RU2693272C1 - Device for high-order quadrature amplitude shift keying signal demodulator recovery - Google Patents

Abstract

FIELD: communication equipment.

SUBSTANCE: invention relates to communication engineering and can be used in digital demodulators of radio relay links operating in decimetre frequency range for demodulation of quadrature amplitude-shift keying (QASK) signals. Device for recovery of carrier frequency of high-order quadrature amplitude shift keying demodulator comprises quadrature amplitude-shift keying, decision unit on received symbol, loop filter, sine and cosine count generator, connected by its output to quadrature multiplier input, which is connected by an output to the input of the decision unit on the received symbol, and an amplitude detector. Device is also equipped with a unit for selecting a weight coefficient, a multiplier, a unit for calculating the accurate error, which together with the quadrature multiplier, a decision unit on the received symbol, a loop filter, a sine and cosine count generator and an amplitude detector, a circuit for accurate frequency estimation, and a coarse frequency estimation circuit comprising a quadrature multiplier, a unit for rough estimation of frequency offset, a loop filter and a sine and cosine count generator.

EFFECT: high accuracy of capturing a phase of a received signal when using high-order constellations QASK64-QASK256, without changing the band of the loop filter, and shorter time for capturing frequency tuning.

1 cl, 3 dwg

Description

The invention relates to the field of radio communications and can be used in digital demodulators of radio relay communication lines operating in the decimeter frequency range for demodulating quadrature amplitude shift keying signals (hereinafter referred to as QAM).

A device is known for restoring the carrier frequency of signals with amplitude-phase shift keying (see patent, RU 2234816, publ. 08/20/2004, H04L 27/34), containing the first and second phase detectors, the first and second integrators, a loop filter, a generator, controlled voltage, phase shifter and arctangent computing unit, the first inputs of the first and second phase detectors are combined and are the device input, the outputs of the first and second phase detectors are connected to the inputs of the first and second integrators, respectively, the output loop loop The filter is connected to the input of a voltage-controlled generator, the output of which is connected to the second input of the second phase detector and the input of the phase shifter, the output of which is connected to the second input of the first phase detector. The first and second analog-to-digital converters, a permanent storage device, the first and second quadrants, an adder, a square root calculator, a multiplier and a subtractor are entered into it, with the outputs of the first and second integrators connected to the inputs of the first and second analog-digital converters, respectively , the output of the first analog-digital converter is connected to the first input of the arctangent calculation unit, the first input of the permanent storage device and the input of the first quad, the output of the second en The A / D converter is connected to the second input of the arctangent calculator, the second input of the permanent storage device and the second quad input, the output of which is connected to the second input of the adder, the first input and output of which are connected, respectively, to the output of the first quad and the square root calculator input, the output of which is connected to the second input of the multiplier, the output and the first input of which are connected respectively to the input of the loop filter and the output of the subtractor, the first and second inputs of which o are connected, respectively, with the output of the arctangent calculation unit and the output of the persistent storage device.

The disadvantage of the analog is its low noise immunity when restoring the carrier frequency of signals with a combined amplitude-phase shift keying, due to the presence of spurious capture points in phase on its discriminatory characteristic.

The closest technical solution (prototype) is a device for restoring the carrier frequency of a demodulator of signals with sixteen-point amplitude-phase shift keying (see patent RU 2550548, publ. 10.05.2015, H04L 27/34, H03D 3/04) containing sequentially connected quadrature multiplier, the first two inputs of which are connected to two inputs of the device, a block for making decisions about received information symbols, two outputs of which are connected to two outputs of the device, an error signal generator, the second two inputs of which are Connected with two outputs of the quadrature multiplier, loop filter, integrator and shaper of sine and cosine samples, two outputs of which are connected to the second two inputs of the quadrature multiplier. It introduced serially connected unit for estimating the mathematical expectation of phase error, which includes an amplitude detector, the first two inputs of the mathematical expectation error estimating unit for phase error are connected to the outputs of the deciding unit for received information symbols, the third input - with the output of the error signal generator, and an adder, which is included in the gap between the integrator and the sine and cosine sampler, i.e. the second input of the adder is connected to the output of the integrator, and its output is connected to the input of the sine and cosine samples shaper.

The disadvantage of the prototype is its low noise immunity, with demodulation of high order constellations KAM64 - KAM256.

The technical result, the achievement of which the invention is directed, is to increase noise immunity and reduce the likelihood of error at the output, by increasing the accuracy of capturing the frequency control blocks when demodulating high-density signals (KAM64-KAM256) without changing the loop filter band. Also, the device for restoring the carrier frequency of a high-order QAM demodulator allows reducing the time required for capturing the frequency offset.

To achieve the above technical result, the device for restoring the carrier frequency of a high-order quadrature amplitude shift keying demodulator contains a quadrature multiplier, a block for deciding the received symbol, a loop filter, a sine and cosine sampler connected to its output from the quadrature multiplier that is connected to the block input by its output the decision on the received symbol, and the amplitude detector. The device is also equipped with a weighting factor selection unit, a multiplier, an exact error calculation unit that forms together with a quadrature multiplier, a received symbol decision block, a loop filter, a sine and cosine sampler, and an amplitude detector, an accurate frequency estimate loop, and a coarse frequency estimate loop, containing a quadrature multiplier, a coarse estimate unit for the frequency offset, a loop filter and a sine and cosine sampler, with the quadrature multiplier input of the coarse estimate contour and the frequency is connected to the input of the device, and the output is connected to the input of the quadrature multiplier circuit for accurate frequency estimation and to the input of the coarse frequency offset estimation unit, the output of which is connected to the input of loop coarse frequency loop loop filter, which is connected with its output to the input of the sine and cosine sampler whose output is connected to the quadrature multiplier input of the coarse frequency evaluation loop, and the quadrature frequency multiplier output of the exact frequency estimation loop is connected to the device output, the amplitude detector input and input The unit for calculating the exact error, the output of the decision block for the received symbol is connected to the first input of the multiplier, and the output of the amplitude detector is connected to the input of the weight selection block, which is connected by its output to the second input of the multiplier, connected to the input of the exact error calculator, which its output is connected to the input of a loop loop filter for accurate frequency estimation, which by its output is connected to the input of a sine and cosine sample former

A device for restoring the carrier frequency of a high-order QAM demodulator is illustrated by the following drawings:

in fig. 1 is a block diagram of a device to restore the carrier frequency of high-order QAM signals;

in fig. 2 is an example of a complex plane for KAM256;

in fig. 3 is a graph comparing the estimated phase error of the prototype and the claimed device.

The device for restoring the carrier frequency of the high-order quadrature amplitude shift keying demodulator (see Fig. 1) consists of a coarse frequency estimate loop, which contains a quadrature multiplier 1 (hereinafter referred to as KU), a coarse estimate unit for frequency offset 2, loop filter 3 (hereafter, FP ) and the driver of the sine and cosine counts 4, and the circuit of the exact frequency estimation, which includes KU 5, the block for deciding on the received symbol 6, the amplitude detector 7, the block for selecting the weighting factor 8, the multiplier 9, the block for calculating the exact o Shibki 10, OP 11 and shaper counts of sine and cosine 12.

The quadrature multiplier 1 is connected to the input of the device by the first input, the second input is connected to the output of the sine and cosine samples, and the output is connected to the first input of the quadrature multiplier 5 and the input of the coarse frequency offset 2, which is connected to the input of loop 3. The loop filter 3 is connected by its output to the input of the sine and cosine counting device 4. The quadrature multiplier 5 is connected to the output of the sine and cosine counting generator 12 by its second input, and the output is connected to the output of the decider about the received symbol 6, the input of the amplitude detector 7, the first input of the exact error calculation unit 10 and the output of the device. The output of the decision block on the received symbol 6 is connected to the first input of the multiplier 9. The output of the amplitude detector 7 is connected to the input of the weight gain selector 8, which is connected to the second input of the multiplier 9 connected to the second input of the exact error calculator by its output. The unit for calculating the exact error 10 is connected by its output to the input of the loop filter 11, the output of which is connected to the input of the sine and cosine sampler 12.

The device works as follows.

First, the signal is processed in a coarse frequency estimate loop. At the entrance of the KU 1 receive a comprehensive signal samples in the form:

r = I + jQ,

where I is the in-phase component of the signal,

j - imaginary unit

Q is the quadrature component of the signal.

With KU 1, the phase of the complex signal is rotated and a decision is made on the coordinates of the received point of the signal constellation.

выходы линии задержки суммируются, затем умножаются на входной отсчет, полученные произведения накапливаются в аккумуляторе, входящем в состав блока грубой оценки смещения частоты 2, при накоплении М отсчетов формируется сигнал ошибки, имеющий вид:Since the frequency compensation value is not known in advance, the complex signal sample passes without changes KU 1 and is fed to the input of the coarse frequency offset unit 2. It works as follows [1]: the input complex signal samples are converted into complex-conjugate signal samples and pass through the delay line long M, where each complex conjugate sample is multiplied by a weighting factor

the outputs of the delay line are summed up, then multiplied by the input sample, the resulting products are accumulated in the battery, which is part of the coarse estimate of the frequency offset 2, and when M samples are accumulated, an error signal is generated that looks like:

where T is the duration of the complex signal samples;

N is the number of analyzed complex signal samples;

M - the length of the delay line, M≤N-1;

k is the number of signal samples used to estimate the gross error, k = 1, 2, ..., M;

r _i - complex signal readout;

- комплексно-сопряженное значение отсчета сигнала. Затем значение сигнала ошибки поступает на вход ФП 3, который преобразует величину

с выхода блока грубой оценки частоты 2 в оценку грубого смещения Δƒ1, поступающую на вход формирователя отсчетов синуса и косинуса 4. Формирователь отсчетов синуса и косинуса 4 преобразует величину Δƒ1, поступающую с выхода ФП 3, в величину компенсации частоты, которая описывается выражением:

- complex-conjugate signal reading value. Then the value of the error signal is fed to the input of the FP 3, which converts the value

from the output of the coarsely estimated frequency block 2 to the coarse bias estimate Δƒ1, which arrives at the input of the sine and cosine samples shaper 4. The sine and cosine 4 samples shaper converts the value Δƒ1, which comes from the output of the AF 3, to the frequency compensation value, which is described by:

(cos (Δƒ1) + jsin (Δƒ1)).

Comprehensive reports of the frequency compensation value are fed to the input of the CG 1 and there they are multiplied by the input complex cues of the signal from the device's input to the CG 1 input.

The complex signal samples, roughly corrected in frequency, are described by the expression:

where (cos (Δƒ1) + jsin (Δƒ1)) is the signal from the output of the sine and cosine 4 shaper of the coarse frequency estimate contour.

Then, the complex signal samples roughly corrected in frequency are processed in the loop of an accurate frequency estimate, operating according to the feedback principle, where the frequency offset offset occurs in such a way that each subsequent complex signal count that arrives at the device output is multiplied by the exact error value calculated from the previous countdown signal.

The complex signal samples, roughly corrected in frequency, are fed to the input of the CU 5 circuit of an accurate frequency estimate, where they are multiplied by the output of the sine and cosine samples shaper 12. At the same time, complex signal samples of the form are formed:

where Δƒ2 is the estimate of the exact detuning of the carrier frequency;

(cos (Δƒ2) + jsin (Δƒ2)) is the signal from the output of the sine and cosine count generator 12.

From the output of the quadrature multiplier 5 circuit accurate assessment of the frequency of the complex signal samples arrive at the output of the device.

At the initial moment of time, when there is no information about the received signal, the error Δƒ2 = 0, and the complex signal readout to compensate for the frequency offset from the output of the sine and cosine sampler 12 takes the value:

(cos (0) + j sin (0)).

с выхода квадратурного умножителя 5 поступают на вход блока вынесения решения о принимаемом информационном символе 6, который принимает решение о значениях координат точки сигнального созвездия, в зависимости от положения на комплексной плоскости точек комплексного отсчета сигнала

(зависит от используемого режима модуляции) и формирует комплексные отсчеты сигнала вида

At the same time, the complex signal samples of the form

from the output of the quadrature multiplier 5 is fed to the input of the decision block on the received information symbol 6, which decides on the coordinates of the point of the signal constellation, depending on the position on the complex plane of the complex reference points of the signal

(depends on the modulation mode used) and forms complex samples of the signal of the form

For example, constellation KAM4 has only 4 points on the signal constellation: -1-1j, -1 + 1j, 1-1j, 1 + 1j, if a point with a value of 0.5-0.75j is accepted, the decision block on the received information symbol 6 will decide that this is a point 1-1j, since the distance to this point is the minimum. If the point is taken 0.1 + 0.1j, then the decision-making unit on the received information symbol 6 will decide that it is a point 1 + 1j, etc.

поступает на вход умножителя 9. В это же время с КУ 5 контура точной оценки частоты на вход амплитудного детектора 7 приходят комплексные отсчеты сигнала вида

В амплитудном детекторе 7 вычисляется мощность принятого комплексного сигнала по формуле:Comprehensive View Signal View

arrives at the input of the multiplier 9. At the same time, complex counts of the signal of the form arrive at the input of the KU 5 contour of an accurate frequency estimate at the input of the amplitude detector 7

In the amplitude detector 7, the power of the received complex signal is calculated by the formula:

к диагональным элементам созвездия. В зависимости от того, принадлежит ли он к диагональным элементам созвездия, блок выбора весового коэффициента 8 формирует весовой коэффициент k, который принимает следующие значения:The weight selection block 8 compares the values of A with the power threshold values (depending on the received constellation) and decides whether the complex reference signal symbol

to the diagonal elements of the constellation. Depending on whether it belongs to the diagonal elements of the constellation, the block for selecting the weight coefficient 8 forms the weight coefficient k, which takes the following values:

FIG. 2 shows an example of a complex plane for KAM256. The symbol "*" marked diagonal points of the constellation on a concentric circle, the symbol "+" - the remaining points. The choice of the weighting coefficient k allows to increase the noise immunity of the exact error calculation unit 10 when working with high order constellations (KAM64-256) [2].

поступивший с выхода блока вынесения решения о принимаемом символе 6 умножается на весовой коэффициент k. Преобразованный комплексный отсчет сигнала вида

с выхода умножителя 9 поступает на вход блока вычисления точной ошибки 10, в котором происходит вычисление сигнала точной частотной ошибки через выделение мнимой части от произведения комплексно-сопряженного значения сигнала

The weighting factor k from the output of the weight selection block 8 is fed to the input of the multiplier 9, in which the complex signal reading

received from the output of the block deciding on the received symbol 6 is multiplied by the weighting factor k. Converted complex view signal count

from the output of the multiplier 9 is fed to the input of the unit for calculating the exact error 10, in which the calculation of the signal of the exact frequency error occurs through the separation of the imaginary part from the product of the complex-conjugate signal value

Then, the signal for estimating the exact frequency error from the output of the exact error calculation unit 10 is fed to the input of the FP 11 of the exact frequency estimate loop, which forms Δƒ2, the estimate of the exact frequency offset of the carrier frequency. This value is fed to the input of the sine and cosine sampler 12, where a complex signal sample is formed for accurate frequency offset offset, which is described by the expression: (cos (Δƒ2) + jsin (Δƒ2)),

, смещенный по частоте относительно входных комплексных отсчетов сигнала на величину Δƒ=Δƒ1+Δƒ2.From the output of the sine and cosine 12 count generator, it is fed to the input of the CG 5, where it is multiplied by a complex sample signal

, shifted in frequency relative to the input complex signal samples by Δ величину = Δƒ1 + Δƒ2.

Further corrected complex samples of the signal of the form:

arrive at the output from the device to restore the carrier frequency of the demodulator of signals of high order quadrature amplitude manipulation.

From the graph comparing the estimate of the phase error (see Fig. 3), it can be seen that the claimed device allows to correctly generate an error signal in the range of angles from -85 ° to 85 °, while the prototype correctly generates an error signal in the range from -10 ° to 10 °.

Thus, a high-order QAM demodulator carrier recovery device has a high accuracy of capturing the phase of a received signal using high-order constellations KAM64-KAM256, without changing the loop filter band, and reduced time to capture the frequency offset. This allows to increase the noise immunity and reduce the probability of an error at the output of the digital QAM constellation demodulator during demodulation of high-density signals.

Information sources:

1. M. Luise and R. Reggiannini, "IEEE Trans. Carrier Frequency Recovery", "IEEE Trans. Communications, pp. 1169-1178, 1995.

2. H. Sari, S. Moridi, “New Phase and Frequency Detectors and QAM Systems,” IEEE Trans. Communications, pp. 1035-1043, 1988.

Claims (1)

A device for restoring the carrier frequency of a high-order quadrature amplitude shift keying demodulator containing a quadrature multiplier, block for deciding on the received symbol, loop filter, sine and cosine sampler, connected by its output to the input of the quadrature multiplier, which is connected by the output to the input of the decider on the received a symbol, and an amplitude detector, characterized in that it is equipped with a weighting factor selection unit, a multiplier, an exact error calculation unit together with a quadrature multiplier, a symbol decision block, a loop filter, a sine and cosine sample former and an amplitude detector, an accurate frequency estimate loop, and a rough frequency estimate loop containing a quadrature multiplier, a rough frequency offset estimate block, a loop filter and the driver of sine and cosine counts, and the input of the quadrature multiplier of the coarse frequency estimate loop is connected to the input of the device, and the output is connected to the input of the quadrature multiplier of the loop for accurate estimation frequency and with the input of the coarse frequency offset estimate unit, the output of which is connected to the input of the loop coarse frequency loop loop filter, which is connected by its output to the input of the sine and cosine sampler, the output of which is connected to the quadrature multiplier loop input, and the quadrature multiplier output An accurate frequency estimate loop is connected to the device output, an amplitude detector input and an accurate error calculation block input, the output of the received symbol decision block is connected to the first m input of the multiplier, and the output of the amplitude detector is connected to the input of the weighting factor selection unit, which is connected to the second input of the multiplier, connected to the input of the exact error calculator, which is connected to the input of the loop filter of the exact frequency estimate, which its output connected to the input of the sine and cosine count generator.

Images