JPH09181693A - Complex distortion estimation method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize distortion estimate and distortion compensation with higher accuracy. SOLUTION: In a known symbol reception timing by channels CH1, CH4, in-phase components r1k, r4k and quadrature components i1k, i4k, of complex distortion are detected by the multiplication with the reciprocal and they are converted into amplitude components A1, A4 and phase components θ1, θ4. Based on them, linear interpolation is carried out to estimate amplitude components A2, A3 and phase components θ2, θ3 of the complex distortion as to channels (CH2, CH3). In a known symbol reception timing by channels CH2, CH3, in-phase components r2k, r3k and quadrature components i2k, i3k of complex distortion are detected by the multiplication with the reciprocal and they are converted into amplitude components A2, A3 and phase components θ2, θ3. Based on them, linear extrapolation is carried out estimate amplitude components A1, A4 and phase components θ1, θ4 of the complex distortion as the channels (CH1, CH4).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is for compensating various complex distortions such as distortions generated at the time of transmission / reception through a fading transmission path and distortions caused by a filter used for connection with a transmission path. The present invention relates to a distortion compensation method, and more particularly to a complex distortion amount estimation method suitable for this type of method.

[0002]

2. Description of the Related Art FIG. 7 shows a structure of a radio slot used for downlink in a digital multi-channel access (MCA) system. In this system, four subcarriers whose frequencies are different from each other are used, and each subcarrier provides a total of four channels CH1 to CH4. The length of each channel CH1 to CH4 (basic slot length) is 1
It is 5.0 ms, and 60 symbols can be transmitted per channel per slot. Of these 60 symbols, the first 3 symbols are synchronization words, which are known symbols used for ensuring synchronization between transmission and reception. Most of the following 57 symbols are data symbols indicating the data to be transmitted.

However, not all symbols other than the sync word are data symbols. Actually, pilot symbols are intentionally mixed in a ratio of 1 symbol to 8 symbols and at a regular period. Also, the time of the pilot symbol in channel CH1 and that in channel CH4 are the same, and that in channel CH2 and that in channel CH3 are the same.
Furthermore, that of channels CH1 and CH4 and that of channels CH2 and CH3 are offset from each other. One of the purposes of mixing pilot symbols in such a manner is to generate various complex distortions such as distortions generated during transmission / reception via a fading transmission path and distortions caused by a filter used in connection with a transmission path. To compensate.

FIG. 8 shows an example of the configuration of a distortion compensation circuit using this pilot symbol (and synchronization word). The circuit shown in this figure roughly has a configuration in which the interpolators 1 and 2 detect or estimate the complex distortion amount in the received symbol, and the distortion compensator 4 compensates the complex distortion in the received symbol based on the result. Is. The delay unit 3 includes the interpolation units 1 and 2
The control section 5 is a means for controlling the components of the circuit shown in FIG. In this figure, it is premised that the signals quadrature-modulated into the I (in-phase) component and the Q (quadrature) component are transmitted through the channels CH1 to CH4.
The symbols received via ~ CH4 each have the complex Cartesian coordinate form I + jQ (j: imaginary unit, and so on). Also, in the figure, it is the received symbol after distortion compensation that has the complex Cartesian coordinate type I '+ jQ'.

Of the two interpolation units provided, the interpolation unit 1
Has the configuration shown in FIG. 9 and is controlled by the control unit 5. First, the control unit 5 synchronizes with the transmission side by detecting the synchronization word in the received symbol. On the other hand,
The symbol arrangement in this system, that is, the arrival order of data symbols whose contents are unknown, pilot symbols and synchronization words whose contents are known, is given in advance as shown in FIG. The control unit 5 maintains the synchronization with the transmission side and based on the symbol arrangement shown in FIG.
The channel CH1 determines whether the symbol input to the interpolation unit 1 at the current time is an unknown symbol or a known symbol, and when the symbol is an unknown symbol, whether the symbol is a pilot symbol or a synchronization word. ~ Determine for each CH4. As a result of the determination, for the channel for which the symbol input to the interpolation unit 1 is a known symbol, the control unit 5
The known symbol is supplied to the corresponding multiplier in the complex multiplication unit 1b by turning on the switch corresponding to the channel among the switches forming the switch 1a. For the other channels, the controller 5 uses the switch 1
Among the switches forming a, the switch corresponding to the channel is kept in the off state.

The control unit 5 supplies the known symbol to the multiplier corresponding to the known symbol receiving channel by the above-mentioned processing. On the other hand, the control unit 5 supplies the reciprocal of the known symbol (corresponding to the complex conjugate when the amplitude of the known symbol is 1) to the multiplier according to the above determination result. Since the system handled here uses three known symbols (two pilot symbols and one synchronization word), the reciprocal number supplied from the control unit 5 to the multiplier is three known. It is the reciprocal of one of the symbols. Each multiplier multiplies the received symbol (known symbol) supplied via the switch 1a by the reciprocal supplied from the control unit 5. When a known symbol that does not include complex distortion is multiplied by its reciprocal number, the result is a real number 1. However, since the received symbol actually input to the multiplier includes complex distortion, The complex number obtained as a result and having the complex Cartesian coordinate type rk + jik is a vector value representing the amount of this complex distortion.

The interpolation calculation unit 1c receives the complex distortion amount rk + jik for the channels CH1 and CH4 from the complex multiplication unit 1b (that is, when a synchronization word is received and when pilot symbols are received via the channels CH1 and CH4). ), The linear distortion interpolation based on these estimates the complex distortion amount in the other channels CH2 and CH3. That is, based on the output of the multiplier corresponding to the channel CH1, that is, the detected value C1k = r1k + ji1k of the complex distortion amount caused by the transmission through the channel CH1 and the value C4k = r4k + ji4k of the channel CH4, the interpolation calculation unit 1c determines the channel CH2. Estimated value of complex distortion amount C2k = r2k + ji2 caused by transmission through the network
k and that of channel CH3 C3k = r3k + ji3k
To

## EQU1 ## C2k = C1k * IN2 + C4k * IN1 C3k = C1k * IN1 + C4k * IN2 where *: Calculated by the formula of multiplication. In this equation, IN1 and IN2 are interpolation coefficients, and IN1 + IN2 = 1 when the difference between adjacent subcarrier frequencies is uniform.

Similarly, the extrapolation calculation unit 1d includes a complex multiplication unit 1d.
b to channel CH2 and CH3 complex distortion amount rk +
When jik is given (that is, when a synchronization word is received and pilot symbols are received through channels CH2 and CH3), linear extrapolation based on these estimates the complex distortion amount in other channels CH1 and CH4. To do. That is, the output value of the multiplier corresponding to the channel CH2, that is, the detected value C2k = r2k + ji2 of the complex distortion amount caused by the transmission through the channel CH2 or the like.
k and that of channel CH3 C3k = r3k + ji3k
Based on the above, the interpolation calculation unit 1c causes the estimated value C1k = r1k + of the complex distortion amount caused by transmission or the like via the channel CH1.
ji1k and that of channel CH4 C4k = r4k + j
i4k,

Equation 2 C1k = C2k * EX2 + C3k * EX1 C4k = C2k * EX1 + C3k * EX2 In this formula, EX1 and EX2 are extrapolation coefficients, and EX2 + EX1 = 1 when the difference between adjacent subcarrier frequencies is uniform.

When the control unit 5 detects that the received symbol is a synchronization word by the above determination, the detected values C1k and C4k from the interpolation calculation unit 1c are changed to the extrapolation calculation unit 1 by the interpolation calculation unit 1c.
The switch 1e is controlled so that the detected values C2k and C3k from d are supplied to the interpolator 2. Further, when it is detected that the received symbols via the channels CH1 and CH4 are pilot symbols, the switch 1e is controlled so that the interpolation calculation unit 1c supplies the detected values and the estimated values C1k to C4k to the interpolation unit 2. To do. Further, when it is detected that the received symbols via the channels CH2 and CH3 are pilot symbols, the extrapolation calculation unit 1d outputs the detected values and the estimated values C1k to C4k to the interpolation unit 2.
The switch 1e is controlled so as to be supplied to.

The detection values or estimated values C1k to C4k of the complex Cartesian coordinate system supplied to the interpolating unit 2 in this manner are the respective channels CH1 to CH4 at the known symbol reception time.
It represents the amount of complex distortion that is (or is likely to be) generated by transmission through the. Therefore, regarding the time other than the known symbol reception time, that is, the time when the data symbols are received from all the channels CH1 to CH4, it is necessary to newly estimate the complex distortion amount. The interpolator 2 estimates the complex distortion amount at other times by performing interpolation calculation based on the detected value or the estimated value C1k to C4k at the known symbol reception time. In this way, the distortion compensator 4
At, a complex distortion amount C4 that can be used for distortion compensation of the received symbol (more strictly, the received symbol whose processing delay has been compensated by the delay unit 3) I + jQ is obtained.

[0011]

However, in such a distortion compensation method, amplitude distortion and phase distortion are newly added to the estimated value of the complex distortion amount. First, for simplification of the explanation, considering the interpolation calculation on the assumption that the in-phase components (I) of the detected values C1k and C4k are equal and the amplitude is constant (I ² + Q ² = constant), As shown in FIG. 10, the estimated values C2k and C obtained by the interpolation calculation
The amplitudes A2 'and A3' of 3k (solid line) are different from the amplitudes A2 and A3 of the estimated value (broken line) that should be originally obtained. Similarly, the in-phase component (I) of the detected values C2k and C3k
Considering the extrapolation calculation under the assumption that the two are equal, as shown in FIG. 11, the amplitudes A1 ′ and A4 of the estimated values C1k and C4k (solid line) obtained by the extrapolation calculation are obtained.
′ Is a value different from the amplitudes A1 and A4 of the estimated value (broken line) that should be originally obtained. These represent the occurrence of amplitude distortion. When the in-phase components of the detected values are not equal, the phase θ of the estimated values C2k and C3k obtained by the interpolation calculation
2 and θ3 and the phases θ1 and θ4 of the estimated values C1k and C4k obtained by extrapolation also show a difference from the originally obtained value, that is, phase distortion. Note that FIG. 10 and FIG.
1 is a diagram based on FIGS. 7 to 9, but the same problem still occurs in other configurations in which the number of subcarriers and the details of symbol arrangement are different.

One of the objects of the present invention is to reduce or prevent distortion occurring at the time of estimation by executing the estimation of the complex distortion amount by interpolation calculation, extrapolation calculation, etc. in polar coordinate expression, that is, in the frequency domain. The result is to ensure that the complex distortion can be compensated more reliably.

[0013]

Means for Solving the Problems and Effects of the Invention In order to achieve such an object, the method for estimating the amount of complex distortion according to the first aspect of the present invention comprises three or more signals which are simultaneously received by carriers of different frequencies. Detection step of detecting the complex distortion amount included in the complex symbols of any two or more of the complex symbols, and performing interpolation or extrapolation based on the detected complex distortion amount and using polar coordinate representation by amplitude and phase. An estimation step of estimating the complex distortion amount included in the remaining complex symbols that are simultaneously received by executing the interpolation calculation,
It is characterized by having. According to this configuration, the interpolation or extrapolation calculation is executed using the polar coordinate expression, and the complex distortion amount is estimated by this. Since the phase of the complex distortion amount detection value expressed in polar coordinates corresponds to the carrier frequency of the complex symbol related to the complex distortion amount, the estimation processing using the polar coordinate expression is equivalent to the estimation processing in the frequency domain. Since interpolation and extrapolation in the frequency domain can be performed without adding distortion to the amplitude and phase, this configuration can reduce or prevent distortion that occurs during estimation, and more accurate complex distortion amount estimation. It is possible to get the value.

In the complex distortion amount estimating method according to the second configuration of the present invention, in the first configuration, the expression type of the three or more complex symbols is a Cartesian coordinate expression with a real part and an imaginary part. The detecting step multiplies the complex symbol, which is considered to be transmitted as a known complex symbol among the above-mentioned three or more complex symbols, by the reciprocal of the known complex symbol to determine the complex distortion amount included in the complex symbol. A complex multiplication step of detecting according to a Cartesian coordinate expression by the part and the imaginary part, and a coordinate conversion step of converting the expression type of the complex distortion amount detected according to the Cartesian coordinate expression from the Cartesian coordinate expression to the polar coordinate expression. Is characterized by. According to this configuration, since the conversion from the rectangular coordinate representation to the polar coordinate representation is performed, the first configuration can be applied to the system that transmits the orthogonally modulated signal.

The distortion compensating method according to the third aspect of the present invention is
A first complex distortion amount estimation step for executing the complex distortion amount estimation method according to the second configuration, based on the above-mentioned three or more complex symbols belonging to the corresponding columns of the respective three or more complex symbol sequences, A second complex distortion amount estimation for estimating a complex distortion amount included in a complex symbol that belongs to the above-mentioned three or more complex symbol sequences and is received at another time based on the complex distortion amount obtained in the complex distortion amount estimation step. And a distortion compensating step for compensating the complex distortion contained in each complex symbol belonging to the above-mentioned three or more complex symbol sequences, based on the complex distortion amount obtained in the first or second complex distortion amount estimating step. And are included. According to this configuration,
In a system for transmitting a quadrature modulated signal, the first
It is possible to realize accurate distortion compensation using the accurate estimated value obtained by the configuration.

The distortion compensating method according to the fourth aspect of the present invention is
In the third configuration, the known complex symbols are mixed in the sequence of three or more complex symbols prior to the transmission so that the time when two or more of the known complex symbols are simultaneously received arrives at a predetermined frequency. It is characterized by setting.
According to this configuration, since the timing for detecting and estimating the complex distortion amount is intentionally generated, it is possible to perform the distortion compensation at a necessary frequency.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings. The same components as those in the conventional technique shown in FIGS. 7 to 11 are designated by the same reference numerals and the description thereof will be omitted. FIG. 1 shows the configuration of a complex distortion amount estimation circuit used as an interpolation unit 1 according to an embodiment of the present invention. The first difference of this circuit from the conventional interpolation unit 1 shown in FIG. 9 is that the interpolation calculation unit 1f and the extrapolation calculation unit 1g.
Is to execute the linear interpolation or extrapolation (interpolation) operation after converting the complex distortion amount detection value given to any two channels from the rectangular coordinate expression to the polar coordinate expression. That is, the interpolation calculation unit 1f and the extrapolation calculation unit 1g
Is the in-phase component rk and the quadrature component ik of the complex distortion amount detection value.
Is converted into an amplitude component A and a phase (frequency) component θ, and then linear interpolation or extrapolation (interpolation) calculation is executed.

More specifically, the interpolation calculation section 1f causes the complex distortion amount detection value r1k + for the channels CH1 and CH4.
ji1k and r4k + ji4k,

## EQU00003 ## A1 = SQRT (r1k ** 2 + i1k ** 2) A4 = SQRT (r4k ** 2 + i4k ** 2) θ1 = arcTAN (r1k, i1k) θ4 = arcTAN (r4k, i4k) However, SQRT (x) : A function that gives a positive square root of x arcTAN (x, y): A function that gives an arctangent of x / y **: Amplitude components A1 and A4 and a phase component θ1 according to the power formula
And θ4. Then, the interpolating calculator 1f relates to the channels CH2 and CH3.

## EQU4 ## A2 = A1 * IN2 + A4 * IN1 A3 = A1 * IN1 + A4 * IN2 θ2 = θ1 * IN2 + θ4 * IN1 θ3 = θ1 * IN1 + θ4 * IN2 According to the formula, amplitude components A2 and A3 and phase component θ2 And θ3 are obtained.

Similarly, the extrapolation calculation unit 1g uses the channel CH.
2 and CH3 complex distortion amount detection value r2k + ji2k
And r3k + ji3k,

A2 = SQRT (r2k ** 2 + i2k ** 2) A3 = SQRT (r3k ** 2 + i3k ** 2) θ2 = arcTAN (r2k, i2k) θ3 = arcTAN (r3k, i3k) A2 and A3 and phase component θ2
And θ3. Then, the extrapolation calculation unit 1g relates to the channels CH1 and CH4.

[Equation 6] A1 = A2 * EX2 + A3 * EX1 A4 = A2 * EX1 + A3 * EX2 θ1 = θ2 * EX2 + θ3 * EX1 θ4 = θ2 * EX1 + θ3 * EX2 Amplitude components A1 and A4 and phase component θ1 And θ4 are obtained.

Thus, from the (I, Q) planes shown in FIGS. 10 and 11 to (A, θ) shown in FIGS. 2 and 3.
In the present embodiment, new distortion is prevented from occurring in the complex distortion amount estimation operation by performing linear interpolation and extrapolation (interpolation) operations after converting into a plane. For example,
In the (I, Q) plane, the amplitude undistorted line becomes a curve, but (A,
Since the θ) plane is a straight line as shown by the broken lines in FIGS. 2 and 3, if linear interpolation is performed on the (A, θ) plane, in principle, distortion with respect to amplitude and phase is unlikely to occur.

The second difference of the circuit shown in FIG. 1 from the conventional interpolating section 1 shown in FIG. 9 is that the switch 1e is replaced with the switch 1h because the linear interpolation is performed in the frequency amplitude space. Is provided and the complex conversion unit 1i is provided at the subsequent stage of the switch 1h. The switch 1h switches in accordance with the control signal from the control unit 5, and outputs the complex distortion amount detection value from the interpolation calculation unit 1f and the extrapolation calculation unit 1g at the synchronization word reception timing and the internal signal at the pilot signal reception timing by the channels CH1 and CH4. The complex distortion amount detection value and the estimated value from the insertion calculation unit 1f are set to the channels CH2 and CH3.
At the pilot signal reception timing by
The complex distortion amount detection value and the estimated value from g are supplied to the complex conversion unit 1i. The complex transformation unit 1i has the following equation

## EQU00007 ## C1k = A1 * e **-j.theta.1 = r1k + ji1k C2k = A2 * e **-j.theta.2 = r2k + ji2k C3k = A3 * e **-j.theta.3 = r3k + ji3k C4k = A4 * e * -j.theta.4 = r4k + ji4k, ji4k. The complex distortion amount detected value or estimated value is converted into a rectangular coordinate type.

The third difference between the circuit shown in FIG. 1 and the conventional interpolator 1 shown in FIG. 9 is that the interpolation calculator 1f and This is because the extrapolation calculation unit 1g executes the frequency replacement process. That is, as shown in FIG. 4 using the interpolation calculation as an example, the phase θ represented with the line of θ = 0 as a reference.
When the sum of 1 and θ4 exceeds 180 degrees due to the shift of the initial phase, etc., when the linear interpolation is executed at the phases θ1 and θ4, the estimated values become the phases θ2 and θ3, and therefore the detected values of the phases θ1 and θ4. And the phase between the estimated values θ2 and θ3 is lost. Therefore, in the present embodiment, when such a condition is satisfied, a frequency replacement process is performed to add or subtract 360 degrees to either of the detected value phases θ1 and θ4. For example, if the frequency reading process is executed for the detected value phase θ4 and linear interpolation is executed based on the detected value phase θ1 and the read phase θ4 ′, the resulting estimated value phases are θ2 ′ and θ3. It becomes ´. As a result, the detected value phases θ1 and θ4 and the estimated value phase θ2 '
And the continuity between θ3 ′ and θ3 ′ can be maintained, and thus the complex distortion amount estimation with higher accuracy can be realized.

Although the above description is based on the digital MCA system, the present invention can be applied to other systems. That is, even in a system in which the symbol arrangement on the radio interface is different from that shown in FIG. 7, the number N of subcarriers (channels) is 3
As described above, the number m of simultaneously inserted known symbols is 2 or more, and N>
The present invention can be applied to any system having a symbol arrangement designed to be m (see, for example, FIGS. 5 and 6). Further, although the above description does not particularly mention the signal modulation method, 16QAM (16-value quadrature amplitude modulation),
The present invention can be applied to any quadrature modulation with a phase modulation component such as 4PSK (four-valued phase shift keying). In a system in which a known symbol is included in the symbols other than the synchronization word and the pilot signal, the symbol can also be used for distortion estimation / distortion compensation according to the present invention. In addition, the present invention can be applied even to a system in which the functional configuration of the distortion compensation circuit on the receiving side is different from that shown in FIG.

[Brief description of the drawings]

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

FIG. 2 is a frequency domain diagram showing the contents of interpolation calculation in this embodiment.

FIG. 3 is a frequency domain diagram showing the contents of extrapolation calculation in this embodiment.

FIG. 4 is an IQ plan view showing the contents of frequency replacement processing in this embodiment.

FIG. 5 is a timing chart showing a symbol arrangement in which the number of subcarriers is 5 and the number of pilot symbols inserted simultaneously is 2 or 3.

FIG. 6 is a frequency domain diagram showing the contents of interpolation calculation in this symbol arrangement.

FIG. 7 is a timing chart showing symbol arrangement in a digital MCA system.

FIG. 8 is a block diagram showing a configuration of a distortion compensation circuit.

FIG. 9 is a block diagram showing a configuration of a circuit according to a conventional technique.

FIG. 10 is an IQ plan view showing the contents of interpolation calculation in this conventional technique.

FIG. 11 is an IQ plan view showing the contents of extrapolation calculation in this conventional technique.

[Explanation of symbols]

1, 2 interpolation unit, 1a, 1h switch, 1b complex multiplication unit, 1f interpolation calculation unit, 1g extrapolation calculation unit, 3 delay unit, 4 distortion compensation unit, 5 control unit, I1 to I4 in-phase components of received symbols, Q1 to Q4 orthogonal components of received symbols,
I1 ′ to I4 ′ In-phase component of received symbol after distortion compensation (real part), Q1 ′ to Q4 ′ Quadrature component of received symbol after distortion compensation (imaginary part), C1k to C4k Complex distortion amount detection at known symbol reception time Value or estimated value, C1 ■ to C4
■ Complex distortion amount detected value or estimated value, r1k to r4k Complex distortion amount detected value at known symbol reception time or in-phase component (real part) of estimated value, i1k to i4k Complex distortion amount detected value or estimated value at known symbol reception time Quadrature component (imaginary part), A1 to A4 amplitude of complex distortion amount detection value or estimated value at known symbol reception time, θ1 to θ4 phase (frequency) of complex distortion amount detected value or estimated value at known symbol reception time, IN1 , IN2 interpolation coefficient, EX1, EX
2 Extrapolation coefficient.

Claims (4)

[Claims] 1. A detection step of detecting a complex distortion amount contained in a complex symbol of any two or more of three or more complex symbols simultaneously received by carriers of different frequencies, and detecting the complex distortion amount. An estimation step of estimating the amount of complex distortion contained in the remaining complex symbols received at the same time by executing interpolation or extrapolation calculation based on the polar coordinate expression based on the amplitude and the phase, and A method for estimating the amount of complex distortion. 2. The complex distortion amount estimation method according to claim 1, wherein the expression type of the three or more complex symbols is an orthogonal coordinate expression by a real part and an imaginary part, and the detection step is the three or more. Of the complex symbols of, the complex symbol that is considered to have been transmitted as a known complex symbol is multiplied by the reciprocal of the known complex symbol to express the complex distortion amount included in the complex symbol in the orthogonal coordinate representation by the real part and the imaginary part. And a coordinate transformation step of transforming the phenotype of the complex distortion amount detected according to the Cartesian coordinate expression into the polar coordinate expression from the Cartesian coordinate expression. Method. 3. A complex distortion amount estimation method according to claim 2, which is based on the three or more complex symbols belonging to corresponding columns of the three or more complex symbol sequences, respectively.
Based on the complex distortion amount estimation step and the complex distortion amount obtained in the first complex distortion amount estimation step, the complex distortion amounts included in the complex symbols belonging to the above three or more complex symbol sequences and received at other times are calculated. , The second complex distortion amount estimating step, and the complex distortion included in each complex symbol belonging to the above-mentioned three or more complex symbol strings, the complex distortion obtained in the first or second complex distortion amount estimating step. A distortion compensation step of compensating based on the quantity; 4. The distortion compensating method according to claim 3, wherein the known complex symbols are set to the three or more complex symbols so that a time at which two or more known complex symbols are simultaneously received arrives at a predetermined frequency. Distortion compensation method characterized in that the columns are mixed before transmission.