CN-117014103-B - SCMA-D2D network low-complexity coding and decoding method, device and equipment

CN117014103BCN 117014103 BCN117014103 BCN 117014103BCN-117014103-B

Abstract

The invention relates to a mixed network coding and decoding technology and discloses a low-complexity coding and decoding method, device and equipment of an SCMA-D2D network, wherein the method comprises the steps of generating a connection factor matrix according to user information of SCMA cellular users and D2D users and a frequency spectrum resource block, and generating a connection factor graph according to the connection factor matrix; the method comprises the steps of constructing a code word generator according to a connection factor graph, calculating code words corresponding to user information by the code word generator, generating mixed network signal data according to the code words corresponding to the user information and frequency spectrum resource blocks, inputting the mixed network signal data transmitted through a Gaussian channel into a relay network to obtain relay signal data, broadcasting the relay signal data to receiving ends corresponding to SCMA cellular users and D2D users, and decoding the relay signal data by a multi-user classification decoder after the receiving ends receive the relay signal data to obtain decoding information. The invention can improve the accuracy of multi-user decoding and reduce the complexity of encoding and decoding.

Inventors

JIANG FANG
ZHOU XINYUAN
WANG YI
XU YAOHUA
BAI NA
LIU YU
XU LEI
FANG HONGYU
WANG GUIZHU

Assignees

安徽大学

Dates

Publication Date: 20260512
Application Date: 20230612

Claims (10)

1. A SCMA-D2D network low complexity codec method, the method comprising: Generating a connection factor matrix according to user information of SCMA cellular users and D2D users and a preset frequency spectrum resource block, and generating a connection factor graph according to the connection factor matrix; constructing a code word generator according to the connection factor graph, and calculating a code word corresponding to the user information by using the code word generator, wherein the code word generator is composed of a plurality of DNN coding units, and the number of output nodes of an output layer in the DNN coding units is set based on the mapping relation of the connection factor graph; generating mixed network signal data according to the code word corresponding to the user information and the frequency spectrum resource block; Inputting the mixed network signal data transmitted by the Gaussian channel into a pre-constructed relay network to obtain relay signal data, and broadcasting the relay signal data to the SCMA cellular user and a receiving end corresponding to the D2D user; and after the receiving end receives the relay signal data, decoding the relay signal data by utilizing a pre-constructed multi-user classification decoder to obtain decoding information corresponding to the SCMA cellular user and the D2D user, wherein the multi-user classification decoder comprises the SCMA cellular user in the receiving end and a DNN decoding unit corresponding to the D2D user.
2. The SCMA-D2D network low complexity codec method of claim 1, wherein the generating the hybrid network signal data from the codeword corresponding to the user information and the spectrum resource block comprises: Determining a target user node connected with the frequency spectrum resource block, and acquiring a target code word of user information corresponding to the target user node; Superposing the target code words to obtain resource block code words corresponding to each frequency spectrum resource block; calculating the resource block code word by using the following output formula to obtain mixed network signal data; the output formula is expressed as: Wherein, the Represented as the hybrid network signal data, Denoted as the first Resource block codewords corresponding to the individual spectrum resource blocks, Expressed as the total number of spectrum resource blocks.
3. The SCMA-D2D network low complexity codec method of claim 1, wherein the connection factor matrix is represented as: Wherein, the Represented as a matrix of connection factors, Expressed as the number of said spectrum resource blocks, Expressed as the number of SCMA cellular users, Expressed as the number of D2D users in the user information.
4. The SCMA-D2D network low complexity codec method of claim 1, wherein before inputting the mixed network signal data transmitted via the gaussian channel into the pre-constructed relay network, further comprising: Constructing a relay sharing layer and initializing a DNN relay unit; Configuring the number of nodes of an input layer in the initialization DNN relay unit according to the dimension of the mixed network signal data transmitted through the Gaussian channel to obtain the DNN relay unit; And generating a relay network according to the relay sharing layer and the DNN relay unit.
5. The SCMA-D2D network low complexity codec method of claim 1, wherein the constructing a codeword generator from the connection factor graph comprises: Respectively configuring corresponding initialization DNN coding units according to user nodes in the connection factor graph; obtaining Shan Re vectors generated by performing single thermal coding on user information corresponding to a user of the user node; setting the number of nodes of an input layer in the initialization DNN coding unit according to the dimension of the single thermal vector, and setting the number of nodes of an output layer in the initialization DNN coding unit according to the number of connecting lines of user nodes in the connection factor graph to obtain a DNN coding unit; And summarizing DNN coding units corresponding to each user node to obtain a code word generator.
6. The SCMA-D2D network low complexity codec method of claim 5, wherein prior to decoding the relay signal data with the pre-constructed multi-user class decoder, further comprising: Configuring a corresponding initialization DNN decoding unit for each user of the receiving end; Setting the number of nodes of an input layer in the initialization DNN decoding unit according to the dimension of the relay signal data, and setting the number of nodes of an output layer in the initialization DNN decoding unit according to the dimension of Shan Re vectors to obtain a DNN decoding unit; And generating a multi-user classification decoder according to the initialized decoding sharing layer and the DNN decoding unit.
7. The SCMA-D2D network low complexity codec method of claim 1, wherein the decoding the relay signal data using a pre-constructed multi-user classification decoder to obtain the user information corresponding to the SCMA cellular user and the D2D user comprises: Decoding the relay signal data by utilizing the multi-user classification decoder to obtain SCMA cellular user decoding data and D2D user decoding data; Performing probability calculation on the SCMA cellular user decoding data and the D2D user decoding data to obtain probability information corresponding to the SCMA cellular user decoding data and the D2D user decoding data; And binary conversion is carried out on the probability information, so that decoding information corresponding to the SCMA cellular user and the D2D user is obtained.
8. The SCMA-D2D network low complexity codec method of claim 1, wherein after obtaining the decoding information corresponding to the SCMA cellular user and the D2D user, further comprising: calculating cross entropy loss according to user information corresponding to the SCMA cellular user and the D2D user and corresponding decoding information; And carrying out joint optimization on the codeword generator, the relay network and the multi-user classification decoder according to the cross entropy loss to obtain an SCMA-D2D hybrid network self-encoding and decoding model formed by the optimized codeword generator, the relay network and the multi-user classification decoder.
9. The SCMA-D2D network low complexity codec method of claim 8, wherein the calculating the cross entropy loss from the user information and the corresponding decoding information of the SCMA cellular user corresponding to the D2D user comprises: Calculating user information and decoding information corresponding to the SCMA cellular user through a preset first cross entropy loss function to obtain first cross entropy loss; the first cross entropy loss function is expressed as: Wherein, the Denoted as the first The cross entropy corresponding to each SCMA cellular user, Denoted as the first The decoding information corresponding to the individual SCMA cellular subscribers, Denoted as the first User information corresponding to the individual SCMA cellular users, Expressed as the number of SCMA cellular users, Expressed as the number of said spectrum resource blocks, Is that Is the first of (2) The term "is used to refer to, Is that Is the first of (2) The term "is used to refer to, Denoted as the SCMA cellular user's corresponding first cross entropy loss, Represented as , Is a set of (a) and (b), Represented as , Is a collection of (3); Calculating user information and decoding information corresponding to the D2D user through a preset second cross entropy loss function to obtain second cross entropy loss; The second cross entropy loss function is expressed as: Wherein, the Denoted as the first The cross entropy corresponding to the individual D2D users, Denoted as the first Decoding information corresponding to the individual D2D users, Denoted as the first User information corresponding to the individual D2D users, Expressed as the number of D2D users, Expressed as the number of said spectrum resource blocks, Is that Is the first of (2) The term "is used to refer to, Is that Is the first of (2) The term "is used to refer to, Represented as a corresponding second cross entropy loss for the D2D user, Represented as , Is a set of (a) and (b), Represented as , Is a set of (3).
10. A SCMA-D2D network low complexity codec apparatus, the apparatus comprising: the connection factor graph generation module is used for generating a connection factor matrix according to user information of the SCMA cellular user and the D2D user and a preset frequency spectrum resource block, and generating a connection factor graph according to the connection factor matrix; The coding module is used for constructing a code word generator according to the connection factor graph, and calculating the code word corresponding to the user information by utilizing the code word generator, wherein the code word generator is composed of a plurality of DNN coding units, and the number of output nodes of an output layer in the DNN coding units is set based on the mapping relation of the connection factor graph; the spectrum resource mapping module is used for generating mixed network signal data according to the code words corresponding to the user information and the spectrum resource blocks; The relay network processing module is used for inputting the mixed network signal data transmitted by the Gaussian channel into a pre-constructed relay network to obtain relay signal data, and broadcasting the relay signal data to the SCMA cellular user and a receiving end corresponding to the D2D user; And the decoding module is used for decoding the relay signal data by utilizing a pre-constructed multi-user classification decoder after the receiving end receives the relay signal data to obtain decoding information corresponding to the SCMA cellular user and the D2D user, wherein the multi-user classification decoder comprises the SCMA cellular user in the receiving end and a DNN decoding unit corresponding to the D2D user.

Description

SCMA-D2D network low-complexity coding and decoding method, device and equipment Technical Field The present invention relates to the field of hybrid network coding and decoding technologies, and in particular, to a method, an apparatus, and a device for coding and decoding a SCMA-D2D network with low complexity. Background With the continuous development of the internet of things (Internet of Things), the number of mass equipment and various service types are required to be higher for the network access technology. Network architecture capable of supporting heterogeneous devices and multiple access techniques with higher spectrum utilization are becoming important research points. The sparse code division multiple access technology (Sparse Code Multiple Access, SCMA) maps the user bit information into specific sparse multidimensional code words through low-density spread spectrum and multidimensional modulation coding at a transmitting end, and different user code words are non-orthogonally overlapped on spectrum resources, so that multi-user access and data transmission are realized. The Device-to-Device (D2D) communication is a short-distance communication technology capable of supporting short-distance users to directly transmit data, and D2D user communication is added into a cellular system, so that the two devices share spectrum resources, the utilization rate of the spectrum resources can be further increased, and the large-scale and diversified application requirements of a communication network are met. However, in a hybrid network system where two communication technologies SCMA-D2D coexist, serious inter-user interference will be caused due to the sharing of resources and the increase of devices, the complexity of encoding and decoding increases, and the accuracy of BER (Bit Error Rate) decreases. Meanwhile, the intelligent mobile terminal in the internet of things often has the characteristics of miniaturization, light weight and the like, and the characteristics require that the coding and decoding algorithm of the intelligent terminal has low complexity while maintaining the bit error rate performance. Therefore, in the application scenario of multi-user access and data transmission, improving the accuracy of multi-user decoding and reducing the complexity of encoding and decoding become a problem to be solved urgently. Disclosure of Invention The invention provides a low-complexity coding and decoding method, device, electronic equipment and computer readable storage medium of an SCMA-D2D network, and mainly aims to solve the problem that the accuracy of multi-user decoding cannot be effectively improved and the complexity of coding and decoding is reduced. In order to achieve the above object, the present invention provides a low complexity encoding and decoding method for SCMA-D2D network, comprising: Generating a connection factor matrix according to user information of SCMA cellular users and D2D users and a preset frequency spectrum resource block, and generating a connection factor graph according to the connection factor matrix; constructing a code word generator according to the connection factor graph, and calculating a code word corresponding to the user information by using the code word generator, wherein the code word generator is composed of a plurality of DNN coding units, and the number of output nodes of an output layer in the DNN coding units is set based on the mapping relation of the connection factor graph; generating mixed network signal data according to the code word corresponding to the user information and the frequency spectrum resource block; Inputting the mixed network signal data transmitted by the Gaussian channel into a pre-constructed relay network to obtain relay signal data, and broadcasting the relay signal data to the SCMA cellular user and a receiving end corresponding to the D2D user; and after the receiving end receives the relay signal data, decoding the relay signal data by utilizing a pre-constructed multi-user classification decoder to obtain decoding information corresponding to the SCMA cellular user and the D2D user, wherein the multi-user classification decoder comprises the SCMA cellular user in the receiving end and a DNN decoding unit corresponding to the D2D user. Optionally, the generating mixed network signal data according to the codeword corresponding to the user information and the spectrum resource block includes: Determining a target user node connected with the frequency spectrum resource block, and acquiring a target code word of user information corresponding to the target user node; Superposing the target code words to obtain resource block code words corresponding to each frequency spectrum resource block; calculating the resource block code word by using the following output formula to obtain mixed network signal data; the output formula is expressed as: Wherein, the Represented as the hybrid network signal data,Denoted as the