WO2002025834A2

WO2002025834A2 - Interference compensator device

Info

Publication number: WO2002025834A2
Application number: PCT/EP2001/009825
Authority: WO
Inventors: Heinrich Schenk
Original assignee: Infineon Technologies Ag
Priority date: 2000-09-20
Filing date: 2001-08-24
Publication date: 2002-03-28
Also published as: WO2002025834A3; DE10046636A1

Abstract

The invention relates to an interference compensator device for compensating for interference signals on two-wire lines of communications systems, whereby a useful signal (S1) and the inverted useful signal (S2) are connected to a transmitter (UE) via a first line (L1) a second line (L2), respectively. The inventive interference compensator device comprises: a first differential signal generating device (D1), which is connected to the transmitter (UE) and which is provided for generating a first differential signal between the signals transmitted on the first and second line (L1, L2); a delay device (30) for delaying the differential signal; a first summation signal generating device (SU1) for generating a first summation signal of the signals transmitted on the first and second line (L1, L2); an interference estimating device (50) for estimating an interference signal, which is contained in the differential signal, based on the first summation signal, and; a second differential signal generating device (D2) for generating a second differential signal between the first differential signal and the estimated interference signal for compensating for the interference signal. The interference estimating device (50) has an adapted non-recursive digital filtering device.

Description

description

Störkompensatorvorrichtung

The present invention relates to a noise compensator device according to the preamble of claim 1, as known from WO 99/63675.

FIG. 3 is a diagram for illustrating the general problems on which the present invention is based, which is illustrated by way of example on a telephone line.

In the case of communication signal transmission via symmetrical wire lines (two-wire lines, e.g. telephone connection lines), interference is usually coupled in from the environment or from neighboring lines. The coupling takes place on both lines in the same way, so that these disturbances remain ineffective with ideal symmetry of the transmission arrangement, since the difference between the two partial signals is passed on to the receiver as the received signal.

According to Fig. 3, the useful signal is in particular transmitted to a transmitter UE, e.g. in the form of a transformer, and the inverted useful signal is applied to the transformer UE via a second line L2.

A pulse-shaped disturbance is superimposed on this useful signal, as outlined in FIG. 3.

Since real transmission systems can never be completely symmetrical, such a coupled interference can never be completely compensated for by simple difference formation. For the evaluation of the interference coupling, the asymmetry damping can be used, which indicates how strongly a signal coupled in common mode is damped at the differential output. This asymmetry damping is generally frequency-dependent. In known interference compensator devices, the sum signal Sl + S2 of the signals Sl, S2 of the two lines L1, L2 is used for interference compensation.

Fig. 4 is a timing chart showing the output of a sum signal generator in a known noise compensator device. FIG. 4 shows how one that is asymmetrically coupled to the signals S1 of the first line L1 and S2 of the second line L2

Fault ΔS1, ΔS2 affects the sum formation. The resulting sum signal S1 + S2 then consists of a disturbance which corresponds to the sum ΔS1 + ΔS2 of the respective disturbance. From this sum signal, the residual disturbance must now be determined or estimated, which is the difference signal after the

Difference formation in the difference signal formation device D1 is still superimposed.

5 is a schematic illustration of a known interference compensator device. In the one shown in Fig. 5

System is connected behind the transmitter UE a first differential signal generating device Dl, which forms a differential signal from the signals transmitted on the first line L1 and second line L2. This difference signal is converted into digital form in the analog / digital converter 10 and delayed in the delay device 30 by a time period T which is sufficient for the disturbance to be estimated in the lower branch. The delayed difference signal and the estimated disturbance are fed to a second difference signal generating device D2, which thus forms the disturbance-compensated output signal.

In the lower branch, the sum signal generating device SU1 is a sum signal from those on the first and second

Line L1, L2 transmitted signals formed and in one second analog / digital converter 20 converted into a digital signal. The fault determination device STE uses this to determine the estimate for the residual fault on the difference signal formed in the upper branch.

In other words, the distortion of the interference due to the asymmetry of the transmission system is simulated in the interference estimation device STE. A problem with this, however, is that the asymmetry is generally not known and can also only be estimated or can be determined by a calibration measurement.

WO 99/63675 discloses an interference compensator circuit for a communication channel with a hybrid device, which is connected to the communication channel, for supplying a differential output signal in accordance with a received signal. A delay unit delays the differential signal in a first branch by a suitable amount of time, so that the determination and subtraction of an estimated disturbance can be carried out in a second branch during this time. A summing device in the second branch extracts a digital sum signal from the transmitted signal, and a disturbance estimation unit creates a disturbance estimation signal as a function of a history or a time profile of the sum signal over a predetermined period of time and over a plurality of frequency bands. The estimated noise signal is subtracted from the delayed differential signal in the first branch to compensate for the noise in the differential signal.

The interference estimation device comprises an analysis filter bank for generating a multiplicity of sub-band signals, which are each at different frequencies, a multiplicity of noise detection circuits, each for Processing a respective one of the plurality of sub-band signals to create a respective component of the estimated interference signal, and a synthesis filter bank for processing the components of the estimated interference signal to form the resulting estimated interference signal.

CA 2,237,460 discloses a method for reducing radio frequency interference in digital subscriber loops, wherein a sum signal is also determined from a two-wire line and interference from a digital one

Noise estimation device is formed based on the sum signal. A large number of narrow-band filters are matched to corresponding passbands, and a fault detection unit detects the most disturbed passband and tunes one of the filters to this passband. The process is repeated for all other passbands to suppress the corresponding radio frequency interference signals in these bands.

From semiconductor circuit technology, U. Tietze, CH. Schenk, 11th edition, 1999, page 1133 ff., The basics of digital filters, in particular non-recursive filters with a finite impulse response, are known.

A disadvantage of the known interference compensation devices has been the fact that they have a complicated structure and in particular require a large number of components for subband splitting.

It is therefore the object of the present invention to provide an improved interference compensator device which can be implemented with less effort. According to the invention, this object is achieved by the interference compensator device specified in claim 1.

The idea on which the present invention is based is that the interference estimation device has an adapted, non-recursive digital filter device.

This has the advantage that the interference compensator device according to the invention is a residual interference after the communication systems that are not completely symmetrical

Differential signal formation can be largely compensated with a simple circuit structure.

Advantageous further developments and improvements of the subject matter of the invention can be found in the subclaims.

According to a preferred development, the adapted, non-recursive digital filter device has a plurality of multipliers, to which the sum signal is supplied digitized and delayed by one sampling period, the multipliers multiplying the signal value supplied in each case by a corresponding coefficient and a second sum signal forming device for forming a supply second sum signal; and the second sum signal is the estimated interference signal.

According to a further preferred development, a coefficient setting device is provided, to which an error signal can be fed which is a measure of the interference signal on the difference signal; and the coefficient setting device can set the coefficients such that the error signal satisfies a predetermined setting criterion.

According to a further preferred development, the predetermined setting criterion is cΛk + \) = c _n (k) -gF {Δx (k), x (kn, t ₀ ))

in which

g is a manipulated variable, in particular the reciprocal of a power of two, k is a time index, n is a coefficient number, F is a suitable function, x is the input function of the non-recursive filter device and tO is a delay in the error signal.

According to a further preferred development, the predetermined setting criterion is the MSE algorithm:

c _n (k + ϊ) = c _n (k) - g ■ Ax (k) ■ x (k -nt ₀ ).

According to a further preferred development, the predetermined setting criterion is the sign algorithm:

c _n (k + l) = c „(k) - g ^■ sga (Ax (k)) -x (knt ₀ )

c _n (k + l) = c _n (k) -g- Ax (k) ■ sga (x (k -nt ₀ ))

c „(k + i) = c _n (k) -g- sgn (Δx ()) • sgn (; t (-nt ₀ ))

where sgn () represents the sign function.

An embodiment of the invention is shown in the drawings and explained in more detail in the following description.

Show it:'

1 shows a schematic illustration of a noise compensator device according to an embodiment of the present invention; FIG. 2 shows a schematic illustration of a sum signal generating device in the embodiment according to FIG. 1;

3 shows a diagram to illustrate the general problems on which the present invention is based;

FIG. 4 shows a timing diagram for illustrating the output signal of a sum signal generating device in a known interference compensator device; and

Fig. 5 is a schematic representation of a known interference compensator device.

In the figures, identical reference symbols designate identical or functionally identical components.

1 is a schematic illustration of a noise compensator device according to an embodiment of the present invention.

The embodiment of the present invention shown in FIG. 1 is distinguished from the known interference compensator device described with reference to FIG. 5 in that it uses a special interference estimation device in the form of an adaptive, non-recursive digital filter device 50.

In addition to the reference symbols already introduced, in FIG. 1 x (k) denote the digitized sum signal of the first sum signal generating device SU1; 101, ... 10n-l delay elements; Ci, c ₂ , ..., Cn-i, c _n adjustable coefficients corresponding to multipliers Ml to Mn of the adaptive filter device 50, SU2 a second sum signal formation device, KE a coefficient setting device, FE a fault lersignalvermittlungseinrichtung for determining an error signal Δx, EM a receiver and ED receive data.

In this exemplary embodiment of the invention, the difference signal in the upper branch is delayed by the sampling device 30 by N sampling intervals T. After the coefficients ci, c ₂ ,,..., C _n -ι, - c _{n have} been set once, the adaptive, non-recursive filter device 50 provides an estimate of the residual interference signal which is superimposed on the difference signal. This estimate assumes that the residual interference signal is essentially linearly related to the sum signal, which has proven to be a good approximation in practice.

The following describes how the coefficients ci, c ₂ , ..., c _n _ι, c _{n of} the non-recursive digital filter device 50 are set.

The error determination device FE extracts from the receiver EM the difference signal generated by the second difference signal generation device D2 when no useful data are being transmitted and sends this signal as an error signal Δx to the coefficient setting device KE.

Based on this error signal Δx, the coefficient setting device KE sets the coefficients ci, c ₂ , ..., c _n -ι, c _{n of} the multipliers Ml to Mn such that the error signal Δx fulfills a predetermined error criterion. However, this coefficient setting is only possible in time segments if an interference signal is actually present.

On the other hand, it would also be possible to put a test interference signal on the line before starting the transmission of the actual useful signal and to set the filter coefficients Δx on the basis thereof. The receiver input signal as error signal .DELTA.x is appropriate if the setting of the coefficients Ci, c ₂ , ..., c _n _ι, c _{n of} the multipliers Ml to Mn can be carried out when the useful signal is switched off, for example during a starting procedure. The coefficient can be set, for example, according to the criterion of the middle square of the error signal.

The setting rule is then:

c „(k + ϊ) = c _n (k) ~ g - F {Ax (k), x (k - n, t ₀ ))

where g is a manipulated variable, in particular the reciprocal of

Power of two, k a time index, n a coefficient number, F a suitable function, x (k) the input function of the non-recursive filter device 50 and tO a possible delay of the error signal.

Depending on the selection of function F, various setting algorithms can be implemented, e.g.

a) the MSE algorithm:

c _n (k + 1) = c _n (k) - g - Ax (k) - x (k - n - t ₀ )

b) the sign algorithm

Here only the sign of either the error or the

Size x (k) or used by both. The following setting relationships are thus obtained:

c _n (k + l) = c _n (k) - g - sgn (Δx (fc)) - x (k - n - t ₀ ) c _n (k + ϊ) = c _n (k) - g - Ax (k) ■ sgn (x (k ~ n - t ₀ ))

c _n (k + 1) = c „(k) - g - sgn (Δx (t)) • sgn (x (£ - n - t ₀ ))

sgn () represents the sign function,

The setting time and the setting accuracy can be defined with the manipulated variable g. For reasons of stability, the manipulated variable must not exceed a certain maximum value.

FIG. 2 is a schematic illustration of a sum signal generating device in the embodiment according to FIG. 1.

2a, a resistor arrangement R, R 'is used to form the sum signal Σ, which is connected to the input side of the transmitter UE. This arrangement is particularly simple because it can be implemented solely through ohmic resistors R, R '.

A further possibility for forming the sum signal is shown in FIG. 2b, where the input-side winding of the transformer UE has two winding sections W, W ¹ , which have a center tap from which the sum signal Σ can be tapped.

Although the present invention has been described above with reference to a preferred exemplary embodiment, it is not restricted to this, but can be modified in a variety of ways.

The error signal Δx can in particular not only be the receiver input signal, but also the difference signal between see the signals at the input and output of the integrated decision maker of the receiver, which makes an estimate of the transmitted data based on the equalized difference signal. In this case, however, a possible delay in setting the coefficient must be taken into account. This criterion can also only be used when the receiver is in the synchronized state, ie the number of incorrectly decided symbols is sufficiently small. This option is practical if the error cannot be determined without a useful signal.

LIST OF REFERENCE NUMBERS

Ll, L2 first, second line

UE transmitter

Dl, D2 first, second difference signal forming device

SU1, SU2 sum signal generating device

10, 20 A / D converter

30 delay device

T sampling period

EM receiver

ED receive data

FE error signal detection device

Δx error signal

101, 10n-l delay device by em. Sampling period T

Ml-Mn multiplier cl-cn coefficients

KE coefficient setting device x (k) digital sum signal

50 non-recursive filter device

R, R ^Λ resistors

W, W winding sections

Σ sum signal

GND mass

Sl, S2 first, second signal

ΔS1, ΔS2 first, second signal interference

STE fault estimation facility

Claims

claims

1. Interference compensator device for compensating interference signals on two-wire lines of communication systems, with a useful signal (Sl) via a first line (Ll) and the inverted useful signal (S2) via a second line (L2) connected to a transmitter (UE) with :

a first difference signal forming device (Dl) connected to the transmitter (UE) for forming a first difference signal between the signals transmitted on the first and second lines (L1, L2);

delay means (30) for delaying the difference signal;

a first sum signal generating device (SU1) for forming a first sum signal of the signals transmitted on the first and second lines (L1, L2);

disturbance estimating means (50) for estimating a disturbance signal contained in the difference signal based on the first sum signal; and

a second difference signal formation device (D2) for forming a second difference signal between the first difference signal and the estimated interference signal for compensating the interference signal;

d a d u r c h g e k e n n z e i c h n e t that

the interference estimation device (50) has an adapted, non-recursive digital filter device (50).

2. interference compensator device according to claim 1, characterized in that the adapted, non-recursive digital filter device (50) has a plurality of multipliers (Ml, M2, ..., Mn-1, Mn), to which the sum signal is digitized, delayed by one sampling period, the multiplier (Ml , M2, ..., Mn-1, Mn) multiply the supplied signal value by a corresponding coefficient (cl, c2, ..., cn-1, cn) and a second sum signal generating device (SU2) to form a second one Supply sum signal; and the second suramen signal is the estimated interference signal.

3. Interference compensator device according to claim 2, so that a coefficient setting device (KE) is provided, which can be supplied with an error signal (Δx) which is a measure of the interference signal located on the difference signal; and the coefficient setting device (KE) can set the coefficients (cl, c2, ..., cn-1, cn) in such a way that the error signal (Δx) satisfies a predetermined setting criterion.

4. interference compensator device according to claim 3, d a d u r c h g e k e n n z e i c h n e t that the predetermined setting criterion

c _n (k + l) = c „(k) -gF {& x (k), x (kn, t ₀ j)

is where

g is a manipulated variable, in particular the reciprocal of a power of two, k is a time index, n is a coefficient number, F is a suitable function, x is the input function of the non-recursive filter device (50) and tO is a delay in the error signal.

5. Interference compensator device according to claim 4, so that the predetermined setting criterion is the MSE algorithm:

c „(k + l) = c _n (k) -g- x (k) ^■ x (k -nt _Q ).

6. Interference compensator device according to claim 4, so that the predetermined setting criterion is the algebraic sign algorithm:

c _n (k + l) = c _n (k) -g- sgn (Δ (/)) -x (knt _Q )

c _n (k + 1) = c _n (k) -g- Ax (k) ■ sga {x (k -nt ₀ ))

c _n (k + \) = c _n (k) -g- sg {Ax (k)) ^■ sga (x (k -nt ₀ ))

where sgn () represents the sign function.