CN110752870B - Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system - Google Patents

Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system Download PDF

Info

Publication number
CN110752870B
CN110752870B CN201911037523.2A CN201911037523A CN110752870B CN 110752870 B CN110752870 B CN 110752870B CN 201911037523 A CN201911037523 A CN 201911037523A CN 110752870 B CN110752870 B CN 110752870B
Authority
CN
China
Prior art keywords
signal
error estimation
error
coefficient
roll
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201911037523.2A
Other languages
Chinese (zh)
Other versions
CN110752870A (en
Inventor
崔霞霞
李新玲
苟晓刚
王少飞
张丽娜
刘一龙
韩中良
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
CETC 54 Research Institute
Original Assignee
CETC 54 Research Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by CETC 54 Research Institute filed Critical CETC 54 Research Institute
Priority to CN201911037523.2A priority Critical patent/CN110752870B/en
Publication of CN110752870A publication Critical patent/CN110752870A/en
Application granted granted Critical
Publication of CN110752870B publication Critical patent/CN110752870B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/14—Relay systems
    • H04B7/15—Active relay systems
    • H04B7/185—Space-based or airborne stations; Stations for satellite systems
    • H04B7/1851—Systems using a satellite or space-based relay
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/14—Relay systems
    • H04B7/15—Active relay systems
    • H04B7/185—Space-based or airborne stations; Stations for satellite systems
    • H04B7/1851—Systems using a satellite or space-based relay
    • H04B7/18519—Operations control, administration or maintenance
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04B—TRANSMISSION
    • H04B7/00—Radio transmission systems, i.e. using radiation field
    • H04B7/14—Relay systems
    • H04B7/15—Active relay systems
    • H04B7/185—Space-based or airborne stations; Stations for satellite systems
    • H04B7/18578—Satellite systems for providing broadband data service to individual earth stations
    • H04B7/18586—Arrangements for data transporting, e.g. for an end to end data transport or check

Abstract

The invention discloses a timing recovery method and a timing recovery device for a broadband satellite transmission system with a variable roll-off coefficient, belonging to the field of broadbandThe technical field of timing recovery of a self-adaptive transmission system with a satellite. Firstly, according to the roll-off coefficient of a forming filter, correspondingly adjusting the roll-off coefficient of a matched filter; then, according to the roll-off coefficient, determining which timing error estimation algorithm is adopted by the signal output by the matched filter; then, the estimated error obtained by different timing error estimation algorithms is adopted, the same loop filter, timing NCO and interpolation filter are adopted, the parameters of the loop filter are adjusted, and the optimal sampling value y is interpolated in 4 times of sampling pointsk. The invention adjusts the roll-off coefficient of the matched filter according to the signaling information according to the characteristics of the broadband satellite self-adaptive transmission system with variable roll-off coefficient, can realize optimal reception, and can ensure the stable operation of the system by adopting different timing algorithms according to the roll-off coefficient.

Description

Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system
Technical Field
The invention relates to the technical field of timing recovery of a broadband satellite self-adaptive transmission system, in particular to a timing recovery method and a timing recovery device of a broadband satellite transmission system with a variable roll-off coefficient, which can be used for timing recovery of a communication system with a variable roll-off coefficient and a high-order modulation mode.
Background
In a broadband satellite self-adaptive transmission system with variable roll-off coefficients, the change of the roll-off coefficients of a shaping filter at a sending end can change transmission bandwidth and influence transmission efficiency. If a general matched filter with a fixed roll-off coefficient is adopted at a receiving end, optimal receiving cannot be achieved, and system performance is seriously influenced. In addition, the general matched filter is tapped when being applied in a wide bandThe coefficient is 64 th order at the maximum, and the roll-off coefficient
Figure DEST_PATH_IMAGE001
The smaller the roll-off coefficient, the greater the impact on system performance. Coefficient of roll off
Figure 921987DEST_PATH_IMAGE001
In time, the error estimation accuracy of the general timing algorithm is deteriorated, so that the timing jitter is too large, the demodulation performance is influenced, and the step-out is easily caused.
The Moeneclaey algorithm is not affected by self-noise and can work in low roll-off coefficient and high-order modulation application systems. However, when the roll off coefficient is
Figure 609845DEST_PATH_IMAGE002
Particularly, when a high-order modulation mode is adopted, the error estimation precision is greatly reduced, and the system performance is increasingly poor.
Disclosure of Invention
In order to solve the defects of the prior art in the matching and timing algorithms of the broadband satellite adaptive transmission system with variable roll-off coefficients, the invention provides a timing recovery method and a timing recovery device of the broadband satellite transmission system with variable roll-off coefficients, which can carry out variable matching filtering of different tap coefficients aiming at different roll-off coefficients and can adopt different timing algorithms in sections according to the roll-off coefficients.
In order to achieve the purpose, the technical scheme adopted by the invention is as follows:
a timing recovery method for a broadband satellite transmission system with a variable roll-off coefficient is applied to a receiving end of the broadband satellite transmission system and used for timing recovery of a signal received by the receiving end from a transmitting end of the broadband satellite transmission system, and comprises the following steps:
(1) matching signals output by a forming filter in a transmitting end of the broadband satellite transmission system, and outputting sampling signals to realize optimal receiving of the signals; when the roll-off coefficient of the sending end is changed, the tap coefficient is changed in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
(2) adopting a corresponding error estimation mode according to the roll-off coefficient of the sending end when the roll-off coefficient
Figure 381492DEST_PATH_IMAGE002
According to the sampling signal output by matched filter, adopting square timing error estimation algorithm to make error estimation, when roll-off coefficient is used
Figure 269813DEST_PATH_IMAGE001
According to the optimal sampling point output by the error compensation filter
Figure DEST_PATH_IMAGE003
Carrying out error estimation by adopting a Moeneclaey algorithm;
(3) sequentially performing loop filtering and timing time-controlled oscillator processing on the error estimation result of the step (2), and outputting a processed error signal;
(4) carrying out timing compensation on the sampling signal output in the step (1) according to the error signal obtained in the step (3), and outputting the optimal sampling point
Figure 152318DEST_PATH_IMAGE003
And completing the timing recovery of the receiving end.
Furthermore, in the step (1), matching is performed in a parallel manner, and a plurality of paths of parallel sampling signals are output, and in each parallel process, the tap coefficient of the process is set in a table look-up manner according to the roll-off coefficient.
Further, in the step (1), there are four parallel processes and four parallel sampling signals are output, and the square timing error estimation algorithm in the step (2) is a parallel square timing error estimation algorithm, and the parallel square timing error estimation algorithm includes the following steps:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
and dividing the real part signal by the imaginary part signal, and obtaining the arc angle through arc tangent to obtain an estimated value of the timing error.
Further, the step (4) of performing timing compensation on the sampling signal output in the step (1) by using an interpolation filter specifically includes the following steps:
adopting corresponding input signals for the interpolation filter according to the error estimation mode selected in the step (2); specifically, if the Moeneclose algorithm is selected for error estimation, the input signal is the multi-path parallel sampling signal output in the step (1), and if the square timing error estimation algorithm is selected for error estimation, the input signal is the effective signal part of the multi-path parallel sampling signal after burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals obtained in the step (3);
selecting the tap coefficient of the interpolation filter by table look-up method according to the error signal, and recovering the optimal sampling point by parallel resampling
Figure 691753DEST_PATH_IMAGE003
。
Further, the moencleay algorithm comprises the following steps:
single-pass optimum sampling value of interpolation filter output using error estimation as 0
Figure 634301DEST_PATH_IMAGE003
As the initial input signal of the Moeneclaey algorithm, the input signal of the subsequent Moeneclaey algorithm is the adjusted optimal sampling value output by the interpolation filter
Figure 9919DEST_PATH_IMAGE003
；
Obtaining an output error estimation signal of a Moeneclaey algorithm according to the following formula
Figure 617487DEST_PATH_IMAGE004
：
Figure DEST_PATH_IMAGE005
，
Wherein the content of the first and second substances,
Figure 27739DEST_PATH_IMAGE006
is that
Figure 141189DEST_PATH_IMAGE003
The signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
In addition, the invention also provides a timing recovery device for the broadband satellite transmission system with the variable roll-off coefficient, which is used for timing recovery of a signal from a broadband satellite transmission system transmitting end received by a broadband satellite transmission system receiving end, and comprises a matched filter, a square timing error estimation module, a Moeneclose timing error estimation module, an error estimation selection module, an error processing module and an error compensation filter, wherein:
the matched filter is used for matching the signal output by the forming filter in the transmitting end of the broadband satellite transmission system, outputting a sampling signal and realizing the optimal receiving of the signal; when the roll-off coefficient of the sending end is changed, the matched filter changes the tap coefficient in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
the square timing error estimation module is used for carrying out error estimation by adopting a square timing error estimation algorithm according to the sampling signal output by the matched filter and outputting an error estimation result;
a Moeneclaey timing error estimation module for estimating the optimal sampling point according to the output of the error compensation filter
Figure 253370DEST_PATH_IMAGE003
Carrying out error estimation by adopting a Moeneclaey algorithm, and outputting an error estimation result;
the error estimation selection module is used for selecting a corresponding error estimation result according to the roll-off coefficient of the broadband satellite transmission system sending end and transmitting the error estimation result to the error processing module; coefficient of roll off
Figure 477678DEST_PATH_IMAGE002
When the error estimation result output by the square timing error estimation module is adopted, the roll-off coefficient is used as
Figure 539175DEST_PATH_IMAGE001
Then, an error estimation result output by a Moeneclaey timing error estimation module is adopted;
the error processing module is used for sequentially carrying out loop filtering and timing time-controlled oscillator processing on the error estimation result and outputting a processed error signal;
the error compensation filter is used for receiving the sampling signal output by the matched filter, carrying out timing compensation on the sampling signal according to the error signal output by the error processing module and outputting the optimal sampling point
Figure 761209DEST_PATH_IMAGE003
And completing the timing recovery of the receiving end.
Furthermore, the matched filter comprises a plurality of sub-filtering modules, each sub-filtering module is provided with an input signal and outputs a sampling signal, the input signal of each sub-filtering module forms a multi-path parallel input signal of the matched filter, the output signal of each sub-filtering module forms a multi-path parallel sampling signal output by the matched filter, and when the roll-off coefficient of a sending end is changed, the tap coefficient of the sub-filtering module is changed in a table look-up mode according to the roll-off coefficient.
Furthermore, the matched filter comprises four sub-filtering modules, and the square timing error estimation module receives four parallel sampling signals output by the matched filter and adopts a parallel square timing error estimation algorithm to carry out error estimation; the parallel square timing error estimation algorithm comprises the steps of:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
and dividing the real part signal by the imaginary part signal, and obtaining the arc angle through arc tangent to obtain an estimated value of the timing error.
Further, the error compensation filter is an interpolation filter, and the processing procedure of the error compensation filter includes the following steps:
using the corresponding input signal for the interpolation filter according to the error estimation result selected by the error estimation selection module; specifically, if the error estimation selection module selects the error estimation result output by the moencleay timing error estimation module, the input signal of the interpolation filter is the multi-path parallel sampling signal output by the matched filter, and if the error estimation selection module selects the error estimation result output by the square timing error estimation module, the input signal of the interpolation filter is the effective signal part of the multi-path parallel sampling signal after the burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals output by the error processing module;
selecting the tap coefficient of the interpolation filter by table look-up method according to the error signal, and recovering the optimal sampling point by parallel resampling
Figure 173736DEST_PATH_IMAGE003
。
Further, the moencleay algorithm comprises the following steps:
single-pass optimum sampling value of interpolation filter output using error estimation as 0
Figure 391615DEST_PATH_IMAGE003
As the initial input signal of the Moeneclaey algorithm, the input signal of the subsequent Moeneclaey algorithm is the adjusted optimal sampling value output by the interpolation filter
Figure 573198DEST_PATH_IMAGE003
；
Obtaining an output error estimation signal of a Moeneclaey algorithm according to the following formula
Figure 700554DEST_PATH_IMAGE004
：
Figure 600377DEST_PATH_IMAGE005
，
Wherein the content of the first and second substances,
Figure 166487DEST_PATH_IMAGE006
is that
Figure 655106DEST_PATH_IMAGE003
The signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
Compared with the background technology, the invention has the following advantages:
(1) the invention adjusts the roll-off coefficient of the matched filter according to the signaling information according to the characteristics of the broadband satellite self-adaptive transmission system with variable roll-off coefficient, thereby realizing the optimal receiving.
(2) The invention is based on the roll-off coefficient
Figure DEST_PATH_IMAGE007
Different timing algorithms are adopted in the range segmentation, so that the stable work of the system can be ensured.
Drawings
Fig. 1 is a block diagram of a timing recovery apparatus according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a sampled signal after burst processing in an embodiment of the invention.
Detailed Description
The invention is further described with reference to the following figures and detailed description.
A timing recovery method for a broadband satellite transmission system with a variable roll-off coefficient is applied to a receiving end of the broadband satellite transmission system and used for timing recovery of a signal received by the receiving end from a transmitting end of the broadband satellite transmission system, and comprises the following steps:
(1) matching signals output by a forming filter in a transmitting end of the broadband satellite transmission system, and outputting sampling signals to realize optimal receiving of the signals; when the roll-off coefficient of the sending end is changed, the tap coefficient is changed in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
(2) adopting a corresponding error estimation mode according to the roll-off coefficient of the sending end when the roll-off coefficient
Figure 953364DEST_PATH_IMAGE002
According to the sampling signal output by matched filter, adopting square timing error estimation algorithm to make error estimation, when roll-off coefficient is used
Figure 340483DEST_PATH_IMAGE001
According to the optimal sampling point output by the error compensation filter
Figure 897235DEST_PATH_IMAGE003
Carrying out error estimation by adopting a Moeneclaey algorithm;
(3) sequentially performing loop filtering and timing time-controlled oscillator processing on the error estimation result of the step (2), and outputting a processed error signal;
(4) carrying out timing compensation on the sampling signal output in the step (1) according to the error signal obtained in the step (3), and outputting the optimal sampling point
Figure 787830DEST_PATH_IMAGE003
And completing the timing recovery of the receiving end.
Furthermore, in the step (1), matching is performed in a parallel manner, and a plurality of paths of parallel sampling signals are output, and in each parallel process, the tap coefficient of the process is set in a table look-up manner according to the roll-off coefficient.
Further, in the step (1), there are four parallel processes and four parallel sampling signals are output, and the square timing error estimation algorithm in the step (2) is a parallel square timing error estimation algorithm, and the parallel square timing error estimation algorithm includes the following steps:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
and dividing the real part signal by the imaginary part signal, and obtaining the arc angle through arc tangent to obtain an estimated value of the timing error.
Further, the step (4) of performing timing compensation on the sampling signal output in the step (1) by using an interpolation filter specifically includes the following steps:
adopting corresponding input signals for the interpolation filter according to the error estimation mode selected in the step (2); specifically, if the Moeneclose algorithm is selected for error estimation, the input signal is the multi-path parallel sampling signal output in the step (1), and if the square timing error estimation algorithm is selected for error estimation, the input signal is the effective signal part of the multi-path parallel sampling signal after burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals obtained in the step (3);
selecting the tap coefficient of the interpolation filter by table look-up method according to the error signal, and recovering the optimal sampling point by parallel resampling
Figure 522568DEST_PATH_IMAGE003
。
Further, the moencleay algorithm comprises the following steps:
single-pass optimum sampling value of interpolation filter output using error estimation as 0
Figure 131404DEST_PATH_IMAGE003
As the initial input signal of the Moeneclaey algorithm, the input signal of the subsequent Moeneclaey algorithm is the adjusted optimal sampling value output by the interpolation filter
Figure 491847DEST_PATH_IMAGE003
；
Obtaining an output error estimation signal of a Moeneclaey algorithm according to the following formula
Figure 502528DEST_PATH_IMAGE004
：
Figure 470484DEST_PATH_IMAGE005
，
Wherein the content of the first and second substances,
Figure 183926DEST_PATH_IMAGE006
is that
Figure 895530DEST_PATH_IMAGE003
The signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
As shown in fig. 1, a timing recovery apparatus for a wideband satellite transmission system with a variable roll-off coefficient is used to perform timing recovery on a signal received by a receiving end of the wideband satellite transmission system and from a transmitting end of the wideband satellite transmission system, and includes a matched filter, a square timing error estimation module, a moencleay timing error estimation module, an error estimation selection module, an error processing module, and an error compensation filter, where:
the matched filter is used for matching the signal output by the forming filter in the transmitting end of the broadband satellite transmission system, outputting a sampling signal and realizing the optimal receiving of the signal; when the roll-off coefficient of the sending end is changed, the matched filter changes the tap coefficient in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
the square timing error estimation module is used for carrying out error estimation by adopting a square timing error estimation algorithm according to the sampling signal output by the matched filter and outputting an error estimation result;
a Moeneclaey timing error estimation module for estimating the optimal sampling point according to the output of the error compensation filter
Figure 698401DEST_PATH_IMAGE003
Carrying out error estimation by adopting a Moeneclaey algorithm, and outputting an error estimation result;
the error estimation selection module is used for selecting a corresponding error estimation result according to the roll-off coefficient of the broadband satellite transmission system sending end and transmitting the error estimation result to the error processing module; coefficient of roll off
Figure 837258DEST_PATH_IMAGE002
When the error estimation result output by the square timing error estimation module is adopted, the roll-off coefficient is used as
Figure 607637DEST_PATH_IMAGE001
Then, an error estimation result output by a Moeneclaey timing error estimation module is adopted;
the error processing module is used for sequentially carrying out loop filtering and timing time-controlled oscillator processing on the error estimation result and outputting a processed error signal;
the error compensation filter is used for receiving the sampling signal output by the matched filter, carrying out timing compensation on the sampling signal according to the error signal output by the error processing module and outputting the optimal sampling point
Figure 857353DEST_PATH_IMAGE003
And completing the timing recovery of the receiving end.
Furthermore, the matched filter comprises a plurality of sub-filtering modules, each sub-filtering module is provided with an input signal and outputs a sampling signal, the input signal of each sub-filtering module forms a multi-path parallel input signal of the matched filter, the output signal of each sub-filtering module forms a multi-path parallel sampling signal output by the matched filter, and when the roll-off coefficient of a sending end is changed, the tap coefficient of the sub-filtering module is changed in a table look-up mode according to the roll-off coefficient.
Furthermore, the matched filter comprises four sub-filtering modules, and the square timing error estimation module receives four parallel sampling signals output by the matched filter and adopts a parallel square timing error estimation algorithm to carry out error estimation; the parallel square timing error estimation algorithm comprises the steps of:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
and dividing the real part signal by the imaginary part signal, and obtaining the arc angle through arc tangent to obtain an estimated value of the timing error.
Further, the error compensation filter is an interpolation filter, and the processing procedure of the error compensation filter includes the following steps:
using the corresponding input signal for the interpolation filter according to the error estimation result selected by the error estimation selection module; specifically, if the error estimation selection module selects the error estimation result output by the moencleay timing error estimation module, the input signal of the interpolation filter is the multi-path parallel sampling signal output by the matched filter, and if the error estimation selection module selects the error estimation result output by the square timing error estimation module, the input signal of the interpolation filter is the effective signal part of the multi-path parallel sampling signal after the burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals output by the error processing module;
selecting the tap coefficient of the interpolation filter by table look-up method according to the error signal, and recovering the optimal sampling point by parallel resampling
Figure 780309DEST_PATH_IMAGE003
。
Further, the moencleay algorithm comprises the following steps:
single-pass optimum sampling value of interpolation filter output using error estimation as 0
Figure 90068DEST_PATH_IMAGE003
As the initial input signal of the Moeneclaey algorithm, the input signal of the subsequent Moeneclaey algorithm is the adjusted optimal sampling value output by the interpolation filter
Figure 347743DEST_PATH_IMAGE003
；
Obtaining an output error estimation signal of a Moeneclaey algorithm according to the following formula
Figure 338833DEST_PATH_IMAGE004
：
Figure 178613DEST_PATH_IMAGE005
，
Wherein the content of the first and second substances,
Figure 846223DEST_PATH_IMAGE006
is that
Figure 138664DEST_PATH_IMAGE003
The signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
A more specific example is as follows.
A timing recovery method of a broadband satellite adaptive transmission system with a variable roll-off coefficient comprises the following steps:
(1) firstly, according to the roll-off coefficient of the adopted shaping filter, correspondingly adjusting the roll-off coefficient of the matched filter;
the carrier rate of broadband transmission is high, 4-time sampling is adopted by a receiving end DA, and tap coefficients of the 4-time sampling matched filter are all unified to 128 orders in comprehensive consideration of the low roll-off transmission characteristic. When the roll-off coefficients of the received data frame are adjusted, the 128 tap coefficients of the matched filter generated thereby are also adjusted accordingly.
(2) Judging which timing error estimation algorithm is adopted by the signal output by the matched filter according to the roll-off coefficient;
coefficient of roll off
Figure 933445DEST_PATH_IMAGE002
A parallel square timing error estimation algorithm is used. As shown in FIG. 1, the input signal of the parallel square timing error estimation algorithm is a four-way parallel signal P output by a variable matched filterk 、Pk+1、Pk+2、Pk+3The phase of the spectral component can be obtained by performing Discrete Fourier Transform (DFT) on the modulo-square-sum sequence of the four parallel signals, and thus the timing error.
Coefficient of roll off
Figure 627732DEST_PATH_IMAGE001
Then, Moeneclaey's algorithm is adopted, and as shown in figure 1, the one-way signal output by the interpolation filter is taken
Figure 469173DEST_PATH_IMAGE003
And
Figure 514489DEST_PATH_IMAGE003
one-way signal delayed by one symbol period
Figure 847382DEST_PATH_IMAGE006
As input signal for the error algorithm, a timing error e is calculatedk。
The parallel square timing algorithm can accurately obtain timing error estimation and roll-off coefficient in low signal-to-noise ratio and high-order modulation mode
Figure 583126DEST_PATH_IMAGE008
And
Figure DEST_PATH_IMAGE009
there is little difference in time from the performance loss, but the parallel square timing algorithm is at the roll-off factor
Figure 671167DEST_PATH_IMAGE010
In time, the estimation error of the timing error can be caused, and the normal work of the system is seriously influenced. The Moeneclaey algorithm works stably when the roll-off coefficient is as low as 0.05, but the roll-off coefficient
Figure 875884DEST_PATH_IMAGE002
The moencleary algorithm performs worse than the parallel square timing algorithm. Therefore, the present invention uses the roll-off coefficient
Figure 74784DEST_PATH_IMAGE008
As a boundary, the parallel square timing algorithm or the moencleay algorithm is thus selected.
(3) The estimation errors obtained by adopting different timing error estimation algorithms adopt the same loop filter, timing NCO and interpolation filter, the parameters of the loop filter are adjusted, and the optimal sampling value y is interpolated in 4 times of sampling pointsk。
The timing error estimation adopts different algorithms, but the timing error adjusting part is shared, and although the Moeneclaey algorithm adopts the output of an interpolation filter, the adjusting part is still the four-way parallel signal P which is 4 times of the sampling of the matched filter outputk、Pk+1、Pk+2、Pk+3. The estimation errors of different algorithms pass through the same loop filter and NCO to control the coefficient of an interpolation filter, and the optimal sampling value y is obtained through calculationkAnd completing timing synchronization. And the adjustment parameters of the loop filter corresponding to different algorithms are different.
In the above embodiment, when the roll-off coefficient is
Figure 665034DEST_PATH_IMAGE002
And meanwhile, a square timing error estimation algorithm is adopted, the algorithm can be suitable for a high-order modulation mode, a parallel timing algorithm is adopted for a parallel matched filter, and 4 paths of sampling signals work in parallel. As shown in FIG. 1, the input signal of the parallel square timing error estimation algorithm is a four-way parallel signal P output by a variable matched filterk 、Pk+1、Pk+2、Pk+3The phase of the spectral component can be obtained by performing Discrete Fourier Transform (DFT) on the modulo-square-sum sequence of the four parallel signals, and the timing error e can be obtainedk。
Coefficient of roll off
Figure 923977DEST_PATH_IMAGE001
And in the process, a Moeneclaey algorithm is adopted, the algorithm is also suitable for a high-order modulation mode, each symbol only needs one sampling point, a single-path signal output by the interpolation filter is taken, and the timing error is calculated through serial work.
The original moencleary algorithm uses the following formula for error estimation:
Figure DEST_PATH_IMAGE011
，
the method is more suitable for a QPSK modulation mode, removes amplitude influence on the basis of the original Moeneclaey algorithm, and changes an error algorithm into a QPSK modulation mode
Figure 615990DEST_PATH_IMAGE012
(
Figure DEST_PATH_IMAGE013
) Therefore, the method is more suitable for a high-order modulation mode.
According to the requirement of timely adjusting the bandwidth of a satellite self-adaptive transmission system, the roll-off coefficient of a square root raised cosine shaping filter at a sending end is adjusted, and the smaller the roll-off coefficient is, the less the occupied bandwidth is; the higher the carrier rate, the greater the roll-off factor has an effect on the bandwidth. The low roll-off factor shaping can increase the self-noise of the system, cause overlarge timing jitter, influence the demodulation performance and is easy to realizeCausing loss of mains supply. Existing general timing algorithm is in roll-off coefficient
Figure 539952DEST_PATH_IMAGE001
When the system performance is poor, the smaller the roll-off coefficient is, the worse the performance is; the Moeneclaey algorithm is not influenced by self-noise and can work in an application system with a low roll-off coefficient, but the roll-off coefficient is larger, particularly for a high-order modulation mode, when the roll-off coefficient is larger
Figure 1020DEST_PATH_IMAGE002
The system performance will get worse. Therefore, a plurality of different high-order modulation modes and roll-off coefficients are adopted according to the broadband satellite adaptive transmission system
Figure 430865DEST_PATH_IMAGE014
And the matched filter adjusts the tap coefficient according to the roll-off coefficient of the current frame indicated by the signaling to realize the optimal receiving. The timing algorithm also makes a selection based on the roll-off factor: coefficient of roll off
Figure 596791DEST_PATH_IMAGE002
The square timing algorithm is adopted, and the roll-off coefficient is measured
Figure 137494DEST_PATH_IMAGE001
The Moeneclaey algorithm is adopted.

Claims (2)

1. A timing recovery method of a broadband satellite transmission system with a variable roll-off coefficient is characterized in that the method is applied to a receiving end of the broadband satellite transmission system and used for timing recovery of a signal received by the receiving end and from a transmitting end of the broadband satellite transmission system, and comprises the following steps:
(1) matching signals output by a forming filter in a transmitting end of the broadband satellite transmission system, and outputting sampling signals to realize optimal receiving of the signals; when the roll-off coefficient of the sending end is changed, the tap coefficient is changed in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
(2) according toThe roll-off coefficient of the transmitting end adopts a corresponding error estimation mode, when the roll-off coefficient alpha is more than or equal to 0.2, the error estimation is carried out by adopting a square timing error estimation method according to the sampling signal output by the matched filter, and when the roll-off coefficient alpha is less than 0.2, the error compensation filter outputs the optimal sampling point y according to the error compensation filterkCarrying out error estimation by adopting a Moeneclaey method;
(3) sequentially performing loop filtering and timing time-controlled oscillator processing on the error estimation result of the step (2), and outputting a processed error signal;
(4) carrying out timing compensation on the sampling signal output in the step (1) according to the error signal obtained in the step (3), and outputting the optimal sampling point ykCompleting the timing recovery of the receiving end;
the step (1) is totally provided with four parallel processes and outputs four paths of parallel sampling signals, and a tap coefficient of each parallel process is set in a table look-up mode according to a roll-off coefficient; the square timing error estimation method in the step (2) is a parallel square timing error estimation method, and the parallel square timing error estimation method comprises the following steps:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
dividing the real part signal by the imaginary part signal, and obtaining an arc angle through arc tangent to obtain an estimated value of the timing error;
in the step (4), the interpolation filter is adopted to perform timing compensation on the sampling signal output in the step (1), and the method specifically comprises the following steps:
adopting corresponding input signals for the interpolation filter according to the error estimation mode selected in the step (2); specifically, if the Moeneclose method is selected for error estimation, the input signal is the multi-path parallel sampling signal output in the step (1), and if the square timing error estimation method is selected for error estimation, the input signal is the effective signal part of the multi-path parallel sampling signal after burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals obtained in the step (3);
selecting the tap coefficient of the interpolation filter by a table look-up mode according to the error signal, and recovering the optimal sampling point y by a parallel resampling modek；
The Moeneclaey method comprises the following steps:
single-pass optimum sampling value y of interpolation filter output using error estimation of 0kAs the initial input signal of the Moeneclaey method, the input signal of the subsequent Moeneclaey method is the adjusted optimal sampling value y output by the interpolation filterk；
The output error estimation signal e (k) of the moencleay method is obtained according to the following formula:
Figure FDA0003068358450000031
wherein, yk+1Is ykThe signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
2. The utility model provides a timing recovery device of variable broadband satellite transmission system of roll-off coefficient, which characterized in that for carry out timing recovery to the signal that the broadband satellite transmission system receiving terminal received is come from broadband satellite transmission system transmitting terminal, including matched filter, square timing error estimation module, moencleay timing error estimation module, error estimation selection module, error processing module and error compensation filter, wherein:
the matched filter is used for matching the signal output by the forming filter in the transmitting end of the broadband satellite transmission system, outputting a sampling signal and realizing the optimal receiving of the signal; when the roll-off coefficient of the sending end is changed, the matched filter changes the tap coefficient in a table look-up mode, and the corresponding change of the roll-off coefficient of the receiving end is realized;
the square timing error estimation module is used for carrying out error estimation by adopting a square timing error estimation method according to the sampling signal output by the matched filter and outputting an error estimation result;
a Moeneclaey timing error estimation module for estimating the optimal sampling point y according to the output of the error compensation filterkCarrying out error estimation by adopting a Moeneclaey method, and outputting an error estimation result;
the error estimation selection module is used for selecting a corresponding error estimation result according to the roll-off coefficient of the broadband satellite transmission system sending end and transmitting the error estimation result to the error processing module; when the roll-off coefficient alpha is larger than or equal to 0.2, adopting an error estimation result output by a square timing error estimation module, and when the roll-off coefficient alpha is smaller than 0.2, adopting an error estimation result output by a Moeneclose timing error estimation module;
the error processing module is used for sequentially carrying out loop filtering and timing time-controlled oscillator processing on the error estimation result and outputting a processed error signal;
the error compensation filter is used for receiving the sampling signal output by the matched filter, carrying out timing compensation on the sampling signal according to the error signal output by the error processing module and outputting the optimal sampling point ykCompleting the timing recovery of the receiving end;
the matched filter comprises four sub-filtering modules, each sub-filtering module is provided with an input signal and outputs a sampling signal, the input signals of the sub-filtering modules form multi-path parallel input signals of the matched filter, the output signals of the sub-filtering modules form multi-path parallel sampling signals output by the matched filter, and when the roll-off coefficient of a sending end is changed, the tap coefficient of the sub-filtering module is changed in a table look-up mode according to the roll-off coefficient;
the square timing error estimation module receives four paths of parallel sampling signals output by the matched filter and adopts a parallel square timing error estimation method to carry out error estimation; the parallel square timing error estimation method comprises the following steps:
taking the clock of each path of sampling signal as a symbol clock;
carrying out burst processing on the sampling signal, and converting the continuous signal into a burst signal in a pilot frequency part of the sampling signal;
multiplying the effective signal part in each path of sampling signal by the conjugate signal thereof, subtracting and accumulating the products of the paths 1 and 3 in parallel to be used as a real part signal, and adding and accumulating the products of the paths 2 and 4 in parallel to be used as an imaginary part signal;
dividing the real part signal by the imaginary part signal, and obtaining an arc angle through arc tangent to obtain an estimated value of the timing error;
the error compensation filter is an interpolation filter, and the processing process of the error compensation filter comprises the following steps:
using the corresponding input signal for the interpolation filter according to the error estimation result selected by the error estimation selection module; specifically, if the error estimation selection module selects the error estimation result output by the moencleay timing error estimation module, the input signal of the interpolation filter is the multi-path parallel sampling signal output by the matched filter, and if the error estimation selection module selects the error estimation result output by the square timing error estimation module, the input signal of the interpolation filter is the effective signal part of the multi-path parallel sampling signal after the burst processing;
adjusting the sequence of the multi-channel parallel signals input to the interpolation filter according to the error signals output by the error processing module;
selecting the tap coefficient of the interpolation filter by a table look-up mode according to the error signal, and recovering the optimal sampling point y by a parallel resampling modek；
The Moeneclaey method comprises the following steps:
single-pass optimum sampling value y of interpolation filter output using error estimation of 0kAs the initial input signal of the Moeneclaey method, the input signal of the subsequent Moeneclaey method is the adjusted optimal sampling value y output by the interpolation filterk；
The output error estimation signal e (k) of the moencleay method is obtained according to the following formula:
Figure FDA0003068358450000051
wherein, yk+1Is ykThe signal is delayed by one symbol clock period, and the superscript indicates taking the conjugate signal.
CN201911037523.2A 2019-10-29 2019-10-29 Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system Active CN110752870B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911037523.2A CN110752870B (en) 2019-10-29 2019-10-29 Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911037523.2A CN110752870B (en) 2019-10-29 2019-10-29 Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system

Publications (2)

Publication Number Publication Date
CN110752870A CN110752870A (en) 2020-02-04
CN110752870B true CN110752870B (en) 2021-08-31

Family

ID=69280793

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911037523.2A Active CN110752870B (en) 2019-10-29 2019-10-29 Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system

Country Status (1)

Country Link
CN (1) CN110752870B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112867135B (en) * 2020-12-31 2022-07-01 京信网络系统股份有限公司 Timing error estimation apparatus, method, medium, and communication device
CN113507324B (en) * 2021-06-17 2022-07-29 西安空间无线电技术研究所 Feedforward timing recovery method and system suitable for high-speed satellite-borne optical communication
CN113542164B (en) * 2021-07-13 2023-03-31 电子科技大学 Communication parameter estimation method and system in blind signal detection
CN115001645B (en) * 2022-06-13 2023-12-26 北京邮电大学 Clock recovery method, clock recovery device, electronic equipment and computer storage medium

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101217355A (en) * 2008-01-11 2008-07-09 清华大学 An iterative timing synchronous method based on low density parity check coding soft decision information
CN108471347A (en) * 2018-02-11 2018-08-31 杭州电子科技大学 A kind of parallel time synchronization method based on accurate loop filtering

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5646609B2 (en) * 2009-05-29 2014-12-24 トムソン ライセンシングＴｈｏｍｓｏｎ Ｌｉｃｅｎｓｉｎｇ Method and apparatus for symbol timing recovery
CN102571033B (en) * 2012-02-01 2014-12-03 成都久鑫电子科技有限公司 Method for estimating forming-filter roll-off coefficient
US8903028B2 (en) * 2012-09-20 2014-12-02 Novelsat Ltd. Timing recovery for low roll-off factor signals
CN103457680B (en) * 2013-08-20 2016-05-11 重庆邮电大学 Timing Synchronization error detection method based on digital reception in satellite communication
DE112015003372T5 (en) * 2014-07-22 2017-05-11 Deutsches Zentrum für Luft- und Raumfahrt e.V. A method of receiving a digital signal in a digital communication system
CN105024799B (en) * 2015-06-19 2018-04-27 北京遥测技术研究所 A kind of band limit timing restoration methods based on p rank squares

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101217355A (en) * 2008-01-11 2008-07-09 清华大学 An iterative timing synchronous method based on low density parity check coding soft decision information
CN108471347A (en) * 2018-02-11 2018-08-31 杭州电子科技大学 A kind of parallel time synchronization method based on accurate loop filtering

Also Published As

Publication number Publication date
CN110752870A (en) 2020-02-04

Similar Documents

Publication Publication Date Title
CN110752870B (en) Timing recovery method and device for roll-off coefficient variable broadband satellite transmission system
CN102546484B (en) Signal channel training method and signal channel training receiver device based on beacon frame
CN111212007A (en) Universal 600Mbps intermediate-speed demodulator implementation method and modem
JP4614829B2 (en) Communication apparatus and communication method
JPH0983590A (en) Demodulator
US9621381B2 (en) Communication apparatus and method for high efficiency satellite service
CN111194077B (en) Timing synchronization method under low sampling rate
WO2012121151A1 (en) On-the-fly compensation of sampling frequency and phase offset at side of receiver executing ultra high-speed wireless communication
US20110013685A1 (en) Channel and Equalization Coefficient Based Timing Recovery
CN111600823B (en) Parallel OQPSK offset quadriphase shift keying demodulator
WO2014194940A1 (en) Coherent optical receiver
CN106603217B (en) Sampling frequency offset suppression method for Bluetooth signal of wireless comprehensive tester
CN108494467B (en) Physical layer self-adaptive ACM synchronization equipment based on satellite communication
CN107135175A (en) A kind of sync equalizing method in high-speed digital transmission system
KR20070110325A (en) Method and system for channel equalization
US10608846B2 (en) Receiving device
CN110401609B (en) Gardner symbol timing recovery method and device for accelerating convergence
CN112468281B (en) High-precision symbol synchronization system
JP3576420B2 (en) Frequency offset correction device
US6993099B2 (en) Communications receiver architectures and algorithms permitting hardware adjustments for optimizing performance
CN107948111B (en) Sampling frequency offset correction method of OFDM system
CN114465691A (en) Low-complexity constant envelope phase modulation signal sampling deviation estimation and compensation method and system
JP2007201729A (en) Adaptive equalizer and receiver
WO2012132185A1 (en) Transmission-reception device
CN104052692A (en) Data Signal Correction Circuit, Receiver, And Data Signal Correction Method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant