KR19990005736A - Digital data detection system - Google Patents

Abstract

The present invention relates to a digital data detection system in which the calculation of the accumulation path of a survival path is an integer when the output of the equalizer is detected using a Viterbi algorithm. The present invention relates to a partial response of a reproduced signal reproduced after being recorded on a transmission channel. A partial response equalization means for sampling into a channel and outputting the analog trivalue signal, an A / D conversion means for digitizing the analog trivalue signal output from the partial response equalization means and outputting a digitized integer signal, and a noise component In order to obtain a signal closest to the signal recorded on the transmission channel by minimizing the input signal, the calculation of the survival path accumulation process for estimating the survival path by inputting the digitized integer signal output from the A / D conversion means is performed. Viterbi detection means performed by.

In the present invention, since the calculation of the survival path accumulation process of the Viterbi combiner is made to be an integer type, the calculation is simplified and thus the circuit configuration is simplified.

Description

Digital data detection system

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital data detection system, and more particularly to a digital data detection system in which the calculation of the survival path accumulation process is an integer type when the output of the equalizer is detected using the Viterbi algorithm.

When a digital signal is transmitted using a channel, for example, magnetic tape, noise is generated by the characteristics of the channel.

In order to reduce noise generated by using a transmission channel such as a magnetic tape, the partial response channel IV is used in a conventional digital data detection system. The partial response channel IV is processed to have the characteristics similar to that of the medium in consideration of the characteristics of the magnetic medium, so that the magnetic channel response is approximated to the ideal partial response IV channel, thereby reducing the error data when the recording data is restored.

A conventional digital data detection system using such a partial response channel IV is a precoder 100, a channel 110, a partial response equalizer (Partial Response Class IV Equalizer) 120, as shown in FIG. It consists of a two-level detector (130).

The input signal a (t) is precoded as a signal b (t) having an easy error detection form in the precoder 100 and then recorded on the channel 110, that is, the magnetic tape. r (t) is loaded onto the magnetic tape together with the output signal b (t) of the precoder 100.

The signal c (t) recorded after being recorded on the magnetic tape 110 is output as a three-value signal by the partial response equalizer 120. That is, the signal output from the channel 110 is sampled into the partial response channel in the partial response equalizer 120, wherein the reproduced digital signal is formed in analog form to set the amplitude of the analog reproduction data signal at an identification point. The magnitude is determined in comparison with the threshold voltage, and is output as one of the ternary signals yn, for example, '+2, 0, -2'.

The signal yn sampled by the nT period in the partial response equalizer 120 is represented by Equation 1 below.

[Equation 1]

Here, yn is a three-valued signal (+2, 0, -2) output from the partial response equalizer 120 compared with a predetermined threshold value as an analog value, and bn is the current value of the precoder 100. B _n-2 is an output signal of the precoder 100 two cycles ago, and rn is noise.

As described above, the three-value signal output from the partial response equalizer 120 is output as a two-value signal from the two-level detector 130. That is, according to the following equations 2 and 3, the three-value signal is decoded into a two-value signal.

[Equation 2]

when

[Equation 3]

when

Such a conventional digital data detection system recovers a digital signal reproduced from a magnetic tape by using a ternary signal detected by contrast with a threshold, and has a problem in that an error is accumulated due to noise.

The conventional digital data detection system, which has been improved to improve the problem of the conventional digital data detection system, uses the Maximum Lillyhood Sequence Detector in conjunction with the partial response equalizer to reduce the error rate.

The highest order detector is an apparatus for estimating the code sequence that is the lightest for an arbitrary received code string. The detector uses a Viterbi algorithm to efficiently search for a close code string using a trellis.

The Viterbi detector maintains the order of data decisions and corresponding Error Metrics by the Survival Sequence for each possible state assumed by the grid at any data time. For example, if there are two states in the grid, they maintain two survival orders and two corresponding error metrics.

At each data time, the equalized received signal values are evaluated to update the error metric for the state of the grid, resulting in an updated survival order. The updated survival order with the minimum error metric is the recorded data signal order. The oldest data in the survival order is produced at the output of the Viterbi detector and the data is output continuously as the above process is repeated for the next data.

Thus, an improved conventional digital data detection system using a Viterbi detector can perform a decoding function with a lower error rate compared to a conventional digital data detection system using a threshold.

However, in the conventional digital data detection system using a Viterbi detector, it is possible to decode the output signal of the partial response channel equalizer using the Viterbi algorithm. However, since the number of states of the grid is four, the calculation including the algorithm becomes complicated. There is a problem that the circuit configuration is complicated.

The present invention for solving the above problems is to provide a digital data detection system for simplifying the calculation and circuit configuration by making the calculation of the accumulation process of the survival path in the integer form in detecting the output of the equalizer using the Viterbi algorithm. The purpose is.

In order to achieve the above object, the digital data detection system according to the present invention comprises a partial response equalization means for sampling a reproduced reproduction signal recorded on a transmission channel and outputting the reproduced signal into a partial response channel and outputting the analog trivalue signal. A / D conversion means for digitizing the analog 3-valued signal outputted from the digital signal and outputting the digitized integer signal, and the A / D conversion to obtain a signal closest to the signal recorded on the transmission channel by minimizing noise components. Viterbi detection means for performing the calculation of the accumulation process of the survival path for the estimation of the survival path with the input of the digitized integer signal output from the means, characterized in that it comprises a.

1 is a block diagram of a general decoding system

2 is a block diagram showing an embodiment of a digital data detection system according to the present invention

Figure 3 is a block diagram showing an embodiment of a digital data detection system according to the present invention

4 is a diagram for explaining a path generated by a state of an input signal;

5 is a diagram for describing a path generated by a state of an input signal;

FIG. 6 is a detailed configuration diagram of the Viterbi detector of FIGS. 2 and 3.

Explanation of symbols for the main parts of the drawings

200, 300: precoder 210, 310: channel

220, 320: Partial response equalizer 230, 330: A / D converter

240, 350, 360: Viterbi detector 241: branches metric generator

242: comparator 243: difference estimate generator

244: survival path determiner 340: demultiplexer

370: Multiplexer

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

One embodiment of the digital data detection system according to the present invention comprises a partial response equalizer 220, an A / D converter 230, and a Viterbi detector 240. As shown in FIG.

The partial response equalizer 220 samples the reproduced playback signal which is precoded and recorded on the transmission channel, and outputs the analog reproduced signal as a partial response channel.

The A / D converter 230 quantizes the analog digit signal output from the partial response equalizer 220 and outputs the digitized integer signal yn.

The Viterbi detector 240 minimizes the noise component rn so as to obtain a signal closest to the signal recorded in the transmission channel. The digitized integer signal yn output from the A / D converter 240 is obtained. The calculation of the accumulation process of the survival path for estimating the survival path is performed as an integer.

The operation of one embodiment of the digital data detection system according to the present invention configured as shown in FIG. 2 will be described.

The pre-coded recording signal precoded by the precoder 200 is recorded on the transmission channel 210 such as a magnetic tape, and then reproduced. ) Is entered. For this purpose, the tri-valued signal is input to the A / D converter 230 and output as a digitized value, that is, a digitized value. At this time, the counted signal (yn) is an integer and the maximum value and the minimum value are determined by the number of bits used in the A / D converter.

The digitized equalizer output yn is used for estimating the survival path in the beater detector 240, and the calculation for estimating the survival path is made of an integer rather than a real number.

The signal passing through the partial response equalizer 220 is sampled every nT intervals in the A / D converter 230 and the result is shown in Equation 1 above.

In Equation 1 above, yn is a three-value signal (+2, 0, -2) output from the partial response equalizer 120 compared with a predetermined threshold as an analog value, and bn is a current precoder ( Bn-2 has an output signal of the precoder 100 two cycles before, and has -1 or +1.

On the other hand, when using the A / D converter 230 of any number of bits (p) yn is considered in consideration of noise , In Has the value of, and bn is , In Has the value At this time, If C is C, bn becomes + C or -C.

On the other hand, in the Viterbi detector, bk 'is an undetermined value. If + C or -C, the signal code string finally selected ( ) Is a signal code string bn 'that minimizes the following expression (4).

[Equation 4]

In Equation 4 above, yk is the same value for all possible signal code sequences, and the squares of bk 'and bk-2' have the same C ² . Therefore, in order to minimize the above Equation 4, the second term on the right side should be maximized.

In the partial response equalizer 220, the even signal code string and the odd signal code string do not influence each other. Thus, the partial response equalizer can be viewed as two independent die code partial response equalizer systems with the PR polynomial 1-D '(D' = 2D). Another embodiment of the digital data detection system using this is a partial response equalizer 320, an A / D converter 330, a demultiplexer 340, two Viterbi detectors 350, 360 as shown in FIG. , And a multiplexer 370.

Here, the partial response equalizer 320 and the A / D converter 330 are the same as in one embodiment, and the demultiplexer 340 outputs an even signal code string and an odd signal code from the output of the A / D converter 330. The two Viterbi detectors 350 and 360 decode the even-numbered code strings and the odd code strings output from the A / D converter 330 through the demultiplexer 340 as inputs. The multiplexer 370 multiplexes the outputs of the two Viterbi detectors 350 and 360 to form one code string.

In this way, when the signal code strings are separated and processed using two independent die code partial response equalizer systems, the evaluation amount can be separated as shown in Equation 5 below, and the even evaluation amount (JE) and the odd evaluation amount (IO) are respectively Can be treated separately.

[Equation 5]

Here, survival metrics may be defined as in Equation 6 below.

[Equation 6]

Since bn 'may be + C or -C, it can be divided into Equation 7 below.

[Equation 7]

Recursively writing Equation 7 above gives the following Equation 8.

[Equation 8]

Since the evaluation amount J continues to increase over time, it cannot be used in an actual apparatus. However, the information needed to determine the survival path is the difference between the two estimates and their sign. Therefore, it is easy to perform the algorithm by calculating and maintaining the difference evaluation amount as shown in Equation 9 for every nT.

[Equation 9]

When the difference evaluation amount is defined as in Equation 9 above, several situations occur. That is, as shown in FIG. 4, first, when bn is + C and bn-2 is + C, second, when bn is + C, and when bn-2 is -C, third, bn is -C, When bn-2 is + C, a fourth case occurs where bn is -C and bn-2 is -C.

First, when bn is + C and bn-2 is + C, it is performed as shown in Equation 10 below.

[Equation 10]

Secondly, when bn is + C and bn-2 is -C, this is performed as shown in Equation 11 below.

[Equation 11]

Third, when bn is -C and bn-2 is + C, it is as shown in Equation 12 below.

[Equation 12]

Fourth, when bn is -C and bn-2 is -C, it is made as shown in the following formula (13).

[Equation 13]

In four cases as above As shown in Figure 4 can be represented in three forms (①, ②, ③), it is possible to find a possible path.

First of all, ③ If is the difference estimate ( ) Is as shown in Equation 14.

[Equation 14]

Second, ② If is the difference estimate ( ) Is as shown in Equation 15 below.

[Equation 15]

Third, ① of If is the difference estimate ( ) Is as shown in Equation 16 below.

[Equation 16]

In summary, the above relationship has a path as shown in FIG. 5 (a) (b) (d) (d).

That is, the difference evaluation amount ( )this In this case, the survival path of FIG. 5 (A) may be formed, where the current (n) + C or -C becomes 2 C before the cycle (n-2).

In addition, the difference evaluation amount ( )this In this case, the survival path of FIG. 5 (b) may be formed, and the state 2 and the current state n are the same. In other words, the current state (n) is + C in the state n-2 of + C, which is + C, or the current state (n) -C is present in the state (n-2), which is -C, which is -C.

In addition, the difference evaluation amount ( )this In this case, the survival path of FIG. 5 (C) may be formed, and at this time, the current (n) + C or -C degrees in a state two times before (n-2) which is -C.

On the other hand, as shown in Fig. 5 (d), the case where the state before two cycles and the current state n are reversed does not occur. In other words, as described above, when (n) -C is present in the state before (n-2) which is + C, or (n) + C is present in the state before (n-2) which is -C before, Will not occur.

The Viterbi detector 240 used in one embodiment and another embodiment of the digital data detection system according to the present invention has respective reference levels of the first and second states having an ideal digital value as shown in FIG. A branch metric generator 241 for calculating a difference between the integer signal yn outputted from the A / D converter 240 and a branch metric representing possible paths for each state, respectively, the first and the first A comparator 242 for outputting first and second selection signals for comparing the magnitude of the difference between the respective metrics in the two states and selecting the branch metric having the small difference as the current error metric for the first and second states; The difference between the current error metrics for the selected first and second states is calculated as an integer and the difference evaluation amount ( ) Is generated and outputted to be used recursively by the comparator 242, and a survival path is input by inputting first and second selection signals under the control of the difference estimation generator 243. Survival path determiner 244 to follow.

As described above, the difference evaluation amount ( ) Is defined by the following equation.

Formula 17

when

when

when

Where ± C is the respective reference level in the first and second states, yn is the current digitized integer signal, and yn-2 is the digitized integer signal two cycles ago.

Equation 17 above summarizes the equations 14, 15, and 16 above.

The operation of the Viterbi detector configured as described above will be described.

In the branch metric generator 241, the reference level (+ C, -C) of the first and second states of the ideal digital value and the integer signal (yn) output from the A / D converter 240 are used. The difference is calculated to produce each branch metric representing the possible paths for each state.

The comparator 242 compares the magnitude of the difference between the respective metrics in the first and second states (+ C, -C) according to the difference evaluation amount output from the difference estimation amount generator 243, thereby reducing a small difference. Outputting a first and second selection signal that selects the branch metric to have as the current error metric for the first and second states. Here, the selection signal is a sign necessary for determining the survival path.

In addition, the state difference evaluation amount generator 243 calculates the difference between the current error metrics for the selected first and second states as an integer type according to Equation 17 above, and calculates the difference evaluation amount ( ). The difference estimate generated at this time ( ) Is input to the comparator 242 to be used recursively.

Under the control of the difference estimation generator 243, first and second selection signals corresponding to codes output from the comparator 242 are input to the survival path determiner 244 to be used to track the survival path. .

As described above, in the digital data detection system according to the present invention, the calculation of the survival path accumulation process of the Viterbi detector is an integer while the calculation of the survival path accumulation process of the Viterbi combiner is real. The calculation is simplified and the circuit configuration is simplified accordingly.

Claims (3)

Partial response equalization means for sampling the reproduced signal reproduced in the transmission channel into a partial response channel and outputting it as an analog 3-valued signal; A / D conversion means for digitizing the analog digit signal output from the partial response equalization means and outputting a digitized integer signal; And a process of accumulating a survival path for estimating a survival path by inputting a digitized integer signal output from the A / D conversion means to minimize a noise component and obtain a signal closest to a signal recorded on the transmission channel. And a Viterbi detection means for performing the calculation in an integer form. 2. The apparatus according to claim 1, wherein the Viterbi detection means calculates a difference between the respective reference levels of the first and second states of the ideal digital value and the integer signal output from the A / D conversion means to each state. A branch metric generator, each generating a branch metric indicative of possible paths for the path; Comparing the magnitudes of the differences between the respective metrics in the first and second states and outputting first and second selection signals for selecting branch metrics having a small difference as current error metrics for the first and second states. Comparator; A difference estimate generator for calculating a difference between the current error metrics for the selected first and second states as an integer, generating a difference estimate, and outputting the difference estimate recursively for use in the comparator; And a survival path determiner that tracks a survival path by inputting first and second selection signals under the control of the difference evaluation amount generator. The method of claim 2, wherein the difference evaluation amount ( )silver when when when Wherein ± C is the respective reference level in the first and second states, yn is the current digitized integer signal, and yn-2 is the digitized integer signal two cycles ago.