JPH03278722A - Data receiver - Google Patents

Abstract

PURPOSE:To attain equalization without deterioration in the effect even when a line complex number impulse response is fluctuated with time by multiplying a set coefficient with an individual complex number impulse response, adding them and giving the result to a delay device so as to output an estimate value therefrom. CONSTITUTION:Complex number impulse response inputted to multipliers 70-7(M-1) are multiplied with AR model coefficients 80-8(M-1) obtained by a coefficient setting device 4 and the results of multiplication of the multipliers 70-7(M-1) are added by an adder 9. The impulse response is inputted to an equalizer 11 as an estimate value 10 via a changeover circuit 5(M-1) and delay devices 61-6(M-1). The equalizer 11 uses the information being the estimate value 10 to equalize a reception signal 1 and to obtain a demodulation signal 12. Thus, even when a line complex number impulse response is fluctuated with time, the deterioration in the performance of equalization is suppressed.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a data receiving device used in digital mobile communications and the like.

Conventional technologyDigital data communication devices have become widely used in recent years because they can perform so-called signal equalization, which compensates for signal waveform distortion on the receiving side, and thereby achieve high fidelity. It has become.

Hereinafter, a conventional data receiving device of this type will be explained based on the drawings.

FIG. 2 is a block diagram showing a schematic configuration of a conventional data receiving device. In FIG. 2, 121 is a received signal, and 122 is a received signal 1 by inputting this received signal 121.
This is an impulse response estimator that estimates a complex impulse response of a line from known signals included in 21 and outputs the complex impulse response.

Further, 123 is an equalizer that equalizes the received signal 121 based on the output of the impulse response estimator 122 and outputs a demodulated signal 124.

Next, the operation of the above conventional example will be explained. In FIG. 2, the impulse response estimator 122 includes a received signal 121
The complex impulse response of the line is estimated from the known signals included in the information, and the information is given to the isoflower 123.

By equalizing the received signal 121 based on this information, the equalizer 123 can remove deterioration due to frequency fading and the like and obtain a demodulated signal 124 with a lower error rate than when the received signal 121 is demodulated as is. can.

In this way, even in the conventional data receiving apparatus described above, the complex impulse response of the line can be estimated by the impulse response estimator 122, and that information can be used.
When there is no temporal variation in the complex impulse response of the line, deterioration due to frequency selective fading and the like can be removed, and a demodulated signal with a good error rate can be obtained.

Problems to be Solved by the Invention However, once a complex impulse response is estimated, in the conventional data receiving device described above, the first estimated complex impulse response is used even for a received signal at a time different from the time when the complex impulse response was estimated. Since the equalization is performed using the same method, there is a problem in that the effect of equalization deteriorates when the complex impulse response of the line fluctuates over time.

The present invention solves these conventional problems,
It is an object of the present invention to provide an excellent data receiving device that can prevent the equalization effect from deteriorating even if the complex impulse response of a line fluctuates over time.

Means for Solving the Problems In order to achieve the above object, the present invention includes a plurality of impulse response estimators that input received signals and estimate complex impulse responses of lines at different times, and an estimation method using the impulse response estimators. A coefficient setter sets the coefficients for modeling the complex impulse response, and the coefficients set by this coefficient setter are multiplied by the complex impulse response at different times set by the impulse response estimator. and a calculation means that calculates an estimated value of a time-varying complex impulse response of the line through a delay device after adding the multiplication results, and an equalizer that equalizes the received signal using the estimated value and outputs a demodulated signal. The structure is as follows.

Therefore, according to the present invention, a plurality of impulse response estimators estimate the complex impulse response of a line at different times, and a coefficient setting device sets coefficients for modeling the complex impulse response. By multiplying the coefficients set by the coefficient setter by the individual complex impulse responses, then adding the respective multiplication results, and passing this addition result through a delay device to output an estimated value. , it is possible to approximate the complex impulse response at different points in time using a fixed model equation and sequentially estimate the complex impulse response at any point in time of the line, so even if there is temporal variation in the complex impulse response of the line, the effect will not deteriorate. Equalization can be performed without

Example Embodiments of the present invention will be described below based on the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a data receiving apparatus according to an embodiment of the present invention. In Fig. 1, 1 is a received signal, and 20 to 2 (M-1) each input the received signal 1, and use known signals included in the received signal 1 to determine the line at a shifted point in time. The complex impulse responses are estimated and the complex impulse responses 30 to 3 (M−
1) (M is the number of impulse response estimators).

4 are the respective impulse response estimators 20 to 2 (M-
1) Complex impulse responses 30 to 3 (M-
Input the initial value of 1) and create each complex impulse response 30~
This is a coefficient setting device that outputs AR model coefficients 80 to 8 (M-1), which are coefficients for modeling 3 (M-1).

The AR model coefficients mentioned here are impulse response 30~
This is a coefficient used when modeling and approximating a stochastic time series such as y(M-1), for example, stochastic time series y(
t) as the model y(t)=-aly(t-1)-azy(t-2)"
・This refers to Ql, a2°..., an in the above equation when generated by -any(t-n)+bov(t).

50-5 (M-1) is the impulse response estimator 20-2
Impulse responses 30 to 3 (M-1) output from (M-1)
-1), the delay devices 61 to 6 (M-
1) side terminal AO-Au-+ to multipliers 70 to 7 (M-
This is a changeover switch that switches between terminals BO to BM-1 on the 1) side.

The delay devices 61 to 6 (M-1) store only the initial values of the complex impulse responses 30 to 3 (M-1) based on the initial estimation of the impulse response estimators 20 to 2 (M-1). Impulse response estimator 20-2 (M-1)
Complex impulse response from second and subsequent estimations 30-3
(M-1) changes the changeover switches 50 to 5 according to this response.
(Ml) is the terminal BO to B on the multiplier 70 to 7 (M-1) side
By switching to M-1, multipliers 70 to 7 (M-
1).

Also, the multiplier 70~? (M-1), the complex impulse responses 30 to 3 (M-1) estimated from the second time onwards by the impulse response estimators 20 to 2 (M-1) are set to the changeover switches 50 to 5 (M-1). AR model coefficients 80 to 8 (M-1) set by the coefficient setter 4 are also input together.

This causes the multipliers 70 to 7 (M-1) to multiply the AR model coefficients 80 to 8 (M-1) and the complex impulse responses 30 to 3 (M-1) estimated from the second time onwards. It has become.

9 adds the multiplication results output from the multipliers 70 to 7 (M-1), and transfers the addition result to the selector switch 5 (M-1).
1), the adder 13 outputs the output to the delay devices 61 to 6 (M-1) through the adder 13, the changeover switches 50 to 5 (M-1), the delay devices 61 to 6 (M-1), and the multiplier 70. 7 (M-1) and an adder 9, and is an arithmetic means for calculating the estimated value 1o.

11 is demodulated by inputting the estimated value 10 based on the addition result inputted to the delay devices 61 to 6 (M-1) via the changeover switch 5 (M-1) and equalizing the received signal. This is an equalizer that outputs a signal 12.

Next, the operation of the above embodiment will be explained. In the above embodiment, the impulse response estimators 20 to 2 (M-1) use known signals included in the received signal 1 to estimate the complex impulse response of the line at the shifted time point, and calculate the complex impulse response h (0;11), h(1;n)
, ...-, h(M-1;n) (n is the number of the impulse response).

Here, if the temporal fluctuation of the complex impulse response h(t, n) can be approximated by the AR model, it can be expressed as and X is A
the order of the R model, a(i, n) is the coefficient of the AR model].

Therefore, by determining the order X of the AR model and finding the coefficient a(i, n) of the AR model, it is possible to estimate the complex impulse response h(t, n) at any point in time. In the case of this embodiment, a plurality of impulse response estimators 20 to 2 (M
−1) by multiple complex impulse responses h(0,n)
', h(1°n), ·, h(M-1, n) is found, so the order X of the AR model becomes M-1 at most.

From the above equation (I), the coefficient a(i, n) of the AR model that minimizes the squared error between both sides is determined by t=M-1,X=
In order to find it at the time of M-1, first, multiply both sides of equation (I) by h'<yt-1-'', n> and take the expected value. Here, the complex impulse response h (0°n), h
Complex impulse responses other than (1, n), -, h (M-1, n') are assumed to be O. Further, K=1 to (M-1). If this is expressed as a matrix, it becomes.

By solving this determinant, the coefficient a(f
, n) are found.

The coefficient calculator 4 performs this calculation and calculates the coefficient a(
0,n) ~a(M-1,Jl) as AR model coefficient 80
~8(M-1).

On the other hand, the initial values of the complex impulse responses 30 to 3 (M-1) output from the impulse response estimators 20 to 2 (M-1) are set to Delay device 61 only when estimating impulse response
~6(M-1) is stored in the delay line.

When estimating the complex impulse response from the second time onward, the changeover switches 50 to 5 (M-1) switch to the delay devices 61 to 6 (M-1).
1) side terminal AO-/ from w-1 to multipliers 70 to 7 (
M-1) side terminals Bo-Bu-s and sequentially multiply the newly estimated complex impulse response by W units 70 to 7 (M-1).
.

The complex impulse responses input to the multipliers 70 to 7 (M-1) are here multiplied by the AR model coefficients 80 to 8 (M-1) obtained by the coefficient setter 4, and each multiplier 70
The multiplication results of .about.7(M-1) are added together in an adder 9. This operation realizes equation (1).

The impulse response obtained in this manner is input to the flower vase 11 as an estimated value 10 via the changeover switch 5 (M-1) and the delay devices 61 to 6 (M-1). Etiquette vase 1
1 equalizes the received signal 1 using the information of the estimated value 10 to obtain a demodulated signal 12.

As described above, according to the above embodiment, the temporal fluctuation of the complex impulse response of the line is predicted, and the AR model coefficients 80 to 8 (
Since the complex impulse response at any point in time can be estimated using M-1), the received signal 1 can be equalized by the equalizer 11 using the estimated value 10, and the complex impulse response of the line can be Even when fluctuations occur, there is an advantage that deterioration in equalization performance can be suppressed and a demodulated signal 12 with good error rate characteristics can be obtained.

In the above embodiment, the complex impulse response was directly estimated by the impulse response estimators 20 to 2 (M-1), but the complex impulse response was calculated as follows: h(t;n)=hi(t:n)+Jhq (
t ; n), and the real part hi (t; n) and the imaginary part h (
t; n) and may be estimated independently. In this case, the configuration and operation are similar to those of the above embodiment.

Also, considering the complex impulse response as follows, the amplitude component (jan) and the phase component θ(t;n)
It may be divided into 2 and estimated independently. In this case as well, the configuration and operation are the same as in the above embodiment.

Effects of the Invention As described above, the present invention includes a plurality of impulse response estimators that input received signals and estimate complex impulse responses of lines at different times, and a plurality of impulse response estimators that estimate complex impulse responses of lines at different times. A coefficient setter sets coefficients for modeling, and the coefficients set by this coefficient setter are multiplied by complex impulse responses at different times set by the impulse response estimator, and the multiplication results are added. The configuration includes a calculation means that calculates an estimated value of a complex impulse response that changes over time of the line through a delay device, and an equalizer that equalizes a received signal using the estimated value and outputs a demodulated signal.

Therefore, a plurality of impulse response estimators are used to estimate the complex impulse response of the line at different times, and a coefficient setter is used to set coefficients for modeling this complex impulse response. Multiply the coefficient set by the individual complex impulse response,
Furthermore, by adding the results of each multiplication, passing this addition result through a delay device, and outputting an estimated value, we can approximate the complex impulse response at different points in time using a fixed model equation, and calculate the complex impulse response at any point in the line. Since impulse responses can be estimated sequentially, equalization can be performed without deterioration of effectiveness even if there is temporal variation in the complex impulse response of the line.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a schematic configuration of a data receiving device according to an embodiment of the present invention, and FIG. 2 is a block diagram showing a schematic configuration of a conventional data receiving device. 20-2(M-1)... Impulse response estimator, 4.
... Coefficient setter, 61-6 (M-1) ... Delay device, 8
0 to 8 (M-1)...AR model coefficient, 11...Equistop, 13...Calculating means.

Claims (1)

[Claims] Multiple impulse response estimators that input received signals to estimate the complex impulse responses of lines at different times, and a coefficient setter that sets coefficients for modeling the complex impulse responses estimated by the impulse response estimators. The coefficient set by this coefficient setter is multiplied by the complex impulse response at different times set by the impulse response estimator, and the multiplication results are added, and then the time-varying complex of the line is transmitted through a delay device. A data receiving device comprising a calculation means for calculating an estimated value of an impulse response, and an equalizer for equalizing a received signal using the estimated value and outputting a demodulated signal.