CN109547370A - The super Nyquist system symbol estimation method that joint equalization and interference are eliminated - Google Patents
The super Nyquist system symbol estimation method that joint equalization and interference are eliminated Download PDFInfo
- Publication number
- CN109547370A CN109547370A CN201910013077.5A CN201910013077A CN109547370A CN 109547370 A CN109547370 A CN 109547370A CN 201910013077 A CN201910013077 A CN 201910013077A CN 109547370 A CN109547370 A CN 109547370A
- Authority
- CN
- China
- Prior art keywords
- symbol
- super nyquist
- nyquist system
- iteration
- intersymbol interference
- 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.)
- Granted
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L25/03012—Arrangements for removing intersymbol interference operating in the time domain
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L25/00—Baseband systems
- H04L25/02—Details ; arrangements for supplying electrical power along data transmission lines
- H04L25/03—Shaping networks in transmitter or receiver, e.g. adaptive shaping networks
- H04L25/03006—Arrangements for removing intersymbol interference
- H04L2025/03433—Arrangements for removing intersymbol interference characterised by equaliser structure
- H04L2025/03439—Fixed structures
- H04L2025/03445—Time domain
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Cable Transmission Systems, Equalization Of Radio And Reduction Of Echo (AREA)
- Digital Transmission Methods That Use Modulated Carrier Waves (AREA)
Abstract
The invention discloses the super Nyquist system symbol estimation method that a kind of joint equalization and interference are eliminated, the steps include: to obtain input symbol；Training balanced device；Judge whether equalizer training number is less than balanced device threshold value；Symbol after calculating is balanced；Calculate the interference excretion factor of super Nyquist system；First time iteration elimination intersymbol interference；Pass through iteration elimination intersymbol interference；Judge whether the number of iterations is less than estimation threshold value；Obtain final estimate symbol.The present invention can be used for eliminating the intersymbol interference in super Nyquist system, and then estimate to send symbol, compared with art methods, even if the present invention in more serious intersymbol interference is it is also possible that super Nyquist system has better performance of BER, and sign estimation complexity of the present invention is low, more practicability.
Description
Technical field
The invention belongs to field of communication technology, further relate to one of wireless communication technology field joint equalization and
Interfere the super Nyquist system symbol estimation method eliminated.It is dry that the present invention can be used for eliminating the intersymbol in super Nyquist system
It disturbs, and then estimates the transmission symbol of super Nyquist system transmitter.
Background technique
When designing legacy communications system, in order to avoid the intersymbol interference of system, communication system follows Nyquist
One criterion.However, the orthogonality between the symbol that Nyquist Transmission system is transmitted without intersymbol interference is to sacrifice spectrum efficiency
For cost.By being artificially introduced intersymbol interference, super Nyquist system can support higher transmission rate and spectrum efficiency.
Correspondingly, super Nyquist system then needs higher complexity to eliminate intersymbol interference, to estimate super Nyquist system
The transmission symbol of transmitter.
Ebrahim Bedeer is in its " A very low complexity successive symbol- that publishes thesis
By-symbol sequence estimator for faster-than-Nyquist signaling " (IEEE Access,
2017,5:7414-7422) it is proposed in a kind of based on the low complex degree symbol estimation method to retract and interference is eliminated.This method
After receiving a symbol, symbol is currently received first with the current sign estimation for receiving symbol and estimating before this, so
The several symbols in the front end for reevaluating current estimate symbol using the estimate symbol of current sign afterwards, finally using reevaluating after
Symbol estimate current sign again.This method selects low-order-modulated mode and slight intersymbol interference in super Nyquist system
In the case where can effectively eliminate the intersymbol interference of super Nyquist system, realize good performance.Deficiency existing for this method
Place is that the interference due to only needing the previous symbol for removing current reception symbol causes its estimated accuracy low, and as super Nai Kuisi
(super Nyquist accelerated factor is smaller or connects when special system selection high-order modulating or in the case of more serious intersymbol interference
Receipts machine matched filter is rolled using smaller by the factor), sign estimation performance is poor.
Ebrahim Bedeer is in its " Low-complexity detection of high-order QAM that publishes thesis
A kind of base is proposed in faster-than-Nyquist signaling " (IEEE Access, 2017,5:14579-14588)
In the symbol estimation method of semidefinite decoding, it is effectively applicable to using high-order orthogonal amplitude modulation QAM (Quadrature
Amplitude modulation) super Nyquist system, using 16-QAM super Nyquist system in show
Excellent performance of BER.Shortcoming existing for this method is that complexity and selected modulation system order are positively correlated,
When super Nyquist system modulation system higher using order, its excessively high complexity causes this method not have practical
Property.
Patent document " bidirectional linked list interference in super Nyquist communication system of the Xian Electronics Science and Technology University in its application
Proposed in removing method " (number of applying for a patent: 201810744483.4, publication number: CN108632182A) it is a kind of for it is super how
The bidirectional linked list intersymbol interference removing method of Qwest's system.This method with truncated Waveform shaping filter to sampled data into
To adding backward bidirectional linked list to interfere elimination before row, i.e., serially done before being eliminated first with the demodulation code element before current symbol
It disturbs, obtains the temporary decision values of demodulated signal, to serial interference after recycling the temporary decision values of demodulated signal to eliminate, obtain most
Whole demodulated signal.This method reduces receiving end while improving super Nyquist receiving terminal of communication system demodulation performance
Complexity, be suitable for slight intersymbol interference situation, however this method is difficult to approximation theory performance limit, it is especially dry in severe inter-symbol
Performance loss is larger in the case of disturbing.
Summary of the invention
It is an object of the invention to be directed to the deficiency of above-mentioned prior art, it is super to propose that a kind of joint equalization is eliminated with interference
Nyquist system symbol estimation method.
Realizing the thinking of the object of the invention is, in more serious intersymbol interference, is using iteration elimination intersymbol interference
When, since cause can not be real successive iterations for super Nyquist system performance of BER difference after first time iterative estimate
Now preferable performance, therefore can first carry out an equalization operation after obtaining receiver matched filter output symbol and obtain
Then symbol after weighing apparatus obtains final estimate symbol by iteration elimination intersymbol interference to the symbol after equilibrium.
Specific steps of the present invention include the following:
(1) input symbol is obtained:
The symbol of real-time reception receiver matched filter output corresponding with super Nyquist system transmitter transmission symbol
Number, by the symbol of super Nyquist system receiver matched filter each moment output, as sign estimation and hair
It penetrates machine and sends the corresponding input symbol of symbol；
(2) training super Nyquist system receiver balanced device:
(2a) according to the following formula, calculates the error amount of super Nyquist system receiver balanced device:
E (k)=[d (k) d (k-1) ... d (k- ψ+1)]T-YT(k)w*
Wherein, e (k) indicates that the error amount of k-th of moment super Nyquist system receiver balanced device, k indicate each defeated
Enter the serial number that symbol corresponds to the moment, d (k) indicates the θ tap of the super Nyquist system receiver balanced device at k-th of moment
In symbol corresponding to known transmission signal,K indicates super Nyquist system receiver equalizer tap system
Several sums, value are odd number,Expression rounds up operation, and ψ indicates the data of super Nyquist system receiver balanced device
Reuse-time, T indicate transposition operation, and Y (k) indicates that the dimension at k-th of moment is the input sign matrix of K × ψ, ykIndicate kth
The input symbol at a moment, w indicate the column vector of super Nyquist system receiver equalizer tap coefficient, initial value zero
Vector, * expression take conjugate operation；
(2b) according to the following formula, calculates super Nyquist system receiver equalizer tap coefficient vector:
Wherein,Indicate that super Nyquist system receiver equalizer tap coefficient vector, μ indicate super Nyquist system
The step-length of receiver, the step-length are the decimals chosen in (0,1) range, and H expression takes conjugate transposition operation, γ table
Show that the constant chosen in (0,1) range, E indicate that dimension is the unit matrix of ψ × ψ, ()-1Indicate inversion operation；
(2c) uses super Nyquist system receiver equalizer tap coefficient vectorEach of element value update this
The element value of corresponding position in super Nyquist system receiver equalizer tap coefficient column vector w before secondary equilibrium；
(3) judge that super Nyquist system receiver balanced device obtains whether current frequency of training is less than balanced device threshold value, if
It is to execute step (2) after current frequency of training is then added 1, otherwise, executes step (4) after completing the training of balanced device；
(4) with input symbolic vector [yk yk-1…yk-K+1] arranged with super Nyquist system receiver equalizer tap coefficient
The conjugate vector of vector w is multiplied, and obtains carrying out the symbol after equilibrium to+1 moment input symbol of kth-θ
(5) according to the following formula, the interference excretion factor of super Nyquist system is calculated:
Gj=gP+jτQ(h)
Wherein, GjIndicate that j-th of interference excretion factor in super Nyquist system, j indicate the sequence of interference excretion factor
Number, the value range of j isIndicate that downward floor operation, P indicate super Nyquist system
The sum of all time domain response coefficients of middle receiver matched filter, τ indicates super Nyquist system accelerated factor, described
Accelerated factor is the decimal chosen in (0,1) range, and Q indicates receiver matched filter of uniting in super Nyquist system
Down-sampling multiple, the down-sampling multiple are the integers chosen in [2,10] range, indicate multiplication operations, g () table
Show that, from convolution operation, h indicates time domain response coefficient sum P and rolling according to receiver matched filter in super Nyquist system
The time domain response coefficient for the super Nyquist system receiver matched filter that the factor generates drops；
(6) first time iteration elimination intersymbol interference:
(6a) according to the following formula, calculates intersymbol interference when first time iteration in each moment input symbol:
Wherein, χk-θ-L+1Intersymbol interference when first time iteration in kth-θ-L+1 moment input symbol is indicated, due to wanting
Eliminate intersymbol interference caused by the symbol of front and rear sides in each symbol, what when each iteration, calculated be current input symbol it
Intersymbol interference in preceding symbol, L are indicated according to the total of the time domain response coefficient of super Nyquist system receiver matched filter
The unilateral symbol lengths of sign estimation are used in all iterative process of number P setting, value range to be Table
Show that the input symbol to the -2L+1 moment of kth-θ carries out the symbol after first time iteration；
(6b) uses the input symbol y at kth-θ-L+1 momentk-θ-L+1Subtract intersymbol interference χ when first time iterationk-θ-L+1,
It obtains eliminating the symbol beta after intersymbol interference when first time iterationk-θ-L+1；
(6c) is to the symbol beta for eliminating intersymbol interference after first time iterationk-θ-L+1Hard decision operation is carried out, is obtained to kth-θ-
The input symbol at L+1 moment carries out the symbol after first time iteration
(7) pass through iteration elimination intersymbol interference:
(7a) according to the following formula, inputs the intersymbol interference in symbol when calculating current iteration:
Wherein, χk-θ-cL+1Indicate intersymbol interference when current iteration in the input symbol at kth-θ-cL+1 moment, c is indicated
The serial number of current iteration,Indicate that the input symbol to the-cL-L+1 moment of kth-θ carries out and current iteration number
Symbol after phase homogeneous iteration,Indicate that the input symbol to the-cL+2 moment of kth-θ is carried out than current iteration time
Symbol after the few an iteration of number；
(7b) uses the input symbol y at kth-θ-cL+1 momentk-θ-cL+1Subtract intersymbol interference when current iteration
χk-θ-cL+1, intersymbol interference is eliminated, after obtaining the input symbol elimination intersymbol interference at kth-θ-cL+1 moment when current iteration
Symbol betak-θ-cL+1；
The input symbol at kth-θ-cL+1 moment eliminates the symbol beta after intersymbol interference when (7c) is to current iterationk-θ-cL+1
Hard decision operation is carried out, obtains carrying out the input symbol at-cL+1 moment of kth-θ the symbol after current iteration
(8) judge whether current iteration number is less than estimation threshold value, if so, executing step after current iteration number is added 1
Suddenly (7) have otherwise eliminated the intersymbol interference in each input symbol, and terminal estimation procedure executes step (9)；
(9) final estimate symbol is obtained:
The estimate symbol after iteration will be terminated, sends the corresponding estimate symbol of symbol as with transmitter, complete it is super how
The symbol estimation procedure of Qwest's system.
Compared with the prior art, the present invention has the following advantages:
First, since the invention firstly uses balanced devices to carry out equilibrium, iteration elimination code then is passed through to symbol after equilibrium
Between interfere, overcome the prior art when super Nyquist system select high-order modulating when or more serious intersymbol interference situation
Under, the problem of sign estimation performance difference so that the present invention has higher estimated accuracy, more can accurately estimate it is super how Kui
The transmitting symbol of this special system, is particularly suitable for using the super Nai Kuisi in the case of high-order modulating, more serious intersymbol interference
Special system.
Second, since the present invention is on the basis of equilibrium, by iteration elimination intersymbol interference, so that intersymbol interference is eliminated
It is unrelated with modulation system, and balanced device complexity is lower, the complexity and modulation system for overcoming the prior art are positively correlated, and are caused
The excessively high problem of complexity when super Nyquist system selects high-order modulating, even if so that the present invention is using high contrast
Also its intersymbol interference can be eliminated with lower complexity in the super Nyquist system of mode processed, there is stronger practicability.
Detailed description of the invention
Fig. 1 is flow chart of the invention；
Fig. 2 is simulation result diagram of the invention.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawing.
Referring to attached drawing 1, step of the invention is described in further detail.
Step 1, input symbol is obtained.
The symbol of real-time reception receiver matched filter output corresponding with super Nyquist system transmitter transmission symbol
Number, by the symbol of super Nyquist system receiver matched filter each moment output, as sign estimation and hair
It penetrates machine and sends the corresponding input symbol of symbol.
The transmission symbol is the symbol generated according to the planisphere of super Nyquist system, the super Nyquist
The planisphere of system refers to, by the distribution maps for sending symbol and forming all after the modulation of super Nyquist system transmitter constellation.
Step 2, training super Nyquist system receiver balanced device.
According to the following formula, the error amount of super Nyquist system receiver balanced device is calculated:
E (k)=[d (k) d (k-1) ... d (k- ψ+1)]T-YT(k)w*
Wherein, e (k) indicates that the error amount of k-th of moment super Nyquist system receiver balanced device, k indicate each defeated
Enter the serial number that symbol corresponds to the moment, d (k) indicates the θ tap of the super Nyquist system receiver balanced device at k-th of moment
In symbol corresponding to known transmission signal,K indicates super Nyquist system receiver equalizer tap system
Several sums, value are odd number,Expression rounds up operation, and ψ indicates the data of super Nyquist system receiver balanced device
Reuse-time, T indicate transposition operation,The dimension for indicating k-th of moment is the defeated of K × ψ
Enter sign matrix, ykIndicate that the input symbol at k-th of moment, w indicate super Nyquist system receiver equalizer tap coefficient
Column vector, initial value be null vector, * expression take conjugate operation.
According to the following formula, super Nyquist system receiver equalizer tap coefficient vector is calculated:
Wherein,Indicate that super Nyquist system receiver equalizer tap coefficient vector, μ indicate super Nyquist system
The step-length of receiver, the step-length are the decimals chosen in (0,1) range, and H expression takes conjugate transposition operation, γ table
Show that the constant chosen in (0,1) range, E indicate that dimension is the unit matrix of ψ × ψ, ()-1Indicate inversion operation.
With super Nyquist system receiver equalizer tap coefficient vectorEach of element value update this
The element value of corresponding position in super Nyquist system receiver equalizer tap coefficient column vector w before weighing apparatus.
Step 3, judge that super Nyquist system receiver balanced device obtains whether current frequency of training is less than balanced device threshold value,
If so, executing step 2 after current frequency of training is added 1, otherwise, step 4 is executed after completing the training of balanced device.
The balanced device threshold value refers to that super Nyquist system receiver balanced device terminates parameter when training, takes
Value are as follows: when the accelerated factor of super Nyquist system be less than or equal to 0.8 or the roll-off factor of receiver matched filter be less than
When equal to 0.3, which takes 400000；When the accelerated factor of super Nyquist system is more than or equal to 0.9 or receiver
When the roll-off factor of matched filter is more than or equal to 0.5, training threshold value takes 10000；Remaining situation, training threshold value take 200000.
Step 4, with input symbolic vector [yk yk-1…yk-K+1] and super Nyquist system receiver equalizer tap system
The conjugate vector of ordered series of numbers vector w is multiplied, and obtains carrying out the symbol after equilibrium to+1 moment input symbol of kth-θ
Step 5, according to the following formula, the interference excretion factor of super Nyquist system is calculated.
Gj=gP+jτQ(h)
Wherein, GjIndicate that j-th of interference excretion factor in super Nyquist system, j indicate the sequence of interference excretion factor
Number, the value range of j isIndicate that downward floor operation, P indicate in super Nyquist system
The sum of all time domain response coefficients of receiver matched filter, τ indicate super Nyquist system accelerated factor, and described adds
The fast factor is the decimal chosen in (0,1) range, and Q indicates to unite under receiver matched filter in super Nyquist system
Multiple is sampled, the down-sampling multiple is the integer chosen in [2,10] range, indicates multiplication operations, and g () is indicated
From convolution operation, h indicates the time domain response coefficient sum P according to receiver matched filter in super Nyquist system and roll-offs
The time domain response coefficient for the super Nyquist system receiver matched filter that the factor generates.
Step 6, first time iteration elimination intersymbol interference.
According to the following formula, intersymbol interference when first time iteration in each moment input symbol is calculated:
Wherein, χk-θ-L+1Intersymbol interference when first time iteration in kth-θ-L+1 moment input symbol is indicated, due to wanting
Eliminate intersymbol interference caused by the symbol of front and rear sides in each symbol, what when each iteration, calculated be current input symbol it
Intersymbol interference in preceding symbol, L are indicated according to the total of the time domain response coefficient of super Nyquist system receiver matched filter
The unilateral symbol lengths of sign estimation are used in all iterative process of number P setting, value range to be Table
Show that the input symbol to the -2L+1 moment of kth-θ carries out the symbol after first time iteration.
With the input symbol y at-L+1 moment of kth-θk-θ-L+1Subtract intersymbol interference χ when first time iterationk-θ-L+1, obtain
The symbol beta after intersymbol interference is eliminated when first time iterationk-θ-L+1。
To the symbol beta for eliminating intersymbol interference after first time iterationk-θ-L+1Hard decision operation is carried out, is obtained to kth-θ-L+1
The input symbol at a moment carries out the symbol after first time iteration
The step of hard decision operates is as follows:
Step 1 calculates the distance between each symbol and symbol after elimination intersymbol interference in planisphere according to the following formula:
Wherein, κiIndicate the distance between i-th of symbol and symbol after elimination intersymbol interference in planisphere, i indicates constellation
The serial number of symbol in figure, value range are [1, ν], and ν indicates the total number of distinct symbols in planisphere, siIt indicates in planisphere
I-th of symbol, β indicate symbol after elimination intersymbol interference,Indicate that sqrt operation, * expression take conjugate operation.
Step 2 is chosen minimum from all distances after symbol each in planisphere and elimination intersymbol interference between symbol
Value, using the symbol in planisphere corresponding to the minimum value as the symbol after hard decision.
Step 7, pass through iteration elimination intersymbol interference.
According to the following formula, intersymbol interference when calculating current iteration in input symbol:
Wherein, χk-θ-cL+1Indicate intersymbol interference when current iteration in the input symbol at kth-θ-cL+1 moment, c is indicated
The serial number of current iteration,Indicate that the input symbol to the-cL-L+1 moment of kth-θ carries out and current iteration number
Symbol after phase homogeneous iteration,Indicate that the input symbol to the-cL+2 moment of kth-θ is carried out than current iteration time
Symbol after the few an iteration of number.
With the input symbol y at-cL+1 moment of kth-θk-θ-cL+1Subtract intersymbol interference χ when current iterationk-θ-cL+1, disappear
Except intersymbol interference, the input symbol for obtaining kth-θ-cL+1 moment when current iteration eliminates the symbol after intersymbol interference
βk-θ-cL+1。
The input symbol at kth-θ-cL+1 moment eliminates the symbol beta after intersymbol interference when to current iterationk-θ-cL+1It carries out
Hard decision operation obtains carrying out the input symbol at-cL+1 moment of kth-θ the symbol after current iteration
The step of hard decision operates is as follows:
Step 1 calculates the distance between each symbol and symbol after elimination intersymbol interference in planisphere according to the following formula:
Wherein, κiIndicate the distance between i-th of symbol and symbol after elimination intersymbol interference in planisphere, i indicates constellation
The serial number of symbol in figure, value range are [1, ν], and ν indicates the total number of distinct symbols in planisphere, siIt indicates in planisphere
I-th of symbol, β indicate symbol after elimination intersymbol interference,Indicate that sqrt operation, * expression take conjugate operation.
Step 2 is chosen minimum from all distances after symbol each in planisphere and elimination intersymbol interference between symbol
Value, using the symbol in planisphere corresponding to the minimum value as the symbol after hard decision.
Step 8, judge whether current iteration number is less than estimation threshold value, if so, executing after current iteration number is added 1
Step 7, otherwise, the intersymbol interference in each input symbol has been eliminated, terminal estimation procedure executes step 9.
The estimation threshold value refer to terminate iteration when parameter, value are as follows: when super Nyquist system acceleration because
For son less than 0.8 and when the roll-off factor of receiver matched filter is less than 0.3, which takes 8；When super Nyquist system
Accelerated factor be more than or equal to 0.9 or receiver matched filter roll-off factor be more than or equal to 0.3 when, iteration threshold takes 3；
Remaining situation, iteration threshold take 6.
Step 9, final estimate symbol is obtained.
The estimate symbol after iteration will be terminated, sends the corresponding estimate symbol of symbol as with transmitter, complete it is super how
The symbol estimation procedure of Qwest's system.
The transmission symbol is the symbol generated according to the planisphere of super Nyquist system, the super Nyquist
The planisphere of system refers to, by the distribution maps for sending symbol and forming all after the modulation of super Nyquist system transmitter constellation.
Effect of the invention is further illustrated below with reference to emulation experiment.
1. simulated conditions:
Emulation experiment of the invention is carried out under MATLAB 2017B software.In emulation experiment of the invention, surpass
All time domain response coefficient sums of receiver matched filter are 161 in Nyquist system and its down-sampling multiple is 10.
In the emulation of the present invention " the super Nyquist system symbol estimation method of joint equalization and interference elimination ", super Nyquist system
Accelerated factor of uniting is 0.8, and receiver matched filter roll-off factor is 0.3 in super Nyquist system.In simulation process, surpass how
Qwest's system noise type is white Gaussian noise.
2. emulation content and interpretation of result:
Emulation experiment is using the present invention and two kinds of existing methods (frequency-domain equilibrium methods, the side eliminated based on rollback and interference
Method) intersymbol interference of super Nyquist system is eliminated respectively, estimates to send symbol, the emulation total bit number of individual bit signal-to-noise ratio
It is 1 × 108, super Nyquist system modulation mode selected four phase shift keying QPSK (quadrature phase shift
Keying), octal system phase-shift keying 8-PSK (8phase shift keying), 16- Amplitude phase shift keying APSK
(amplitude phase shift keying) and 32/64/128/256-APSK.
In order to verify the effect of emulation experiment, the property of the present invention and two kinds of existing methods is assessed using bit error rate curve
Energy.The acquisition methods of the bit error rate curve are as follows: comparison super Nyquist system is issued in a bit state of signal-to-noise
Send bit data corresponding to symbol after symbol and estimation, the sum of statistics wherein different bits, and divided by total bit number can be with
Emulation bit error rate of the super Nyquist system under the state of signal-to-noise is obtained, and passes through 10 different bit noises of emulation
The different bit error rate than obtaining 10, and then draw bit error rate curve, the simulation result of emulation experiment such as attached drawing 2.
Horizontal axis in Fig. 2 indicates the bit signal-to-noise ratio of super Nyquist system, and unit is decibel dB (decibel), indulges
The bit error rate of axis expression super Nyquist system.Shared in Fig. 27 solid lines, 7 with the curves of cross mark, 7 with diamond shape
The curve of mark and 7 curves with circle mark, wherein solid line indicates that the theoretical bit error rate of super Nyquist system is bent
Line, the curve are drawn according to the corresponding theoretical bit error rate of 7 kinds of modulation systems, and the theoretical bit error rate refers to from reason
By the optimal bit error rate being above derived by.It is indicated with the curve that cross indicates, is estimated using the frequency-domain equilibrium method of the prior art
The bit error rate curve of super Nyquist system when meter sends symbol.The base using the prior art is indicated with the curve that diamond shape indicates
In rollback and interference elimination method, the bit error rate curve of super Nyquist system when estimation sends symbol.With circle mark
Curve indicate use the method for the present invention, estimation send symbol when super Nyquist system bit error rate curve.
Abscissa in fixed Fig. 2, observation comparison use four phase shift keying QPSK 4 curves (solid line, with cross mark
The curve shown, the curve indicated with diamond shape and the curve indicated with circle), it is known that the situation identical in bit signal-to-noise ratio
Under, the position of point corresponding on the bit error rate curve of super Nyquist system when estimating to send symbol using the method for the present invention
Lower than use two kinds of prior arts (frequency-domain equilibrium method, based on retract and interference elimination method) estimation send symbol when it is super how
The position of corresponding point on the bit error rate curve of Qwest's system.Remaining 6 kinds are used using the observation comparison of identical method
4 curves of any one modulation system in modulation system use point corresponding on the bit error rate curve of the method for the present invention
Also it is below point corresponding on the bit error rate curve using two kinds of prior arts.This shows can be using the method for the present invention
More accurate estimation sends symbol under the more serious intersymbol interference scene of super Nyquist system, so that super Nyquist system has
Better performance of BER.
Claims (5)
1. the super Nyquist system symbol estimation method that a kind of joint equalization and interference are eliminated, which is characterized in that obtain it is super how
An equalization operation is carried out after Qwest's system receiver matched filter output symbol and is equalized rear symbol, after equilibrium
By iteration elimination intersymbol interference, the step of this method includes the following: symbol
(1) input symbol is obtained:
The symbol of real-time reception receiver matched filter output corresponding with super Nyquist system transmitter transmission symbol, will
The symbol of each moment output of super Nyquist system receiver matched filter, as sending out with transmitter for sign estimation
Send symbol corresponding input symbol；
(2) training super Nyquist system receiver balanced device:
(2a) according to the following formula, calculates the error amount of super Nyquist system receiver balanced device:
E (k)=[d (k) d (k-1) ... d (k- ψ+1)]T-YT(k)w*
Wherein, e (k) indicates that the error amount of k-th of moment super Nyquist system receiver balanced device, k indicate each input symbol
The serial number at number corresponding moment, d (k) are indicated in the θ tap of the super Nyquist system receiver balanced device at k-th of moment
Known transmission signal corresponding to symbol,K indicates super Nyquist system receiver equalizer tap coefficient
Sum, value are odd number,Expression rounds up operation, and ψ indicates the data reusing of super Nyquist system receiver balanced device
Number, T indicate transposition operation, and Y (k) indicates that the dimension at k-th of moment is the input sign matrix of K × ψ, ykWhen indicating k-th
The input symbol at quarter, w indicate super Nyquist system receiver equalizer tap coefficient column vector, initial value be zero to
Amount, * expression take conjugate operation；
(2b) according to the following formula, calculates super Nyquist system receiver equalizer tap coefficient vector:
Wherein,Indicate that super Nyquist system receiver equalizer tap coefficient vector, μ indicate that super Nyquist system receives
The step-length of machine, the step-length are the decimals chosen in (0,1) range, and H expression takes conjugate transposition operation, and γ is indicated
The constant chosen in (0,1) range, E indicate that dimension is the unit matrix of ψ × ψ, ()-1Indicate inversion operation；
(2c) uses super Nyquist system receiver equalizer tap coefficient vectorEach of element value update this equilibrium
The element value of corresponding position in preceding super Nyquist system receiver equalizer tap coefficient column vector w；
(3) judge that super Nyquist system receiver balanced device obtains whether current frequency of training is less than balanced device threshold value, if so,
Step (2) are executed after current frequency of training is added 1, otherwise, execute step (4) after completing the training of balanced device；
(4) with input symbolic vector [yk yk-1…yk-K+1] and super Nyquist system receiver equalizer tap coefficient column vector
The conjugate vector of w is multiplied, and obtains carrying out the symbol after equilibrium to+1 moment input symbol of kth-θ
(5) according to the following formula, the interference excretion factor of super Nyquist system is calculated:
Gj=gP+jτQ(h)
Wherein, GjIndicate that j-th of interference excretion factor in super Nyquist system, j indicate the serial number of interference excretion factor, j's
Value range is Indicate that downward floor operation, P indicate receiver in super Nyquist system
The sum of all time domain response coefficients of matched filter, τ indicate super Nyquist system accelerated factor, the accelerated factor
It is the decimal chosen in (0,1) range, Q indicates the down-sampling times of system receiver matched filter in super Nyquist system
Number, the down-sampling multiple are the integers chosen in [2,10] range, indicate multiplication operations, and g () is indicated from convolution
Operation, h indicate raw according to the time domain response coefficient sum P and roll-off factor of receiver matched filter in super Nyquist system
At super Nyquist system receiver matched filter time domain response coefficient；
(6) first time iteration elimination intersymbol interference:
(6a) according to the following formula, calculates intersymbol interference when first time iteration in each moment input symbol:
Wherein, χk-θ-L+1Intersymbol interference when first time iteration in kth-θ-L+1 moment input symbol is indicated, due to eliminate
Intersymbol interference caused by the symbol of front and rear sides in each symbol accords with before what when each iteration calculated is current input symbol
Intersymbol interference in number, L expression are set according to the total P of the time domain response coefficient of super Nyquist system receiver matched filter
The unilateral symbol lengths of sign estimation are used in fixed all iterative process, value range to be Expression pair
The input symbol at -2L+1 moment of kth-θ carries out the symbol after first time iteration；
(6b) uses the input symbol y at kth-θ-L+1 momentk-θ-L+1Subtract intersymbol interference χ when first time iterationk-θ-L+1, obtain
The symbol beta after intersymbol interference is eliminated when first time iterationk-θ-L+1；
(6c) is to the symbol beta for eliminating intersymbol interference after first time iterationk-θ-L+1Hard decision operation is carried out, is obtained to kth-θ-L+1
The input symbol at moment carries out the symbol after first time iteration
(7) pass through iteration elimination intersymbol interference:
(7a) according to the following formula, inputs the intersymbol interference in symbol when calculating current iteration:
Wherein, χk-θ-cL+1Indicate intersymbol interference when current iteration in the input symbol at kth-θ-cL+1 moment, c indicates current
The serial number of iteration,It is identical as current iteration number to indicate that the input symbol to the-cL-L+1 moment of kth-θ is carried out
Symbol after secondary iteration,It is fewer than current iteration number to indicate that the input symbol to the-cL+2 moment of kth-θ is carried out
Symbol after an iteration；
(7b) uses the input symbol y at kth-θ-cL+1 momentk-θ-cL+1Subtract intersymbol interference χ when current iterationk-θ-cL+1, eliminate
Intersymbol interference, the input symbol for obtaining kth-θ-cL+1 moment when current iteration eliminate the symbol beta after intersymbol interferencek-θ-cL+1；
The input symbol at kth-θ-cL+1 moment eliminates the symbol beta after intersymbol interference when (7c) is to current iterationk-θ-cL+1It carries out
Hard decision operation obtains carrying out the input symbol at-cL+1 moment of kth-θ the symbol after current iteration
(8) judge whether current iteration number is less than estimation threshold value, if so, executing step after current iteration number is added 1
(7), the intersymbol interference in each input symbol otherwise, has been eliminated, terminal estimation procedure executes step (9)；
(9) final estimate symbol is obtained:
Estimate symbol after termination iteration is completed into super Nai Kuisi as estimate symbol corresponding with transmitter transmission symbol
The symbol estimation procedure of special system.
2. the super Nyquist system symbol estimation method that joint equalization according to claim 1 and interference are eliminated, special
Sign is: transmission symbol described in step (1), step (9) is the symbol generated according to the planisphere of super Nyquist system,
The planisphere of the super Nyquist system refers to, by all transmission symbols after the modulation of super Nyquist system transmitter constellation
The distribution map of composition.
3. the super Nyquist system symbol estimation method that joint equalization according to claim 1 and interference are eliminated, special
Sign is: balanced device threshold value described in step (3) refers to that super Nyquist system receiver balanced device terminates ginseng when training
Number, value are as follows: when super Nyquist system accelerated factor be less than or equal to 0.8 or receiver matched filter roll-off because
When son is less than or equal to 0.3, which takes 400000；When super Nyquist system accelerated factor be more than or equal to 0.9 or
When the roll-off factor of receiver matched filter is more than or equal to 0.5, training threshold value takes 10000；Remaining situation, training threshold value take
200000。
4. the super Nyquist system symbol estimation method that joint equalization according to claim 1 and interference are eliminated, special
Sign is: the step of hard decision described in step (6c), step (7c) operates is as follows:
The first step calculates the distance between each symbol and symbol after elimination intersymbol interference in planisphere according to the following formula:
Wherein, κiIndicate the distance between i-th of symbol and symbol after elimination intersymbol interference in planisphere, i indicates to accord in planisphere
Number serial number, value range be [1, ν], ν indicate planisphere in distinct symbols total number, siIt indicates in planisphere i-th
Symbol, β indicate symbol after elimination intersymbol interference,Indicate that sqrt operation, * expression take conjugate operation；
Second step chooses minimum value from all distances after symbol each in planisphere and elimination intersymbol interference between symbol,
Using the symbol in planisphere corresponding to the minimum value as the symbol after hard decision.
5. the super Nyquist system symbol estimation method that joint equalization according to claim 1 and interference are eliminated, special
Sign is: estimation threshold value described in step (8) refers to parameter when terminating iteration, value are as follows: when super Nyquist system
Accelerated factor less than 0.8 and when the roll-off factor of receiver matched filter is less than 0.3, which takes 8；When it is super how Kui
The accelerated factor of this special system is more than or equal to 0.9 or when the roll-off factor of receiver matched filter is more than or equal to 0.3, iteration
Threshold value takes 3；Remaining situation, iteration threshold take 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910013077.5A CN109547370B (en) | 2019-01-07 | 2019-01-07 | Symbol estimation method of super-Nyquist system based on joint equalization and interference cancellation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910013077.5A CN109547370B (en) | 2019-01-07 | 2019-01-07 | Symbol estimation method of super-Nyquist system based on joint equalization and interference cancellation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109547370A true CN109547370A (en) | 2019-03-29 |
| CN109547370B CN109547370B (en) | 2020-06-05 |
Family
ID=65834455
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910013077.5A Active CN109547370B (en) | 2019-01-07 | 2019-01-07 | Symbol estimation method of super-Nyquist system based on joint equalization and interference cancellation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109547370B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110149287A (en) * | 2019-06-18 | 2019-08-20 | 西安电子科技大学 | Super Nyquist system and its symbol estimation method based on linear predictive coding |
| CN112235217A (en) * | 2020-10-16 | 2021-01-15 | 兰州理工大学 | Self-adaptive pre-equalization method based on wireless optical fiber to the home (FTN) communication system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106549892A (en) * | 2016-11-22 | 2017-03-29 | 北京理工大学 | A kind of joint time-frequency doubly selective channel is estimated and super Nyquist signal detecting method |
| US20180034591A1 (en) * | 2016-07-29 | 2018-02-01 | Electronics And Telecommunications Research Institute | Apparatus for receiving signal based on faster-than-nyquist and method for using the same |
-
2019
- 2019-01-07 CN CN201910013077.5A patent/CN109547370B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180034591A1 (en) * | 2016-07-29 | 2018-02-01 | Electronics And Telecommunications Research Institute | Apparatus for receiving signal based on faster-than-nyquist and method for using the same |
| CN106549892A (en) * | 2016-11-22 | 2017-03-29 | 北京理工大学 | A kind of joint time-frequency doubly selective channel is estimated and super Nyquist signal detecting method |
Non-Patent Citations (2)
| Title |
|---|
| EBRAHIM BEDEER；HALIM YANIKOMEROGLU；MOHAMED HOSSAM AHMED: "REDUCED COMPLEXITY OPTIMAL DETECTION OF BINARY FASTER-THAN-NTQUIST SIGNALING", <2017 IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS(ICC)> * |
| 王志峰 白勇 唐啸宇 黄梦醒: "多径信道下基于频域均衡的超奈奎斯特传输", 《高技术通讯》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110149287A (en) * | 2019-06-18 | 2019-08-20 | 西安电子科技大学 | Super Nyquist system and its symbol estimation method based on linear predictive coding |
| CN110149287B (en) * | 2019-06-18 | 2021-06-25 | 西安电子科技大学 | Linear precoding-based super-Nyquist system and symbol estimation method thereof |
| CN112235217A (en) * | 2020-10-16 | 2021-01-15 | 兰州理工大学 | Self-adaptive pre-equalization method based on wireless optical fiber to the home (FTN) communication system |
| CN112235217B (en) * | 2020-10-16 | 2023-02-03 | 兰州理工大学 | Self-adaptive pre-equalization method based on wireless optical fiber to the home (FTN) communication system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109547370B (en) | 2020-06-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113890796B (en) | High-speed channel estimation device and method based on OTFS (optical transport System) modulation and demodulation | |
| CN110266617B (en) | Multipath channel estimation method of super-Nyquist system | |
| CN101945066B (en) | Channel estimation method of OFDM/OQAM system | |
| CN101136884B (en) | Channel estimation method for TDS-OFDM system | |
| CN110149287A (en) | Super Nyquist system and its symbol estimation method based on linear predictive coding | |
| CN101662434B (en) | WiMAX channel estimation method designed by utilizing pilot frequency format | |
| CN105099968A (en) | Communication system at super-nyquist rate in multi-path channel | |
| CN109246039A (en) | A kind of Soft Inform ation iteration receiving method based on two-way time domain equalization | |
| CN107483373B (en) | Anti-multipath iterative weighting LMMSE channel estimation method and device | |
| CN109861939B (en) | OQPSK frequency domain equalization wireless data transmission method | |
| CN108768905B (en) | Wireless communication method based on chaotic forming and chaotic matched filter | |
| CN113852580A (en) | MIMO-OTFS symbol detection method based on multi-level separation | |
| US20040193965A1 (en) | Error adjustment in direct conversion architectures | |
| CN106161304B (en) | A kind of transmitting terminal IQ imbalance compensation method of joint channel estimation | |
| CN109547370A (en) | The super Nyquist system symbol estimation method that joint equalization and interference are eliminated | |
| US20030115061A1 (en) | MPSK equalizer | |
| CN113037671A (en) | Low-complexity efficient SOQPSK symbol timing and phase joint synchronization algorithm | |
| CN102340466B (en) | Method for designing adaptive decision feedback equalizer based on support vector machine | |
| TWI603603B (en) | In the downlink transmission system to solve the imperfect joint estimation of pre-compensation method | |
| CN113078966B (en) | High-precision 5G large-bandwidth signal testing method | |
| CN110753011A (en) | Single carrier equalization method for GMSK signal | |
| US20220140866A1 (en) | Wireless communication system, wireless communication method, transmitting station device and receiving station device | |
| CN106656879B (en) | A kind of high-speed and High-order variable-step self-adaptive equalization methods | |
| CN109818894B (en) | GMSK signal detection method and detection device under multipath channel | |
| EP2507957A1 (en) | Bit soft value normalization |
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 |