CN111835365A - Comb filtering Turbo equalization algorithm suitable for short-wave communication - Google Patents
Comb filtering Turbo equalization algorithm suitable for short-wave communication Download PDFInfo
- Publication number
- CN111835365A CN111835365A CN202010749832.9A CN202010749832A CN111835365A CN 111835365 A CN111835365 A CN 111835365A CN 202010749832 A CN202010749832 A CN 202010749832A CN 111835365 A CN111835365 A CN 111835365A
- Authority
- CN
- China
- Prior art keywords
- substep
- wave communication
- turbo equalization
- equalization algorithm
- comb filtering
- 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.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 56
- 238000001914 filtration Methods 0.000 title claims abstract description 39
- 238000004891 communication Methods 0.000 title claims abstract description 29
- 239000011159 matrix material Substances 0.000 claims abstract description 34
- 238000000034 method Methods 0.000 claims abstract description 12
- 239000013598 vector Substances 0.000 claims description 40
- 238000000926 separation method Methods 0.000 claims description 14
- 230000004927 fusion Effects 0.000 claims description 7
- 230000014509 gene expression Effects 0.000 claims description 7
- 239000000126 substance Substances 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 4
- 238000010923 batch production Methods 0.000 claims description 3
- 238000010276 construction Methods 0.000 claims description 3
- 230000009191 jumping Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 12
- 239000000284 extract Substances 0.000 abstract description 7
- 238000004364 calculation method Methods 0.000 abstract description 5
- 108091006146 Channels Proteins 0.000 description 56
- 238000004088 simulation Methods 0.000 description 11
- 238000005562 fading Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000005433 ionosphere Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000010363 phase shift Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/25—Error detection or forward error correction by signal space coding, i.e. adding redundancy in the signal constellation, e.g. Trellis Coded Modulation [TCM]
- H03M13/258—Error detection or forward error correction by signal space coding, i.e. adding redundancy in the signal constellation, e.g. Trellis Coded Modulation [TCM] with turbo codes, e.g. Turbo Trellis Coded Modulation [TTCM]
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03M—CODING; DECODING; CODE CONVERSION IN GENERAL
- H03M13/00—Coding, decoding or code conversion, for error detection or error correction; Coding theory basic assumptions; Coding bounds; Error probability evaluation methods; Channel models; Simulation or testing of codes
- H03M13/25—Error detection or forward error correction by signal space coding, i.e. adding redundancy in the signal constellation, e.g. Trellis Coded Modulation [TCM]
- H03M13/256—Error detection or forward error correction by signal space coding, i.e. adding redundancy in the signal constellation, e.g. Trellis Coded Modulation [TCM] with trellis coding, e.g. with convolutional codes and TCM
Abstract
The invention belongs to the field of short-wave communication, and discloses a comb filtering Turbo equalization algorithm suitable for short-wave communication
Each signal can extract soft information
Thereby achieving a diversity effect; secondly, all the soft information is merged to obtain L (x), and the L (x) is input into a decoder for decoding; then, based on the soft information L output by the decoderout(x) Constructing average information
Combined with known multipath channel parameters to model the received signal
Finally, the comb filter utilizes the analog received signal
Extracting the multipath signal again from the real received signal r; the algorithm is simple and easy to implement, the inverse characteristic of a channel does not need to be solved, complex matrix inversion operation is avoided, the calculation complexity is greatly reduced, and the method has the advantages of simplicity, high convergence speed, easiness in implementation and the like.
Description
Technical Field
The invention relates to the field of short-wave communication, in particular to a comb filtering Turbo equalization algorithm suitable for short-wave communication.
Background
In mobile communication, the position of a receiver is constantly changed, and due to the diversity of communication environments, electromagnetic waves are reflected when encountering obstacles or ionosphere, and diffuse reflection is generated when encountering obstacles with uneven surfaces. Therefore, in actual communication, the receiver receives the superposition of signals from different paths, and even if the noise interference is small, the normal demodulation cannot be performed. The presence of multipath can lead to Inter-symbol interference (ISI), which degrades the performance of the communication system and creates an error floor. The front and rear symbols are superposed together when viewed from the time domain, and frequency selective fading is caused when viewed from the frequency domain, that is, deep fading is generated for a certain frequency, and part of detail information is completely lost.
Equalization is an effective method for solving multipath effects, and conventional equalization can be divided into time domain equalization and frequency domain equalization. With the idea of iterative decoding of modern coding and decoding (such as Turbo/LDPC codes), people gradually recognize that an iterative manner can bring extra performance gain to a communication system, and thus an idea of joint iterative decoding is proposed; the essence is that the information is continuously cycled between the equalizer and the decoder until the decoding is successful or the maximum number of iterations is reached. Since the operation mechanism of the process is very similar to the decoding mechanism of Turbo codes, joint iterative decoding is also called Turbo equalization.
The publication CN102185617A simplified Turbo equalization algorithm proposes a method for reducing the computational complexity of Turbo equalization. The variance of the sent symbol is basically the same during the first iteration, the coefficient of the equalizer is fixed at the moment, when the iteration reaches a certain number of times, the variance of the sent modulation symbol is very small, the relationship between the coefficient of the equalizer and the sent symbol is very small, the coefficient can be considered not to change any more, and therefore the inverse operation of the matrix is avoided once every time one symbol is calculated.
The main idea of the existing equalization technology, whether in the frequency domain or the time domain, is to obtain the inverse of the channel through some kind of calculation, and to counteract the multipath effect of the channel by the interaction of the received aliasing signal and the inverse of the channel. A transversal filter is usually used in the time domain to approximate the inverse of the channel, and the more the order of the filter is, the closer to the true inverse channel, the better the equalization performance. However, good performance comes at the expense of computational complexity, and thus the prior art has largely focused on how to solve the inverse of the channel in noisy environments quickly and with low complexity. The disadvantages mainly include the following three points:
1) most Turbo equalization techniques use frequency domain equalization, requiring Fast Fourier Transform (FFT) FFT/Inverse Fast Fourier Transform (IFFT) to increase system complexity. Meanwhile, due to the introduction of FFT, the data length can only select specific values (such as 512, 1024 and the like) which accord with the FFT rule, and the length of the data frame is greatly limited, so that the data frame is inflexible in actual use; 2) in engineering practice, filters of finite order are usually used to approximate the inverse of the channel, so that there is a truncation effect; a tradeoff between computational complexity and performance is required; 3) the time domain Turbo equalization algorithm inevitably needs to perform channel matrix inversion operation, the operation amount of the inversion operation increases in a cubic mode along with the increase of the order of the transverse filter, and when the order is large, the engineering realization is almost impossible.
Comb filters are devices commonly used in the field of digital signal processing, and are composed of a plurality of pass-band or stop-band filters arranged at equal intervals in frequency, so as to allow only signals in certain specific frequency ranges to pass or not pass, and the spectral characteristics of the pass-band filters are like a comb, so that the pass-band filters are called comb filters. The invention uses the concept of comb filter in the field of digital signal processing, however, the working principle is completely different.
Disclosure of Invention
Aiming at the problems in the prior art, the invention aims to provide a comb filtering Turbo equalization algorithm suitable for short wave communication, aiming at effectively resisting various problems caused by multi-path propagation through a certain equalization algorithm, under the premise of acquiring multi-path channel parameters, a comb filter is used for separating signals which are mixed together and extracting soft information, so that a plurality of versions of useful signals subjected to different fading are acquired, the diversity effect is realized, and the reliability and the effectiveness of a system are further improved.
The main idea of the invention is as follows: the comb filter separates received aliasing signals path by path based on soft information generated by the decoder, extracts a plurality of noise adding versions of useful signals to realize diversity effect, the decoder decodes the enhanced signals and feeds back the decoded information to the comb filter. The process is iterated repeatedly, and finally the purpose of improving the performance of the communication system is achieved.
In order to achieve the purpose, the invention is realized by adopting the following technical scheme.
A comb filtering Turbo equalization algorithm suitable for short-wave communication comprises the following steps:
transmitting end transmits data symbol xprefixAfter the data symbol is transmitted through a multipath channel, a receiving end receives a signal r;
Further, in step 1, the high-order modulation is BPSK modulation, QPSK modulation, or 8PSK modulation.
Further, in step 1, the data symbol xprefixComprises the following steps:
xprefix＝(x-l,…,x-1,x0,x1,…,xj,…,xn-1,xn,…,xn+l-1)
wherein (x)-l,…,x-1) As a front guard band, (x)n,…,xn+l-1) The front guard band and the rear guard band are arbitrary sequences formed by +/-1, and the lengths of the front guard band and the rear guard band are respectively more than or equal to p + q;
the mathematical expression of the receiving value of the receiving signal at the t-th moment is as follows:
rt＝h-p(t)xp+t+…+h-1(t)x1+t+h0(t)xt+h1(t)x-1+t+…+hq(t)x-q+t+wt，(-p<＝t<n+q)
wherein, wtAnd the sampling value of the t moment follows normal distribution with the mean value of 0 and the variance of w-is obtained.
Further, step 2 comprises the following substeps:
substep 2.1, initialization: setting the current iteration number L as 0, setting the maximum iteration number L of the joint decoding, and setting the soft information vector output by the initial decoder as a full 0 vector, namely Lout(x)＝(0,…,0,…,0)；
Substep 2.2, separating the received aliasing signals path by path based on a signal separation algorithm to obtain a path separation matrix DdispartAnd constructing a mean matrix MdispartAnd variance matrix Vdispart；
Substep 2.3, calculating the log-likelihood ratio L (x) for each equivalent received valuej,k)；
Substep 2.4, obtaining a log-likelihood ratio vector L (x) by soft information self-adaptive fusion;
substep 2.5, sending log-likelihood ratio vector L (x) to decoder for decoding, and outputting soft information Lout(x)；
Substep 2.6, adding 1 to an iteration variable L, and entering substep 2.2 if L is less than L; otherwise, entering the substep 2.7;
substep 2.7 of decoding the soft information L output by the decoderout(x) Making decision to obtain decoding estimation code word
Further, sub-step 2.2 comprises the sub-steps of:
substep 2.2.1, based on the soft information vector L output by the decoderout(x) Constructing average information
Sum variance information
Wherein the content of the first and second substances,
wherein, tanh represents hyperbolic tangent;
is reflected by the decoder pair symbol xj(ii) an estimate of (d);
substep 2.2.2, let the soft information bit number j equal to 0, if j < n, go to substep 2.2.2.1; otherwise, entering the substep 2.2.3;
substep 2.2.2.1, order
Calculating an analog output vector at time j
Wherein the content of the first and second substances,
substep 2.2.2.2. construction of the connection at time jReceive vector rj＝(rj-p,…,rj,…,rj+q) And difference vector dj＝(dj,-p,…,dj,0,…,dj,q) Wherein, in the step (A),
substep 2.2.2.3, calculating dj,kMean value of
And variance
Substep 2.2.2.4, adding 1 to the variable j, and jumping to substep 2.2.2;
substep 2.2.3, constructing a path separation matrix Ddispart：
Constructing a mean matrix MdispartAnd variance matrix Vdispart：
Further, in sub-step 2.3, the log-likelihood ratio L (x) for each equivalent received valuej,k) Comprises the following steps:
further, in sub-step 2.4, the soft information adaptive fusion obtains a log-likelihood ratio vector l (x) as:
L(x)＝(L(x0),…,L(xj),…,L(xn-1) In a batch process), wherein,
further, in sub-step 2.5, soft information L is outputout(x) Comprises the following steps:
Lout(x)＝(Lout(x0),…,Lout(xj),…,Lout(xn-1))
wherein L isout(xj) Representing decoder output with respect to symbol xjIs defined as a log-likelihood ratio of
rjIs the received value at the j time; if L isout(xj) If the j bit is more than or equal to 0, judging the j bit as 0; otherwise, it is judged as 1.
Further, sub-step 2.7, the estimated codeword
Wherein the criterion of the judgment is
Compared with the prior art, the invention has the beneficial effects that:
the comb filtering Turbo equalization algorithm suitable for short-wave communication adopts a comb filter to separate multipath aliasing signals from a time domain and extract soft information, so that a plurality of versions of useful signals of a user subjected to different fading are obtained, and the diversity effect is realized; the decoder decodes the enhanced signal and feeds back the decoding information to the comb filter; this process iterates over and over. The comb filtering Turbo equalization algorithm does not need matrix inversion operation, greatly reduces the calculation complexity, and has the advantages of simplicity, fast convergence, easy realization and the like.
Drawings
The invention is described in further detail below with reference to the figures and specific embodiments.
FIG. 1 is a block diagram of comb filter Turbo equalization logic in accordance with the present invention;
FIG. 2 is a performance diagram of the comb filtering Turbo equalization algorithm and the MMSE Turbo equalization algorithm of the present invention under different channels; wherein, the graph (a) is a performance graph under a time-invariant channel; figure (b) is a graph of performance under a time-varying channel;
FIG. 3 is a graph of the impact of different iteration times on performance of the comb filter Turbo equalization algorithm of the present invention;
FIG. 4 is a diagram illustrating the effect of different paths extracted by the comb filtering Turbo equalization algorithm on performance.
Detailed Description
Embodiments of the present invention will be described in detail below with reference to examples, but it will be understood by those skilled in the art that the following examples are only illustrative of the present invention and should not be construed as limiting the scope of the present invention.
Referring to the logic block diagram of comb filtering Turbo equalization of fig. 1, equalization is performed in an iterative manner. Firstly, the comb filter separates the multipath signals path by path, and extracts clear and hierarchical multi-path user useful signals
Each signal can extract soft information
Thereby achieving a diversity effect; secondly, all the soft information is merged to obtain L (x), and the L (x) is input into a decoder for decoding; then, based on the soft information L output by the decoderout(x) Constructing average information
Combined with known multipath channel parameters to model the received signal
Finally, the comb filter utilizes the analog received signal
And realityThe received signal r extracts the multipath signal again.
It should be noted that, in fig. 1, the multipath channel parameters are usually obtained by a "channel estimation" module, and the accuracy of estimation directly affects the performance of the Turbo equalization system. Because the invention is mainly focused on the comb filtering Turbo equalization algorithm, the receiving end is assumed to completely acquire the channel parameters for the sake of simplicity, and a Binary Phase Shift Keying (BPSK) modulation mode is adopted to describe the technical scheme.
Specifically, the comb filtering Turbo equalization algorithm suitable for short-wave communication of the invention comprises the following steps:
transmitting end transmits data symbol xprefixAfter the data symbol is transmitted through the multipath channel, the receiving end receives the signal rt。
Specifically, an information transmission model is established
Suppose that the channel has p + q +1 orders, the main path is preceded by p orders and followed by q orders, and the multipath channel parameter at the t-th time is h (t) ═ h-p(t),…,h0(t),…,hq(t)). In view of the complexity of the actual communication environment, the communication system must have multipath resistance. Therefore, guard bands of length l are added before and after the transmission of information, i.e.
xprefix＝(x-l,…,x-1,x0,x1,…,xj,…,xn-1,xn,…,xn+l-1)
Wherein (x)-l,…,x-1) As a front guard band, (x)n,…,xn+l-1) A rear guard band. The front and back guard bands may be any sequence of + -1, and their lengths may be different, but they should be equal to or greater than p + q. For convenience, the lengths of the front and back guard bands are equal, are all l (l is more than or equal to p + q) and are all +1 vectors. The mathematical expression of the received value at the t-th time can be known from the parameters as
rt＝h-p(t)xp+t+…+h-1(t)x1+t+h0(t)xt+h1(t)x-1+t+…+hq(t)x-q+t+wt，(-p<＝t<n+q)
Wherein, wtAnd the sampling value of the t moment follows normal distribution with the mean value of 0 and the variance of w-is obtained. For the sake of convenience in describing the comb filter principle, j (0 ≦ j) is defined here<n) the received vector r at time instantj＝(rj-p,…,rj,…,rj+q)。
a) the comb filter principle is as follows:
the core part of the invention is a comb filter, which has the capability of separating multipath aliasing signals path by path and extracting clear and hierarchical user information. This section will describe the working principle of the comb filter in detail. Suppose the user soft information vector output by the decoder is Lout(x) (this information is not completely correct), the jth mapping symbol is taken as an example to illustrate how to extract the equivalent output of each path. The extraction method comprises the following steps:
1. soft information vector L based on decoder outputout(x) Constructing average information
And variance information
wherein, tanh represents hyperbolic tangent;
is reflected by the decoder pair symbol xjWhen the estimation is very accurate, there is
Denotes the symbol xjEvaluating the degree of deviation (dispersion) of the results if
Larger indicates greater dispersion.
2. Structure of j (0. ltoreq. j)<n) analog output vector at time instant
Order to
Respectively calculating the analog output at the j-p, …, j, … j + q time
3. Structure of j (0. ltoreq. j)<n) time rjAnd
difference vector d ofj＝(dj,-p,…,dj,0,…,dj,q)
From the above expressions, it can be seen that the mathematical expressions of all differences have the same form, consisting of three parts.
Wherein the first part is the received value obtained by transmitting (fading) the j-th bit information of the user through a certain path, such as h-p(j-p)xjDenotes xjVia channel component h-p(j-p) the obtained reception value; h is0(j)xjDenotes xjVia channel component h0(j) The obtained received value; h isq(j+q)xjDenotes xjVia channel component hq(j + q) the resulting received value, and so on.
The second part is the interference of other paths to the first term (useful signal), and it can be seen from the expression that the interference is smaller if the average information output by the decoder is more reliable.
The third part is the channel-induced noise sample values.
For the first portion, the second portion and the third portion are both interfering. So that each difference dj,kShould be considered as a random variable.
The first part is dj,kIs taken to be the ideal value (i.e., the value that would be obtained without interference). The sum of the second part and the third part is dj,kError (degree of deviation) from the ideal value.
The ideal value only corresponds to the transmitted symbol xjAnd channel characteristics h at time j + kk(j + k). For comb filters xjIs an unknown quantity, while the symbol x is not known when calculating the log-likelihood ratio of the jth mapped symboljSo that the channel characteristic h at the j + k th time can be obtainedk(j + k) as an ideal value (mean), i.e.
Average value in general
Sum variance
Respectively as follows:
4. run through the index variable j, (0)<＝j<n) to obtain a difference matrix Ddispart：
On average of information
Difference matrix D under very accurate and very small channel superposition noisedispartIs equivalent to the symbol xjRespectively pass through the path h-p(j-p),…,h0(j),…,hqThe transmission result of (j + q) (this can be taken from d)j,kSee in mathematical expression of (a) i.e.:
based on this, a difference matrix DdispartReferred to as a path separation matrix.
Meanwhile, the mean matrix M after path separation can be obtaineddispartAnd variance matrix Vdispart：
Matrix Ddispart、Mdispart、VdispartThe dimensions of (A) are (p + q +1) × n.
The 4 steps described above are similar to "combing" the aliased signal r with a "comb", extracting the signal (viewed from the column) from the time domain after each symbol has been transmitted (faded) via various paths. It can also be seen as a comb filter that transforms aliased multipath time-varying channels into a parallel (as observed from the row) of separate channels. The comb filter is used for separating multipath aliasing signals from a time domain to obtain a plurality of noisy versions of user useful signals, and is similar to the method for combing the multipath signals by using a special comb to extract clear and layered user signals.
b) The signal separation algorithm is as follows:
in general, the multipath channel parameter h (t) ═ h is known-p(t),…,h0(t),…,hq(t)), white Gaussian noise variance
Soft information vector L output by decoderout(x)＝(Lout(x0),…,Lout(xj),…,Lout(xn-1) R ═ r (r), received value after multipath channel transmission-p,…r-1,r0,…,rn-1,…,rn+q-1) (ii) a The target is as follows: the aliased signals are separated path by path. The specific algorithm for separating the aliased signals path by path is as follows:
1. soft information vector L based on decoder outputout(x) Constructing average information
Sum variance information
Wherein the content of the first and second substances,
2. let j equal 0. The following loop is entered when j < n:
2.1. order to
Calculating an analog output vector at time j
Wherein
2.2. Constructing a received vector r at time jj＝(rj-p,…,rj,…,rj+q) And difference vector dj＝(dj,-p,…,dj,0,…,dj,q) Wherein, in the step (A),
2.3. calculating dj,kMean value of
And variance
2.4. Adding 1 to the variable j, and jumping to the substep 2;
3. constructing a path separation matrix Ddispart：
Constructing a mean matrix MdispartAnd variance matrix Vdispart：
c) Multipath soft information fusion
Separating received aliasing signals path by path to obtain a separation matrix DdispartThen, based on the mean matrix MdispartAnd variance matrix VdispartComputingThe corresponding log-likelihood ratio L (x) is obtainedj,k)。
Because the log-likelihood ratio vector corresponding to each path contains the same information, the log-likelihood ratio vector of any path can be independently input into a decoder for decoding. It should be noted that although the log-likelihood ratio vectors corresponding to each path contain the same information, their reliability (reliability) is different, and some have higher reliability and some have lower reliability. Intuitively speaking, the reliability is related to the corresponding channel coefficient, and the reliability with high channel coefficient is high.
However, to obtain better decoding performance, the log-likelihood ratio vectors of the paths are usually added/fused. Multipath soft information fusion generally employs a direct addition method, as follows:
d) comb filtering Turbo equalization algorithm
The fused soft information is used as the input of a decoder to carry out a new round of decoding, thereby obtaining a more accurate soft information vector Lout(x) And then fed back to the comb filter to separate the aliased signals again. Therefore, the following comb filtering Turbo equalization algorithm is available:
it is known that: multipath channel parameter h (t) ═ h-p(t),…,h0(t),…,hq(t)) and a white gaussian noise variance w-;
initial soft information vector Lout(x)＝(Lout(x0),…,Lout(xj),…,Lout(xn-1))；
Receiving value r after multipath channel transmission-p,…r-1,r0,…,rn-1,…,rn+q-1)。
The target is as follows: turbo equalizing the received multipath signalTo codeword estimation
0. Initializing, setting an iteration variable L to be 0, and setting the maximum iteration number L of the joint decoding;
1. construction of a path separation matrix D based on a signal separation algorithmdispartMean matrix MdispartAnd variance matrix Vdispart；
2. Calculating a log-likelihood ratio L (x) for each equivalent received valuej,k)
3. The soft information self-adaptive fusion obtains a log-likelihood ratio vector L (x) ═ L (x)0),…,L(xj),…,L(xn-1) In a batch process), wherein,
4. sending L (x) to decoder for decoding and outputting soft information Lout(x)；
Let the soft information vector output by the decoder be Lout(x)＝(Lout(x0),…,Lout(xj),…,Lout(xn-1)). Wherein, the symbol Lout(xj) Representing decoder output with respect to symbol xjIs defined as a log-likelihood ratio of
Wherein r isjIs the received value at time j. If L isout(xj) If the j bit is more than or equal to 0, judging the j bit as 0; otherwise, it is judged as 1. Obviously, if Lout(xj) The larger the | is, the more reliable the result of the determination is.
5. Adding 1 to an iteration variable L, if L is less than L, entering the step 1, and otherwise, entering the step 6;
6. judging the soft information output by the decoder to obtain the decoded code word
The decision criterion is
At the initial moment of iteration, the decoder cannot know any soft information about the user and cannot provide any useful information to the mean information constructing module, and the initial soft information vector is usually set to be an all-0 vector, namely Lout(x)＝(0,…,0,…,0)。
It should be emphasized and pointed out that the present invention only uses BPSK modulation as an example to illustrate the comb filtering Turbo equalization algorithm and the main idea, and the idea of comb filtering is also applicable to other modulation schemes (such as QPSK, 8PSK, etc.).
Performance simulation
In order to verify the performance of the comb filtering Turbo time domain equalization algorithm provided by the invention, the LDPC code is used as an error correcting code to simulate the BPSK modulation baseband. The signal-to-noise ratio in simulation is defined as
The unit is dB. Wherein E issFor the average symbol energy received,
is gaussian white noise variance. Generally let E in simulation s1. The simulation mainly looks at the following 3 points:
A. the performance of the comb filtering Turbo time domain equalization algorithm in a multipath environment;
B. the influence of the iteration times on the time domain equalization performance of the comb filtering Turbo;
C. the influence of the number of extraction paths on the overall performance.
All simulation parameters were set as follows:
LDPC code: selecting an LDPC code (516,1032) adopted by a mobile broadband wireless access standard (IEEE 802.16e), wherein the code rate is 0.5, the base matrix is 12 multiplied by 24, and the expansion factor is 43; the LDPC decoding Algorithm is Sum-Product decoding Algorithm (SPA), and the iteration number is 50;
baud rate (transmit symbol rate): 2000 symbols/sec.
Note that the time-invariant channel means that the channel characteristics at each moment are the same and do not change with time variation;
the time-varying channel means that the channel characteristics are different at each time, but the characteristics of the two previous and next times have certain correlation. The double-diameter 2ms means that the interval of two channels is 2ms before and after the existence of the two channels; fading 1Hz refers to doppler frequency shift caused by relative motion of transmitting and receiving dual-emission, and the doppler spectrum has a width of 1Hz, and the wider the width, the more severe the channel changes.
It should be noted that, the following simulation focuses on investigating the performance of the Turbo equalization algorithm, so that, assuming that the receiving end knows the parameters of the multipath channel, the channel estimation algorithm can be used to obtain the parameters of the multipath channel in practical applications.
Simulation A
And (5) inspecting the performance of the comb filtering Turbo time domain equalization algorithm under the channels 1 and 2. For comparison, the paper "Turbo Equalization" is also given: performance of MMSE Turbo equalization algorithm described in Principles and New Results, with transversal filter order of 25. The joint iteration times of the two algorithms are both 5 times; the simulated performance is shown in fig. 2.
As can be seen from fig. 2, both equalization algorithms have nearly the same performance in either a time-invariant or a time-variant channel. For example, when BER is 10-5Then, the required signal-to-noise ratios of the MMSE Turbo equalization algorithm and the comb filter Turbo equalization algorithm under the time-invariant channel are 2.13dB and 2.16dB, respectively, as shown in fig. 2 (a); the required signal-to-noise ratios are 12.2dB and 11.9dB, respectively, for time-varying channels, as shown in fig. 2 (b). However, the MMSE Turbo equalization process involvesAnd the inverse operation to the channel cyclic matrix, and the calculation amount is increased along with the increase of the order of the transverse filter, and the complexity and the calculation amount are far higher than those of a comb filtering Turbo equalization algorithm. Therefore, the comb filtering Turbo equalization algorithm can greatly reduce the complexity of engineering implementation on the premise of hardly losing performance.
Simulation B
And (5) examining the influence of the iteration times on the balance performance of the comb filtering Turbo. As shown in fig. 3, the effect of different iteration numbers of the algorithm on the performance under channel 1 is shown. It can be seen from the figure that:
1) the performance of the comb filtering Turbo equalization algorithm is gradually improved along with the increase of the iteration times;
2) the convergence speed of the comb filtering Turbo equalization algorithm is high, and the stable performance state can be basically achieved only by 2 iterations. For example, when BER is 10-5The signal to noise ratio required for 2 and 5 iterations is 2.15 and 2.19dB, respectively, with only a 0.04dB separation.
Simulation C
As described above, the comb filter separates the multipath aliasing signals path by path, and extracts a plurality of noisy versions of the useful signal, and because the information contained in each path is the same, the log-likelihood ratio vector of any path can be independently input into a decoder for decoding. The simulation examined the effect of extracting the number of different paths under channel 1 on the overall performance. The number of extraction is 1 path (path 2), 2 paths (paths 1 and 2) and all paths. The simulation results are shown in fig. 4. It can be seen from the figure that:
1) the performance is better as the number of extraction paths is larger. For example, when BER is 10-5Then, the signal-to-noise ratios required for extracting 1 path, 2 paths and all paths are respectively 4.40dB, 2.85dB and 2.15 dB;
2) the comb filter can effectively separate aliasing signals so as to obtain a plurality of noisy versions of useful signals, and the diversity effect is realized. For example, the performance of extracting 2 paths for combining in the comb filter is far better than that of extracting 1 path, and the BER is 10-5When the temperature of the water is higher than the set temperature,there is about a 1.55dB difference between the two. When all paths are extracted, the comb filtering Turbo equalization algorithm achieves the optimal performance.
Although the present invention has been described in detail in this specification with reference to specific embodiments and illustrative embodiments, it will be apparent to those skilled in the art that modifications and improvements can be made thereto based on the present invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (9)
1. A comb filtering Turbo equalization algorithm suitable for short-wave communication is characterized by comprising the following steps:
step 1, setting p + q +1 orders of short wave communication channel, wherein, p orders are arranged in front of a main path, q orders are arranged behind the main path, and the multi-path parameter at the t moment is h (t) ═ h-p(t),…,h0(t),…,hq(t)), the code word of the user message u to be transmitted after channel coding is c ═ c0,c1,…,cj,…,cn-1),cjE {0,1}, and then performing high-order modulation to obtain a mapping symbol x ═ x (x)0,x1,…,xj,…,xn-1) (ii) a Wherein x isj＝1-2cj,(xj± 1), n is the codeword length;
transmitting end transmits data symbol xprefixAfter the data symbol is transmitted through a multipath channel, a receiving end receives a signal r;
step 2, the receiving end adopts the comb filtering Turbo equalization algorithm to equalize and decode the received signal and output the estimated code word
2. The comb filtering Turbo equalization algorithm for short wave communication according to claim 1, wherein in step 1, the high order modulation is BPSK modulation, QPSK modulation, 8PSK modulation.
3. The comb filtering Turbo equalization algorithm for short wave communication according to claim 1, wherein in step 1, the data symbol x isprefixComprises the following steps:
xprefix＝(x-l,…,x-1,x0,x1,…,xj,…,xn-1,xn,…,xn+l-1)
wherein (x)-l,…,x-1) As a front guard band, (x)n,…,xn+l-1) The front guard band and the rear guard band are arbitrary sequences formed by +/-1, and the lengths of the front guard band and the rear guard band are respectively more than or equal to p + q;
the mathematical expression of the receiving value of the receiving signal at the t-th moment is as follows:
rt＝h-p(t)xp+t+…+h-1(t)x1+t+h0(t)xt+h1(t)x-1+t+…+hq(t)x-q+t+wt，(-p<＝t<n+q)
wherein, wtAnd the sampling value of the t moment follows normal distribution with the mean value of 0 and the variance of w-is obtained.
4. The comb filtering Turbo equalization algorithm for short wave communication according to claim 1, wherein the step 2 comprises the following sub-steps:
substep 2.1, initialization: setting the current iteration number L as 0, setting the maximum iteration number L of the joint decoding, and setting the soft information vector output by the initial decoder as a full 0 vector, namely Lout(x)＝(0,…,0,…,0)；
Substep 2.2, separating the received aliasing signals path by path based on a signal separation algorithm to obtain a path separation matrix DdispartAnd constructing a mean matrix MdispartAnd variance matrix Vdispart；
Substep 2.3, calculating the log-likelihood ratio L (x) for each equivalent received valuej,k)；
Substep 2.4, obtaining a log-likelihood ratio vector L (x) by soft information self-adaptive fusion;
substep 2.5, sending log-likelihood ratio vector L (x) to decoder for decoding, and outputting soft information Lout(x)；
Substep 2.6, adding 1 to an iteration variable L, and entering substep 2.2 if L is less than L; otherwise, entering the substep 2.7;
substep 2.7 of decoding the soft information L output by the decoderout(x) Making decision to obtain decoding estimation code word
5. Comb filtering Turbo equalization algorithm for short wave communication according to claim 4 characterized by the sub-step 2.2 comprising the sub-steps of:
substep 2.2.1, based on the soft information vector L output by the decoderout(x) Constructing average information
Sum variance information
Wherein the content of the first and second substances,
wherein, tanh represents hyperbolic tangent;
is reflected by the decoder pair symbol xj(ii) an estimate of (d);
substep 2.2.2, let the soft information bit number j equal to 0, if j < n, go to substep 2.2.2.1; otherwise, entering the substep 2.2.3;
substep 2.2.2.1, order
Calculating an analog output vector at time j
Wherein the content of the first and second substances,
substep 2.2.2.2. construction of a received vector r at time jj＝(rj-p,…,rj,…,rj+q) And difference vector dj＝(dj,-p,…,dj,0,…,dj,q) Wherein, in the step (A),
substep 2.2.2.3, calculating dj,kMean value of
And variance
Substep 2.2.2.4, adding 1 to the variable j, and jumping to substep 2.2.2;
substep 2.2.3, constructing a path separation matrix Ddispart：
Constructing a mean matrix MdispartAnd variance matrix Vdispart：
6. The method of claim 5The comb filtering Turbo equalization algorithm suitable for short-wave communication is characterized in that in the substep 2.3, the log likelihood ratio L (x) of each equivalent received valuej,k) Comprises the following steps:
7. the comb filtering Turbo equalization algorithm for short wave communication according to claim 6, wherein in sub-step 2.4, the soft information adaptive fusion obtains a log-likelihood ratio vector L (x) as:
L(x)＝(L(x0),…,L(xj),…,L(xn-1) In a batch process), wherein,
8. the comb filtering Turbo equalization algorithm for short wave communication according to claim 1, wherein in sub-step 2.5, soft information L is outputtedout(x) Comprises the following steps:
Lout(x)＝(Lout(x0),…,Lout(xj),…,Lout(xn-1))
wherein L isout(xj) Representing decoder output with respect to symbol xjIs defined as a log-likelihood ratio of
rjIs the received value at the j time; if L isout(xj) If the j bit is more than or equal to 0, judging the j bit as 0; otherwise, it is judged as 1.
9. The comb filtering Turbo equalization algorithm for short wave communication according to claim 8, wherein sub-step 2.7, the estimated codeword
The judgment criterion is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010749832.9A CN111835365A (en) | 2020-07-30 | 2020-07-30 | Comb filtering Turbo equalization algorithm suitable for short-wave communication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010749832.9A CN111835365A (en) | 2020-07-30 | 2020-07-30 | Comb filtering Turbo equalization algorithm suitable for short-wave communication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111835365A true CN111835365A (en) | 2020-10-27 |
Family
ID=72920422
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010749832.9A Pending CN111835365A (en) | 2020-07-30 | 2020-07-30 | Comb filtering Turbo equalization algorithm suitable for short-wave communication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111835365A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3991277A (en) * | 1973-02-15 | 1976-11-09 | Yoshimutsu Hirata | Frequency division multiplex system using comb filters |
| WO2007137484A1 (en) * | 2006-05-11 | 2007-12-06 | Shanghai Jiao Tong University | A channel estimation method and the device thereof |
| CN101741786A (en) * | 2009-12-18 | 2010-06-16 | 中国人民解放军理工大学 | Ultra-broadband receiver for digital communication system and signal processing method thereof |
| CN110572340A (en) * | 2019-09-26 | 2019-12-13 | 西安烽火电子科技有限责任公司 | turbo time domain equalization method for short wave communication |
-
2020
- 2020-07-30 CN CN202010749832.9A patent/CN111835365A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US3991277A (en) * | 1973-02-15 | 1976-11-09 | Yoshimutsu Hirata | Frequency division multiplex system using comb filters |
| WO2007137484A1 (en) * | 2006-05-11 | 2007-12-06 | Shanghai Jiao Tong University | A channel estimation method and the device thereof |
| CN101741786A (en) * | 2009-12-18 | 2010-06-16 | 中国人民解放军理工大学 | Ultra-broadband receiver for digital communication system and signal processing method thereof |
| CN110572340A (en) * | 2019-09-26 | 2019-12-13 | 西安烽火电子科技有限责任公司 | turbo time domain equalization method for short wave communication |
Non-Patent Citations (1)
| Title |
|---|
| 张凯: "短波衰落信道下Turbo 均衡算法研究", 《无线电通信技术》, vol. 25, no. 3, 15 March 2019 (2019-03-15), pages 296 - 300 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7583763B2 (en) | Multi input multi output wireless communication reception method and apparatus | |
| Davis et al. | Joint MAP equalization and channel estimation for frequency-selective and frequency-flat fast-fading channels | |
| US7333540B2 (en) | Equalisation apparatus and methods | |
| US20070248151A1 (en) | Inter-carrier interference cancellation method and receiver using the same in a MIMO-OFDM system | |
| EP1135908A1 (en) | Adaptive channel characterization using decoded symbols | |
| EP1217798B1 (en) | Adaptive equalization method and adaptive equalizer | |
| CN110572340A (en) | turbo time domain equalization method for short wave communication | |
| Pukkila | Channel estimation modeling | |
| CN112866151B (en) | Underwater sound MPSK signal blind Turbo equalization method based on channel blind estimation | |
| GB2407007A (en) | Adaptive Space Time Decoding using soft decisions | |
| US20060156207A1 (en) | Signal decoding methods and apparatus | |
| Kong et al. | Sequence detection and channel state estimation over finite state Markov channels | |
| Omidi et al. | Joint data and Kalman estimation for Rayleigh fading channels | |
| CN107171990B (en) | single carrier signal time domain equalization method based on fractional interval and iterative algorithm | |
| CN109547369B (en) | Self-adaptive channel parameter estimation method | |
| CN111835365A (en) | Comb filtering Turbo equalization algorithm suitable for short-wave communication | |
| CN111901262A (en) | High-order modulation Turbo time domain equalization algorithm suitable for short-wave communication | |
| Kalaycioğlu et al. | A robust threshold for iterative channel estimation in OFDM systems | |
| JP4187968B2 (en) | Adaptive equalization method and adaptive equalizer | |
| Özbek et al. | Pilot-symbol-aided iterative channel estimation for OFDM-based systems | |
| Wo et al. | Graph-based iterative Gaussian detection with soft channel estimation for MIMO systems | |
| Matsumoto | Iterative (turbo) signal processing techniques for MIMO signal detection and equalization | |
| Si et al. | An Iterative Receiver for Coded Faster-than-Nyquist Signaling over Rayleigh Fading Channels | |
| Mao et al. | Soft decision directed dual-layer channel estimation for time-varying MIMO channels | |
| Mai et al. | Iterative channel estimation for turbo equalization |
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 |