JPH01200831A - Decision feed back type equalizer - Google Patents

Abstract

PURPOSE:To have an adaptability sufficiently following the fluctuation of the transmission characteristic even under a high-speed phasing and to execute an application to a mobile communication by providing a decision feed back type equalizing circuit, a correlation device and a tap coefficient initial value setting circuit, etc. CONSTITUTION:For a signal to be received, a receiving level is smoothed by an RF band AGC circuit 18, and it is detected by a quasi-synchronization detector 19. For the output of the detector 19, the level is made fixed by a base band AGC circuit 22, and thereafter, it is inputted to a decision feedback type equalizing circuit 23. Next, a correlation device 24 estimates the complex profile of a receiving wave based on a training pulse inserted into the receiving weave. A tap coefficient initial value setting circuit 25 decides the initial value of the tap coefficient of the decision feedback type equalizing circuit 23 by using the complex profile obtained by the correlation device 24. Thus, since an output error can be made small and, simultaneously, the fluctuation of a transmission line can be followed at high speed, the adaptability to sufficiently follow the fluctuation of the transmission characteristic can be obtained even under the high-speed phasing, and the application to the mobile communication be realized.

Description

Detailed Description of the Invention "Industrial Application Field" The present invention suppresses deterioration of transmission characteristics when the transmission band characteristics fluctuate quickly and deteriorate significantly under high-speed fading such as in digital mobile communications. The present invention relates to a decision feedback equalizer used for this purpose.

"Prior Art" A signal sent out from a transmitter is subjected to linear distortion due to the band characteristics of a transmission device and a transmission medium on a signal transmission path. Therefore, in digital signal transmission, code amount interference occurs and reception characteristics deteriorate.

Various equalizers have been devised to reduce the deterioration of transmission characteristics due to this type of linear distortion, and a decision feedback type equalizer is known as a typical one with high performance.

FIG. 4 shows the configuration of a conventional decision feedback equalizer.

In FIG. 4, a certain signal sequence is input from an input terminal 1 to a unit delay unit 3 in a tapped delay circuit 2 and a first stage tap coefficient multiplier 5. The output of each unit delay device 3 in the tapped delay circuit 2 is inputted to a multiplier 5 as a multiplicand.

On the other hand, the multipliers (tap coefficients) of the multiplier 5 are values C8 to C1 determined by the tap coefficient changing circuit shown in FIG.
is input. The output of each multiplier 5 is input to an adder 6, and the output thereof is input as the minuend of a subtracter 7.

The output of the subtracter 7 is subjected to sign determination by a determiner 8, and the result is outputted from the output terminal II and simultaneously inputted to the first unit delay unit 3 of the second tapped delay circuit 9. The output of each unit delay device 3 in the tapped delay circuit 9 is multiplied by the tap coefficients D1 to D by the multiplier 5 in the same manner as in the case of the tapped delay circuit 2, and the result is multiplied by the tap coefficients D1 to D, and the result is sent to the adder lO. and the addition result is input as the subtracted number of the subtracter 7.

To use a decision feedback equalizer to compensate for a fluctuating transmission path, it is necessary to perform adaptive control to make the tap coefficients follow changes in the characteristics of the fluctuating transmission path, and this is performed by a tap coefficient changing circuit.

FIG. 5 shows the configuration of a tap coefficient changing circuit of a conventional decision feedback equalizer.

In FIG. 5, the input and output of the determiner 8 in FIG. 4 are input to a subtracter 12 that calculates the difference between them. The output of the subtracter 12 is multiplied by a constant by a constant multiplier 13, then multiplied by the output of the tapped delay circuit 2 by a multiplier 14, and then added to an accumulator 15 consisting of a subtracter 16 and a register 17. The output of the accumulator 15 becomes the multiplier of the multiplier 5 in FIG. 4, that is, the tap coefficient C3-C□.

Further, the multipliers of the multiplier 5 connected to the tapped delay circuit 9, that is, the tap coefficients D1 to DJ, are similarly determined.

Using this adaptive algorithm, tap coefficients C8 to C, .
By changing D, ~DJ, the characteristics of the equalizer can be made to follow changes in the characteristics of the transmission path.

``Problem to be Solved by the Invention'' The conventional decision feedback equalizer described above has problems when the fluctuation of the delay profile is relatively slow compared to the speed of the transmission signal, as in wired communication and fixed wireless communication. Adaptive control using a tap coefficient changing circuit alone has become sufficient for practical use.

However, when the transmission path characteristics fluctuate faster than the signal transmission speed, such as in mobile communications, tap coefficient C3
-Ci, D1 to D become difficult to follow.

As a means to quickly obtain tap coefficients, a complex profile of a delayed wave is obtained using a correlator from a training pulse with a fixed pattern that is inserted into the transmission signal at a fixed period (for example, 1 ms period), and based on this, Levinson's
A method is known in which a general solution for tap coefficients is obtained by solving an algorithm.

However, this method requires complicated calculations, and the tap coefficients only take into account the complex profile of the delayed wave, and there are many solutions for making the code amount interference zero.

Some of these solutions have more noise in the equalizer output than other solutions.

In other words, it has the disadvantage that it is not optimized in terms of noise.

From the above, when applying an equalizer to mobile communications, a new method is required for setting tap coefficients.

SUMMARY OF THE INVENTION An object of the present invention is to provide a decision feedback equalizer having adaptability that can sufficiently follow fluctuations in transmission characteristics even under high-speed fading, for application to mobile communications.

"Means for Solving the Problems" In order to solve the above problems, the present invention has tapped delay circuits before and after the determiner, multiplies each tap output of the tapped delay circuits by a tap coefficient, and adds the results. In a decision feedback equalizer configured to equalize a received wave based on the addition result, a main wave having the highest level among the plurality of discrete components included in the received wave; When there is a preceding wave that temporally precedes the main wave, a correlator that detects the time interval, amplitude ratio, and phase difference between the main wave and the preceding wave using a training pulse; There are two modes: a leading wave mode in which a judgment is made based on the main wave and the leading wave is canceled out by a leading wave with a low level, and a main wave mode in which a judgment is made based on the main wave and the leading wave is canceled out by a leading wave with a high level. and a tap coefficient initial value setting circuit that selects a mode in which the SN ratio of the equalizer output is larger among the modes, and sets a value corresponding to the selected mode as an initial value of the tap coefficient. It is characterized by

"Operation" According to the above means, when setting the tap coefficient, the correlator calculates the complex profile of the received wave, that is, the time interval between the main wave and the preceding wave, the amplitude ratio, and the training pulse inserted into the received signal. Determine the phase difference and select either the leading wave mode or the main wave mode from the CNR (carrier-to-noise power ratio) of the received wave, the level ratio of the leading wave and the main wave, and the number of taps of the equalizer. Select and initialize the tap coefficient. This makes it possible to immediately set optimal tap coefficients.

Note that interference caused by delayed waves arriving after the main wave is not directly related to the present invention, but the reason will be described later.

In the conventional technology, the tap coefficients were changed sequentially according to changes in the input signal, but in the method according to the present invention, the optimum value of the tap coefficients is determined every time a training pulse is received and set as the initial value. are different.

"Example" Hereinafter, the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

In the figure, 18 is an RF band AGC circuit, 19 is a quasi-synchronous detector, 20 is an APC circuit, 21 is a reference carrier wave generator, 2
2 is a baseband band AGC circuit, 23 is a decision feedback Sugi equalization circuit, 24 is a correlator, and 25 is a tap coefficient initial value setting circuit.

The reception level of the received signal is smoothed by the RF band AGC circuit 18 and detected by the quasi-uniform detector 19. Quasi-synchronous detector 19 consisting of in-phase signals and quadrature signals
The level of the output is made constant by the baseband AGC circuit 22, and then inputted to the decision feedback equalization circuit 23.

The AFC circuit 20 controls the frequency of the reference carrier wave in addition to the quasi-synchronous detector 19, and suppresses the difference between the frequency of the received signal and the reference carrier wave within a certain range.

The correlator 24 estimates the complex profile of the received wave based on the training pulse inserted into the received wave. That is, when there is a main wave with the highest level among a plurality of discrete components included in the received wave and a preceding wave that temporally precedes this main wave, the correlator 24 detects the main wave and the preceding wave. The time interval, amplitude ratio, and phase difference (these are referred to as a complex profile) from the training pulse are detected from the training pulse. The above training pulse is
For example, it is a pulse train of a fixed pattern consisting of a 32-bit bit string, and is sent at fixed intervals (for example, 1 m5).

The tap coefficient initial value setting circuit 25 determines the initial value of the tap coefficient of the decision feedback equalization circuit 23 using the complex profile obtained by the correlator 24, and the determination method will be described in detail below. .

A specific method for determining the initial setting value of the tap coefficient will be explained using a two-wave model consisting of a leading wave and a delayed wave.

In the following explanation, to simplify the explanation, the time interval between the preceding wave and the delayed wave is assumed to be l time slots, and the amplitude of the preceding wave is assumed to be W. , let Wl be the amplitude of a delayed wave delayed by one time slot from the preceding wave. Here, wo”+ w,”= 1% ξ= W (/W
.

The larger of these two waves will be called the main wave. That is, when the ratio ξ>1, the delayed wave becomes the main wave, and when the ratio ξ < l, the leading wave becomes the main wave. This invention is
This is effective in the former case, that is, when the main wave arrives following the preceding wave, and the reason for this will be explained when explaining FIGS. 2 and 3.

Now, the received signal sequence is 1, A-t, A-1. Ao, A 1. A t,”, this series is Wo A n+ v, A n-.

Assume that the transmission path passes through a transmission line with the following characteristics.

At this time, the data in the tapped delay circuit 2 of FIG. 4 is as follows (see FIG. 4).

B −r = Wo A i ” W+ A i−+B
-+ ” W6 A 1 + 11 A.

Bo ” Wo Ao + W+ A −+ ””
” (1) Also, since the data in the tapped delay circuit 9 in FIG. 4 is input from the output of the determiner 8, when the error rate is 0, B, = A-1 B x = A -x B, = A-, (2).

Therefore, the output Y of the decision feedback equalization circuit 23 is as follows.

−Σ DkA−1 + (c ov+ −DI) A −1
-Σ D k A-1... (3) In equation (3), if Y o = A o, the tap coefficients 00 to C1, D, and -DJ must satisfy the following conditions. It won't happen.

Clwo = 0 -- (4) CkW
o 10Ck++ w, = 0 (1≦to≦i −1)
--(5) Co Wo + C+ 9r+ = 1
""" (6) CoW, -D1= 0
・・・・・・ (7) Dk-〇 (2≦to≦i)
...... (8) When solving this equation, a judgment is made based on the leading wave, and a leading wave mode in which the leading wave cancels out the main wave with a high level, and a judgment based on the main wave, which cancels the main wave with a high level. There is a dominant wave mode that cancels the small preceding wave.

(A) Case of leading wave mode In this case, if the number of terms of the tap coefficient C1 is finite, the tap coefficient is ci-0 from the above equation (4). By substituting this into equation (5), tap coefficients C~C1-0 are obtained. By further substituting this into equation (6), the following equation is obtained.

Ck=Dri=0 (1≦≦i, 2≦Q≦j)...
- (9) (B) Case of main wave mode In this case, the tap coefficient Ck decreases in a geometric progression and continues indefinitely. Therefore, by assuming that the tap coefficient CI is infinite, and assuming that the tap coefficient RI is an arbitrary value, and solving these equations, we obtain the following relationship from equations (7) and (6) above. It will be done.

D, = arbitrary, D(= O(2≦(!≦j)...
(10) Next, assuming that the noise contained in the received wave has no correlation, the power of the noise output from the decision feedback equalizer 23 is the absolute value of the tap coefficients C3-C1, D1 to D, It is proportional to the sum of squares. Therefore, when the sum of squares of the absolute values of the tap coefficients is evaluated as a function of the tap coefficient D1, D + = wo/
The minimum value is '91+. At this time, the convergence value of tap coefficient 00 to CISD ID J is (4) to (
8) From the equation, we get the following.

DQ=upper'-, DQ=O(2≦g≦j)･1 (11) Also, when these modes are used, the error that appears in the output of the decision feedback equalization circuit 23 is as follows, assuming that the power of the noise included in the received wave is N1 and the number of taps is i+1.

First, as mentioned above, if we set w, )"+ W1'= l w+/wo-ξ, the following formula holds true. Therefore, for each case of the leading wave mode and the main wave mode, the equalizer Find the error appearing in the output of .In finding this error, tap coefficients D1~
Since D is a tap coefficient at the subsequent stage of the determiner 8, it is assumed that it does not include noise, and attention is focused only on the tap coefficients 00 to C.

(1) In the case of leading wave mode In this case, from equation (9), the tap coefficients other than Co and D+ are zero, so the error C is proportional to the sum of the squares of the tap coefficients.
Only 0′・NT appears. Therefore, the error is as follows.

(2) Main wave mode In this case, an error due to the finite number of taps and an error proportional to the sum of squares of the tap coefficients appear.

(a) Error due to tap truncation If the number of taps is i+1, the error is (Cnwo)” becomes.

Note that the error due to tap truncation is Cn C2C,
co D, D.

Leading wave woA n woA t WoA 1woA
oA -1 main wave w+A n-1w+A I w+A
In the signal sequence o v + A -1, the main wave and the preceding wave diagonally above it (for example, WIAO of the main wave and W.A of the preceding wave) cancel each other out, so it corresponds to the tap coefficient co. The leading wave component W. This is caused by only A remaining.

Here, the above tap coefficients are as follows.

D, -W-! l! - Dt=0 (b) The error due to the sum of squares of the tap coefficients is Go''+Σ Ck! Therefore, from the above (13a) and (13b), the overall error is as follows.

Figure 2 shows the amount of error in the output when the number of taps of the decision feedback equalization circuit 23 is kept constant and the power of the noise is varied, and the received wave with a CNR of 20 dB or 1 OdB is determined by decision feedback equalization. SNR on the output side when input to the device 23
This shows that.

In the case of the leading wave mode, the main wave having a high level is canceled by the leading wave or delayed wave having a low level, so that the amount of SNR deterioration increases as the ratio ξ increases.

Here, as mentioned above, the ratio ξ is the amplitude of the delayed wave divided by the amplitude w0 of the preceding wave, so when the absolute value of the ratio ξ is 1ξI<1, the preceding wave becomes the main wave. , when the ratio Iξl>1, the delayed wave becomes the main wave.

As can be seen from the figure, in the case of 1ξ1<l, the leading wave mode always has a better SNR, and this mode is selected.

On the other hand, when the absolute value of the ratio Iξl>1, the delayed wave becomes the main wave, and the main wave mode has a better SNR.
There are cases where the leading wave mode is better.

In other words, if 1ξ1<1, there is no choice, but
In the case of IF5>l, by selecting the one with better SNR, it is possible to obtain effects that cannot be obtained conventionally.

As can be seen from FIG. 2, the absolute value of the ratio 1ξ1 is approximately 1.
In the case of 5 or more, SNH in the main wave mode is better, but as this absolute value approaches 1, the convergence of the tap coefficient becomes worse, so the effect of truncating at a finite number of taps and the input noise The amount of SNR deterioration increases due to the influence of However, near the absolute value of the ratio 1ξIh<1,
Since the first term C1 becomes smaller, the amount of deterioration of the SNR becomes smaller, and matches the case of the leading wave mode when 1ξl=1.

Figure 3 shows the amount of error in the output when the noise power is held constant and the number of taps of the decision feedback equalizer circuit 23 is varied, and as the number of taps i increases, the SNR improves. I understand that.

As can be seen from these figures, depending on the number of taps of the CNR1 equalizer of the received wave and the level ratio of the leading wave and delayed wave, the two modes, leading wave mode and main wave mode, are switched and the initial values of the tap coefficients are set. By setting , an output with small error can be obtained.

"Effects of the Invention" As explained above, according to the present invention, when initializing tap coefficients of a decision feedback equalizer, a correlator calculates a complex profile of a delayed wave from a training pulse inserted into a signal sequence. Estimate this and the CNR of the received wave
, Depending on the number of taps of the equalizer, one of the two modes that reduces the noise of the equalizer output is selected, and the tap coefficient at that time is set as the initial value, so the output error is small. In addition, an equalizer that can quickly follow fluctuations in the transmission path can be obtained.

By using the present invention, an equalizer can be applied to an application field such as digital mobile communication, where characteristics of a transmission path change rapidly under a fast fading environment.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention, and FIG.
The figure shows the CNR of the received signal and the noise characteristics of the output of the decision feedback equalizer when this embodiment is used. Figure 3 shows the number of taps of the decision feedback equalizer when this embodiment is used. Figure 4 is a block diagram showing the configuration of the main parts of a conventional decision feedback equalizer, and Figure 5 is the configuration of a tap coefficient changing circuit of a conventional decision feedback equalizer. FIG. l... Input terminal, 2... Delay circuit with tap, 3... Unit delay device, 4... Tap coefficient, 5... Tap coefficient multiplier, 6...
...Adder, 7...Subtractor, 8...
Determiner, 9... Delay circuit with tap, lO...
... Adder, 11 ... Output terminal, 12 ...
...Subtractor, 13...Constant multiplier, 14.
... Multiplier, 15 ... Accumulator, 16 ...
...Subtractor, 17...Register, 18...
...RF band AGC circuit, 19 ... Synchronous detector, 20 ... AFC circuit, 21 ... Reference carrier generator, 22 ... Base Band band AFC circuit, 23...
... Decision feedback equalizer, 24 ... Correlator, 2
5... Tap coefficient initial value setting circuit, 26...
...Output terminal. Applicant Nippon Telegraph and Telephone Corporation Figure 2 Relationship between CNR and output difference (number of taps 1 and 7) Figure 3

Claims (1)

[Claims] A tapped delay circuit is provided before and after the determiner, each tap output of the tapped delay circuit is multiplied by a tap coefficient and added, and the received wave is equalized based on the addition result. In the decision feedback equalizer configured as follows, the main wave having the highest level among the plurality of discrete components included in the received wave and the preceding wave that temporally precedes this main wave are At some point, a correlator detects the time interval, amplitude ratio, and phase difference between the main wave and the preceding wave using a training pulse, and a determination is made based on the preceding wave. There are two main wave modes: a leading wave mode that cancels out the leading wave, and a main wave mode that makes a determination based on the main wave and cancels the leading wave with a main wave of a high level.
a tap coefficient initial value setting circuit that selects a mode in which the SN ratio of the equalizer output is larger from among the two modes, and sets a value corresponding to the selected mode as an initial value of the tap coefficient. A decision feedback equalizer characterized by: