JPS6041490B2 - automatic equalizer - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to modification of a transversal automatic equalizer in multilevel data transmission using polyphase phase modulation.

Automatic equalizers for polyphase phase modulation have conventionally used two systems of tapped delay lines whose input signals are the two output signals of a two-way synchronous detector. In many cases, a method is used in which the signals taken out from two taps are converted using a two-dimensional bridge type circuit, which will be described later.

The circuit shown in Figure 1 is a two-dimensional bridge type circuit, with two
A pair of tap gains t, u for two input signals r, s
This is a circuit that obtains outputs V and W by acting in a bridge type manner as shown in the figure.

Further, a variable attenuator using this circuit is called a two-dimensional bridge type variable attenuator. A method for correcting the tap gain of such an equalizer is to create a pair of error signals created from the difference between a pair of brushed output signals and an estimated value obtained from that pair of equalized output signals. A method is used in which the information to be corrected is extracted using the same two-dimensional bridge circuit as shown in Figure 1, and the input signal to the pair of taps to be corrected is extracted, and the tap gain is corrected. I came here. When this method is used, the estimated value must be used to correct the automatic equalizer, so if there is large intersymbol interference at the beginning of automatic equalization, the correct estimated value cannot be obtained and the automatic equalizer tap The drawback was that the gain diverged.

Figure 2 shows the two output signals of the two-axis synchronous detector, respectively.
It is a complex vector diagram with real and imaginary axes, and the signal sampled at an interval equal to the data transmission interval T is as shown by diagonal lines around the original data points due to intersymbol interference. It is received with a circular spread.

As is clear from Figure 2, in polyphase phase modulation, data points are dense in the phase direction, and if intersymbol interference occurs in the shaded area, there is margin for discrimination in the amplitude direction. However, a situation in which identification errors occur in the phase direction is finally occurring. The purpose of the present invention is to use the fact that polyphase phase modulation has discrimination margin in the amplitude direction to correct the automatic equalizer using only the error in the amplitude direction, and to correct the automatic equalizer using only the error in the amplitude direction. The present invention also provides an automatic equalizer that does not cause tap gain divergence due to identification errors.

The principle of the present invention will be explained below.

The outputs x(t) and y(t) of two-rut synchronous detection in the automatic equalizer for polyphase phase modulation are expressed by the following equations.

x(t)=Z {Acosaip(t-iT)-Funami noiq(t-iT)} ,. ,tl'y(t
) = Z {Acosoiq (t-iT) Ichiro ina
ip(t-iT)} ...■ However, oi is the amplitude of the transmitted information A, T is the data interval, and p(t)
is the overall impulse response from the input point of the transmission data Acosai to the x-side output point of the two-way synchronous detection output, q
(t) is the total impulse response from the input point of the transmission data Acosoi to the y-side output point of the two-axis synchronous detection output. Assuming that the timing has been appropriately set and evaluated, consider the response at t=nT. x(iT),y
When (iT), etc. are expressed by xi, etc., the equation '1} and the equation '2) become as follows. Yanagi = Toru's {Ac
os81pn-i-shelf n81qn-i} Ten
...glue; Z {Acosalq
n-i ten$inolpn-i}
...{4'Here, let the mutually orthogonal components be Xn
:xnJujyn...{5)p
n=pnjqn...{
6) Ane=Aejan
...If we decide to express it as a complex number like {7', the expressions '3' and '4) can be combined into ×n 12
AnPn-1,. , [8}.

This is expressed as a complex number tap gain CI=cl+i
When equalization is performed using a transversal filter having dl...{9' as a variable attenuator, the output is a signal Gn+jhn, . . . expressed as a complex number.
It is expressed as follows using OQ.

Gn=science≧3C1Xn” ---(11) Here, the number of taps is N.

In equation (2), applying the attenuation amount CI expressed as a complex number to the signal Xn-1 expressed as a complex number can be realized by the two-dimensional bridge type circuit shown in FIG. Here, consider the following modification formula for sequentially modifying Ck. Ckn+1=C and one QX*-IE small・
(12) However, Q is a correction coefficient and Ckn is Ck
*' means the value at time n, and *' is a complex conjugate.

Further, En follows the following formula. En=static {one Gn'-A}
．． (13) Here, A is a real number and is a reference level representing the amplitude of phase modulation, and therefore the term {ICnl - A} is an amplitude error.

The unique equalized error signal En is obtained by allocating the amplitude error to the proportion of each equalized output. Also, (13 method can be rewritten as follows.En=Gn-Aejane
j〆... (n=m with one heart "It's a plexus.

In a state in which equation (12) is considered to have sufficiently converged, the expected value of Ckn11 and the expected value of C are considered to be equal, and ei〆 is considered to approach a constant value ej.

Therefore, the expected value E (X ice - IEn) of X ice - IEn must be 0. X * 1 IE ice = Xn-1 {Nki1CIXn-・-^ei8ne's n} ‐-
-(15)2=0, so E(×Ice-IEn)=0 (16)
Yorichi lcl Interval Pi*pi+k-1=P*-Ke
Rino Z=o (K=〇, 1,...N-1) (17)
The solution for Ce in the above equation gives the tap gain.

Comparing this with the normal case of normal self-equalization, we get
In this case, the error signal En is En=Gn-An
(Since it is given by 1$, a satisfactory solution for the tap gain is N-1 10 by the same process as shown in equations (15) and (16).

OZ CI Z Pi*Pi tenk-1=P-k2=o(
k go o, 1,...N-1) (obtain 190.

Comparing equations (17) and (19), it can be seen that the solution to equation (17) gives a solution with a rotated phase of e-'' relative to the solution to equation (19). Therefore, it is found that by adding phase control to the equalized output signal and removing the phase rotation -i, these two methods can be converged to a completely equivalent state. Based on the above principle, the present invention corrects automatic equalization using only amplitude error information, thereby making it possible to secure a wider convergence area for various distortions and to maintain characteristics equivalent to normal automatic equalization. It is optimally used for equalization from an arbitrary initial state of polyphase phase modulation. Embodiments of the present invention will be described below with reference to the drawings.

FIG. 3 is a diagram showing an embodiment of the present invention. In FIG. 3, two mutually orthogonal baseband signals demodulated by a two-way synchronous detector enter terminals 1 and 2. A delay line in which delay elements 3-1 and 3-2 are connected in series is connected to terminal 1, and delay elements 4-1 and 4 are connected to terminal 2.
-2 connected in series is connected. Delay element 3
The delay times -1, 3-2, 4-1 and 4-2 are all equal to the data interval T. Two pairs of signal lead-out lines 6-1 and 7-1, 6-2 and 7-2, and 6-3 and 7-3 are drawn out from the connection line connecting each delay element, and the above-mentioned The two-dimensional bridge type circuits 5-1, 5-2, 5-3 shown in FIG. 2, 17
The tap gains output from 13 and 18-3 act. As the output of each two-dimensional bridge type circuit, the signals flowing through the lines 21-1, 21-2, 21-3 are accumulated in the accumulator 8, and the signals flowing on the lines 22-1, 22-2 are accumulated in the accumulator 9. , 22-3 are accumulated, and the signals flowing to terminals 10 and 11 are accumulated, respectively.
An equalized output signal is output. The signals output to terminals 10 and 11 are simultaneously input to an amplitude error generation circuit 13, which generates an amplitude error signal as shown later. The amplitude error signal is multiplied by two equalized signals in a multiplier 14-1 and a multiplier 14-2, respectively, to output two automatically equalized error signals to a line 12-1 and a line 12-2. The signals output to the lines 12-1 and 12-2 are connected to the paired lead lines 6-1 and 7-1, 6-2 and 7-2, 6-
3, 7-3, and the two-dimensional bridge type circuit 19 shown in Figure 1.
11, 19-2, 19-3, and the output pair signal is attenuated by 15-1, 16-1, 15-2, 16-2
, 15-3, 16-3, and the integrators 17-1, 18-1 are multiplied by a constant at a rate of once per data interval T.
, 17-2, 18-2, 17-3, 18-3, a new tap gain is obtained.
By repeating this operation every time one piece of data is output, the tap gain of the automatic equalizer can be corrected to an optimal value. FIG. 4 shows an example of an amplitude error creation circuit, which can be easily realized with the illustrated configuration.

In FIG. 4, two equalized output signals come in from terminals 131 and 132, are squared by squaring circuits 133 and 134, respectively, and the results are added together by adder 135. The read-only memory 136 is adapted to read out the value of {1-unit} of the female organ 135, and can therefore output an amplitude error signal to the terminal 137.

As described above, the present invention constitutes an optimal adaptive automatic equalizer for polyphase phase modulation, but it can also be applied to polyphase phase modulation.

Also, since phase information is not required for automatic equalizer correction,
It also provides effective means as an automatic equalizer for analog and digital mixed modulation in which analog values correspond to each other in the phase direction.

[Brief explanation of the drawing]

FIG. 1 is a block diagram for explaining the operation of the two-dimensional bridge type circuit used in the present invention, FIG. FIG. 4 is a block diagram showing an embodiment of the invention, and is a diagram of an amplitude error generating circuit used in the invention. In the figure, 3-1, 3-2, 4-1, 4-2 are delay elements, 5-1, 5-2, 5-3 are two-dimensional IJ type circuits shown in FIG. 9 is an accumulator, 13 is a judgment circuit, 1
4 is a subtracter circuit, 15-1, 1-2, 15-3, 16-
1, 16-2, 16-3 are attenuators, 17--, 17-2
, 17-3, l8-1, 18-2, 18-3 are integrators,
19-1, 19-2, 19-3, 20-1, 20-2,
20-3 is a multiplier. Figure 1 Figure 2 Figure 3 Figure 4

Claims (1)

[Claims] 1. In a demodulator used for data transmission by polyphase phase modulation, first and second delay lines each receiving two mutually orthogonal baseband signals demodulated by a two-axis synchronous detector; A set of a plurality of signals given different delays of integral multiples of a fixed time drawn from the first delay line, and a set of signals given different delays of integral multiples of a fixed time drawn out from the second delay line. a plurality of two-dimensional bridge type variable attenuators that act on each pair of signals included in both sets and delayed by the same time with respect to the plurality of signals delayed by the plurality of signals; Means for collecting only the first signal out of the two signals outputted by the two-dimensional variable attenuator and calculating the sum of the two signals; an amplitude error generation circuit that outputs an amplitude error signal from the two equalized output signals obtained by the means for determining the two sums; and the amplitude error signal and the two equalized output signals. two automatic equalization error signal generation circuits obtained by multiplying each attenuation amount of all the variable attenuators with the set of all output signals of the first and second delay lines. , and means for correcting by using the automatic equalization error signal, the automatic equalizer eliminating intersymbol interference.