WO2008098417A1 - Procédé de multiplexage codé et de transmission multi-adresse - Google Patents

Procédé de multiplexage codé et de transmission multi-adresse Download PDF

Info

Publication number
WO2008098417A1
WO2008098417A1 PCT/CN2007/000536 CN2007000536W WO2008098417A1 WO 2008098417 A1 WO2008098417 A1 WO 2008098417A1 CN 2007000536 W CN2007000536 W CN 2007000536W WO 2008098417 A1 WO2008098417 A1 WO 2008098417A1
Authority
WO
WIPO (PCT)
Prior art keywords
coding
matrix
parallel
code
coding matrix
Prior art date
Application number
PCT/CN2007/000536
Other languages
English (en)
Chinese (zh)
Inventor
Daoben Li
Original Assignee
Daoben Li
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 Daoben Li filed Critical Daoben Li
Priority to PCT/CN2007/000536 priority Critical patent/WO2008098417A1/fr
Priority to CN200780050698.6A priority patent/CN101632248B/zh
Publication of WO2008098417A1 publication Critical patent/WO2008098417A1/fr

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/004—Arrangements for detecting or preventing errors in the information received by using forward error control
    • H04L1/0056—Systems characterized by the type of code used
    • H04L1/0059—Convolutional codes

Definitions

  • the present invention relates to the field of mobile communications, and in particular, to the problem of parallel multiplexing or multiple access transmission in the field of mobile communications, and in particular to a method for coding multiplexing and multiple access transmission .
  • ITU International Telecommunication Union
  • the basic information theory states: When there is a "channel capacity" after the channel is given, that is, the maximum transmission rate C, the actual system's transmission rate can only be approached to C and cannot be exceeded (:. But this is only for single source and Single-address user case. When it comes to the sharing problem of channel capacity by multi-source or multi-address users, the basic information theory also points out that parallel multiplexing (or multiple access) transmission is an important factor to improve system capacity and frequency efficiency. The only way to have the best “Waveform Division Multiplexing (Multiple Access)" method. Because of the "waveform, it is usually generated by code, usually called code or Code division multiplexing (or multiple access).
  • CDM code division multiple access CDMA
  • code division multiplexing or code division multiple access CDMA
  • it can be regarded as a kind of multiplexing (or multiple access) transmission technology of the coding domain
  • the orthogonal relationship between the address codes and the internal pursuit is no coding constraint relationship.
  • it should belong to a multiplexing (or multiple access) approach that uses waveform orthogonality to allocate (decompose) channel capacity.
  • It belongs to code multiplexing (or multiple access) without coding gain. Strictly speaking, it is quite different from the true code multiplexing (or multiple access) property of the present invention which can share channel capacity.
  • Time division multiplexing TDM (or time division multiple access TDMA). It is a multiplexing (or multiple access) transmission technique in the time domain, but unfortunately there is no overlap between its sub-slots, and the pursuit is an orthogonal relationship with no coding constraints. In theory, it still belongs to a multiplexing (or multiple access) method that uses waveform orthogonality to allocate (decompose) channel capacity, and also belongs to simple code multiplexing (or multiple access) without coding gain.
  • Frequency division multiplexing FDM Frequency Division Multiple Access (FDMA)
  • OFDM Orthogonal Frequency Division Multiplexing
  • FDMA Frequency Division Multiple Access
  • OFDM Orthogonal Frequency Division Multiplexing
  • FDMA Frequency Division Multiple Access
  • OFDM Orthogonal Frequency Division Multiplexing
  • OFDM Orthogonal Frequency Division Multiplexing
  • OFDM OFDM
  • OFDM orthogonal frequency division multiplexing
  • Physical space division multiplexing SDM or physical space division multi-address SDMA
  • statistical space division multiplexing also known as multi-input, multi-out, multi-antenna in rich scattering environment
  • SDM Physical space division multiplexing
  • SDM physical space division multi-address SDMA
  • statistical space division multiplexing also known as multi-input, multi-out, multi-antenna in rich scattering environment
  • MIMO transmission technology Although they can greatly improve the capacity and spectral efficiency of the system, they are very demanding on the conditions of channel propagation.
  • the former requires that the channel must exhibit poor scattering, or the angular spread of the channel is very small, close to 0°.
  • the latter requires that the channel must exhibit rich scattering, or that the angular spread of the channel is very large, close to 36 0 ⁇ . Otherwise their air separation gain will increase with the channel angle, the latter will follow the letter The corner spread is reduced and gradually reduced until it is lost.
  • a multiplexing technique in which all of the plurality of sources share one channel is a linear code multiplexing (multi-access) having a constraint length of 1 and no coding gain, and
  • CDM CDMA
  • the traditional multi-level modulation techniques with high-frequency transmission efficiency such as pulse amplitude modulation PAM, multi-phase modulation PM, quadrature amplitude modulation QAM, partial response modulation, etc., are in fact all higher than the code rate.
  • a constraint length of 1 simple linear coding multiplexing without coding gain.
  • the steady state number of the system is common . They are: ⁇ +, + ⁇ ; ⁇ +, -); ⁇ - + ⁇ ; (- -); The initial and final states are all zero states, ie ⁇ 0,0 ⁇ , and the system has no transition state.
  • Each stable (including initial) state can be shifted to all other four stable states, and the four stable states can be transferred to the final state ⁇ 0, 0 ⁇ .
  • the corresponding Trdlis diagram is shown in Figure 1.
  • ⁇ ⁇ x) ⁇ , ⁇ ui ⁇ x) ⁇ contains 4 elements, namely:
  • the coded output set ⁇ v ° ) ⁇ , ⁇ Vl also has 4 elements, namely:
  • Parallel coding structure Coded output: ( 6 ) Obviously, after such parallel coding, if there is a one-to-one correspondence between the input and output of the code, the spectral efficiency of the coded output is doubled compared with the coded input. The number of steady state systems is common
  • the initial and final states are all ⁇ 0,0,0,0 ⁇ ; there are four types of pre-transition states, namely:
  • the initial state ⁇ 0,0,0,0 ⁇ can be transferred to all four pre-transition states ⁇ 0,0,X,X ⁇ .
  • the pre-transition and each steady state can be transferred to the other four stable states, every 4
  • the steady state can be transferred to the same post-transition state ⁇ X, X, 0, 0 ⁇ , and the four post-transition states can only be transferred to the final state ⁇ 0, 0, 0, 0 ⁇ . Move. Its corresponding Trellis diagram is shown in Figure 2.
  • One object of the present invention is to provide a coding multiplexing transmission method which uses code rate high data transmission, thereby greatly improving system capacity and frequency borrowing efficiency.
  • a code multiplexing transmission method for performing coded multiplexing transmission using parallel linear or nonlinear coding with a code rate higher than one When the code rate of the parallel coding is higher than 1, there is a corresponding relationship between the parallel coded input sequence and the parallel coded output sequence.
  • the parallel coding described includes: any code other than a linear or nonlinear generalized or narrowly concatenated convolutional convolutional code with a code rate higher than one)
  • the parallel coding tap coefficients are: samples of complex or real Gaussian distribution random variables; or samples of uniformly distributed random variables within a certain range of complex or real planes; or ⁇ +, ⁇ , ⁇ ⁇ , ⁇ , ⁇ or other finite Non-integral real numbers; or parallel coding tap coefficients are located on: the unit circle or the real or imaginary axis or the real and imaginary axes.
  • Each of the coding tap coefficients is in a different or mutually rotating or mutually misaligned domain; the domain refers to: different or mutually rotated or mutually misaligned coding domains, different or mutually rotated or mutually misaligned A spatial domain, a frequency domain that is different or mutually rotated or has a misalignment with each other, a time domain that is different or overlapped with each other, or a mixed domain of the above various domains.
  • Coded taps with parallel coding polynomials of mutual impedance coefficients shall ensure that the code they encode has the greatest free separation.
  • the code multiplexing transmission method includes the following steps: constructing an encoding matrix B, the matrix B containing K coding vectors; forming K-channel data for parallel transmission, the K-channel data corresponding to the K coding vectors; One-way data is linearly or nonlinearly convolutionally encoded with the code vector corresponding to the path data, and the coding constraint length is L; the linear or nonlinear convolutional coding result of the K-way data is added to obtain an N-dimensional coded output vector; The N-dimensional coded output vector is described and detected; wherein, the K, N, and L are parallel coding basic parameters.
  • Another object of the present invention is to provide a multiple access transmission method, which uses a multi-address transmission of data with a code rate higher than 1, so as to greatly improve the capacity and spectral efficiency of the system.
  • the parallel coding tap polynomials ( x ), o, ⁇ ..., - L are linearly independent.
  • the parallel encoding includes: linear or nonlinear generalized or narrow convolutional coding with a code rate higher than 1, or other types of generalized or narrowly parallel coding with a code rate higher than one.
  • the parallel coding tap coefficients are: samples of complex or real Gaussian distributed random variables; or samples of uniformly distributed random variables within a certain range of complex or real planes; or ⁇ + , ⁇ , ⁇ ⁇ , ⁇ , ⁇ ,...) Or other finite non-holonomic real numbers; or parallel coding tap coefficients are located on: a unit circle or a real or imaginary axis or a real axis and an imaginary axis.
  • Each of the coding tap coefficients is in a different or mutually rotating or mutually misaligned domain; the domain refers to: different or mutually rotated or mutually misaligned coding domains, different or mutually rotated or mutually misaligned Spatial domains, different or mutually overlapping or mutually misaligned frequency domains, different or overlapping time domains of each other or a mixed domain of the above various domains.
  • An encoding tap with a parallel coding polynomial of linear uncorrelated coefficients should ensure that the code it has has the largest free distance.
  • the method comprises the steps of: constructing a linear or non-linear coding matrix B, the matrix B comprising K coding vectors; forming K-channel data for parallel transmission, the K-channel data corresponding to the K coding vectors;
  • the one-way data is convolutionally encoded with the coding vector corresponding to the path data, and the coding constraint length is L; the convolutional coding result of the K-way data is added to obtain an N-dimensional coded output vector; and the N-dimensional coded output vector is received and The test is performed; wherein, the K, N, and L are parallel codes 1: .
  • the present invention only considers the destructive factors external to the system as interference, and any overlap between signals (symbols) within the system is regarded as a coding constraint relationship. And the present invention uses parallel coding with a code rate greater than 1 for data transmission, by finding a correspondence between the input sequence and the output sequence. System, realizes parallel transmission of multiple data, greatly improving system capacity and spectrum efficiency.
  • Example 1 is a coded trellis diagram of Example 1 in the prior art
  • Example 2 is a coded trellis diagram of Example 2 in the prior art
  • FIG. 3 is a simulation diagram of an embodiment of the present invention.
  • Figure 5 is a diagram showing the coding structure of Figure 3.
  • the present invention employs generalized or narrowly concatenated convolutions with code rates above one or other types of generalized or narrowly parallel coding to substantially increase system capacity and spectral efficiency.
  • the overlapping convolutional code multiplexing (multiple access) of the present invention employs the following generalized convolutional coding operations:
  • V n F ⁇ [ ⁇ U ⁇ BJ ⁇
  • each element in the Amy bb set has its corresponding negative - w , conjugate, negative conjugate one element, the size (total number of elements) is 2Q (Q is the bit loaded by each symbol number);
  • the size is 2QK, which is a KQ heavy binary data set or Q heavy Amy set;
  • This is a generalized nonlinear vector convolutional coding model. It is a typical parallel convolutional code with a code rate of ⁇ / ⁇ when ⁇ ⁇ ⁇ , > 1, and it is a typical code rate when L l.
  • TDM Time Division Multiple Access TDMA
  • FDM Frequency Division Multiplexing
  • OFDM Orthogonal Frequency Division Multiplexing
  • CDM Code Division Multiple Access CDMA
  • L coding constraint length
  • the linear coding multiplexing system of the present invention is the overlapping time division multiplexing OVTDM system proposed in the prior invention patent.
  • Model [3] Similarly, if the encoding is performed in the spatial or frequency domain, the linear coding multiplexing system of the present invention is the overlapping space division multiplexing OVSDM or overlapping frequency division multiplexing OVFDM system model proposed in the prior invention patent [ 4].
  • CDM CDMA
  • TDM TDM
  • FDMA FDM
  • OFDM OFDM
  • SDMA OFDM
  • MIMO MIMO
  • symbol interference [SI channel , O
  • code multiplexing (multi-access) is better because its coding constraint length is greater than 1, especially its swing space is much larger, and the best performance in the corresponding domain should of course be excellent.
  • the polynomial representation of linear overlapping convolutional coded multiplexing (multiple access) is as follows:
  • This coding model like traditional convolutional coding, can also be represented by polynomials as follows:
  • overlapping convolutional code multiplexing must also require a correspondence between the input sequence and the output sequence, ie an output sequence cannot be combined with two or More than two input sequences correspond, and vice versa.
  • code rates are less than or equal to 1
  • there is no doubt that there is a one-to-one correspondence unless a bad coding vector such as all 0s is selected).
  • ⁇ since the total code rate is also less than or equal to 1, there is a corresponding relationship should also be unquestionable (unless a bad coding matrix such as 0 or a line of dissatisfaction is selected).
  • the overlapping convolutional coding system has a stable state of 2QK when expressed in binary, which are:
  • the initial and final states of the I KQ(L- ⁇ ) , KQ(L-l) , , KQ(L-l) J systems are all zero (0) states, ie
  • the state of the front (left) side of the state ⁇ 1, 2 ,..., - 2)
  • the binary data is all 0s is called the pre-transition state
  • the state in which l, 2, ... - 2) binary data is all 0 is called the post transition state.
  • the front transition state and the steady state can be transferred to other 2 QK pre-transition or steady state
  • the steady state and the post-transition state can only be transferred from the front 2 QI "steady state or post-transition state, initial state and pre-transition state.
  • the state can only be shifted backwards, and the final state and the post-transition state can only be transferred from the front.
  • the state transition relationship is:
  • a , b , —, c , d , e are all K-dimensional vectors (QK-dimensional binary vectors) containing K-dimensional 0 vectors (depending on the pre-transition or post-transition state), a / 1 ⁇ 6 indicates the new input
  • the K-dimensional vector, a new K-dimensional vector input inevitably leads to the departure of one of the oldest (early) K-dimensional vectors, and the representation of a corresponding branch code (tributary metric).
  • the system can be greatly improved by using convolutional code multiplexing (multiple access) with a code rate higher than 1. Frequency efficiency.
  • ⁇ u k (x) ⁇ is a set of symmetric functions
  • ⁇ ⁇ W ⁇ is also necessarily a set of symmetric functions, so it is sufficient to study only the semi-sets of ⁇ ⁇ ⁇ in the set ⁇ ⁇ ⁇ , because the set ⁇ ⁇ ⁇ middle The other elements must be negative for the elements in ⁇ ⁇ ⁇ +.
  • Voo(x) l + 2x + 3x 2 + 2x 3 + x 4
  • voi(x) l + 2x + 2 -x 4
  • Vn (x) 1 ⁇ e J2nl3 x - 2x 2 + ⁇ e j2 ⁇ - ⁇ /3 ) ⁇ 3 - ⁇ /3 ⁇ 4
  • Vi 2 ( ) 1 + (e J27l3 - l)x + (1 + e ⁇ j2Kl ' ⁇ e ⁇ ' ⁇ x 1 + f ⁇ e' ⁇ ' ⁇ x 3 + ⁇ 2 ⁇ /3 ⁇ "
  • ⁇ 3 ( ) 1—(1 ⁇ e J2!tl3 )x ⁇ (e ⁇ /3 ⁇ ⁇ 31 ⁇ 1 ' + 1) ⁇ 2 + ⁇ 3 - e Jlnl
  • V2 (x) 1 - (1 - e ⁇ j2 ⁇ )x + (e J2 ⁇ - - l)x 2 - ⁇ e ⁇ n + e in x" -
  • the relationship must be - corresponding.
  • the input of the code is not binary but quaternary QPSK data, ie M £ ⁇ +1 , a 1 , one, due to its data polynomial and ⁇ ), ( ⁇ At this time, it is not acceptable, and its input-output relationship must also be a "" ⁇ corresponding. Therefore, the encoding is not only for binary input data, but also for the quaternary input data is an effective three-fold increase in spectral efficiency. .
  • code rate including code rate higher than 1
  • Figure 3 is a simulation result.
  • V n exp ⁇ j[ ⁇ ⁇ ,] ⁇
  • Non-linear encoding of / o , ( 12).
  • the coded input is binary data - ⁇ 1 ), which uses double-encoding multiplexing (multi-access) with a constraint length of three.
  • FIG. 3 shows the properties of BPSK and QPSK
  • the left curve shows the nature of OVCDM
  • the abscissa of Figure 3 is the signal-to-noise ratio (in dB) and the ordinate is the bit error rate.
  • Figure 5 is a diagram showing the coding structure of Figure 3.
  • Linear coding with 0 (14).
  • the spectral efficiency and coding gain of the overlap coding multiplexing (multiple access) are determined by the selection of the coding matrix ⁇ after given the basic parameters K, N, L of the code multiplexing (multiple access).
  • the optimal coding matrix ⁇ should ensure that the coded code has the largest free distance ⁇ , and the coefficients of the preceding items of the generator function polynomial should be as small as possible.
  • the complexity of the search depends on the number of stable states of the code ⁇ ⁇ —in general, the workload of the search will be very difficult after U. But in fact, when the code multiplexing (multi-access) parameter ⁇ ' ⁇ is given, the relationship between the state and the state transition of the system is determined, and various types of closed paths are also determined.
  • Computer search can be performed primarily around the relationship between the minimum Euclidean distance between these closed paths and the coding matrix ⁇ . To reduce computational complexity, the search can be performed only around a few shorter closed paths. Since the Euclidean distance of the shorter closed path generally increases, the Euclidean distance of the non-shortest closed path will also increase. The present invention will then give a number of search algorithms.
  • High-order coding matrices can also be constructed from low-order coding matrices: for example, ⁇ and ⁇ are high-order respectively
  • the coding matrix can be generated by the following methods:
  • a ( + K 2 , N, 2L X + - 1) higher-order coding matrix can be generated by the following method
  • 0 ⁇ 0 2 , 0 3 are X - ⁇ , XL 2 , K 2 X (2L, -1) order all zero matrix t N2007/000536
  • the correspondence must be:
  • each coding tap coefficient is not in a domain at all, such as: TDM (TOMA), FDM (FDMA), SDM (SDMA), etc.; or linear irrelevant or Independent, such as: TDM (TDMA), FDM (FDMA), OFDM (OFDMA), CDM (CDMA), statistical space division multiplexing MIMO, etc.
  • TDM TOMA
  • FDM FDMA
  • SDMA SDM
  • linear irrelevant or Independent such as: TDM (TDMA), FDM (FDMA), OFDM (OFDMA), CDM (CDMA), statistical space division multiplexing MIMO, etc.
  • each coding tap coefficient is in a different "domain” or although it is in the same "domain” but is actually rotated or offset from each other to make the coding tap coefficients the most prime.
  • the method of the cartridge allows the respective code taps to be in different time or space, or in a mixed time, in a time, space, frequency or mixed domain, and allows strong overlap between them. In this way, a better way to implement OVTDM, OVSDM, OVFDM, and OVHDM with code is actually found.
  • Step 1 According to the actual design of the system is a reuse system or multiple access system and specific requirements for frequency efficiency, etc.
  • K represents the index of the system decoding detection complexity. The larger the KLQ, the higher the decoding complexity of the system, and the greater the delay of decoding detection.
  • Step 2 Determine the value "domain" of the coding matrix B and the value constraint according to the specific index requirements of the actual system.
  • the elements of the coding matrix B may be located in different "coding domains" that are rotated or displaced relative to each other; the elements of the coding matrix B in a multi-antenna system may be located in different or mutually rotated or relatively misaligned "spaces" In the "domain”, and often do not have any special requirements for the relative magnitude and phase between different "row” or “column” elements in the encoding matrix B, but for different elements in the same "row” or “column” in B
  • the relative amplitude and phase allow strict requirements; as in the non-orthogonal multi-carrier system, the elements in the coding matrix B can be located in different or mutually rotated or relatively misaligned "frequency domains", and often do not
  • the actual value of the element in B can be:
  • 2 is located on the unit circle or inside the circle; 3 is located on the real axis and the imaginary axis or one of them;
  • 5 is a sample of a random variable of some complex or real Gaussian (normal) distribution
  • the channel capacity C is the maximum value (discrete channel) or upper bound (continuous channel) of the mutual information-to-channel input distribution between the channel input and output sequences.
  • AWGN additive white Gaussian noise
  • AWGN additive white Gaussian noise
  • the channel input obeys the binomial distribution; while the QPSK signal obeys the quadratic distribution; multivariate PSK, QAM signals, etc. essentially obey the polynomial distribution. Therefore, for conventional modulation, for the additive white Gaussian noise (AWGN) channel, the input X does not obey the optimal distribution and must have a gap with the channel capacity.
  • AWGN additive white Gaussian noise
  • the two-level BPSK signal becomes K+1 level after K-fold overlap, and the larger the K, the larger the number of levels.
  • the QPSK signal is input, the levels of the in-phase I channel and the quadrature Q channel of the overlapping output also become multi-level.
  • the larger the overlap number K the larger the number of levels.
  • the input distribution tends to be complex or real Gaussian (normal) distribution, which is the inventor's previously invented OVTDM.
  • OVFDM OVFDM
  • each input symbol is multiplied by a complex weight and shifted to superimpose, so that the statistical characteristics of the channel input information can be changed in part.
  • each tap is a Gaussian (normal) distributed random variable, multiplied by the input symbol, the Gaussian (normal) distribution of the random variable is followed by a Gaussian (normal) random variable.
  • the channel input is a Gaussian (normal) variable that is likely to approach the channel capacity.
  • the coding tap coefficients are complex or real Gaussian
  • Gallager has demonstrated the performance bounds of linear block codes using the random coding method and the Gallager community. Although there is a possibility that multiple information sequences correspond to the same codeword sequence in the proof process, that is, the minimum Hamming Distance between these information sequences is 0, but this does not affect the final coding rate less than the channel capacity. Under the premise, the bit error rate is exponentially attenuated. Because there must be a coding sequence that makes the Hamming distance between different information sequences much larger, and if the good coding sequence and the bad coding sequence are evenly distributed in the extended binary domain ⁇ ( ), the statistical average effect is better. . This may be due to the fact that a good coding sequence is much more than a bad coding sequence.
  • Step 3 Search for the optimal coding matrix B according to the constraints of step two:
  • the first recommended optimal coding matrix B search algorithm is the first recommended optimal coding matrix B search algorithm.
  • node error event types There are only 2 types of path lengths in the node error event. There are L-2 types with lengths of £+1, + 2 , ⁇ , 2 - 1, and node error event types of length L+1. for:
  • the tributaries caused by the total length L+1 are arranged in order.
  • the tributaries caused by the total length of L+2 are arranged in order.
  • the metrics of the branches caused by the total length of 2L-1 are arranged in order:
  • the interval between error vectors can vary arbitrarily from 0 to L-2, but cannot be greater than L-2, otherwise the node error event ends.
  • the optimal coding matrix 6 ⁇ 6 ⁇ 1 ... ⁇ - 1 ] should ensure that the minimum Euclidean distance between the various closed paths in the Trellis diagram, that is, the error path of the various nodes from the correct path, is also called the free distance ⁇ maximum, where The definition of free distance is:
  • the probability of occurrence of a shorter node error event is greater, so it is sufficient to search for the optimal coding matrix B only for the shortest at most plus a number of shorter node error events.
  • Sub-step 1 According to the closedness of the linear code distance, a correct path can be arbitrarily selected. It is assumed that the node error starts from any "first node” and starts from the shortest closed path of a vector symbol error, for all possible vector errors. Symbol VX - (common type), search for the optimal coding matrix B under the constraints of step two to satisfy
  • Sub-step 2 Bring the first preferred coding matrix B' searched in sub-step 1 into a closed path with two symbol errors, and calculate
  • Sub-step 3 Bring the "best" coding matrix B jointly searched in sub-steps 1 and 2 into a closed path with three symbol errors, and calculate
  • the maximum free distance does not necessarily guarantee the best performance of the error probability of the system, but also the distribution of the different distance paths, that is, the coefficients of the generated elements, especially the coefficients of the previous items. .
  • the smaller the coefficient of the front item the better the performance of the system. Since the generation function is not easy to display directly from the process of searching for the maximum free distance (not impossible), and directly solving the generator function is very difficult when K, N, and L are large, so the final is also in the free distance. In the same coding matrix, the simulation performance is ultimately used to determine who is really the best coding matrix.
  • the first ( Z : 0 , 1 , ..., - nodes of the code multiplexing system, the ⁇ ( ⁇ ⁇ ) path is defined as the previous branch of the Euclidean modulus set as
  • Constraint: b i that is, the squared Euclidean modulus of b is as small as possible, and the nodes/numbers of ⁇ 2 — are as many as possible, and the difference is as large as possible; so continue until sub-step K: search for the best b -!
  • ⁇ - ' ⁇ 0 constraint: b _i b _i, that is, the squared Euclidean modulus of b - 1 is as small as possible, and the number of nodes of 1 - ⁇ - 2 ⁇ 0 is as much as possible, and the difference is as large as possible;
  • Step K+1 Change the initial b if necessary. , repeat step 1 ⁇ :.
  • the biggest advantage of the algorithm is that it can easily start from the known order optimal coding matrix, further increase the number of rows K, and improve the spectral efficiency of the system.
  • the number of channels transmitted in parallel is adaptively increased or decreased within K according to channel conditions, and is always guaranteed to be a good code.
  • Step 4 Construct a high-order coding matrix from a low-order coding matrix:
  • high-order coding matrices can be formed from known low-order coding matrices.
  • ⁇ ⁇ '' and two known low-order coding matrices of ( , and ( 2, ⁇ , ), respectively, a K , N ' L high-order coding matrix can be generated by:
  • ⁇ ⁇ 2 ⁇ 1 ® ⁇ 2
  • ⁇ ⁇ are the XA -1, W 2 ⁇ (2 ⁇ -1) order all-zero matrix.
  • Step 5 Generate more address codes needed in the "Wave Split Multiple Access" system
  • the order ⁇ XL of the best coding matrix ⁇ searched by the optimal coding matrix search method is not very large at present.
  • K and L are relatively large, the complexity of decoding is often difficult to accept.
  • a large number of address codes are often required.
  • the optimal coding matrix B is ⁇ " and uses the following spanning tree method to generate a large number of address codes:
  • the thus generated multiple access code is just an orthogonal address code with the conventional CDMA, and can only participate in the allocation and cannot share the channel capacity with other addresses.
  • Steps 4 and 5 do not necessarily need to be implemented in the actual system design. One step can be implemented according to the specific requirements of the actual system, and even one step is not implemented.
  • Step 6 Form a desired multi-address code according to the coding matrix searched or constructed in the previous steps, and design a desired multiplex or multi-address communication system with high frequency efficiency according to the selected multiple address code Transmitter.
  • Step 7 According to the transmitter designed in step 6, the receiver of the multiplex or multi-address communication system with high spectral efficiency is specifically designed.
  • Step 8 The state diagram or tree diagram or Trdlis diagram of the system is made in the receiver according to the state and state transition relationship of the selected coding matrix. And performing Maximum Likelihood or Maximum A Posterior Probability or other Sequential Decoding Detection Algorithm on the received signal, in the state diagram or tree diagram of the system or Trellis The path with the shortest Euclidean distance (Shortest Euclidian Distance) is detected in the figure.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Error Detection And Correction (AREA)

Abstract

La présente invention concerne un procédé de multiplexage codé. Le procédé assure la transmission en multiplexage codé au moyen d'un code linéaire parallèle ou non linéaire dont le débit est supérieur à 1. Lorsque le débit est supérieur à 1, il existe une relation correspondante biunivoque entre la séquence d'entrée de code parallèle et la séquence de sortie de code parallèle. Le code parallèle s'écarte de la zone limite et tous les ensembles multinomiaux de codes parallèles sont co-primaires entre eux. Le procédé comprend les étapes suivantes: la construction d'une matrice de code B, la matrice B comprenant K vecteurs de code; la formation de K ensembles de données transmis en parallèle, les K ensembles de données correspondant aux K vecteurs de code; l'exécution d'un code de convolution pour chaque ensemble de données et le vecteur de code correspondant de l'ensemble de données, L étant la longueur de contrainte du code; l'ajout des résultats du code de convolution des K ensembles de données de manière à obtenir N vecteur de sortie de code de dimension; la réception de N vecteur de sortie de code de dimension et l'exécution d'une vérification des séquences effectuée sur ces derniers; K, N et L étant des paramètres de base du code parallèle. Cette invention peut améliorer la capacité du système et son efficacité fréquentielle du fait de l'utilisation de la transmission de code linéaire ou non linéaire d'un débit supérieur à 1.
PCT/CN2007/000536 2007-02-14 2007-02-14 Procédé de multiplexage codé et de transmission multi-adresse WO2008098417A1 (fr)

Priority Applications (2)

Application Number Priority Date Filing Date Title
PCT/CN2007/000536 WO2008098417A1 (fr) 2007-02-14 2007-02-14 Procédé de multiplexage codé et de transmission multi-adresse
CN200780050698.6A CN101632248B (zh) 2007-02-14 2007-02-14 一种编码复用与多址传输方法

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2007/000536 WO2008098417A1 (fr) 2007-02-14 2007-02-14 Procédé de multiplexage codé et de transmission multi-adresse

Publications (1)

Publication Number Publication Date
WO2008098417A1 true WO2008098417A1 (fr) 2008-08-21

Family

ID=39689613

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2007/000536 WO2008098417A1 (fr) 2007-02-14 2007-02-14 Procédé de multiplexage codé et de transmission multi-adresse

Country Status (2)

Country Link
CN (1) CN101632248B (fr)
WO (1) WO2008098417A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110103236A1 (en) * 2009-09-13 2011-05-05 Research Institute Of Tsinghua University In Shenzhen Transmission method of code division multiplexing and multiple access
CN109831276A (zh) * 2017-11-23 2019-05-31 深圳超级数据链技术有限公司 信号处理方法和装置
CN115514453A (zh) * 2022-08-31 2022-12-23 北京航空航天大学 格码多址***及收发机处理方法

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2018014738A1 (fr) * 2016-07-22 2018-01-25 深圳超级数据链技术有限公司 Procédé et dispositif de décodage rapide adaptés à un système ovxdm, et système ovxdm
CN107968697B (zh) * 2016-10-19 2021-05-14 南通海嘉智能科技有限公司 重叠复用***的译码方法和装置
CN109831397B (zh) * 2017-11-23 2022-10-11 唐山蓝湛环保科技有限公司 重叠复用***及其处理方法和装置、存储介质、处理器
CN109831277A (zh) * 2017-11-23 2019-05-31 深圳超级数据链技术有限公司 重叠复用***及其处理方法和装置、存储介质、存储器
CN115378591B (zh) * 2022-07-18 2023-04-07 咚咚数字科技有限公司 一种基于融合的匿名生物特征密钥传输方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5691995A (en) * 1994-04-05 1997-11-25 Sony Corporation Transmission of data by using convolutional coding of different code rates and the encoded data reception including decoding of the received data
US20040028021A1 (en) * 2002-07-25 2004-02-12 Prabodh Varshney CDMA telecommunication system
US20060285605A1 (en) * 2005-06-16 2006-12-21 Qualcomm Incorporated Coding and modulation for multiple data streams in a communication system

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5691995A (en) * 1994-04-05 1997-11-25 Sony Corporation Transmission of data by using convolutional coding of different code rates and the encoded data reception including decoding of the received data
US20040028021A1 (en) * 2002-07-25 2004-02-12 Prabodh Varshney CDMA telecommunication system
US20060285605A1 (en) * 2005-06-16 2006-12-21 Qualcomm Incorporated Coding and modulation for multiple data streams in a communication system

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110103236A1 (en) * 2009-09-13 2011-05-05 Research Institute Of Tsinghua University In Shenzhen Transmission method of code division multiplexing and multiple access
CN109831276A (zh) * 2017-11-23 2019-05-31 深圳超级数据链技术有限公司 信号处理方法和装置
CN109831276B (zh) * 2017-11-23 2022-06-03 广州禄德投资有限公司 信号处理方法和装置
CN115514453A (zh) * 2022-08-31 2022-12-23 北京航空航天大学 格码多址***及收发机处理方法

Also Published As

Publication number Publication date
CN101632248B (zh) 2013-11-06
CN101632248A (zh) 2010-01-20

Similar Documents

Publication Publication Date Title
WO2008098417A1 (fr) Procédé de multiplexage codé et de transmission multi-adresse
Jafar et al. Generalized degrees of freedom of the symmetric Gaussian $ K $ user interference channel
KR101719969B1 (ko) 희소 코드 다중 액세스를 위한 시스템 및 방법
Shirvanimoghaddam et al. Near-capacity adaptive analog fountain codes for wireless channels
US8051355B2 (en) Multilevel low density parity-check coded modulation
Bao et al. Bit-interleaved coded SCMA with iterative multiuser detection: Multidimensional constellations design
Yang et al. On the optimal design and performance of linear physical-layer network coding for fading two-way relay channels
US20110103236A1 (en) Transmission method of code division multiplexing and multiple access
EP3324583B1 (fr) Procédé, appareil et système de traitement de données
Li et al. Full-diversity binary frame-wise network coding for multiple-source multiple-relay networks over slow-fading channels
Cheng et al. Capacity analysis for non-orthogonal overloading transmissions under constellation constraints
Sharma et al. SCMA codebook based on optimization of mutual information and shaping gain
Harshan et al. A novel power allocation scheme for two-user GMAC with finite input constellations
ES2683084T3 (es) Método de ecualización Turbo y sistema de ecualización Turbo
Roth et al. Coplanar Turbo‐FSK: A Flexible and Power Efficient Modulation for the Internet‐of‐Things
Yu et al. Uniquely decodable multi-amplitude sequence for grant-free multiple-access adder channels
Abadi et al. Zero-error codes for multi-type molecular communication in random delay channel
Kim et al. Adaptive sliding-window coded modulation in cellular networks
Arora et al. Novel Energy Optimized LDPC Codes for Next-Generation MIMO OFDM Systems
Wen et al. Joint sparse graph for FBMC/OQAM systems
Zhang et al. A novel scheme for secure network coding using one-time pad
Zhong et al. A novel SCMA codebook construction based on extended factor graph design
WO2009131252A1 (fr) Systèmes et procédés pour fournir une protection de message inégale
Zhu et al. Design of a physical layer encryption scheme for rate compatible modulation
Griot et al. Nonlinear trellis codes for binary-input binary-output multiple-access channels with single-user decoding

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 200780050698.6

Country of ref document: CN

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 07710957

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 07710957

Country of ref document: EP

Kind code of ref document: A1