SCMA user transmission method compatible with regular and irregular mapping

Abstract

The invention relates to a SCMA user transmission method compatible with regular and irregular mapping, which belongs to the field of communication technology. In the traditional SCMA user transmission mechanism, the number of resource blocks occupied by each user is the same. When the number of users cannot meet the rule mapping access system, it cannot adapt well to such massive connections. The present invention is a SCMA user transmission method compatible with regular and irregular mapping. Aiming at the problem that traditional SCMA user transmission can only adapt to regular mapping, the overload formula, LDPC and degree distribution theory are combined to construct a factor matrix and a mapping matrix, and establish an SCMA user transmission mechanism that adapts to regular and irregular mapping, and is suitable for a constellation matrix with a variety of access user numbers. This method can not only improve the applicability of the SCMA system, but also be applied to 5G mobile communications in massive connection scenarios, and improve the performance of the SCMA system. This study is applicable to 5G mobile communication systems using SCMA technology.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular to an SCMA (Sparse Code Multiple Access) user transmission method, specifically an SCMA user transmission method compatible with regular and irregular mapping, and an SCMA user transmission method that can satisfy both regular and irregular mapping. Background Art

[0002] SCMA is a popular multiple access method for 5G mobile communications. It evolves from low-density signals and combines sparse spread spectrum with high-dimensional modulation. The bit stream in the link is directly mapped to the multi-dimensional codeword in the pre-set codebook, thereby solving the overload problem of a large number of user accesses and improving spectrum utilization. However, sparse spread spectrum can only handle a limited number of users. In the scenario of massive connections, the number of links will increase by 10 times or 100 times and change dynamically over a large range. However, the number of resource blocks occupied by each user in the traditional SCMA user transmission mechanism is the same. When the number of users cannot meet the regular mapping access system, it cannot adapt well to such massive connections, which limits its application. Therefore, it is urgent to design a user transmission method for SCMA that can adapt to irregular mapping, which will have a significant impact on improving the applicability of the SCMA system and increasing the scope of application. Summary of the invention

[0003] The SCMA user transmission method compatible with regular and irregular mapping aims to solve the problem that traditional SCMA user transmission can only adapt to regular mapping. By combining the overload formula, LDPC (Low Density Parity Check Code) and degree distribution theory, the factor matrix and mapping matrix are constructed to establish a SCMA user transmission mechanism that adapts to regular and irregular mapping, and is suitable for constellation matrices with a variety of access user numbers. This method can not only improve the applicability of the SCMA system, but also be applied to 5G mobile communications in massive connection scenarios, and improve the performance of the SCMA system.

[0004] The technical solution adopted is:

[0005] Firstly, according to the overload condition of the SCMA system, a compatible mapping mechanism is established to obtain the parameters required for the SCMA system to allocate user information.

[0006] Then, the factor graph matrix is ​​redesigned according to LDPC and degree distribution theory to allocate user information to the set resource blocks.

[0007] Secondly, the designed factor graph matrix is ​​split to represent the mapping mechanism of each user, and the mapping matrix is ​​calculated.

[0008] Finally, a sub-constellation selection mechanism is established. Based on the factor graph matrix, a constellation matrix is ​​designed to sparsely map user information to the complex domain.

[0009] Its advantages are:

[0010] The SCMA user transmission method compatible with regular and irregular mapping aims to solve the problem that traditional SCMA user transmission can only adapt to regular mapping. By combining the overload formula, LDPC and degree distribution theory, the factor matrix and mapping matrix are constructed to establish an SCMA user transmission mechanism that adapts to regular and irregular mapping, and is suitable for constellation matrices with a variety of access user numbers. This method can not only improve the applicability of the SCMA system, but also be applied to 5G mobile communications in massive connection scenarios, and improve the performance of the SCMA system. This study is applicable to 5G mobile communication systems using SCMA technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 It is a flow chart of the SCMA user transmission method compatible with regular and irregular mappings of the present invention.

[0012] Figure 2 It is the relationship between users and resource blocks corresponding to the factor graph matrix F in the embodiment. DETAILED DESCRIPTION

[0013] Step 1: Establish a compatible mapping mechanism; according to user requirements, set the given number of users to J and the given number of orthogonal resource blocks to K. The SCMA transmitter spreads the information transmitted by J users to these K resource blocks in the form of sparse spreading codes for data transmission. During the data transmission process, each user transmits information on N resource blocks at the same time, and each resource block is superimposed with transmission d f The information of each user makes the number of users J greater than K, thereby improving the system capacity and resource utilization. This process is called overloading of the SCMA system.

[0014] From the above SCMA system overload conditions, we can know that the parameters J, K, N and d f Satisfies formula (1):

[0015] JN=Kd f , J>K(1).

[0016] Among them, N represents the number of resource blocks used by each user to transmit information. To ensure the transmission quality, each user needs to have at least two transmission nodes to exchange information. Usually, N = 2. In the traditional SCMA system, N is a fixed value, which leads to the limitation of the number of users connected to the existing SCMA system.

[0017] For some access users J, user information cannot be regularly allocated to each resource block in the traditional way. For example, when the number of users J = 7, K = 5, N = 2, d f =3, it is impossible to find a parameter N that satisfies equation (1) and conforms to physical meaning.

[0018] Therefore, in order to solve the problem that the traditional mapping mechanism cannot satisfy the flexible mapping, a compatible mapping mechanism is established to allocate user information to each resource block according to the new mechanism, making resource allocation more flexible.

[0019] The number of resource blocks occupied by the jth user is N j represents, j∈[1,J]. According to formula (1), N j The calculation formula is shown in formula (2):

[0020]

[0021] Combined with formula (2), it can be seen that according to the new resource allocation method, the information transmitted by J users can be allocated to K resource blocks, so that each resource block is superimposed with the transmission d f The information of each user is shown in formula (3).

[0022]

[0023] Step 2: Design the factor graph matrix F; In order to specifically allocate users to specific resource blocks and enable the system to use the LDPC-SCMA joint decoding strategy to improve the SCMA system error performance when transmitting information, the factor graph matrix F is designed based on LDPC and degree distribution theory to represent the relationship between each user and resource block.

[0024] The factor graph matrix F has K rows and J columns, and each column has N j Elements 1, each row has d f When the user transmits data on the resource block, let the element Fk,j in the kth row and jth column in the factor graph matrix = 1; when the user does not transmit data on the resource block, let F k,j = 0. The row weight of the matrix F is equal to the number of superimposed symbols of the resource block d f , the column weight is equal to the number of resource blocks occupied by the jth user N j .

[0025] Split F into column vectors by column, representing the resource allocation of each user separately:

[0026] F=[f1,f2...,fj...,fJ](4).

[0027] Among them, f j is the j-th column vector, indicating the resource allocation of the j-th user.

[0028] Example 1: Based on the above example, use parameters J = 7, K = 5, d f =3, N1=3, N2=N3=N4=N5=N6=N7=2.

[0029] The factor graph matrix F established according to LDPC and degree distribution theory is:

[0030]

[0031] The relationship between users and resource blocks corresponding to the factor graph matrix F is specifically expressed as follows Figure 2 :

[0032] Step 3: Design the mapping matrix V j ; Based on the design of the factor graph matrix F, in order to map the information transmitted by each user in the system into the user codebook, the mapping matrix V is designed j The user transmission information mapping method is represented, and the element 1 in the matrix represents information transmission, and 0 represents no information transmission.

[0033] Split the factor graph matrix F according to formula (4) to obtain the resource allocation of the jth user f j , and then f j The mapping matrix can be obtained by performing the following processing:

[0034] V j =diag(f1)(5).

[0035] V j is a K-row N j The matrix of N columns is composed of N diagonal elements all set to 1. j Insert KN into the diagonal matrix j The positions of these all-zero rows correspond to the resource blocks on which user j does not transmit information.

[0036] Since the mapping matrix V j The same mapping relationship is shown with the factor graph matrix F, so they can be converted to each other, as shown in formula (6):

[0037]

[0038] Example 1: Based on the factor graph matrix F in the above example, find the jth column of F and expand the rows with elements 1 to N j dimensional identity matrix, and pad the rows with zero elements with N j -1 0 element expands to an all-zero row vector.

[0039] The mapping matrices of each user are:

[0040]

[0041] According to the position of element 1 of mapping matrix V1, it can be known that the information of user 1 is transmitted on the second resource block, the fourth resource block and the fifth resource block. Similarly, according to mapping matrix V2, it can be known that the information of user 2 is transmitted on the first resource block and the third resource block. In general, there are 3 users transmitting information on the resource blocks corresponding to each row.

[0042] Step 4: Design the constellation matrix MC; In the SCMA system, the transmission signals of each user are superimposed on the same resource block for transmission. The greater the power difference between these signals, the easier it is for the receiving end to separate them. Therefore, when designing the SCMA codebook, the constellation points used to transmit information on each resource block are often divided into d f different sub-constellations, mapping the transmission d f To obtain the information of each user, it is necessary to establish a constellation matrix MC based on the factor graph matrix F to explain the sub-constellation selected by the user.

[0043] Since this method expands the selection range of the number of access users J, the number of resource blocks N also increases. The traditional constellation matrix design method designs different sub-constellations for each resource block. This leads to the need to select different sub-constellations for each user on different resource blocks when designing the traditional constellation matrix. It is also necessary to ensure that the Euclidean distance between the constellation points on each resource block is small, that is, it is necessary to design constellation points, Kd f The design process of the constellation matrix MC is relatively complicated and is not suitable for irregular SCMA systems. Therefore, the constellation matrix MC uses the same resource block constellation diagram in different resource blocks, that is, only the design of The codebook complexity is greatly reduced.

[0044] The design mechanism of the constellation matrix MC is: d is selected on each resource block f Different sub-constellations are superimposed, that is, the elements of each row in the corresponding constellation matrix MC cannot be repeated; at the same time, in order to increase the decoding reliability of the user, the same user should use different sub-constellations when transmitting information on different resource blocks.

[0045] According to the above mechanism, we get the set of sub-constellation ranges selected by users to transmit information on resource blocks, and use Γ m,n In other words, m represents the order in which users are superimposed on resource blocks, n represents the order in which users select resource blocks for information transmission, and the sub-constellations selected by df users on each resource block are The sub-constellations selected by each user to transmit information on Nj resource blocks are

[0046]

[0047] According to the resource block restriction conditions, the sub-constellation value set is obtained and used express:

[0048]

[0049] According to the resource user restriction conditions, the sub-constellation value set is obtained and used express:

[0050]

[0051] According to formula (7) and formula (8), the sub-constellation range set is calculated The constellation matrix MC is derived by combining the factor graph matrix as shown in formula (9). m,n It indicates the range set of sub-constellations selected by the mth user transmitting information on a certain resource block on the nth resource block.

[0052]

[0053] After obtaining the constellation matrix MC representing the mapping mechanism according to the above steps, the user can select the required user sub-constellation according to the needs, and then use the constellation matrix MC of this method to map the information of J users to K resource blocks to obtain the required specific codebook.

[0054] Example 1: Based on the above example, use the parameters J=7, K=5, d f =3, N1=3, N2=N3=N4=N5=N6=N7=2. Assume that user 1 selects sub-constellations S1, S3, and S2 on resource blocks 2, 4, and 5, respectively, and user 2 selects sub-constellations S2 and S3 on resource blocks 1 and 3, respectively. According to formulas (6) and (7), the range of sub-constellations selected by user 3 on resource block 1 is Γ 2,1 ={1,3}, assuming that S1 is selected, then user 3 selects the range of sub-constellations on resource block 2 Γ 2,2 ={2,3}. Similarly, user 4 selects S2 on resource block 3 and S1 on resource block 4; user 5 selects S3 on resource block 1 and S2 on resource block 4; user 6 selects S3 on resource block 2 and S1 on resource block 5; user 7 selects S1 on resource block 3 and S3 on resource block 5.

[0055] Combined with the factor graph matrix generated by formula (6), the constellation matrix MC is established as:

[0056]

[0057] Based on this, the user can use the user sub-constellation of his / her choice and generate the corresponding codebook using the MC mapping of this method. Taking the interleaved code sub-constellation when M=4 as an example, the sub-constellation is:

[0058] S1 = {3.0000 + 3.0000i 1.0000 + 1.0000i - 1.0000 - 1.0000i - 3.0000 - 3.0000i} S2 = {-0.7071 - 1.2247i 2.1213 + 3.6742i - 2.1213 - 3.6742i 0.70710 + 1.2247i} S3 = {1.0980 + 4.0980i 0.3660 + 1.3660i - 0.36602 - 1.3660i - 1.0980 - 4.0980i} According to the constellation matrix MC, combined with the mapping matrix V j The user codebook can be obtained as:

[0059]

[0060]

Claims

1. A SCMA user transmission method compatible with regular and irregular mappings, characterized in that The following steps are involved: In the SCMA system, according to user needs, the given number of users is set to J, and the given number of orthogonal resource blocks is set to K. The SCMA transmitter spreads the information transmitted by J users to these K resource blocks in the form of sparse spreading codes for data transmission. During the data transmission process, each user transmits information on N resource blocks at the same time, and each resource block is superimposed with transmission d f The information of users makes the number of users J greater than K, and the parameters J, K, N and d f Satisfies formula [1]: JN=Kdf,J>K[1]; Where N represents the number of resource blocks used by each user to transmit information. To ensure the transmission quality, each user needs at least two transmission nodes to exchange information; The specific implementation process of the method is: Step 1: Establish a compatible mapping mechanism; allocate user information to each resource block according to this mechanism; The number of resource blocks occupied by the jth user is N j Indicates, j∈[1,J]; according to formula [1], N j The calculation formula is shown in formula [2]: Combined with formula [2], it can be seen that according to this resource allocation method, the information transmitted by J users can be allocated to K resource blocks, so that each resource block is superimposed with the transmission d f The information of each user is shown in formula [3]; Step 2: Design the factor graph matrix F. The factor graph matrix F is designed based on LDPC and degree distribution theory to represent the relationship between each user and resource block. The factor graph matrix F has K rows and J columns, and each column has N j Elements 1, each row has d f When the user transmits data on the resource block, let the element F in the kth row and jth column of the factor graph matrix k,j =1; when the user does not transmit data on this resource block, let F k,j =0; the row weight of the matrix F is equal to the number of superimposed symbols of the resource block d f , the column weight is equal to the number of resource blocks occupied by the jth user N j ; Split F into column vectors by column, representing the resource allocation of each user separately: F = [f1,f2...,fj...,fJ][4]; Among them, f j is the jth column vector, indicating the resource allocation of the jth user; Step 3: Design the mapping matrix V j ; Based on the design of the factor graph matrix F, the mapping matrix V is designed j The user transmission information mapping method is represented, and the element 1 in the matrix represents information transmission, and 0 represents no information transmission; Split the factor graph matrix F according to formula [4] to obtain the resource allocation of the jth user f j , and then f j The mapping matrix can be obtained by performing the following processing: Vj = diag(f1)[5]; V j is a K-row N j The matrix of N columns is composed of N diagonal elements all set to 1. j Insert KN into the diagonal matrix j The positions of these all-zero rows correspond to the resource blocks on which user j does not transmit information; Mapping Matrix V j The same mapping relationship is shown as the factor graph matrix F. j and F are converted to each other, as shown in formula [6]: Step 4: Design the constellation matrix MC; when designing the SCMA codebook, divide the constellation points used to transmit information on each resource block into df different sub-constellations, and map the transmission d f The information of each user is then used to build a constellation matrix based on the factor graph matrix F. Constellation points: Each resource block uses d f Different sub-constellations are superimposed. At the same time, the same user should use different sub-constellations when transmitting information on different resource blocks. The design mechanism of the constellation matrix MC is: d is selected on each resource block f Different sub-constellations are superimposed, that is, the elements of each row in the corresponding constellation matrix MC cannot be repeated; at the same time, in order to increase the decoding reliability of the user, the same user should use different sub-constellations when transmitting information on different resource blocks; According to the constellation matrix MC design, the set of sub-constellation ranges selected by users for transmitting information on resource blocks is obtained, which is represented by Γm,n, where m represents the order in which users are superimposed on resource blocks, and n represents the order in which users select resource blocks for transmitting information. Let d carried on each resource block f The sub-constellations selected by the users are Each user in N j The sub-constellations selected for transmitting information on resource blocks are According to the resource block restriction conditions, the sub-constellation value set is obtained and used express: According to the resource user restriction conditions, the sub-constellation value set is obtained and used express: According to formula [7] and formula [8], the sub-constellation range set is calculated Combining the factor graph matrix, the constellation matrix MC is derived as shown in formula [9]; where Γ m,n Indicates the range set of sub-constellations selected by the mth user transmitting information on a certain resource block on the nth resource block; After obtaining the constellation matrix MC representing the mapping mechanism according to the above steps, the user can select the required user sub-constellation according to the needs, map the information of J users to K resource blocks, and obtain the required specific codebook.

Images