIG. 12, the digital control unit 40 first sets the amplification gain of the second equalization filter 20 to the minimum (S101), sets “reference signal = -reference signal middle”, and then sets the size comparison unit 30 ) Collect the N-time sampling results (S102). If the sampling results (i.e., comparison values of the size comparison unit 30) in step S102 are not all high, the digital control unit 40 moves to the equalization filter characteristic observation step (S200).

If the sampling results (i.e., comparison values of the size comparison unit 30) in step S102 are all high (“Yes” in S103), the digital control unit 40 sets “reference signal = + reference signal middle”. After setting it to , the N-time sampling results from the size comparison unit 30 are collected (S104). If the sampling results (i.e., comparison values of the size comparison unit 30) in step S104 are all low (“Yes” in S105), the digital control unit 40 determines whether the current amplification gain is maximum ( S106). If the current amplification gain is not the maximum, the digital control unit 40 increases the current amplification gain (S107) and then returns to the above-described step S102 and repeats the operation from the corresponding step.

If the sampling results (i.e., comparison values of the size comparison unit 30) are not all low in step S105, the digital control unit 40 moves to the equalization filter characteristic observation step (S200) and amplifies in step S106. When the gain is maximum, it moves to the waiting stage (S300).

In this way, the magnitude of the input signal can be compared by gradually increasing the amplification gain of the second equalization filter 20 and finding a gain that is greater than the reference signal.

Meanwhile, referring to FIG. 13, the digital control unit 40 first sets the reference signal range in the size comparison unit 30 to the maximum (S111), sets it to “reference signal = -reference signal middle”, and then compares the size. The N-time sampling results from unit 30 are collected (S112). If the sampling results (i.e., comparison values of the size comparison unit 30) in step S112 are not all high, the digital control unit 40 moves to the equalization filter characteristic observation step (S200).

If the sampling results (i.e., comparison values of the size comparison unit 30) in step S112 are all high (“Yes” in S113), the digital control unit 40 sets “reference signal = + middle of reference signal”. After setting, the N-time sampling results from the size comparison unit 30 are collected (S114). If the sampling results (i.e., comparison values of the size comparison unit 30) in step S114 are all low (“Yes” in S115), the digital control unit 40 determines whether the range of the current reference signal is minimum. Do it (S116). If the range of the current reference signal is not the minimum, the digital control unit 40 reduces the current reference signal range (S117), returns to the above-described step S112, and repeats the operation from the corresponding step.

If the sampling results (i.e., the comparison values of the size comparison unit 30) in step S115 are not all low, the digital control unit 40 moves to the equalization filter characteristic observation step (S200), and the reference value in step S116. If the signal range is minimal, move to the waiting step (S300).

In this way, by adjusting the reference signal range of the size comparison unit 30 and allowing the input signal to be larger than the reference signal range, the size of the input signal can be compared.

Figures 14 to 16 are diagrams for explaining the equalization filter characteristic observation step shown in Figure 6. Since the first equalization filter 10 and the second equalization filter 20 have the same characteristics, it is possible to observe the characteristics of the first equalization filter 10 just by observing the characteristics of the second equalization filter 20. It can be something.

Referring to FIG. 14, when the reference signal (i.e., reference voltage (Vref)) is below the input signal range, everything is Low, so the count value is 0 (zero), and when the reference voltage is within the input signal range, the corresponding comparison data is By counting, the count value increases little by little. If the reference voltage is above the input signal range, everything is high, so the count value becomes the number of sampling times. The counted value takes the form of a cumulative density function (CDF), and when the difference with adjacent count values is calculated, it is expressed in the form of a probability density function (PDF). The probability density function (PDF) represents the probability that the input signal stays between each reference voltage. Basically, the digital data of the size comparison unit 30 has a high or low value, so there is a high probability that it will remain at a high or low voltage. As a result, the probability density function (PDF) has a peak value as shown in FIG. 14.

Here, the size of the peak in the probability density function (PDF) varies depending on the excess/lack of the high frequency component, as illustrated in FIG. 15. Figure 15 shows how the adaptive equalization device finds the most optimal state depending on whether the data transition speed is slow, too fast, or optimal. Here, when the data transition speed is slow, it is under-equalized in FIG. 15, and when the data transition speed is too fast, it is over-equalized in FIG. 15, and when the data transition speed is optimal (appropriate). is the case of Optimum-equalized in FIG. 15.

For example, if the high-frequency component is insufficient (i.e., in the case of under-equalized), the data transition moves more slowly than in the optimal state (i.e., in the case of optimal-equalized), so the probability of staying at the intermediate value increases. As the probability density function (PDF) between high and low increases, the peak decreases. Conversely, if the high-frequency component is excessive (i.e., in the case of over-equalized), it protrudes more than High/Low during data transition than in the optimal state (i.e., in the case of Optimum-equalized), resulting in high ) and the probability density function (PDF) outside the low increases, the peak decreases.

In this way, by changing the high-frequency band gain of the second equalization filter 20, the largest peak value on the probability density function (PDF) becomes optimal.

According to the equalization filter characteristic observation step (S200) in the present invention, the amount of stored data can be significantly reduced. In the prior art (Korean Patent No. 10-1074454), probability density functions (PDF) were stored for all equalizer control codes and finally compared, but in the embodiment of the present invention, after sampling for each round, the previous counting value was compared. This is because after calculating the difference, if it is the maximum, it is saved (recorded), and if it is not the maximum, it is immediately discarded.

In addition, in the prior art (Korean Patent No. 10-1074454), the optimal value was found and terminated only once, but in the embodiment of the present invention, it was changed to cycle continuously. In an embodiment of the present invention, continuous equalization filter adaptation is possible by adding a second equalization filter 20 dedicated to monitoring.

Next, the above-described equalization filter characteristic observation step (S200) will be described in more detail with reference to FIG. 16.

First, the digital control unit 40 sets K = 0 (zero), S = 0 (zero), maximum peak = 0 (zero), and optimal code = 0 (zero) (S201), and the second equalization filter (20) ) Enter the Kth equalization filter high frequency band gain code (S202). Here, K may be the equalization filter high-frequency band gain code (i.e., the equalizer monitoring code applied to the second equalization filter 20), and S may be the reference signal control code.

Afterwards, the digital control unit 40 inputs the S reference signal control code to the size comparison unit 30 (S203). Accordingly, the size comparison unit 30 will generate a reference signal (reference voltage) corresponding to the S reference signal control code (i.e., the 0th reference signal control code).

Then, the digital control unit 40 collects the X number of sampling results from the size comparison unit 30, counts the high number of the collected sampling results, and temporarily stores the corresponding counting value (S204). For example, when the 0th reference signal control code is applied to the size comparison unit 30, the size comparison unit 30 compares the input signal with the reference signal corresponding to the 0th reference signal control code. The counting value (value calculated by counting the high number) will be temporarily stored in the corresponding digital control unit 40.

Next, the digital control unit 40 determines whether the absolute value of the difference between the current counting value and the previous counting value is greater than the maximum peak (S205).

If the absolute value of the difference between the current counting value and the previous counting value is greater than the maximum peak, the digital control unit 40 replaces the maximum peak with the absolute value of the difference between the current counting value and the previous counting value. (S206).

Then, the digital control unit 40 sets the optimal code as the current K (S207) and sets the previous counting value as the current counting value (S208).

Meanwhile, as a result of the determination in step S205, if the absolute value of the difference between the current counting value and the previous counting value is not greater than the maximum peak, the digital control unit 40 moves to step S208. In other words, after sampling X times, the difference from before is calculated, and if it is the maximum, the temporarily stored counting value is recorded. If it is not the maximum, it is immediately discarded.

Afterwards, the digital control unit 40 determines whether “S = N” (S209). Here, N means the maximum value of the reference signal control code. For example, N may be preset to “16”.

If not “S = N”, the digital control unit 40 performs “S = S + 1” (S210) and then returns to step S203 and repeats the operation from the corresponding step. If the reference signal control code increases sequentially, the size comparison unit 30 will generate a reference signal (reference voltage) that increases sequentially.

Conversely, if “S = N”, the digital control unit 40 determines whether “K = M” (S211). Here, M means the maximum value of the equalization filter high frequency band gain code. For example, M may be preset to “8”.

If not "K = M", the digital control unit 40 performs "K = K + 1" (S212) and then returns to step S202 and repeats the operation from the corresponding step.

Conversely, if "K = M", the digital control unit 40 sets the optimal code as the equalizer control code most optimal for the current state of the first equalization filter 10 and reflects it in the first equalization filter 10 (S213). . Accordingly, when the corresponding equalizer control code is applied to the first equalization filter 10, the adaptive equalization device according to an embodiment of the present invention can be controlled to have an optimal equalization gain in real time.

As described above, the optimal embodiment has been disclosed in the drawings and specifications. Although specific terms are used here, they are used only for the purpose of describing the present invention and are not used to limit the meaning or scope of the present invention described in the claims. Therefore, those skilled in the art will understand that various modifications and other equivalent embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the attached claims.

10: 1st equalization filter
20: 2nd equalization filter
30: Size comparison unit
40: digital control unit

Claims (4)

A first equalization filter that compensates for and outputs components in the high frequency band of the received signal, a second equalization filter installed in parallel with the first equalization filter and monitoring the received signal, and the second equalization filter for each cycle of the asynchronous clock signal. A magnitude comparison unit that samples the size of the monitoring signal from the equalization filter, and the equalizer monitoring code to be provided to the second equalization filter and the reference signal of the magnitude comparison unit are changed to collect and compile the comparison data from the magnitude comparison unit. An adaptive equalization method of an adaptive equalization device including a digital control unit that finds the optimal equalizer control code based on the comparison data and provides it to the first equalization filter,
Setting the equalizer monitoring code K to 0 (zero), the reference signal control code S to 0 (zero), the maximum peak to 0 (zero), and the optimal code to 0 (zero);
inputting the equalizer monitoring code set in the second equalization filter;
Generating a reference signal corresponding to the S reference signal control code, collecting sampling results from the size comparison unit, and counting the high number of the collected sampling results;
determining whether the absolute value of the difference between the current counted value and the previous counted value is greater than the maximum peak;
If the absolute value of the difference between the current counting value and the previous counting value is greater than the maximum peak, the maximum peak is replaced with the absolute value of the difference between the current counting value and the previous counting value, and the optimal code is monitored by the current equalizer. Replacing the value of the code and replacing the previous counting value with the current counting value;
determining whether the value of the reference signal control code is the maximum value of the preset reference signal control code;
If the value of the reference signal control code is the maximum value of the reference signal control code, determining whether the value of the equalizer monitoring code is the maximum value of the preset equalizer monitoring code; and
If the value of the equalizer monitoring code is the maximum value of the preset equalizer monitoring codes, setting the optimal code as an optimal equalizer control code and reflecting it in the first equalization filter,
Adaptive equalization method, wherein if the absolute value of the difference between the current counting value and the previous counting value is not greater than the maximum peak, the previous counting value is replaced with the current counting value without replacing the maximum peak and the optimal code.
delete According to clause 1,
As a result of the determination in the step of determining whether the value of the reference signal control code is the maximum value of the preset reference signal control code, if the value of the reference signal control code is not the maximum value of the reference signal control code, the reference signal control code The value is +1, generating a reference signal corresponding to the S reference signal control code, collecting the sampling results from the size comparison unit, and moving to the step of counting the high number of the collected sampling results. Adaptive equalization method. According to clause 1,
As a result of the determination in the step of determining whether the value of the equalizer monitoring code is the maximum value of the preset equalizer monitoring code, if the value of the equalizer monitoring code is not the maximum value of the preset equalizer monitoring code, the equalizer is monitored. An adaptive equalization method characterized by increasing the code K by 1 and inputting it to the second equalization filter.

Images