CN107395547B - Compensation method and system under IQ branch imbalance condition - Google Patents

Compensation method and system under IQ branch imbalance condition Download PDF

Info

Publication number
CN107395547B
CN107395547B CN201710546984.7A CN201710546984A CN107395547B CN 107395547 B CN107395547 B CN 107395547B CN 201710546984 A CN201710546984 A CN 201710546984A CN 107395547 B CN107395547 B CN 107395547B
Authority
CN
China
Prior art keywords
message
factor
branch
node
parameter
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
CN201710546984.7A
Other languages
Chinese (zh)
Other versions
CN107395547A (en
Inventor
王忠勇
王玮
崔建华
葛伟力
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.)
Zhengzhou University
Original Assignee
Zhengzhou University
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 Zhengzhou University filed Critical Zhengzhou University
Priority to CN201710546984.7A priority Critical patent/CN107395547B/en
Publication of CN107395547A publication Critical patent/CN107395547A/en
Application granted granted Critical
Publication of CN107395547B publication Critical patent/CN107395547B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/32—Carrier systems characterised by combinations of two or more of the types covered by groups H04L27/02, H04L27/10, H04L27/18 or H04L27/26
    • H04L27/34—Amplitude- and phase-modulated carrier systems, e.g. quadrature-amplitude modulated carrier systems
    • H04L27/36—Modulator circuits; Transmitter circuits
    • H04L27/362—Modulation using more than one carrier, e.g. with quadrature carriers, separately amplitude modulated
    • H04L27/364—Arrangements for overcoming imperfections in the modulator, e.g. quadrature error or unbalanced I and Q levels
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L27/00—Modulated-carrier systems
    • H04L27/26—Systems using multi-frequency codes
    • H04L27/2601—Multicarrier modulation systems
    • H04L27/2647—Arrangements specific to the receiver only

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

A compensation method and system under IQ branch imbalance condition, the method includes: s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced; s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter; s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter. According to the compensation method and the compensation system under the IQ branch imbalance condition, the factor graph is constructed, and the edge posterior probability information of each unknown parameter in the OFDM system is received through a message transmission method in the factor graph, so that the problem that the error rate performance of a receiver is reduced under the condition that the in-phase and quadrature branches are unbalanced is solved, and the reliable error rate performance is obtained with low complexity.

Description

Compensation method and system under IQ branch imbalance condition
Technical Field
The present invention relates to the field of wireless communication technologies, and in particular, to a compensation method and system under an IQ branch imbalance condition.
Background
Currently, high-rate wireless transmission usually suffers from severe frequency selective fading, and the channel length may even be tens or hundreds of data symbol periods. At this time, the conventional time domain equalization technology is difficult to use due to its high complexity, and Orthogonal Frequency Division Multiplexing (OFDM) technology is considered as one of the key technologies of the current and future high-speed wireless mobile communication systems.
In the OFDM technique, signals transmitted by a transmitting end allow overlapping of the channel spectrums of subcarriers, so that a receiving end can correctly separate each sub-data stream by using orthogonality between subcarriers. Since OFDM technology allows for spectral overlap between subchannels, its spectral utilization is high. However, in wireless communication, the channel condition is complicated, radio waves are affected by the propagation modes such as reflection, refraction and scattering due to the non-uniformity of the transmission medium during transmission, and the doppler effect is caused by the relative motion between mobile terminals, which all cause the fading of signals during transmission.
Meanwhile, due to the performance limitation of analog devices, the two orthogonal branches of the down-conversion part are not necessarily phase shifts with a phase difference of 90 °, and the amplitude gains of the branches are not completely the same, so that IQ (In-phase and Quadrature-phase) branch imbalance In the receiver is generated. The low-pass filter, the digital-to-analog converter and the amplifier in the IQ-branch are also not completely identical, which also causes IQ-branch imbalance. The IQ branch imbalance destroys the orthogonality of the I branch and the Q branch, and if the OFDM system is not compensated, the original transmission signal on each subcarrier is interfered by the signal on the corresponding mirror image subcarrier, so that the error rate performance of a receiver is seriously reduced under the condition of IQ branch imbalance, and the performance of the OFDM system is seriously deteriorated.
In the prior art, a Least Square (LS) method is generally adopted to compensate an OFDM system, and the LS method estimates IQ branch imbalance parameters of the OFDM system by adding a plurality of preamble sequences in front of a data block and using a transmission rule of the preamble sequences, and then compensates the system under an IQ imbalance condition by combining channel parameters and IQ imbalance parameters. However, the LS method adds a preamble training sequence, which results in a decrease in throughput of the system, and requires an inversion operation on a matrix when estimating channel parameters and IQ imbalance parameters, which results in a low compensation efficiency.
Disclosure of Invention
The present invention provides a compensation method and system under IQ-branch imbalance condition that overcomes or at least partially solves the above mentioned problems.
According to a first aspect of the present invention, a compensation method under an IQ branch imbalance condition includes:
s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced;
s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter;
s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter.
Wherein, step S1 is preceded by:
and constructing the factor graph model, wherein the factor graph model comprises factor nodes and variable nodes.
Wherein the constructing of the factor graph model, the factor graph model including factor nodes and variable nodes, comprises:
factorizing the joint probability density function of the OFDM system to obtain each factor node;
setting a plurality of corresponding variable nodes based on the channel parameters, the IQ branch imbalance parameters and the symbol parameters;
and dividing each factor node into three disjoint subsets according to the corresponding relation between each factor node and the variable node.
Wherein, step S1 specifically includes:
and acquiring initial values of all channel parameters in the OFDM system based on the pilot signals in the OFDM system.
Wherein, step S2 includes:
s21, based on the subset category of each factor node, using the joint message transfer algorithm to obtain the message received or sent by each factor node to each corresponding connected variable node according to the connection relationship between the factor node and the variable node in the preset factor graph, and performing Gaussian mapping on the message to obtain the estimation value of the IQ branch imbalance parameter;
wherein the joint message delivery algorithm comprises a joint confidence BP delivery algorithm, an expected EP delivery algorithm and an average field MF delivery algorithm;
s22, according to the BP message transmission algorithm, the OFDM system is de-mapped, decoded and channel time domain parameter calculated, and the time domain parameter and bit information of the channel are obtained.
Wherein, step S21 specifically includes:
calculating factor nodes according to BP and EP message transfer algorithm
Figure BDA0001343400690000031
To variable node pkMessage and factor node
Figure BDA0001343400690000032
To variable node QkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure BDA0001343400690000033
To variable node zkThe message of (2);
computing factor nodes according to MF message passing algorithm
Figure BDA0001343400690000034
A message to variable node λ;
computing factor nodes according to MF message passing algorithm
Figure BDA0001343400690000035
To variable node zkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure BDA0001343400690000036
To variable node pkMessage and factor node
Figure BDA0001343400690000037
To variable node QkThe message of (2);
calculating factor nodes according to BP and EP message transfer algorithm
Figure BDA0001343400690000038
To variable node xk、hkMessage of A and factor node
Figure BDA0001343400690000039
To variable node x-k、h-kB, wherein the
Figure BDA00013434006900000310
Figure BDA00013434006900000311
For a factor node preset in the factor graph, the pk、Qk、zk、λ、xk、hk、x-k、h-kA, B are variable nodes preset in the factor graph.
Wherein, step S3 includes:
converting the channel time domain parameters into frequency domain parameter messages, and mapping bit information into symbol messages;
and repeatedly acquiring the estimated value of the IQ branch imbalance parameter based on the frequency domain parameter message and the symbol message until the preset acquisition times are reached. According to a second aspect of the present invention, a compensation system under an IQ branch imbalance condition is characterized by comprising:
the acquisition module is used for acquiring initial values of all channel parameters under the condition that an in-phase quadrature IQ branch of an OFDM system is unbalanced;
the estimation module is used for acquiring an estimation value of the IQ branch imbalance parameter by using a combined message transmission algorithm based on a preset factor graph model and the initial value of each channel parameter;
and the compensation module is used for compensating the IQ branch of the OFDM system based on the estimated value of the IQ branch imbalance parameter.
According to a third aspect of the present invention, there is provided a computer program product comprising program code for performing a compensation method for an IQ leg imbalance condition as described above.
According to a fourth aspect of the invention, there is provided a non-transitory computer readable storage medium storing the computer program as described above.
According to the compensation method and the compensation system under the IQ branch imbalance condition, the factor graph is constructed, and the edge posterior probability information of each unknown parameter in the OFDM system is received through a message transmission method in the factor graph, so that the problem that the error rate performance of a receiver is reduced under the condition that the in-phase and quadrature branches are unbalanced is solved, and the reliable error rate performance is obtained with low complexity.
Drawings
Fig. 1 is a flowchart of a compensation method under an IQ branch imbalance condition according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a factor graph model provided by an embodiment of the invention;
fig. 3 is a schematic diagram illustrating bit error rate comparison of simulation results according to an embodiment of the present invention;
FIG. 4 is a schematic diagram illustrating bit error rate comparison of another simulation result according to an embodiment of the present invention;
fig. 5 is a structural diagram of a compensation system under an IQ branch imbalance condition according to an embodiment of the present invention.
Detailed Description
The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of a compensation method under an IQ branch imbalance condition according to an embodiment of the present invention, as shown in fig. 1, including:
s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced;
s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter;
s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter.
The factor graph is a bipartite graph model for expressing the global function decomposition structure of the system, and an effective message scheduling mechanism is designed by utilizing the factor graph and using various message transfer algorithms, so that the edge probability density function can be effectively calculated and the algorithm complexity can be reduced.
It can be understood that the preset factor nodes and the preset variable nodes in the factor graph are set according to the integral posterior probability distribution factorization of the OFDM system.
The method comprises the steps of obtaining an initial estimation value of each channel parameter of the OFDM system and an estimation value of an in-phase quadrature IQ branch imbalance parameter, calculating the posterior probability distribution of a variable node in the message transmission process to obtain the estimation value of the parameter, wherein the variable posterior probability is in direct proportion to the product of all information obtained by the variable, and obtaining bit information of the OFDM system through the posterior probability distribution of the bit information so as to obtain the error code performance of the system. .
Specifically, the compensation method under the IQ leg imbalance condition provided in the embodiment of the present invention mainly takes joint Message transfer as a basic framework, and takes into consideration information interaction between modules such as IQ leg imbalance parameter estimation and compensation, multipath channel parameter estimation and equalization, symbol detection and decoding, etc. on the whole, so as to enable the overall OFDM system to achieve optimal or near-optimal performance, where the Message transfer algorithm is designed based on a representation factor graph of a system joint probability density function, and the Message transfer algorithm includes a Belief Propagation (BP) algorithm, an Expectation Propagation (EP), a variation Message transfer (spatial Message Passing, VMP) algorithm, a Mean Field (MF) Message transfer algorithm, and various joint Message transfer algorithm rules.
The compensation method under the IQ branch imbalance condition provided by the invention obtains the posterior probability information of each parameter to be measured in the OFDM system by constructing a factor graph and using a combined message transfer algorithm, solves the problem of reduced error rate performance of a receiver under the condition of in-phase and quadrature branch imbalance, and thus obtains reliable error rate performance with lower complexity.
On the basis of the embodiment shown in fig. 1, step S1 further includes:
and constructing the factor graph model, wherein the factor graph model comprises factor nodes and variable nodes.
Wherein the constructing of the factor graph model, the factor graph model including factor nodes and variable nodes, comprises:
factorizing the joint probability density function of the OFDM system to obtain each factor node;
setting a plurality of corresponding variable nodes based on the channel parameters, the IQ branch imbalance parameters and the symbol parameters;
and dividing each factor node into three disjoint subsets according to the corresponding relation between each factor node and the variable node.
Specifically, one data block in the OFDM system is set as one OFDM symbol, and assuming that the length of one OFDM symbol is N, it is expressed as
Figure BDA0001343400690000061
Each symbol is obtained by inverse discrete Fourier transform (IFFT)
Figure BDA0001343400690000062
After parallel-to-serial conversion, after passing through a multipath channel
Figure BDA0001343400690000063
Adding white Gaussian noise v to reach a receiving end; receiving a signal
Figure BDA0001343400690000064
Obtaining received data block through IQ imbalance influence
Figure BDA0001343400690000065
And obtaining y through discrete Fourier transform (FFT) after serial-to-parallel conversion.
The time domain model of the received signal when the IQ branch is unbalanced is as follows:
Figure BDA0001343400690000066
A＝cos(θ/2)+jαsin(θ/2)，
B＝αcos(θ/2)-j sin(θ/2)，
α, θ represents the amplitude and phase imbalance between the I, Q branches, and ideally α is 0, θ is 0, i.e., a is 1 and B is 0.
The k sub-carrier signal y after FFTkComprises the following steps:
Figure BDA0001343400690000071
wherein, wkIs the noise signal on the k-th subcarrier after the conversion of Gaussian white noise v, when k is 1, N/2+1, h-k＝hk,x-k＝xk(ii) a k is not equal to 1, when N/2+1, h-k＝hN-k+2,x-k＝xN-k+2。
Fig. 2 is a schematic diagram of a factorial graph model provided in the embodiment of the present invention, and as shown in fig. 2, factorization is performed according to a joint probability density function of a system, and is represented as:
Figure BDA0001343400690000072
where p (y | h, x, λ, A, B) represents a likelihood function given y;
Figure BDA0001343400690000073
a prior probability density function representing a channel parameter,representing the constraint relation between the symbol x and the code word c and the information bit b;
Figure BDA0001343400690000075
and
Figure BDA0001343400690000076
respectively representing IQ imbalance parameters AAnd the prior probability density function of B;
Figure BDA0001343400690000077
a prior probability density function representing the noise accuracy λ.
Let p bek＝Ahkxk，
Figure BDA0001343400690000078
Wherein, when k is 1, N/2+1, h-k＝hk,x-k＝xk(ii) a When k is not equal to 1, N/2+1, h-k＝hN-k+2,x-k＝xN-k+2，
p(yk|hk,xk,λ,A,B)＝p(yk|zk,λ)p(zk|pk,Qk)p(pk|hk,xk,A)p(Qk|h-k,x-k,B)，
Wherein the factor
Figure BDA0001343400690000079
Denotes given ykA likelihood function of (a); factor(s)
Figure BDA00013434006900000710
Represents the variable zk，pk，QkA deterministic relationship between; factor(s)
Figure BDA00013434006900000711
Represents the variable pk，xk，hkA deterministic relationship between A; factor(s)
Figure BDA00013434006900000712
Represents the variable Qk，x-k，h-kAnd a deterministic relationship between B.
And dividing each factor node into three disjoint subsets according to the corresponding relation between each factor node and the variable node.
In particular, the factor nodes and their associated variable nodes are based on their characteristicsAnd dividing the factor nodes into:
Figure BDA00013434006900000713
a subset,
Figure BDA00013434006900000714
A subset,
Figure BDA00013434006900000715
A subset.
On the basis of the foregoing embodiment, step S1 specifically includes:
and acquiring initial values of all channel parameters in the OFDM system based on the pilot signals in the OFDM system.
Specifically, first, an IQ branch imbalance parameter in the OFDM system needs to be initialized, where the initialized IQ branch imbalance parameter is a value under an ideal condition, that is, a is 1, and B is 0; initialization message
Figure BDA0001343400690000081
And
Figure BDA0001343400690000082
calculating initial values of channel parameters based on the inserted pilots
Figure BDA0001343400690000083
On the basis of the above embodiment, step S2 includes:
s21, based on the subset category of each factor node, using the joint message transfer algorithm to obtain the message received or sent by each factor node to each corresponding connected variable node according to the connection relationship between the factor node and the variable node in the preset factor graph, and performing Gaussian mapping on the message to obtain the estimation value of the IQ branch imbalance parameter;
wherein the joint message delivery algorithm comprises a joint confidence BP delivery algorithm, an expected EP delivery algorithm and an average field MF delivery algorithm;
s22, according to the BP message transmission algorithm, the OFDM system is de-mapped, decoded and channel time domain parameter calculated, and the time domain parameter and bit information of the channel are obtained.
According to the factor graph model diagram shown in fig. 2, the transfer process of step S21 can be specifically divided into:
calculating factor nodes according to BP and EP message transfer algorithm
Figure BDA0001343400690000084
To variable node pkMessage and factor node
Figure BDA0001343400690000085
To variable node QkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure BDA0001343400690000086
To variable node zkThe message of (2);
computing factor nodes according to MF message passing algorithm
Figure BDA0001343400690000087
A message to variable node λ;
computing factor nodes according to MF message passing algorithm
Figure BDA0001343400690000088
To variable node zkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure BDA0001343400690000089
To variable node pkMessage and factor node
Figure BDA00013434006900000810
To variable node QkThe message of (2);
calculating factor nodes according to BP and EP message transfer algorithm
Figure BDA00013434006900000811
To variable node xk、hkMessage of A and factor node
Figure BDA00013434006900000812
To variable node x-k、h-kB, wherein the
Figure BDA00013434006900000813
Figure BDA00013434006900000814
For a factor node preset in the factor graph, the pk、Qk、zk、λ、xk、hk、x-k、h-kA, B are variable nodes preset in the factor graph.
Firstly, factor nodes are calculated according to BP-EP message transmission rules
Figure BDA00013434006900000912
To variable node pkSpecifically, the following information is obtained by calculation according to the BP information rule:
Figure BDA0001343400690000091
obtaining variable p according to the above calculation formulakThe confidence message is:
Figure BDA0001343400690000092
the variable p iskConfidence messages are mapped to gaussian messages:
Figure BDA0001343400690000093
wherein the mean and variance of the Gaussian message are respectively
Figure BDA0001343400690000094
And
Figure BDA0001343400690000095
wherein
Figure BDA0001343400690000096
Figure BDA0001343400690000097
Then, factor nodes are obtained through calculation according to EP message rules
Figure BDA00013434006900000913
To variable node pkThe message of (2):
Figure BDA0001343400690000098
wherein the content of the first and second substances,
Figure BDA0001343400690000099
satisfies the mean and variance of
Figure BDA00013434006900000910
Similarly, factor nodes are calculated according to BP-EP message passing rules
Figure BDA00013434006900000911
To variable node QkSpecifically, the following information is obtained by calculation according to the BP information rule:
Figure BDA0001343400690000101
obtaining variable Q according to the above formulakThe confidence message is:
Figure BDA0001343400690000102
the variable QkConfidence messages are mapped to gaussian messages:
Figure BDA0001343400690000103
wherein the mean and variance of the Gaussian message are respectively
Figure BDA0001343400690000104
And
Figure BDA0001343400690000105
wherein
Figure BDA0001343400690000106
Figure BDA0001343400690000107
Then, factor nodes are obtained through calculation according to EP message rules
Figure BDA0001343400690000108
To variable node QkThe message of (2):
Figure BDA0001343400690000109
wherein the content of the first and second substances,
Figure BDA00013434006900001010
satisfies the mean and variance of
Figure BDA00013434006900001011
Further, factor nodes are calculated according to BP message transmission rules
Figure BDA00013434006900001013
To variable node zkThe message of (1) is:
Figure BDA00013434006900001012
wherein the Gaussian message
Figure BDA0001343400690000111
Satisfies the mean and variance of
Figure BDA0001343400690000112
It will be appreciated that further, factor nodes are computed according to the MF messaging rules
Figure BDA00013434006900001117
Obtaining an iterative estimation value of noise precision by a message of a variable node lambda, which specifically comprises the following steps:
computing factor nodes according to MF messaging rules
Figure BDA00013434006900001118
Messages to variable node lambda
Figure BDA0001343400690000113
Comprises the following steps:
Figure BDA0001343400690000114
wherein
Figure BDA0001343400690000115
And
Figure BDA0001343400690000116
are respectively a variable zkConfidence message b (z)k) Assuming that the prior probability density function of the noise precision lambda is uniformly distributed, the posterior probability density function of the lambda can be obtained as
Figure BDA0001343400690000117
The iteration estimation value of the noise precision obtained according to the MAP rule is as follows:
Figure BDA0001343400690000118
by this, the message passing process from right to left is completed.
Specifically, messages are transmitted from left to right, and further, factor nodes are calculated according to MF message transmission rules
Figure BDA0001343400690000119
To variable node zkThe message of (1) is:
Figure BDA00013434006900001110
Figure BDA00013434006900001111
wherein the observed value ykAnd an estimate of the variance of the noise
Figure BDA00013434006900001112
Are respectively Gaussian messages
Figure BDA00013434006900001113
Mean and variance of.
Further, factor nodes are calculated according to BP message transmission rules
Figure BDA00013434006900001114
To variable node pkOf a message
Figure BDA00013434006900001115
Comprises the following steps:
Figure BDA00013434006900001116
wherein the Gaussian message
Figure BDA0001343400690000121
Satisfies the mean and variance of
Figure BDA0001343400690000122
Calculating factor nodes according to BP message transmission rule
Figure BDA0001343400690000123
To variable node QkOf a message
Figure BDA0001343400690000124
Comprises the following steps:
Figure BDA0001343400690000125
wherein the Gaussian message
Figure BDA0001343400690000126
Satisfies the mean and variance of
Figure BDA0001343400690000127
It can be understood that, in the embodiment of the present invention, the BP-EP sub-graph message combines the BP and EP message passing rules, and performs gaussian approximation on the IQ imbalance parameter A, B message to obtain an estimated value thereof.
Finally, factor nodes are calculated according to the combined BP-EP message transmission rule
Figure BDA00013434006900001212
To variable node xk、hkAnd a.
Calculation according to BP rules
Figure BDA0001343400690000128
Comprises the following steps:
Figure BDA0001343400690000129
computing messages according to EP rules
Figure BDA00013434006900001210
Comprises the following steps:
Figure BDA00013434006900001211
wherein the content of the first and second substances,
Figure BDA0001343400690000131
gauss message
Figure BDA0001343400690000132
Satisfies the mean and variance of
Figure BDA0001343400690000133
Calculation according to BP rules
Figure BDA0001343400690000134
Comprises the following steps:
Figure BDA0001343400690000135
it will be appreciated that, to reduce complexity, a gaussian mapping is performed,
Figure BDA0001343400690000136
wherein the content of the first and second substances,
Figure BDA0001343400690000137
gauss message
Figure BDA0001343400690000138
Respectively of mean and variance of
Figure BDA0001343400690000139
Computing factor nodes according to joint BP-EP message passing rules
Figure BDA00013434006900001310
To variable node x-k、h-kB messages, in particular, calculated according to BP rules
Figure BDA00013434006900001311
Comprises the following steps:
Figure BDA00013434006900001312
computing messages according to EP rules
Figure BDA0001343400690000141
Comprises the following steps:
Figure BDA0001343400690000142
wherein the content of the first and second substances,
Figure BDA0001343400690000143
gauss message
Figure BDA0001343400690000144
Satisfies the mean and variance of
Figure BDA0001343400690000145
Calculation according to BP rules
Figure BDA0001343400690000146
Comprises the following steps:
Figure BDA0001343400690000147
also, to reduce complexity, a gaussian mapping is performed,
Figure BDA0001343400690000148
wherein the content of the first and second substances,
Figure BDA0001343400690000149
gauss message
Figure BDA00013434006900001410
Respectively of mean and variance of
Figure BDA00013434006900001411
The embodiment of the invention takes the joint probability density function of the system as a basic starting point, carries out factorization on the joint probability density function and divides factor nodes into three disjoint subsets: MF, BP-EP subset, has constructed the factor graph; on the basis, the message iteration updating is carried out by adopting the message transmission rules of combining MF, BP and BP-EP to approximate the posterior probability density function of the edge of the variable to be estimated, so that the algorithm complexity is reduced, and the reliable error code performance is obtained.
On the basis of the above embodiment, step S3 includes:
converting the channel time domain parameters into frequency domain parameter messages, and mapping bit information into symbol messages;
and repeatedly acquiring the estimated value of the IQ branch imbalance parameter based on the frequency domain parameter message and the symbol message until the preset acquisition times are reached.
It can be understood that, the edge posterior probability density function of the variable to be estimated, which is calculated in the process of the above embodiment, is not an accurate edge posterior probability density function in general, and in order to make the error code performance of the OFDM system approach to the optimum, it is necessary to ensure the convergence of the algorithm and ensure the performance of the receiver.
The compensation method under the IQ branch imbalance condition provided by the embodiment of the invention is established under the early-stage condition of IQ branch imbalance of an OFDM system, and can be understood that the error code performance of the existing receiving algorithm can meet the requirement if the IQ branch of the OFDM system is balanced.
In addition, the compensation method under the IQ branch imbalance condition provided by the embodiment of the invention can respectively estimate the channel and the IQ parameters under different IQ imbalance conditions and channel information conditions without a pre-training sequence, thereby obtaining a lower error rate and higher reliability and spectral efficiency.
The bit error rate comparison graph of the simulation result is shown in fig. 3 and fig. 4, in fig. 3, the number of subcarriers of the OFDM system is 512, the subcarrier spacing is 15kHz, a pilot frequency is inserted into every 32 subcarriers, QPSK is used for pilot modulation, the number of taps of a multipath channel is 32, the code rate of information bits is 1/2, and the generator polynomial is (23,35)8The convolutional code is encoded and mapped into a 16QAM channel symbol after random interleaving, IQ amplitude and phase imbalance parameters are set to α -1 dB, and theta is set to 2 degrees, IQ is not compensated under different signal-to-noise ratios, IQ does not exist under ideal conditions, and a compensation algorithm based on a message passing algorithm comprises a bit error rate comparison graph under the condition that IQ parameters and pilot frequency joint data are simultaneously estimated.
The setting of simulation parameters in fig. 4 is the same as that in fig. 3, p1 in fig. 4 represents the pilot used in the first iteration when estimating the IQ imbalance parameters A, B, and the other pilot used and data used for joint estimation; p2 denotes pilot for the first two iterations in A, B estimation, and other joint estimates with pilot and data; and so on. Fig. 4 shows a relationship diagram of the bit error rate and the iteration number of the compensation algorithm based on the joint message passing algorithm under the condition that the signal-to-noise ratio is 8dB, IQ is not compensated, and ideally, IQ does not exist in complete synchronization.
As can be seen from fig. 3 and 4, after the technical scheme provided by the embodiment of the invention is adopted, the receiver has lower bit error rate and more reliable performance; in addition, the best iteration effect cannot be achieved only by using the pilot frequency in the iteration process or by using the pilot frequency and the data in each iteration, and the better effect can be achieved by using the pilot frequency firstly and then using the combined data.
Fig. 5 is a structural diagram of a compensation system under an IQ branch imbalance condition according to an embodiment of the present invention, which is characterized in that the compensation system includes: an acquisition module 1, an estimation module 2 and a compensation module 3, wherein,
the acquisition module 1 is used for acquiring initial values of parameters of each channel under the condition that an in-phase quadrature IQ branch of an OFDM system is unbalanced;
the estimation module 2 obtains an estimation value of the IQ branch imbalance parameter by using a joint message transmission algorithm based on a preset factor graph model and the initial value of each channel parameter;
and the compensation module 3 is used for compensating the IQ branch of the OFDM system based on the estimated value of the IQ branch imbalance parameter.
For a specific receiver design system receiving calculation process of the OFDM system, reference may be made to the above embodiments, and details of the embodiments of the present invention are not repeated herein.
The compensation method under IQ branch imbalance condition provided by the invention adopts a formal design method, takes the joint probability density function of an OFDM system as a basic starting point, carries out factorization on the joint probability density function and divides factor nodes into three disjoint subsets: MF, BP-EP subset, has constructed the factor graph; on the basis, message iteration updating is carried out by adopting a combined MF, BP and BP-EP message transfer rule to approximate to a posterior probability density function of the edge of the variable to be estimated; the method solves the problem of the performance reduction of the error rate of the receiver under the condition of the unbalance of the in-phase and quadrature branches, thereby obtaining reliable error code performance with lower complexity.
The present embodiment provides a compensation system under an IQ branch imbalance condition, including: at least one processor; and at least one memory communicatively coupled to the processor, wherein:
the memory stores program instructions executable by the processor, the processor calls the program instructions to perform the methods provided by the method embodiments, for example, including: s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced; s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter; s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter.
The present embodiment discloses a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method provided by the above-mentioned method embodiments, for example, comprising: s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced; s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter; s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter.
The present embodiments provide a non-transitory computer-readable storage medium storing computer instructions that cause the computer to perform the methods provided by the above method embodiments, for example, including: s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced; s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter; s3, compensating the IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter.
Those of ordinary skill in the art will understand that: all or part of the steps for implementing the method embodiments may be implemented by hardware related to program instructions, and the program may be stored in a computer readable storage medium, and when executed, the program performs the steps including the method embodiments; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, the method of the present application is only a preferred embodiment and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (5)

1. A compensation method under IQ branch imbalance condition, comprising:
s1, acquiring initial values of each channel parameter under the condition that an in-phase quadrature IQ branch of the OFDM system is unbalanced;
s2, acquiring an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial value of each channel parameter;
s3, compensating the in-phase and quadrature IQ branch of the OFDM system based on the estimation value of the IQ branch imbalance parameter;
wherein, step S1 is preceded by:
constructing the factor graph model, wherein the factor graph model comprises factor nodes and variable nodes, and comprises the following steps:
factorizing the joint probability density function of the OFDM system to obtain each factor node;
setting a plurality of corresponding variable nodes based on the channel parameters, the IQ branch imbalance parameters and the symbol parameters;
dividing each factor node into three disjoint subsets according to the corresponding relation between each factor node and the variable node;
step S2 specifically includes:
s21, based on the subset category of each factor node, using the joint message transfer algorithm to obtain the message received or sent by each factor node to each corresponding connected variable node according to the connection relationship between the factor node and the variable node in the preset factor graph model, and performing Gaussian mapping on the message to obtain the estimation value of the IQ branch imbalance parameter;
the combined message transfer algorithm comprises a BP message transfer algorithm, an EP message transfer algorithm and an MF message transfer algorithm;
s22, according to the BP message transmission algorithm, carrying out de-mapping, decoding and channel time domain parameter calculation on the OFDM system to obtain the time domain parameters and bit information of the channel;
step S3 specifically includes:
converting the channel time domain parameters into frequency domain parameter messages, and mapping bit information into symbol messages;
and repeatedly acquiring the estimated value of the IQ branch imbalance parameter based on the frequency domain parameter message and the symbol message until the preset acquisition times are reached.
2. The method according to claim 1, wherein step S1 specifically comprises:
and acquiring initial values of all channel parameters in the OFDM system based on the pilot signals in the OFDM system.
3. The method according to claim 1, wherein step S21 specifically comprises:
calculating factor nodes according to BP message transfer algorithm and EP message transfer algorithm
Figure FDA0002468796890000021
To variable node pkMessage and factor node
Figure FDA0002468796890000022
To variable node QkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure FDA0002468796890000023
To variable node zkThe message of (2);
computing factor nodes according to MF message passing algorithm
Figure FDA0002468796890000024
A message to variable node λ;
computing factor nodes according to MF message passing algorithm
Figure FDA0002468796890000025
To variable node zkThe message of (2);
calculating factor nodes according to BP message transfer algorithm
Figure FDA0002468796890000026
To variable node pkMessage and factor node
Figure FDA0002468796890000027
To variable node QkThe message of (2);
calculating factor nodes according to BP and EP message transfer algorithm
Figure FDA0002468796890000028
To variable node xk、hkMessage of A and factor node
Figure FDA0002468796890000029
To variable node x-k、h-kB, wherein the
Figure FDA00024687968900000210
Figure FDA00024687968900000211
For the factor nodes preset in the factor graph model, the pk、Qk、zk、λ、xk、hk、x-k、h-kA, B are variable nodes preset in the factor graph model.
4. A compensation system under IQ branch imbalance conditions, comprising:
the acquisition module is used for acquiring initial values of all channel parameters under the condition that an in-phase quadrature IQ branch of an OFDM system is unbalanced;
the estimation module obtains an estimation value of the IQ branch imbalance parameter by using a joint message transfer algorithm based on a preset factor graph model and the initial values of the channel parameters, and specifically comprises: based on the subset category of each factor node, using the joint message transfer algorithm to obtain the message received or sent by each factor node to each corresponding connected variable node according to the connection relationship between the factor node and the variable node in the preset factor graph model, and performing Gaussian mapping on the message to obtain the estimated value of the IQ branch imbalance parameter;
the combined message transfer algorithm comprises a BP message transfer algorithm, an EP message transfer algorithm and an MF message transfer algorithm;
according to a BP message transfer algorithm, performing demapping, decoding and channel time domain parameter calculation on the OFDM system to obtain time domain parameters and bit information of a channel;
a compensation module, configured to compensate the IQ branch of the OFDM system based on the estimated value of the IQ branch imbalance parameter, specifically including:
converting the channel time domain parameters into frequency domain parameter messages, and mapping bit information into symbol messages;
and repeatedly acquiring the estimated value of the IQ branch imbalance parameter based on the frequency domain parameter message and the symbol message until the preset acquisition times are reached.
5. A non-transitory computer-readable storage medium storing computer instructions that cause a computer to perform the method of any one of claims 1 to 3.
CN201710546984.7A 2017-07-06 2017-07-06 Compensation method and system under IQ branch imbalance condition Active CN107395547B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710546984.7A CN107395547B (en) 2017-07-06 2017-07-06 Compensation method and system under IQ branch imbalance condition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710546984.7A CN107395547B (en) 2017-07-06 2017-07-06 Compensation method and system under IQ branch imbalance condition

Publications (2)

Publication Number Publication Date
CN107395547A CN107395547A (en) 2017-11-24
CN107395547B true CN107395547B (en) 2020-06-30

Family

ID=60334383

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710546984.7A Active CN107395547B (en) 2017-07-06 2017-07-06 Compensation method and system under IQ branch imbalance condition

Country Status (1)

Country Link
CN (1) CN107395547B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108337201B (en) * 2018-02-06 2019-12-27 电子科技大学 IQ imbalance compensation method for large-scale multi-antenna system
CN116244571A (en) * 2023-02-08 2023-06-09 南方海洋科学与工程广东省实验室(湛江) Orthogonal unbalance compensation method, system, device and medium

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102340479A (en) * 2011-10-25 2012-02-01 北京华力创通科技股份有限公司 In-phase and quadrature (IQ) unbalance compensation device and method
CN104486285A (en) * 2014-12-23 2015-04-01 南京理工大学 Compensation method for IQ imbalance existing in OFDM system receiver
CN104980377A (en) * 2014-04-03 2015-10-14 华为技术有限公司 IQ imbalance estimation and correction equipment, system and method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7139536B2 (en) * 2003-12-02 2006-11-21 Mediatek Inc. Method and apparatus for I/Q imbalance calibration of a transmitter system
US8290458B2 (en) * 2008-06-30 2012-10-16 Entropic Communications, Inc. System and method for IQ imbalance estimation using loopback with frequency offset

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102340479A (en) * 2011-10-25 2012-02-01 北京华力创通科技股份有限公司 In-phase and quadrature (IQ) unbalance compensation device and method
CN104980377A (en) * 2014-04-03 2015-10-14 华为技术有限公司 IQ imbalance estimation and correction equipment, system and method
CN104486285A (en) * 2014-12-23 2015-04-01 南京理工大学 Compensation method for IQ imbalance existing in OFDM system receiver

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
A BP–MF–EP Based Iterative Receiver for Joint Phase Noise Estimation, Equalization, and Decoding;Wei Wang,Zhongyong Wang,etc.;《IEEE SIGNAL PROCESSING LETTERS, VOL. 23, NO. 10, OCTOBER 2016》;20161031;全文 *
基于因子图和联合消息传递的无线网络协作定位算法;崔建华，王忠勇等;《计算机应用》;20170510;全文 *

Also Published As

Publication number Publication date
CN107395547A (en) 2017-11-24

Similar Documents

Publication Publication Date Title
US10833785B1 (en) Processing communications signals using a machine-learning network
RU2322762C1 (en) Method for iterative estimation and decoding of channels and interferences
US8259864B2 (en) Methods and systems for fourier-quadratic basis channel estimation in OFDMA systems
US20160254937A1 (en) Low complexity scma/lds detection systems and methods
US8548101B2 (en) Demodulation method and demodulator for orthogonal frequency multiplexing—multiple input multiple output system
US20160036561A1 (en) Orthogonal Frequency Division Multiplexing Based Communications Over Nonlinear Channels
KR102010562B1 (en) Apparatus and method for generating decoding metric for frequency-quadrature amplitude modulated signals in wireless communication system
Murin et al. Low complexity estimation of carrier and sampling frequency offsets in burst‐mode OFDM systems
Zhao et al. Block sparse Bayesian learning-based channel estimation for MIMO-OTFS systems
US8903025B1 (en) Low complexity technique for detecting data in a received multiple-input-multiple-output (MIMO) signal
KR101098686B1 (en) Quadrature imbalance estimation using unbiased training sequences
CN107395547B (en) Compensation method and system under IQ branch imbalance condition
CN112054975A (en) Joint estimation and detection method based on BP-MF framework and VAMP
CN115943395A (en) Radio receiver, transmitter and system for pilot-free OFDM communication
CN117397215A (en) Generation and reception of pre-coded signals based on codebook linearization
US20160065329A1 (en) Single carrier communications harnessing nonlinearity
Osinsky et al. Data-aided ls channel estimation in massive mimo turbo-receiver
CN114172779B (en) Channel estimation method, device, equipment and storage medium
US9929884B2 (en) Systems and methods for detecting data in a received multiple-input-multiple-output (MIMO) signal
Pokamestov et al. Adapting SCMA codebooks to channel state
KR101255635B1 (en) Method and apparatus of estimating frequency offset based on partial periodogram in wireless communication system
Chen et al. Construction and application of polar codes in OFDM underwater acoustic communication
Şahin et al. Majority vote computation with complementary sequences for distributed UAV guidance
Shi et al. Joint phase noise estimation and decoding in OFDM-IM
Ma et al. Low-Complexity Factor Graph-Based Joint Channel Estimation and Equalization for SEFDM Signaling

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