KR101214660B1 - An method for adaptively controlling compensation of equalizer - Google Patents

Abstract

The present invention relates to a method for adaptively determining and controlling the compensation magnitude of an equalizer, and more particularly, to controlling the coefficient of the equalizer from jitter components appearing depending on the data pattern.
According to the adaptive method for controlling the compensation amount of the equalizer by extracting jitter information depending on the data pattern proposed in the present invention, accurate compensation of the equalizer is possible. Further, according to the present invention, by using only a few clock phases irrespective of the clock duty ratio, a structure combined with a phase detector having a similar function can have a small implementation area and lower power consumption.

Description

How to adaptively determine and control the equalizer's compensation magnitude {AN METHOD FOR ADAPTIVELY CONTROLLING COMPENSATION OF EQUALIZER}

The present invention relates to a method for adaptively determining and controlling the compensation size of an equalizer. More particularly, the present invention relates to a method for adaptively determining and controlling the compensation size of an equalizer from jitter components that depend on data patterns. will be.

Due to the physical characteristics of the transmission line medium (skin effect, loss of dielectric constant), the frequency bandwidth for use as the medium of the transceiver is limited, which is an important factor in determining the data transmission speed of the transmission line. Due to the existing physical characteristics, the equalization technique is used to improve the data transmission speed by extending the bandwidth of the limited transmission line. Accurate prediction is difficult because the physical characteristics of the medium change according to the environment (transmission line length, temperature, deterioration) .However, when the loss of the transmission line is measured and the amount of compensation is determined by adaptive algorithm. The compensation magnitude of the equalizer can be optimized. Representative adaptive algorithms include Least Mean Square (LMS) and Jitter Measurement. As the data rate increases, the jitter component of the received data affects the performance of the receiver, such as the clock and data recovery circuitry, which can improve the time margin rather than the vertical margin. Reduction algorithms are becoming more important.

The jitter measurement algorithm technique that controls the equalizer's compensation magnitude by measuring the minimum pulse width of the data compensated by the conventional equalizer, compares the minimum pulse width of the ideal data with the minimum pulse width of the ideal data and the degree of compensation. Judge. The ideal pulse width as a reference is generated using multiple phases of the clock. If it is larger than the pulse width of the ideal data, it indicates excessive compensation state, and if it is small, it indicates excessive compensation state. The measured equalization value is used to control the equalizer coefficient to maintain the optimized pulse width.

The conventional method for detecting the data minimum pulse width has the following problems. The generation of the ideal reference data pulse width varies due to the clock's duty ratio being inaccurate at 50% or due to delay skew between clock phases due to layout mismatch, which may result in a large amount of compensation. This is a factor that hinders optimization. In addition, multiple clock phases (multi-phases) are required to generate an ideal reference pulse width, and the pulse width detection technique according to the data pattern requires a predetermined data pattern. Training sequence process time is required, there is a problem that the size of the circuit is increased and the power consumption is increased by using a plurality of phase clocks.

The present invention is proposed to solve the above problems of the conventionally proposed methods, by measuring the loss of the transmission line from the jitter component appearing depending on the data pattern and adjusting the equalizer coefficient, It is an object of the present invention to propose an adaptive algorithm for maintaining an optimized compensation by operating irrespective of duty ratio.

In addition, another object of the present invention is to enable a small area, low power algorithm implementation by using only a few clock phases.

In order to achieve the above object, a method for adaptively determining and controlling the compensation magnitude of an equalizer according to a feature of the present invention,

The coefficient of the equalizer may be controlled from the jitter component appearing depending on the data pattern.

Preferably, the jitter component appearing depending on the data pattern is

The data pattern may be obtained from phase position information detected in a section in which data transition occurs after continuous data.

Preferably,

The phase detector may be used to detect jitter components appearing depending on the data pattern.

Preferably,

(1) synchronizing a phase between input data and a clock used in phase detection through a phase detector and outputting a voltage pulse (PUP / PDN) for controlling the voltage controlled oscillator whenever a phase difference occurs;

(2) generating a CEN (Compensation Enable) signal by detecting a section in which data transition occurs after continuous data from the pattern detector to detect a section in which jitter components depending on the data pattern appear;

(3) when the CEN signal is outputted, the PUP / PDN signal output from the phase detector is selected as a signal having phase difference information dependent on the data pattern and used as an input signal; And

(4) controlling the coefficients of the equalizer by converting the input signal into a digital or analog signal to determine the magnitude of the compensation.

Preferably, in said step (1)

The clock for synchronizing the phase with the input data may be characterized in that it is independent of the clock duty ratio.

According to the adaptive method for controlling the compensation amount of the equalizer by extracting jitter information depending on the data pattern proposed in the present invention, accurate compensation of the equalizer is possible.

In addition, according to the present invention, only a small number of clock phases irrespective of the clock duty ratio are used, whereby a structure combined with a phase detector having a similar function can reduce the implementation area and lower the power consumption.

1 is a block diagram of an adaptive algorithm for controlling the compensation magnitude of a conventional equalizer.
2 is a diagram illustrating a transient compensation state by a method of measuring a pulse width of 1 bit data in a conventional jitter measurement algorithm.
3 is a diagram illustrating an undercompensated state by a method of measuring a pulse width of 1 bit data in a conventional jitter measurement algorithm.
4 is a diagram illustrating a receiver to which the adaptive method proposed in the present invention is applied.
5 shows jitter simulation results according to an applied data pattern.
6 is a diagram illustrating the operation of a receiver in which an adaptive method proposed in the present invention is implemented.
7 is a diagram showing a specific configuration of the adaptive method proposed in the present invention.
FIG. 8 is a diagram showing an under-compensation state comparing the operation of the adaptive method proposed by the present invention when inputting data having D [012] = 001 and D [234] = 101. FIG.
FIG. 9 is a diagram illustrating an over-compensation state by comparing the operation of the adaptive method proposed by the present invention when inputting data having D [012] = 001 and D [234] = 101. FIG.
FIG. 10 is a diagram showing an under-compensation state comparing the operation of the adaptive method proposed by the present invention when inputting data with D [012] = 001 and D [456] = 011. FIG.
FIG. 11 is a diagram illustrating an over-compensation state in which an operation of the adaptive method proposed by the present invention is compared with data inputs D [012] = 001 and D [456] = 011. FIG.
12 is a view showing a configuration example of an adaptive method proposed in the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings, in order that those skilled in the art can easily carry out the present invention. In the following detailed description of the preferred embodiments of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In the drawings, like reference numerals are used throughout the drawings.

In addition, in the entire specification, when a part is referred to as being 'connected' to another part, it may be referred to as 'indirectly connected' not only with 'directly connected' . In addition, the term 'comprising' of an element means that the element may further include other elements, not to exclude other elements unless specifically stated otherwise.

1 to 3 are diagrams of a conventional adaptive algorithm. 1 is a diagram showing the configuration of an adaptive algorithm for controlling the compensation magnitude of the conventional equalizer 1. As shown in FIG. 1, conventionally, the coefficient of the equalizer is determined through a method of detecting a minimum pulse width of data. 2 is a diagram illustrating a transient compensation state as a method of measuring a pulse width of 1 bit data in a conventional jitter measurement algorithm, and FIG. 3 is a magnitude of a pulse width of 1 bit data in a conventional jitter measurement algorithm. Is a diagram illustrating an undercompensated state by a method of measuring. That is, conventionally, the compensation magnitude of the equalizer 1 is controlled by detecting the pulse width. The pulse width of the data is measured by the method illustrated in FIGS. 2 and 3, and the equalizer 1 is measured using the measured result value. By changing the coefficient of), it is controlled to maintain the optimized pulse width.

4 is a diagram illustrating a receiver to which the adaptive method proposed in the present invention is applied. As shown in FIG. 4, according to the present invention, the coefficient of the equalizer may be controlled by detecting a phase difference appearing depending on the data pattern through data pattern detection. The role of each block shown in FIG. 4 is as follows. The equalizer 1 may serve as a circuit for compensating high frequency and low frequency components lost through the transmission line. The phase detector 2 is a circuit which outputs phase difference information between the received data and the restored clock. The voltage / current converter 3 is a circuit for converting the signal into a current signal proportional to the time of the voltage output generated by the phase detector 2. The loop filter 4 is a filter for controlling the voltage-controlled oscillator 5 by storing the signal converted into the current signal. Voltage-controlled oscillator 5 is a circuit for generating a recovery clock. The coefficient control 6 is a circuit which controls the coefficient which determines the magnitude of compensation of the equalizer 1. The pattern detector 7 is a circuit for detecting a data section '001' / '110' in which a transition occurs after continuous two-bit signal input.

5 to 7, the operation of the adaptive method proposed in the present invention will be described. Regarding jitter information according to the data pattern, FIG. 5 is a diagram illustrating jitter simulation results according to an applied data pattern. When data is continuously transitioned to '1010' and applied as shown in section 2 of FIG. 5, jitter (phase difference between data) does not occur, but when data transition occurs after continuous data such as section 1 The phase difference of data occurs according to the compensation degree of lost data. If accurate compensation is made, it appears as the same phase as the ideal data. If the compensation is small, the phase is lag, and if the compensation is excessive, it appears as a phase lead.

6 is a diagram illustrating the operation of a receiver in which the adaptive method proposed in the present invention is implemented. The data received through the transmission line is compensated by the equalizer 1. The phase detector 2 synchronizes the phase of the input data with the clock used in the phase detector 2. Whenever a phase difference occurs, a voltage pulse (PUP / PDN) for controlling a voltage-controlled oscillator 5 is output. In order to detect a section in which jitter components depend on the data pattern, a section '001' or '110' where data transition occurs after two consecutive bits from the pattern detector 7 is detected and proposed in the present invention. Generate a CEN (Compensation Enable) signal to operate the adaptive method. When the CEN signal is output, the PUP / PDN signal output from the phase detector 2 is selected as a signal having phase difference information dependent on the data pattern and used as an input signal. The output of the phase detector 2 generated in the section where the CEN signal is not output is used for the phase control 8 having only the phase difference information between the normal data and the clock. The phase control 8 consists of a voltage / current converter 3, a loop filter 4, and a voltage controlled oscillator 5. The input signal is converted into digital and analog signals to control (6) the coefficients of the equalizer 1 to determine the magnitude of the compensation.

7 is a diagram showing a specific configuration of the adaptive method 11 proposed in the present invention. As shown in FIG. 7, the adaptive method 11 proposed in the present invention is composed of a phase detector 2, a pattern detector 7, and a logic 9. While the CEN signal is ON, the output signal of the phase detector 2 is used as an input signal for controlling the coefficient of the equalizer 1. Continuous data detection ('001', '110') is implemented using flip-flops and XOR logic (9).

As an operation example of the adaptive method proposed by the present invention, the output state table of the phase detector 2 and the adaptive method according to the input data is shown in Table 1 below.

As another operation example of the adaptive method proposed by the present invention, the operation of each data input is compared and described. 8 and 9 are diagrams comparing the operation of the adaptive method proposed by the present invention when inputting data with D [012] = 001 and D [234] = 101. 8 illustrates an under-compensation state, and FIG. 9 illustrates an over-compensation state. As shown in Figs. 8 and 9, when data D [012] = '001' is received by the phase detector 2, the adaptive method proposed in the present invention is activated and the output of the phase detector 2 ( PUP and PDN are applied simultaneously with phase compensation and the adaptive method (CUP, CDN) proposed in the present invention. When D [234] = '101', the algorithm activation signal CEN is not generated, so the outputs PUP and PDN of the phase detector 2 are not reflected in the adaptive method proposed in the present invention. Only the phase control 8 of (2) is reflected. Although the adaptive method proposed in the present invention may be activated even if the actual input data is not the transition period after the continuous data, the adaptive method proposed in the present invention is proposed because there is no output of the phase detector 2. It does not affect the method.

10 and 11 are diagrams comparing the operation of the adaptive method proposed by the present invention when inputting data with D [012] = 001 and D [456] = 011. 10 shows the undercompensated state, and FIG. 11 shows the overcompensated state. As shown in Figs. 10 and 11, when data D [456] = '011' is received, the compensation algorithm activation signal CEN is generated even if not two consecutive consecutive identical bits (D [45] = '01'). The exceptions that occur are also raised. However, since the output of the phase detector 2 does not occur, it does not affect the adaptive method proposed in the present invention.

12 is a view showing an example of the configuration of the adaptive method proposed in the present invention. It is a configuration diagram in which the pattern detector 7 can detect the continuous data pattern '001' / '110'.

The present invention may be embodied in many other specific forms without departing from the spirit or essential characteristics of the invention.

1: equalizer 2: phase detector
3: voltage / current converter 4: loop filter
5: voltage controlled oscillator 6: coefficient control
7: pattern detector 8: phase control
9: Logic 10: Pulse Width Detection
11: Adaptive method proposed by the present invention

Claims (5)

A method of adaptively determining and controlling the compensation magnitude of an equalizer,
Control the equalizer coefficients from jitter components that depend on the data pattern,
(1) synchronizing the phase between the input data and the clock used for phase detection through a phase detector and outputting a voltage pulse for controlling the voltage controlled oscillator whenever a phase difference occurs;
(2) generating a CEN (Compensation Enable) signal by detecting a section in which data transition occurs after continuous data from the pattern detector to detect a section in which jitter components depending on the data pattern appear;
(3) when the CEN signal is outputted, the PUP / PDN signal output from the phase detector is selected as a signal having phase difference information dependent on the data pattern and used as an input signal; And
And (4) converting the input signal into a digital or analog signal to control the coefficients of the equalizer to determine the magnitude of the compensation. .
The jitter component of claim 1, wherein the jitter component appears depending on the data pattern.
A method for adaptively determining and controlling the compensation amount of an equalizer, characterized by obtaining from phase position information detected in a section in which a data transition occurs after continuous data in a data pattern.
The method of claim 1,
A method for adaptively determining and controlling the compensation magnitude of an equalizer, characterized by detecting a jitter component that depends on the data pattern using a phase detector.
delete 2. The method according to claim 1, wherein in the step (1)
And said clock synchronizing phase with input data is independent of a clock duty ratio.

Images