JPH098718A - Diversity receiver - Google Patents

Abstract

PURPOSE: To improve the S/N at a diversity receiver and to prevent the dispersing phenomenon of an equalizer at the receiver. CONSTITUTION: A discriminate data signal outputted from a discrimination feedback type equalizer (DFE) 105 and a signal before inputting to the DFE 105 are switched by a changeover switch 106 and corresponding to this selected signal, a weight coefficient in the case of weighting a diversity received signal to be synthesized at correlators from 103-1 to 103-N is controlled. With this switching, a DATA CORR method is usually used and even under a multi-path environment, the synthesization of maximum ratio to a main wave is enabled at all times. Besides, since an MRC method is used when a signal is hit by receiving severe fading, the diversity receiver is not permanently dispersed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a diversity receiver, and more particularly to a diversity receiver having a self-restoration capability when an input signal is momentarily cut.

[0002]

2. Description of the Related Art As a diversity receiver which has been conventionally used, a decision data correlation method (DATA CORR
Law). FIG. 2 shows a block diagram of the basic configuration of the plurality of diversity antennas 20.
The signals respectively received by 1-1 to 201-N are converted into complex multipliers 202-1 to 202-N and correlators 203-1 to 20-20.
After passing through the weighting controller composed of 3-N, the combiner 204 performs diversity combining. In addition, the synthesized output is a decision feedback equalizer (DFE: Decis).
Ion Feedback Equalizer) and then the sampling data signal
And correlator 203-1 and 203-2 using this.
A configuration for controlling the weighting coefficient (tap coefficient) of 03-N is adopted. The biggest reason is that the maximum ratio combining is always possible with respect to the desired wave (main wave) even in a multipath environment. Maximum ratio combining is the signal-to-noise ratio (SN
R: Signal to Noise Ratio) is maximized.

In order to simplify the description, a case where the number of diversity branches is 2 will be described below. Normally, microwave lines are subject to severe multipath fading, so leading and lagging waves occur, and they are dispersed on the time axis. Therefore, the reception signals r _n ¹ and r _n ² of each route received by the diversity antennas 201-1 and 201-2 can be set as follows.

[0004]

[Equation 1]

Here, h _i ¹ and h _i ² are impulse responses at time i, and a _ni is transmission data at time (n−i). The received signals r _n ¹ and r _n ² are uncorrelated with each other,
Each of the complex multipliers 202-
1, 202-2 is given a unique weighting. In this case, the tap coefficient in the DATA CORR method is obtained by correlating the determination data signal a _n (main wave) after passing the DFE and the complex conjugate of the received signal. The following equation shows that.

[0006]

[Equation 2]

Here, E [a _i ^* a _j ] ≡δ However, if δ = [1 (i = j), 0 (i ≠ j)] is used, w ₁ = h ₀ ^{1 *} · E [a _i ^* a _j ] = h ₀ ^{1 *} (4) This means i = 0, the complex conjugate of the impulse response for the dominant wave (a _n ).

Similarly, when w ₂ is also obtained, w ₂ = h ₀ ^{2 *} (5)

From the above, the received signals r _n ¹ and r _n ² are respectively convoluted by the complex multiplier and combined to obtain the following equation. Y _n = r _n ¹ · w ₁ + r _n ² · w ₂ = (h ₀ ^{1 *} · h ₀ ¹ ) a _n + (h ₀ ^{2 *} · h ₀ ² ) a _n (6)

In the equation (6), h ₀ ^{1 *} · h ₀ ¹
, And h ₀ ^{2 *} · h ₀ ² , which is a complex conjugate multiplication, and as a result, Amplitude (of Impulse Resp
onse) ... Square Phase ... in-phase. This is the so-called maximum ratio combination.

Further, at the same time, a signal having a high level has a large tap coefficient, and a signal having a low level is regarded as noise, and operates so as to be 0. Therefore, it is also possible to incorporate unnecessary noise and distortion. There is. That is, even in a multipath environment, maximum ratio combining is possible for the highest level signal (which may be considered as the desired wave (main wave)). This is the greatest advantage of the DATA CORR method and is a factor that is often used.

On the other hand, another method used for diversity combining is MRC (Maximal Ratio Comb).
iner) method. FIG. 3 is a block diagram thereof. In this configuration, similar to the DATA CORR method, the signals respectively received by the plurality of diversity antennas 301-1 to 301-N are converted into complex multipliers 302-1 to 30-30.
2-N and the correlator 303: 1 to 303-N is passed through the maximum ratio controller, and then the combiner 304 performs diversity combining. Then, the composite output is DF
Before passing E, AGC (Automatic Gain)
Control) 309, and the tap coefficient of the correlators 303-1 to 303-N is controlled by using this amplified signal.

This MRC method also performs maximum ratio combining, but there is a big difference in the method of determining the tap coefficient, and there is a difference in performance from the DATA CORR method. That is, it is assumed that the received signal can be expressed by the equations (1) and (2) as in the DATA CORR method. When the tap coefficients of each diversity route are w ₁ and w ₂ , the formula after combining is as follows.

[0014]

(Equation 3)

Here, as easy as w ₁ · h _i ¹ + w ₂
Assume h _i ² = 1. At this time, the MRC method is AGC3.
Considering that the tap coefficient is determined by correlating the combined signal whose amplitude is controlled to 1 by 09 and the received signal,
The value of the tap coefficient in the MRC method is as follows. The same applies to other routes, except that the subscripts change.

[0016]

(Equation 4)

The meaning of this equation is DATA C
Although it is easy to understand by comparing with the equation (4) of the ORR method, the tap coefficient does not take only the complex conjugate of the impulse response at the same time as the main wave, but becomes an integrated value including other routes. Therefore, at the time of convolution, only one route having a high signal level is not passed as in the DATA CORR method, and even if the level is low, all signals are regarded as signals and in-phase synthesis is performed.

[0018]

By the way, in such a diversity receiver, when the received signal is severely faded in the propagation line and is in a momentary interruption state, the signal level is lowered and the DFE divergence phenomenon occurs. become. That is, a forward equalizer (FE) that is a component of DFE is
Since this is a linear filter that uses the correlation value between the input signal to itself and the determination data signal as a tap coefficient, when the level of the input signal becomes extremely low, the coefficient becomes a meaningless random pattern numerical value.

Since this depends on the Gaussian distribution, the tap coefficient becomes 0 on average, and in this state, F
E does not pass the signal to the subsequent stage. Therefore, another component of the DFE, the backward equalizer (BE: Back).
A normal equalizer cannot operate normally, and converges to 0 like FE. For this reason,
Since these FE and BE cut off the signal, the determination data signal remains permanently 0 even if the line is restored and the signal level rises. Strictly speaking, one tap which is the standard of FE always takes 1V, and when the signal is restored, FE is also restored, but BE prevents this.

Therefore, as described above, D which takes the correlation value with the decision data signal which is the output of the DFE as the tap coefficient
Since the same can be said for the ATA CORR method,
The DATA CORR method has a problem that a divergence phenomenon of DFE occurs.

On the other hand, in the MRC method, since the signal immediately before combining before DFE input is used as the correlation value for tap coefficient determination, it is possible to recover from the situation where the tap coefficient converges to 0 when the line is restored. . However, this M
In the RC method, when the convolution is performed, DATA is used as described above.
Unlike the CORR method, the signal is not passed through only one route with a high signal level, and even if the level is low, all signals are regarded as signals and in-phase synthesis is performed. Therefore, in a multipath route environment, they can become interference waves for the main wave. Since the maximum ratio of forward and delayed waves is also combined, the total SNR is
There is a problem that it gets worse.

[0022]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a diversity receiver capable of improving the SNR and preventing the occurrence of the divergence phenomenon.

[0023]

DISCLOSURE OF THE INVENTION According to the present invention, in a diversity receiver which weight-controls a plurality of diversity reception signals, combines these signals, and removes distortion by DFE, the determination output from the DFE is made. A feature is that a data signal and a signal before being input to the DFE are selected, and a tap coefficient for weighting is controlled by the selected signal.

That is, in the present invention, a level detector for detecting the level of the combined signal before being input to the DFE, and a changeover switch for switching between the judgment data signal and the signal before input by the detection output of this level detector. It has and. Here, the level detector is configured to switch the changeover switch so as to select the determination data signal when the level of the combined signal is equal to or higher than a predetermined level and to select the pre-input signal when the level is equal to or lower than the predetermined level.

Further, in the present invention, it is also possible to have a frame synchronization circuit for detecting the coincidence of the synchronization frame included in the judgment data signal, and to switch the changeover switch by the coincidence detection output of this frame. Further, it is preferable to have a switch for shutting off BE that constitutes a part of the DFE, and to operate the switch to bring the BE into a shut-off state when switching to the pre-input signal.

[0026]

The received signals are combined after the weighting control is performed by the tap coefficient and the distortion is removed by the DFE. In setting the tap coefficient at that time,
It is performed by the signal selected by the changeover switch. Usually, since maximum ratio combining can be performed on the main wave even in a multipath environment, the determination data signal from the DFE is selected, and thereby the tap coefficient is set by the DATA CORR method with a large gain.

Further, when the received signal undergoes severe fading in the propagation line and is in a momentary interruption state, and the signal level is below a predetermined level, the changeover switch is changed to D
The pre-input signal of FE is selected. As a result, the tap coefficient is set by the circuit of the MRC method,
D in case of sudden interruption due to severe fading
Dispersion of FE is prevented.

Further, when switching to the circuit of the MRC method, D
Although the FE is in a diverged state, it is possible to restore the FE by switching to the circuit of the MRC method and cutting off the BE at the same time. In addition, it is determined whether the synchronization frames included in the determination data signal match, and if they match, DATA C
The circuit of the ORR method is operated, and if they do not match, the circuit of the MRC method is operated.
At the same time, the switch disconnects BE.

[0029]

Next, the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of an embodiment of the present invention. The signals respectively received by the plurality of diversity antennas 101-1 to 101-N are converted into complex multipliers 102-1 to 102-
-N and the correlator 103-1 to 103-N are passed through the weighting controller, and then the combiner 104 performs diversity combining. Then, the combined output is BE1.
13, the FE 114, the adder 115, and the determiner 116 are input to the DFE 105, where multipath distortion is removed.

A changeover switch 106 is inserted in the signal path for controlling the tap coefficients of the correlators 103-1 to 103-N, and is configured to be changed over by the controller 107. That is, the determination data signal after passing through the DEF 105 is input to one input of the changeover switch 106 through the sample circuit 108. A composite signal before being input to the DFE 105 is input to the other input through the AGC 109. Then, the detection output of the level detector 110 that detects the level of the combined signal before being input to the DFE 105 is input to the controller 107 that switches the changeover switch 106.

In this embodiment, the controller 107 is also supplied with the output of the frame synchronization circuit 111 synchronized with the synchronization frame included in the determination data signal after passing through the DFE 105.

Further, a part of the input of the controller 107 is input to a switch 112 provided at the output end of the BE 113 which constitutes a part of the DFE 105 so that the switch 112 is turned on / off based on the controller input. Is composed of.

The operation of such a diversity receiver is as follows. Diversity antenna 101-1 ~
The signal received by 101-N is correlator 103-1.
.About.103-N and complex multipliers 102-1 to 102-N receive different weights. The tap coefficient at this time is set by the signal selected by the changeover switch 106. Normally, even in a multipath environment, maximum ratio combining is possible for the main wave, so DFE
The determination data signal from 105 is selected, and thereby the tap coefficient is set by the DATA CORR method with a large gain.

As a changeover standard of this changeover switch,
As described above, the signal level immediately after synthesis is used. This is because when the tap coefficient is controlled by the normal DATA CORR method, the received signal undergoes severe fading in the propagation line, and when the instantaneous interruption occurs, the DFE 105 does not operate normally and the above-mentioned problem occurs. Therefore, the signal level is detected by the level detector 110 before the DFE 105 is input, and when the signal level is equal to or higher than a predetermined level, the tap coefficient control by the DATA CORR method is performed.

When the signal level becomes lower than a predetermined level, the controller 107 causes the changeover switch 1
06 to switch the signal before input to the DFE 105 to AGC.
The signal amplified in 109 is selected. This allows MR
The tap coefficient is set by the C method, and the DF when the instantaneous interruption occurs due to severe fading
E105 divergence is prevented. Therefore, the decision point of the level detector 110 is set to the minimum level in the common part of the signal level at which the DFE 105 is functioning normally and the signal level at which it can operate normally immediately after switching to the circuit of the MRC method.

The reason for this is that even if the DFE 105 is normal, if the MRC circuit immediately after switching does not operate normally, it will end up falling down and that the circuit of the DATA CORR method with the highest possible gain is used. .
If there is no common part, select the latter. In this case DF
Although the E105 has already diverged, the FE 114 can be restored because the input signal to the FE 114 has not diverged by switching to the circuit of the MRC method.

However, as described above, since the BE 113 tries to prevent it, in this case, the switch 112
By turning off and disconnecting BE113,
This prevents the FE 114 from being disturbed and enables the FE 114 to be restored. Then, when the FE 114 is recovered, the switch 11
2 may be turned on and the BE 113 may be reconnected.

The circuit of the MRC method is used only as an initial pull-in until the circuit of the DATA CORR method works normally. However, the fact that the in-phase synthesis can be performed regardless of the divergence of the DFE 105 is advantageous in compensating the drawback of the DATA CORR method.

Further, in this embodiment, as described above, the synchronization frame included in the determination data signal is also used as the switching reference of the changeover switch 106. Usually, a predetermined bit string (frame) is often inserted in the transmission data in order to achieve clock synchronization or to confirm whether the data is correctly transmitted. Therefore, the frame synchronization circuit 111 determines the frame matching by using the signal which is not distorted by the multipath fading after passing through the DFE. If they match, the operation of the DATA CORR method circuit is performed, and if they do not match, a flag is set or the signal itself is EX-ORed and the result is used. Controls the controller 107, and the changeover switch 1
According to 06, the operation in the circuit of the MRC method is performed.
At the same time, by disconnecting the BE 113 with the switch 112, the FE 114 can be restored as described above.

[0040]

As described above, the diversity receiver according to the present invention selects the decision data signal output from the DFE and the signal before input to the DFE and weights the selected signal. Since the tap coefficient control is performed, the DATA CORR method is normally used, and maximum ratio combining can always be performed on the main wave even in a multipath environment. Further, when the signal is momentarily interrupted due to severe fading, the MRC method is used, so that the diversity receiver does not diverge permanently.

When switching to the circuit of the MRC method, DF
Although E is in a diverged state, it is possible to restore FE by switching to the circuit of the MRC method and cutting off BE at the same time. Further, it is determined whether or not the synchronization frame included in the determination data signal matches, and if they match, D
The ATA CORR method circuit is operated, and if they do not match, the MRC method circuit is operated, and at the same time, the BE is disconnected by the switch, and as described above, the maximum for the main wave is increased. It is possible to perform ratio combining and prevent divergence of the receiver.

[Brief description of drawings]

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

FIG. 2 is a block diagram of a receiver that employs a conventional DATA CORR method.

FIG. 3 is a block diagram of a receiver adopting a conventional MRC method.

[Explanation of symbols]

 101-1 to 101-N diversity antenna 102-1 to 102-N complex multiplier 103-1 to 103-N correlator 104 combiner 105 DFE 106 changeover switch 107 controller 108 sample circuit 109 AGC 110 level detector 111 frame synchronization Circuit 112 Switch 113 BE 114 FE

Claims (5)

[Claims] 1. A diversity receiver that weight-controls a plurality of diversity reception signals, combines these signals, and removes distortion by a decision feedback equalizer, and outputs from the decision feedback equalizer. A diversity receiver characterized in that a decision data signal to be determined and a signal before being inputted to the decision feedback equalizer are selected, and weighting coefficient control for the weighting is performed by the selected signal. 2. A level detector for detecting the level of a composite signal before being input to a decision feedback equalizer, and a changeover switch for switching between the decision data signal and the pre-input signal by the detection output of this level detector. The diversity receiver of claim 1, comprising: 3. The level detector is configured to perform a switching operation of the changeover switch so as to select the determination data signal when the level of the combined signal is equal to or higher than a predetermined level and to select the pre-input signal when the level of the combined signal is equal to or lower than the predetermined level. The diversity receiver according to claim 2. 4. The diversity receiver according to claim 2, further comprising a frame synchronization circuit for detecting a match of a sync frame included in the determination data signal, and switching the changeover switch by a match detection output of the frame. 5. A decision feedback equalizer having a switch for shutting off a rear equalizer forming part of the equalizer, and when the pre-input signal is switched to, the switch is operated to shut off the rear equalizer. The diversity receiver according to any one of claims 1 to 4.