SU1405119A1 - Device for adaptive correction of frequency distortion of signals of m-times phase modulation - Google Patents

Abstract

The invention relates to telecommunications and m. used in data transmission equipment. The purpose of the invention is to improve the accuracy of the correction in the absence of synchronism with the carrier oscillation. The device contains a correction filter 1, 2N + 1 control units 2, an error calculation unit 3. The device also includes an arc tangent calculation unit 4, five delay blocks 5-9, and M frequency doubling blocks 10. The proposed device allows adaptive correction of the frequency distortion of the signal in the absence of synchronism between the carrier and reference oscillations, since the generation of the error signal necessary to generate control signals for adjusting the gain factors of the corrective filter regulators 1 is not associated with the carrier oscillation synchronization system . Device m. implemented on the basis of processor signal processing. 3 il. with (L

Description

1 o ate

;about

The invention relates to electrical and telecommunications and can be used in data transmission equipment.

The aim of the invention is to improve the correction power in the absence of synchronism in the carrier oscillation.

I Figure 1 shows the structural I electrical circuit of the device for adapting the frequency distortion of the M-phase phase modulation signal; 2 is a structural electrical circuit of a frequency doubling unit; fig. 3 - the same, control unit.

The adaptive frequency distortion correction device of the M-time phase modulation signal contains correction filter 1, 2N + 1 control units 2, error calculation unit 3, arc tangent calculation unit 4, first 5 - fifth fifth delay blocks, M frequency doubling blocks 10.

The frequency doubling unit 10 comprises multipliers 11-13 and adders 14 and 15.

Control block 2 contains multipliers 16-25, adders 26-31.

The device for adaptive correction of the frequency distortion of the M-phase phase modulation signal works as follows.

The received samples of the in-phase XD (p) and quadrature components of the complex signal X (p) are fed to the input of the correction filter 1. From the output of the correction filter 1, the corrected samples of the in-phase -Vg (n) and quadrature Vj. (N) components of the complex signal V (n) is fed to the inputs of the first frequency doubling unit 10. The frequency doubling operation can be described as follows:

V4n) (V (n).

e

27

 V (n).

j (WnT + f -K (n) + q (n))

j 2o3-n-T + -K (n) + 2cf (n) lv (n) he CV5. (n) 4-jVc (n) (n) -V2 (n) +

+ j. 2 Vg (n) YC (n) z; (n) + j z (n),

where if is the current phase of the signal;

ω is the carrier frequency; n is the number of the clock interval; T - clock interval; M is the multiplicity of phase modulation. K (n) is a discrete random variable taking the values

0

five

0

0

five

0

five

0

five

0,1,2, ..., s (where S is the volume of the alphabet) and determining the informative value of the phase of the signal at the nth clock interval;

cf (n) is the phase deviation due to frequency distortions in the communication channel;

Z is the in-phase component of the signal reading at twice the frequency;

Z (; is the quadrature component of the signal with double frequency.

After passing the signal through the M frequency doubling unit 10 by means of multipliers 11-13 and adders 14 and 15, we obtain

v%) 4v (nf..T4.2irK (n) -2% (n), lv (n) l2. .M.n.T. (n)) .. m

+ jZc (n).

Thus, the manipulation is removed and the phase of the signal depends only on M times the doubled carrier frequency and phase error.

Next, in-phase Z counts and

U M M

quadrature Z, the constituent signals are received respectively at the first and second inputs of the block and the calculation of the arc tangent, where the current phase of the signal is determined in accordance with the expression

 ) C (n) arctg,

where Zg and Zg are the in-phase and quadrature components of the signal, respectively. From the output of the arctangent calculation unit 4, the current phase of the signal is fed to the first input of the error calculation unit 3 and to the input of the fifth delay unit 9, from the output of which the current phase delayed by the clock is fed to the second input of the error calculation unit 3, in which is the formation of the error signal in accordance with the expression dM ((p) -f (p-1) (p) -Cf (nl),

where the first term is the phase shift of the signal per clock interval and is a constant value at any clock interval for each type of transmission, the second term characterizes the frequency distortions jbv

The IRS of the signal, while sigia.h errors are not. depends on the synchronization on the carrying oscillation.

The signal from the output of the 3 V1 unit. The number of errors goes to the ninth inputs of the control units 2, in which the control signals for are generated. stochastic adjustment of the correction filter gain factors

1 in accordance with the expression

C; (n + 1) C; (n) - / u.P; (P); , N, where С (п + 1), С. (п) is the complex gain of the i-ro regulator of the correction filter at the (n + 1) -th and pth adjustment step, respectively; / К - weight coefficient;

P (p) is a complex control signal at the output of the i-ro of control unit 2.

The control signals are formed by the criterion of the minimum of the time-averaged error signal and, taking into account the stochastic approximation of the gradient, are described by the following expression:

R; (n) Ps; (n) + jPc; (n)) - (h) -Xc (i + 1) -V5 (n-1) + Xs (i + J) "Vc (n-1) VgCn) + Xci-VoCn) - -Xgd + i; -Vs (.n-1) -Xc (i + 1) -VcCn-l),

de and

  c;

in-phase and quadrature components of the control signal 40 at the outputs of the multipliers 24, 25 i-r of the control unit .2, respectively; 45 respectively, the in-phase and quadrature component of the signal samples in the i-th branch of the delayed-gQ delay filter 1;

in-phase and quadrature components of the signal in the (i +) -th retraction of the delay line of the correction filter 1, respectively;

50

five

0

five

0 5 Q

five

9

Vg (n-l) and

V (n-l) is the in-phase and quadrature components of the signal from the output of the correction filter 1, respectively, at the previous clock interval.

In each control unit 2, the signals Xgj are fed to the first inputs of the multipliers 16 and 21, the signals X .; - to the first inputs of multipliers 17, 20, signals X (i + 1) - to the first inputs of multipliers 18, 19, signals X (i + 1) - to the first inputs of multipliers 22, 23; signals US (n) to the second inputs of the multipliers 16, 20, signals V (n) to the second inputs of multipliers 17, 21. Signals Vg (nl) and V (.. (n-1) are formed at the outputs of blocks 6 and 8 delay and go to the second inputs of the multipliers 19, 23 and 18, 22, respectively.

The summation of the signals generated at the outputs of the multipliers 16-23 is carried out by means of adders 26-31. The signal Ler arrives at the combined inputs of the multipliers 24.25.

The signals X (N + 1) and) are generated at the outputs of delay blocks 7 and 5, respectively.

The present invention allows for the adaptive correction of the frequency distortion of the signal in the absence of synchronism between the carrier and reference oscillations, since the generation of the error signal necessary for generating control signals for adjusting the gain factors of the corrective filter regulators 1 is not associated with the synchronization system for the carrier oscillation.

The device can be implemented on the basis of processor signal processing.

Claims (1)

Invention Formula An adaptive correction of the frequency distortion of the M-phase modulation signal, containing a correction filter, the in-phase and quadrature information inputs and outputs of which are the corresponding in-phase and quadrature information inputs and outputs of the device, an error calculator, 2N + 1 frequency adaptive correction blocks dw W about 7 Phie. 2 H i s h .