JP2984556B2 - Cross-polarization interference compensator - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a cross-polarization interference compensator, and more particularly to a transversal filter in a receiver in a transmission system for transmitting a digitally modulated signal using two orthogonal polarizations having the same frequency. And a cross-polarization interference compensator for removing interference between cross-polarizations.

[0002]

2. Description of the Related Art FIG. 5 shows an example of a conventional cross-polarization interference compensator used on the receiving side of a dual-polarization transmission system for transmitting a digital modulation signal using horizontal and vertical polarizations orthogonal to each other at the same frequency. This will be described with reference to the block diagram of FIG.

A self-polarized received signal 102 and a differently polarized received signal 101 are both digitally modulated on the transmitting side, transmitted using mutually orthogonal horizontal and vertical polarized waves of the same radio frequency, and demodulated on the receiving side. A digital signal after analog-digital conversion. Then, the received signal 101 of the different polarization is input to the transversal filter 1.

The transversal filter 1 has a delay line 11, which has a delay time equal to half the transmission symbol interval,
12, weighting devices 13, 14, 15 for weighting the tap outputs 101, 101 ', 101 "based on tap weighting factors C1, C2, C3, and weighting devices 1
And a synthesizer 16 for synthesizing the outputs of 3, 14, and 15.

[0005] The different polarization reception signal 101 is input to a delay line 11, and is output as a first tap output by a weighting device 13.
Is also input to the tap weighting coefficient generator 4. The output 101 ′ of the delay line 11 is input to the delay line 12 and also to the weighting device 14 and the tap weighting coefficient generator 5 as a second tap output. Output 10 of delay line 12
1 "is input to the weighter 15 and the tap weighting coefficient generator 6 as a third tap output.

Weighters 13, 14, and 15 multiply the input signals by tap weighting coefficients C1, C2, and C3 generated by tap weighting coefficient generators 4, 5, and 6, respectively. An arithmetic operation for summing the outputs of 13, 14, and 15 is performed, and a cross-polarization interference compensation signal 103 is output.

[0007] The cross-polarization interference compensation signal 103 is
And is combined with the self-polarized reception signal 102. When the own-polarized reception signal 102 is subjected to cross-polarization interference, the cross-polarization interference compensation signal 103 is a signal having the same amplitude and opposite polarity as the cross-polarization interference component, and the output signal 1 output from the coupler 2.
04 is a signal from which cross-polarization interference has been removed.

The error component detector 3 receives the output signal 104 as input, detects residual error components remaining there, and outputs the error signal 10.
Output as 5. This error signal 105 is input to tap weighting coefficient generators 4, 5, and 6, respectively. The tap weighting factor generators 4, 5, and 6 perform a correlation operation between the tap outputs 101, 101 ', and 101 "of the transversal filter 1 and the error signal 105, and assign tap weighting factors C1, C2, and C3 to weighting devices, respectively. 13,14,1
Output to 5

The tap weighting coefficient generator 4 has a first integrator 41 to which the differently polarized reception signal 101 and the error signal 105, which are the first tap output, are input, and an output of the first integrator 41. A second integrator 42 for weighting with the determined coefficient -α, and an output of the second integrator 42 as one input and the output as a tap weighting coefficient C1.
An adder 4 that outputs to 3 and uses it as its other input
And 4. The other tap weighting coefficient generators 5 and 6 have the same configuration.

[0010]

The tap weighting factor generators 4, 5, 6 of the above-mentioned conventional cross-polarization interference compensator.
Then, assuming that a tap weighting coefficient at time t is C (t), an error signal is E (t), and a tap output signal is D (t), C (t + 1) = C (t) −α ｛E (t) D (t)｝... (1) A tap weighting coefficient is generated based on the following equation.
This equation means -α {E (t) .D (t)}.
Is integrated from time 0 to time t, and the past data is continuously added without disappearing.

Therefore, the integrand value E (t) · D
In the case where a small bias is superimposed on (t) due to the influence of hardware imperfections or the like, even if the original tap weighting coefficient should be 0, a long time integration results in a small bias. , The tap weighting coefficient continues to grow, overflowing, and correct tap weight control may not be performed.

As described above, when the correct tap weight control cannot be performed, it is desired that the cross-polarization interference compensation signal 103 be at a minute level close to zero in a favorable propagation state in which there is almost no cross-polarization interference. Even in such a case, a problem may occur in that the cross-polarization interference compensation signal 103 is output at a level sufficient to cause a code error in the output signal 104.

An object of the present invention is to provide a cross-polarization interference compensator capable of performing stable tap weight control while preventing the growth of tap weight coefficients due to accumulation of minute bias levels.

[0014]

According to the present invention, a cross-polarized signal is transmitted using a transversal filter in a receiver in a transmission system for transmitting a digitally modulated signal using two orthogonal polarizations having the same frequency. A cross-polarization interference compensator for removing mutual interference,
Each of the weighting coefficient generation means of each tap of the transversal filter, an integrator that generates a signal corresponding to an integrated value of each tap output and an error signal extracted from the signal after cross-polarization interference compensation, a first adder for adding a predetermined correction value to the accumulated output, pressurizing of said first adder
A second adder that further adds the calculation power and its own added output to obtain the latest tap weighting coefficient, and generates the correction value having a polarity corresponding to the polarity of the added output of the second adder. A cross-polarization interference compensator characterized by having a correction value generator as an addition input of the first adder.

[0015]

The tap weighting coefficient at a certain time t is represented by C
(T), the error signal after interference compensation is E (t), and the tap output signal is D (t), C (t + 1) = C (t) −α {E (t) · D (t)} − βsgn {C (t)} (2) The tap weighting coefficient generation control is performed based on the following equation.

That is, -α in the above equation (1)
{E (t) · D (t)} term (integrated value from time 0 to t)
Is extinguished with time as -βsgn {C (t)} as shown in equation (2). still,
In the equation (2), sgn {X} means a function indicating the polarity of X.

[0017]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of an embodiment of the present invention, and the same parts as those in FIG. 5 are denoted by the same reference numerals.

Both the self-polarized received signal 102 and the hetero-polarized received signal 101 are digitally modulated on the transmitting side, transmitted using orthogonal horizontal and vertical polarized waves of the same radio frequency, and demodulated on the receiving side. A digital signal after analog-digital conversion. Then, the received signal 101 of the different polarization is input to the transversal filter 1.

The transversal filter 1 has delays 11 and 12 having a delay time of 1/2 of a transmission symbol interval.
And weighting devices 13, 14, 15 for weighting the tap outputs 101, 101 ′, 101 ″ based on separately provided tap weighting factors C1, C2, C3, and outputs of the weighting devices 13, 14, 15 It is composed of a synthesizer 16.

The different polarization reception signal 101 is input to the delay line 11 and is output as a first tap output by the weighting device 13.
Is also input to the tap weighting coefficient generator 4. The output 101 ′ of the delay line 11 is input to the delay line 12 and also to the weighting device 14 and the tap weighting coefficient generator 5 as a second tap output. Output 10 of delay unit 12
1 "is input to the weighter 15 and the tap weighting coefficient generator 6 as a third tap output.

The weighters 13, 14, and 15 multiply the input signals by the tap weighting coefficients C1, C2, and C3 generated by the tap weighting coefficient generators 4, 5, and 6, respectively. A calculation is performed to obtain the sum of the outputs of the cross-polarization interference compensation signals 10, 14, and 15.
3 is output. The cross-polarization interference compensation signal 103 is combined with the self-polarized reception signal 102 by the coupler 2.

When the self-polarized reception signal 102 is subjected to cross-polarization interference, the cross-polarization interference compensation signal 103 has the same amplitude and opposite polarity as the cross-polarization interference component, and is output from the coupler 2. The output signal 104 is a signal from which cross-polarization interference has been removed.

The error component detector 3 receives the output signal 104 as input, detects residual error components remaining there, and outputs the error signal 10.
Output as 5. Error signal 105 is input to tap weighting coefficient generators 4, 5, and 6. The tap weighting coefficient generators 4, 5, and 6 perform a correlation operation between each tap output of the transversal type filter 1 and the error signal 105, and determine tap weighting coefficients C1, C2, and C3 respectively.
Output to 3,14,15.

The tap weighting coefficient generator 4 weights the output of the first integrator 41 with the input of the different polarization received signal 101 and the error signal 105 by a predetermined coefficient -α. And a first adder 43 having the output of the second integrator 42 as one input.
The output of the first adder 43 is used as one input, the output is output to the weighting device 13 as the tap weighting coefficient C1, and the polarity of the tap weighting coefficient C1 is determined. And a discriminator 45 that outputs a predetermined negative value of −β when the polarity is positive, and outputs a positive value of + β as the other input of the first adder 43 when the polarity is negative. . The second and third tap weighting coefficient generators 5 and 6 have the same configuration.

FIG. 2 shows a second embodiment of the tap weighting coefficient generators 4, 5 and 6 of the cross polarization interference compensator. An example in which the present invention is applied to the tap weighting coefficient generator 4 will be described. Like the tap weighting coefficient generator 4 of the first embodiment, the tap weighting coefficient generator 4 of the second embodiment also includes the first and second integrators 41 and 42 and the first and second adders 4.
3,44.

The polarity of the discriminator 45 in FIG. 1, that is, the tap weighting coefficient C1 is determined.
When β is negative, the discriminator 45 that outputs + β as the remaining input of the first adder 43 is replaced with +1 when the polarity of the tap weighting coefficient C1 is positive, and −1 when negative. It has a discriminator 45 for outputting.

The output of the discriminator 45 is integrated by a newly provided third integrator 46 with a predetermined value of -β,
The output of the integrator 46 is one input of the first adder 43.

Here, "growth of tap coefficient by accumulation of minute bias level" cited as a problem of the prior art.
Is an ideal method for solving C (t +
There is a method in which 1) is not the integration of -α · {E (t) · D (t)} from time 0 to time t, but the integration from time t−δT to time t. When this is expressed by an equation, C (t + 1) = C (t) -αα {E (t)） D (t)}
−C (t−δT). However, when this method is implemented by a digital circuit, the integrand from time 0 to time t must be held in the circuit, which complicates the circuit configuration.

Therefore, in the method of the present invention, -C (t-δ
T) is replaced with -βsgn {C (t)} in equation (2),
The object is achieved by the simple circuit configurations shown in FIGS.

For example, since C (0) is 0, at time 1, C (1) = 0−α · {E (0) · D (0)} − βsgn {0} = − α · ｛E (0) D (0)｝

At time 2, C (2) = C (1) −α · {E (0) · D (0)} − βsgn {C (1)} = − α · {E (0) · D ( 0) + E (1) ・ D (1)｝ − βsgn ｛−α ｛{E (0) ・ D (0)}.

−α · {E (0) · D (0)} − βsgn
Focusing on the term {−α · {E (0) · D (0)},
Integrand value accumulated in the past-α · ｛E (0) · D
(0) The effect of｝ is reduced.

At time 3, C (3) = C (2) −α · {E (2) · D (2)} − βsgn {C (2)} = − α · {E (0) · D ( 0) + E (1) · D (1)｝ − βsgn ｛−α · {E (0) · D (0)} − α · {E (2) · D (2)} − βsgn [−α · {E (0) · D (0) + E (1) · D (1)} − βsgn {−α · {E (0) · D (0)}].

Then, -α. {E (0) .D (0)}-
The term βsgn ｛−α · {E (0) · D (0)} is as described above, and −α · {E (1) · D (1)} −βsgn ｛−α · {E (0 ) ・ D (0) + E (1) ・ D (1)｝ − βsgn ｛−α ｛{E (0) ・ D (0)} E (1) ・ D
(1)｝ and -α ｛E (0) · D remaining at time 2
(0) The effect of｝ is reduced. Thus, it is possible to eliminate the influence of the integrand value accumulated in the past.

FIG. 3 shows a third embodiment of the tap weighting coefficient generators 4, 5, and 6 of the cross polarization interference compensator. An example in which the present invention is applied to the tap weighting coefficient generator 4 will be described.

The tap coefficient generator 4 of the third embodiment has the same first and second tap weighting coefficient generators 4 as that of the first embodiment.
Second integrators 41 and 42, first and second adders 43 and 4
4

The discriminator 45 receives the tap weighting coefficient C1 as an input, and outputs a logical level "1" when the polarity of the tap weighting coefficient C1 is positive, and outputs a logical level "0" when the polarity of the tap weighting coefficient C1 is negative. The discriminator 4 as an input signal in the circuit 47
5 is inverted.

The pulse generator 48 has a predetermined interval T
Outputs a pulse of logic level "1", and outputs a signal of logic level "0" when no pulse of logic level "1" is output.

The gate circuit 49 outputs "+1" when the output of the pulse generator 48 is at the logical level "1" and the output of the NOT circuit 47 is at the logical level "1", and the output of the NOT circuit 47 is the logical level "0". When the output of the pulse generator 48 is at the logical level "0", "0" is output regardless of the state of the output of the NOT circuit 47.

FIG. 4A shows an example of this gate circuit 49, and FIG. 4B is a truth table showing its input / output relationship.
Now, it is assumed that the first adder 43 in FIG. 3 requests a binary number indicating a positive and negative signed value by using a sign bit indicating positive and negative polarities and a bit indicating an absolute value as an input signal from the gate circuit 49. . That is, a binary number “00” or “10” is used to represent “0”, a binary number “11” is used to represent “−1”, and a binary number “0” is used to represent “+1”.
It is assumed that an expression method using 1 "is required.

The gate circuit 49 receives the outputs of the NOT circuit 47 and the pulse generator 48 as inputs, and performs a NAND operation on both inputs.
ND) gate 49a, outputs the MSB, that is, the sign bit, and outputs the pulse generator output as it is as a bit indicating the absolute value.

When the output of the pulse generator 48 is at the logic level "0", the sign bit of the gate circuit output is at the logic level "1" regardless of the output of the NOT circuit 47, and the bit indicating the absolute value is at the logic level "0". 0 ", and the binary number" 10 ", that is," 0 "is output.

When the output of the pulse generation circuit 48 is at the logical level "1", the bit indicating the absolute value of the output of the gate circuit 49 is "1", and the sign bit is that the output of the NOT circuit 47 has the logical level "0". At "1", the logic level "0"
In the case of, "1" is output respectively. That is, when the tap weighting coefficient C1 has a positive polarity, the output of the negation circuit 47 outputs a logic level "0", and when the tap weighting coefficient C1 has a negative polarity, a logic level "1" is output. "11" and "01", respectively, "-1" and "+1"
Will be output.

With the above configuration, the gate circuit 49 outputs "+" only when a pulse is output from the pulse generator 48.
"1" or "-1" is output. Then, C (t) corresponds to the integral of -α · {E (t) · D (t)}.
It takes some time for the value to change significantly. By setting the pulse generation interval T of the pulse generator 48 to a time shorter than the time required for a large change of C (t),-(1 / T) sgn {C (t)} is set to the gate circuit 49.
Output from.

Therefore, C (t) = C (t) −α · {E (t) · D (t)} − β
This is equivalent to performing the operation represented by the expression of sgn {C (t)}.

In the example of FIG. 4, it is assumed that the absolute value of the output of the second integrator 42 is one or more. FIG.
In the above example, the first adder 43 has shown an example in which the sign indicating the polarity of the input signal and the bit indicating the absolute value are requested. However, the negation circuit 47 is omitted, and the gate circuit 49 is replaced by another. It is also possible to realize the above configuration. Also, even when the first adder 43 requires an input signal in another code format such as a complement, the negation circuit 47 can be omitted or the gate circuit 4 can be omitted.
9 can be easily realized.

Further, it is a matter of course that, instead of each tap output D (t), a signal generated at the same timing as the tap output D (t) may be a signal generated by a shift register or the like.

[0049]

As described above, according to the present invention, the growth of the tap weighting coefficient due to the accumulation of the minute bias level due to the term of the integration of the tap weighting coefficient from time 0 to time t is -βsgn ｛C ( t) Since it is suppressed by the correction term 項, there is an effect that stable tap weight control becomes possible.

[Brief description of the drawings]

FIG. 1 is a block diagram of one embodiment of the present invention.

FIG. 2 is a block diagram of another embodiment of the present invention.

FIG. 3 is a block diagram of another embodiment of the present invention.

FIG. 4A is a diagram showing an example of the gate circuit of FIG. 3;
(B) is a truth table showing the input / output relationship.

FIG. 5 is a block diagram of a conventional cross-polarization interference compensator.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Transversal type filter 2 Coupler 3 Error signal detector 4-6 Tap weight coefficient generator 11,12 Delay line 13-15 Weighter 16 Synthesizer 41,42,46 Integrator 43,44 Adder 45 Classifier 47 Inverter 48 Pulse generator 49 Gate circuit

──────────────────────────────────────────────────続 き Continuation of front page (72) Inventor Tadashi Shirato 1-6, Uchisaiwaicho, Chiyoda-ku, Tokyo Nippon Telegraph and Telephone Corporation (56) References JP-A-1-125135 (JP, A) JP-A 4-247733 (JP, A) JP-A-4-271534 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) H04J 11/00 H04B 3/06

Claims (3)

(57) [Claims] A receiver for transmitting a digitally modulated signal using two orthogonal polarizations having the same frequency in a transmission system using a transversal filter to eliminate cross-polarization interference. An inter-polarization interference compensator, wherein each of the weighting coefficient generation means for each tap of the transversal filter is extracted from each tap output and a signal after cross-polarization interference compensation.
An accumulator for generating a signal corresponding to an integrated value of the obtained error signal , a first adder for adding a predetermined correction value to the integrated output, and an added output of the first adder and its own A second adder for further adding the addition output to obtain the latest tap weighting coefficient; and generating the correction value having a polarity corresponding to the polarity of the addition output of the second adder to generate the first correction value. A cross-polarization interference compensator, comprising: a correction value generator serving as an addition input of the adder; 2. The method according to claim 1, wherein the correction value generator includes means for determining the polarity of the addition output of the second adder, and means for generating a constant correction value having a polarity according to the determination polarity. The cross-polarization interference compensator according to claim 1, wherein: 3. The correction value generator determines a polarity of an addition output of the second adder, and outputs a signal of a first logic level when positive and a signal of a second logic level when negative. Respectively, a means for generating a pulse of a predetermined period, and an input of this pulse and the output signal of the polarity determination means, and when the pulse is absent, regardless of the output signal level of the polarity determination means "0", "1" when the output signal of the polarity determination means is at the first logic level in the presence of the pulse, and "-" when the output signal is at the second logic level.
And a means for generating 1 "respectively.