JPH06120774A - Data reception equipment - Google Patents

Abstract

PURPOSE:To provide the data reception equipment which gives an excellent error rate improvement effect to the equalizer which sets a tap coefficient in accordance with the condition of a line and changes the transmission characteristic to compensate the waveform distortion and improves the error rate with respect to digital communication. CONSTITUTION:A reception signal 1 is inputted to a correlation calculator 2, and obtained correlation values are inputted to a maximum value detector 3 to detect the component having a maximum absolute value out of correlation values. The initial value of the tap coefficient of an equalizer 5 is set to a value, which is easy to converge, in a main tap setter 4 based on this information and correlation values by training. Thus, the tap coefficient of high precision is stably set with a small operation volume regardless of short training which does not degrade the transmission efficiency, and the excellent error rate improvement effect for reception data 6 after equalization is obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a data receiving device in a digital communication receiver or the like.

[0002]

2. Description of the Related Art In recent years, the development of digital mobile communications has been urgently needed in Japan, the United States and Europe, and services will soon start. Above all, the European system has a higher transmission rate than the US and Japanese systems and is therefore easily affected by the delayed wave, and the waveform distortion must be compensated by the equalizer in the receiver. Since the line condition of the equalizer changes from moment to moment, different tap coefficients must be set for each time slot in time division multiple access communication (TDMA communication), and this must be set by initial synchronization (training). Since a long initial synchronization sequence (training sequence) is required to perform the operation, the transmission efficiency is significantly reduced. Therefore, a method of setting the initial value of the tap coefficient with a short training sequence has been considered.

A conventional data receiving apparatus will be described below. 5 and 6 each show a schematic configuration of a conventional data receiving device. In FIGS. 5 and 6, 1 is a received signal, 2 is a correlation calculator, and 3 is a maximum value detector. The received signal 1 is input to the correlation calculator 2. The correlation calculator 2 takes a correlation with known data, and the correlation value is input to the maximum value detector 3, and the sum of squares of the real part and the imaginary part of the obtained correlation value, that is, the component with the maximum absolute value of the correlation value. Is detected.
5 is an equalizer, and this equalizer 5 is a DFE (Decision).
It is a Feedback Equalizer).

FIG. 7 shows the internal structure of the equalizer 5. In FIG. 7, 7 is a plurality of delay elements, and the delay time of one delay element is T. A portion 8 surrounded by a dotted line is a linear portion. Reference numeral 9 denotes a first plurality of weighting units, which weights only the coefficients of the linear unit 8 to determine tap coefficients. Reference numeral 10 is an adder and 11 is a discriminator from which received data is output. 12 is an error output. A portion 13 surrounded by another dotted line is a decision feedback unit. 1
Reference numeral 4 denotes a second plurality of weighting devices, which weights only the coefficients of the decision feedback unit 13 to determine tap coefficients.

FIG. 5 shows a conventional data receiving apparatus in which the tap coefficient of the decision feedback section 13 of the equalizer 5 shown in FIG. 7 is set first, 15 is the decision feedback section setter, and 16 is the linear section setter. Is. The decision feedback unit setter 15 sets the decision feedback unit tap coefficient by the second plurality of weighters 14 shown in FIG.
The linear part setter 16 sets the linear part tap coefficient from the information and the correlation value. The reception data 17 is output from the equalizer 5.

FIG. 6 shows a conventional data receiving apparatus in which the tap coefficient of the linear section 8 of the equalizer 5 is set first. The linear section setting unit 16 is used to set the first plurality of weighting units 9 shown in FIG. Then, the linear feedback section tap coefficient is set, and the decision feedback section setting unit 15 sets the decision feedback section tap coefficient from the information and the correlation value. The reception data 17 is output from the equalizer 5.

The operation of the conventional data receiving apparatus configured as described above will be described below. First,
Since the training sequence, which is known data, is inserted in the transmission data, and the reception data 1 has the transmission signal convoluted with the impulse response of the line, the correlation calculator 2 correlates the training sequence with the training sequence. The impulse response of is obtained as the correlation value. The maximum value detector 3 detects the component having the maximum absolute value among the obtained correlation values and establishes synchronization. When the decision feedback unit tap coefficient is first set by the second plurality of weighting units 14 as shown in FIG. 5, first, the correlation value obtained by the decision feedback unit setting unit 15 is normalized with the component having the maximum absolute value. The decision feedback part tap coefficient is set by the second plurality of weighting devices 14 by converting it into a linear equation, and the linear part setting device 16 formulates a normal equation from the information and the obtained correlation value. The linear part tap coefficient is set by the plurality of weighting devices 9 of. When the linear part tap coefficient is set first by the first plurality of weighting devices 9 as shown in FIG. 6, a normal equation is first established from the correlation value obtained by the linear part setting device 16, and this is solved. Set the linear part tap coefficient and set the decision feedback part setter 1
The second plurality of weighting devices 14 determine by convolving the linear part tap coefficients obtained by the first plurality of weighting devices 9 with the correlation value obtained in 5 and normalizing with the maximum component. Set the feedback section tap coefficient. The equalizer 5 performs equalization with the tap coefficient set as described above, and outputs received data 17 with an improved error rate.

[0008]

However, in the above conventional data receiving apparatus, since the correlation value may deviate from the true impulse response due to the influence of noise or quantization of input data, when solving the normal equation, As the calculation is repeated many times, errors accumulate and the solution becomes unstable,
There are times when a solution that is not optimal for the line can be obtained, and as a countermeasure to this, it has been judged whether the solution is unstable or performing training after setting the tap coefficient. There is a problem that the amount of calculation is large and the solution is determined and the training is performed only by solving the equation, and thus the amount of calculation is very large and realization is difficult.

The present invention solves the above-mentioned conventional problems, and provides an excellent data receiving apparatus capable of stably and accurately setting an initial value of a tap coefficient in an equalizer with a short training sequence and a small amount of calculation. With the goal.

[0010]

SUMMARY OF THE INVENTION To achieve the above object, the present invention detects a correlation calculator for calculating the correlation between a received signal and a known signal, and a component of the maximum absolute value of the obtained correlation values. The maximum value detector, the main tap setter that sets the initial value of the tap coefficient to a value that easily converges during training from that information and the correlation value, and starts the training from that tap coefficient, and the excellent error rate improvement after training It has the structure of an equalizer having an effect.

[0011]

Therefore, according to the present invention, the tap coefficient of the equalizer is set so that the component having the maximum absolute value of the correlation value becomes 1 after the equalization, and then the training is performed. The tap coefficient of the equalizer is set accurately.

[0012]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the structure of a data receiving apparatus according to an embodiment of the present invention. In FIG.
Reference numerals 1 to 3 and 5 are the same as those shown in FIGS. Further, the internal configuration of the equalizer 5 is the same as that shown in FIG. 7 in the conventional example, and therefore the description thereof will be omitted.

In FIG. 1, reference numeral 4 denotes a main tap setter, which sets only the tap coefficient relating to the component having the maximum absolute correlation value among the tap coefficients of the equalizer 5, and performs training in a short training sequence. Equalization is performed. Reference numeral 6 is received data, which is obtained by the equalizer 5 performing equalization.

2 to 4 are diagrams for explaining the operation. FIG. 2 shows a burst format. As shown in FIG. 2, a training sequence, which is known data, is inserted in the transmission data. FIG. 3 shows the relationship between the training time (number of symbols) and the square error. FIG. 4 shows an example of the relationship between the relative time and the absolute value of the impulse response correlation value.

The operation of the data receiving apparatus configured as described above will be described below. First, a training sequence (known data) is inserted in the transmission data as shown in FIG. 2, and the reception data 1 has the transmission signal convoluted with the impulse response of the line. By correlating with, the impulse response of the line is obtained as the correlation value. The maximum value detector 3 detects the component having the maximum absolute value among the obtained correlation values and establishes synchronization. In the example shown in FIG. 4, since E is the absolute value of the maximum correlation value, this is selected. The main tap setting device 4 is the first
Of the linear part tap coefficients obtained by the plurality of weighting units 9 of, only the taps related to the component having the maximum absolute value of the correlation value are set as the reciprocal of the component (since the correlation value is a complex number, it is complex. All other linear part tap coefficients obtained by the first plurality of weighting devices 9 and the decision feedback part tap coefficients obtained by the second plurality of weighting devices 14 are set to 0. By doing so, although there is some deviation due to the influence of noise or the like, the component with the maximum correlation value becomes approximately 1 after equalization. Compared with the operation of solving a normal equation in which the operation is repeated many times while successively using the previous value, this operation is less affected by noise and the like, and stable setting can be performed. The equalizer 5 increases the accuracy of the tap coefficient by short training from the tap coefficient set by the main tap setter 4, then equalizes the data, and outputs received data 6 with an improved error rate. .

The operation of the data receiving apparatus having the above-mentioned structure will be described with reference to FIGS. When training, as shown by the dotted line in FIG. 3, if it starts from a state where the squared error is small in advance, the convergence will be fast, so it is possible to perform training in a short training sequence, the amount of calculation is small, and the data It does not have to lower transmission efficiency. If the tap coefficient is determined so that the maximum component (E shown in FIG. 4) of the correlation value becomes 1, the absolute values of the correlation values of the other components (A to D, F to H) are smaller than the maximum component. , Even if the tap coefficient of the equalizer 5 is set so that only E is set to 1, the training can be started from a state where the error is considerably small. Therefore, even if training is performed in a short training sequence, the accuracy of the tap coefficient is sufficient. Can be raised.

As described above, according to the above-described embodiment, the maximum absolute value detector 3 detects the component having the maximum absolute value among the correlation values obtained by the correlation calculator 2, and the main tap setting unit 4 detects the first absolute value. Of the linear part tap coefficients obtained by the plurality of weighting devices 9, the reciprocal of the correlation value is set only for the tap having the largest absolute value of the correlation value, and the first plurality of weighting devices are set. 9
The other linear part tap coefficients obtained by the above and the decision feedback part tap coefficients obtained by the second plurality of weighters 14 are set to 0, and then training is performed in a short training sequence to sufficiently improve the accuracy of the tap coefficients. Therefore, there is an effect that it is possible to stably realize an excellent error rate improvement characteristic for the received data 6 with a small amount of calculation without lowering the transmission efficiency.

[0019]

As is apparent from the above embodiments, the present invention provides a correlation calculator for calculating the correlation between a received signal and a known signal, and a maximum value for detecting the maximum absolute value component of the obtained correlation values. From the detector, its information and the correlation value, the main tap setter that sets the initial value of the tap coefficient to a value that easily converges during training, and the training starts from that tap coefficient, and an excellent error rate improvement effect is obtained after training. By providing with the equalizer that has, without reducing the transmission efficiency,
It is possible to realize an excellent data receiving apparatus that can realize an excellent error rate improvement characteristic for received data in a stable manner and with a small amount of calculation.

[Brief description of drawings]

FIG. 1 is a schematic block diagram of a data receiving apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a burst format.

FIG. 3 is a diagram showing the relationship between training time (number of symbols) and square error.

FIG. 4 is a diagram showing a relationship between relative time and absolute value of impulse response correlation value.

FIG. 5 is a schematic block diagram of a conventional data receiving device (when the decision feedback unit is set first).

FIG. 6 is a schematic block diagram of a conventional data receiving device (when a linear part is set first).

FIG. 7 is a schematic block diagram of an equalizer (DFE).

[Explanation of symbols]

 1 Received signal 2 Correlation calculator 3 Maximum value detector 4 Main tap setting device 5 Equalizer

Claims (1)

[Claims] 1. A correlation calculator for calculating a correlation between a received signal and a known signal, a maximum value detector for detecting a component of a maximum absolute value among correlation values obtained by the correlation calculator, and a maximum value thereof. Data including a main tap setter that sets the initial value of the tap coefficient to a value that easily converges in training from the information obtained by the detector and the correlation value, and an equalizer that starts the training from the tap coefficient Receiver.