WO2018176824A1

WO2018176824A1 - Non-orthogonal multiple access method based on scheduling and competition fine-grained fusion

Info

Publication number: WO2018176824A1
Application number: PCT/CN2017/107772
Authority: WO
Inventors: 杨懋; 李波; 白志城; 梁玉生; 闫中江; 左晓亚
Original assignee: 西北工业大学
Priority date: 2017-04-01
Filing date: 2017-10-26
Publication date: 2018-10-04
Also published as: CN106912111A

Abstract

Provided is a non-orthogonal multiple access method based on scheduling and competition fine-grained fusion, which relates to the technical field of communications. A resource area and a resource position occupied by a scheduling-competition fusion access mode are calculated, and a resource is allocated to a scheduling user; a central node sends a resource allocation indication frame to a user equipment of the present cell; and finally, information sent by the scheduling user and information sent by a competition user are superposed on a time frequency resource by means of scheduling-competition fusion access, and a receiving end parses the superposed information by means of multi-user combined detection. By using the method of scheduling access and competition access sharing the same frequency spectrum resource, the present invention achieves fine-grained fusion of two access modes, thereby facilitating the matching of a dynamically changed demand and a resource, facilitating a multi-user diversity gain, and thus significantly improving a network capacity, a frequency spectrum utilization rate and a service quality.

Description

Non-orthogonal multiple access method based on scheduling and competition fine-grained fusion

This application claims priority to Chinese patent application filed on April 1, 2017, the Chinese Patent Office, application number 201710211241.4, and the application name is "a non-orthogonal multiple access method based on scheduling and competition fine-grained fusion" The entire content of which is incorporated herein by reference.

Technical field

The present invention relates to the field of communication technologies, and in particular, to a non-orthogonal multiple access method.

Background technique

In recent years, with the rapid development of mobile wireless communication technology and computer science, the mobile Internet is profoundly affecting people's work and learning, and continues to penetrate into various fields. As the core of mobile wireless communication technology, multiple access technology, also known as Media Access Control (MAC), is responsible for how to efficiently share and use increasingly low-cost wireless among multiple User Equipments (UEs). Spectrum resources. Therefore, the efficiency of MAC technology directly affects the key performance indicators (KPIs) of mobile wireless networks: network capacity, network connection and spectrum utilization.

The next-generation mobile wireless network needs to meet the needs of massive user connections and ultra-large-scale services, which poses great challenges to traditional wireless networks. Therefore, the industry is striving to significantly increase the number of user connections, network capacity, and spectrum utilization. technology. In recent years, the non-orthogonal multiple access technology represented by Sparse Code Multiple Access (SCMA) enables different user equipments to share the same spectrum resources, that is, resources are no longer exclusive. Therefore, the number of user connections, network capacity, and spectrum utilization can be significantly improved under the premise of limited spectrum resources, which is favored by industry and academia, and is expected to be standardized by the fifth generation mobile communication system (5G). For non-orthogonal multiple access techniques, refer to "Dai L, Wang B, Yuan Y, et al. Non-orthogonal multiple access for 5G: solutions, challenges, opportunities, and future research trends [J]. IEEE Communications Magazine, 2015, 53(9): 74-81.".

In terms of the main body and mode of access, multiple access technologies can generally be divided into two categories: 1) scheduling-based multiple access technology, often applied to cellular networks, the advantage of which is the use of resources between users. Without conflicts, resources can be better optimized and allocated, and transmission rate requirements are more easily protected; the disadvantage is that the central control node needs less flexibility, real-time guarantee is poor, and signaling overhead is large; 2) based on random competition Multiple access technology, often used in wireless local area networks, has the advantages of flexibility and robustness, good real-time guarantee, and simple implementation. The disadvantage is that data packets may be sent between multiple users, and transmission rate requirements are difficult to obtain. Very good protection. For convenience of description, the scheduling-based multiple access technology and the random access-based multiple access technology are simply referred to as scheduling access and contention access, respectively, without causing ambiguity.

"Au K, Zhang L, Nikopour H, et al. Uplink contention based SCMA for 5G radio access [C]//Globecom Workshops (GC Wkshps), 2014. IEEE, 2014: 900-905." The inbound mode and the contention-based access mode are simultaneously introduced into the SCMA uplink multiple access. However, in this idea, the spectrum resources occupied by the scheduling and the competition are independent and isolated, that is, a part of the spectrum resources are used for scheduling access, and another A part of the spectrum resources are used for contention access. The two parts of the resources are strictly separated and there is no overlapping of any spectrum resources. This will bring the following defects: 1) The requirements of user equipment using scheduling access and contention access are dynamically changed. Strictly dividing the available resources of the two access methods will result in dynamic user requirements and the number of available resources. Matching, which will result in waste of resources and intensified conflicts; 2) Strictly separating the resources of the scheduled access and the contention access will reduce the number of resources available to the user equipment, that is, the users using the scheduled access cannot use the contention access. Resources and vice versa, so multi-user diversity capabilities are affected.

In summary, the existing non-orthogonal multiple access technology introduces scheduling access and contention access, but the available resource partitioning of the two access modes limits network capacity, spectrum utilization, and quality of service (Quality of Service, QoS) lacks a more sophisticated and efficient multiple access method for converged scheduling access and contention access.

Summary of the invention

In order to overcome the deficiencies of the prior art and solve the problem of the prior art restricting network capacity, spectrum utilization and service quality, the present invention proposes a non-orthogonal multiple access method based on scheduling and competition fine-grained fusion, and the method The resources used for scheduling access and the resources that the user contends for are not split, so that the resources of the two access modes are completely shared, and some user equipments are scheduled to be transmitted while other user equipments are allowed to use the same spectrum resources. Random access to access, which is conducive to the dynamic change of demand and resource matching, is conducive to multi-user diversity gain, and thus significantly improve network capacity, spectrum utilization and quality of service.

The detailed steps of the technical solution adopted by the present invention to solve the technical problem thereof are as follows:

Step 1: Resource allocation phase

The central node in the wireless network calculates the resource area and the resource location occupied by the scheduling-competitive convergence access mode, and allocates resources for the scheduling user in the resource area, and proceeds to step 2;

The central node refers to a base station (BS) in a cellular network, and refers to a wireless access point (AP) in a wireless local area network;

The resource refers to a time-frequency two-dimensional resource, the resource size is measured by the number of time-frequency resource blocks, and the resource location is determined by the j-th frequency unit of the ith time slot;

Step 2: Resource indication phase

Step 2.1: The central node sends a Resource Allocation Indication Frame (RAIF) to the user equipment of the cell;

Step 2.2: In the RAIF, indicate the resource location of the scheduling-competition and convergent access mode of the cell, and the specific indication mode is that the jth frequency of the i th slot is used for the scheduling-competition and convergent access mode, and the indicated resource location is scheduled to be used by the user. Share with competing users;

Step 2.3: In the resource location of the scheduling-competition and converged access mode, the RAIF indicates the resource allocation location for the scheduling user, and the specific indication manner is that the jth frequency of the i th slot is allocated to the user equipment u, and the process proceeds to step 3. ;

Step 3: Scheduling-competitive convergence access

Step 3.1: The scheduled user equipment implements uplink access by using a non-orthogonal multiple access technology at the scheduled resource location;

Step 3.2: The contention access user performs random competition in the resource location of the scheduling-competitive convergence access mode. If the competition is successful, the non-orthogonal multiple access technology is used to implement uplink access; if the competition fails, the current access is abandoned;

The manner of random competition includes but is not limited to the backoff mode and the p probability mode;

Step 3.3: The scheduled users and competing users in steps 3.1 and 3.2 will share the same time-frequency resources, and the transmitted information is superimposed on these shared time-frequency resources, and proceeds to step 4;

Step 4: Receiver User Separation Phase

The receiving end parses the information of the multiple users that are superimposed by the multi-user joint detection method, and the process ends.

The receiver user separation phase of the non-orthogonal multiple access method based on scheduling and contention fine-grained fusion uses the SCMA method, and the information resolution steps of the BS receiver are as follows:

Step a): The original codebook corresponding to all the scheduled user equipments constitutes a deterministic codebook set, the set is not only the codebook is known, but the codebook must be active, and the active means that the user must use the codebook to send Data, go to step b);

Step b): adding a constellation point (0, 0) corresponding to each competing user, which is called an incremental codebook, and forming an incremental codebook of all competing users into an active to be determined codebook set. The codebook in the set is known, but the activity is unknown, and proceeds to step c);

Step c): combining the deterministic codebook set and the active to-be-determined codebook set to form an SCMA receiving codebook set, and then implementing a multi-user detection by using a message passing algorithm, if the determined codebook corresponding constellation is (0, 0) If it is determined to be inactive, otherwise it is determined to be active and the flow ends.

The invention has the beneficial effects that the fine-grained fusion of the two access modes is realized by adopting the method that the scheduling access and the competing access share the same spectrum resource, which is beneficial to the dynamic change of the demand and the resource matching, and is beneficial to multiple users. Diversity gain, which in turn significantly increases network capacity, spectrum utilization, and quality of service.

DRAWINGS

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a configuration diagram of a first embodiment of the present invention.

Fig. 2 is a structural diagram of a first embodiment of the present invention.

Fig. 3 is a structural diagram of a second embodiment of the present invention.

Figure 4 is a structural diagram of a third embodiment of the present invention.

Fig. 5 is a structural diagram of a fourth embodiment of the present invention.

Figure 6 is a structural diagram of Embodiment 5 of the present invention.

Fig. 7 is a structural diagram of a sixth embodiment of the present invention.

Figure 8 is a structural diagram of a seventh embodiment of the present invention.

Figure 9 is a structural diagram of an eighth embodiment of the present invention.

In the figure, STA refers to a station, and IDMA refers to Interleave-Division Multiple Access.

detailed description

The invention will now be further described with reference to the drawings and embodiments.

Specific embodiment 1

As shown in FIG. 1, the embodiment side re-described the implementation of the present invention in an SCMA-based cellular network. The proposed scheduling and competition fine-grained fusion method, in order to ensure compatibility, the radio frame structure of this embodiment follows the specification of the fourth generation mobile communication system (4G).

Step 1: each radio frame scheduling period start time, the BS first sets the uplink and downlink time slot ratio, and proceeds to step 2;

Step 2: The BS calculates the location of the time-frequency resource (Resource Block, RB) that needs to be occupied by the scheduling-competition and convergence access mode in the uplink time slot, and allocates the uplink RB to the scheduled user equipment, and the allocated RB and scheduling- The RBs occupied by the competitive convergence access method overlap, and the process proceeds to step 3;

Step 3: The BS broadcasts the resource allocation result through the physical downlink control channel (PDCCH), indicates the RB location occupied by the scheduling-competing and converged access mode, and indicates the RB resource of the scheduled user equipment, for the scheduled The user proceeds to step 4, and proceeds to step 5 for the contention access user;

Step 4: The scheduled user equipment sends data according to the allocated RB resources by SCMA, and proceeds to step 6;

Step 5: The contention access user equipment selects the access according to the probability p, and selects no access according to the (1-p) probability. If the access is selected, the SCMA layer consisting of several RBs is randomly selected, and the data is sent by using the SCMA method. And go to step 6; if you choose not to access, then give up this transmission opportunity, the process ends;

The probability p is periodically set by the base station by using downlink control signaling;

Step 6: The BS separates the information of each uplink user by using the multi-user joint detection method from the superimposed scheduling user and the competing user signal, and the process ends.

FIG. 2 is a diagram of a BS receiver design of the first embodiment. In the first embodiment, the RB resource corresponding to the scheduling-competition and convergence access mode is allocated to a part of the scheduling user and to a part of the contention users, and the scheduling user codebook is active. And the known, but the competing user's codebook activity is not known, therefore, the BS needs to separate the signals of the two types of user equipment from the superimposed signals.

Step c): combining the deterministic codebook set and the active to-be-determined codebook set to form an SCMA receiving codebook set, and then implementing a multi-user detection by using a message passing algorithm, if the determined codebook corresponding constellation is (0, 0) If it is determined to be inactive, otherwise it is determined to be active. The process ends.

Specific embodiment 2

As shown in FIG. 3, the second embodiment is based on the implementation method of the first embodiment. The difference is that the third embodiment introduces the uplink and downlink full-duplex technology. Therefore, in the third embodiment, the uplink scheduling, the downlink scheduling, and the competing users share the same RB. Resources.

Step 2: The BS calculates the RB position that needs to be occupied by the scheduling-competition and converged access mode in the uplink time slot, and allocates the uplink RB and the downlink RB for the scheduled user equipment, and the allocated RB and the scheduling-competition convergence access mode. The occupied RBs overlap and go to step 3;

Step 3: The BS broadcasts the resource allocation result, indicates the RB location occupied by the scheduling-competing and converged access mode, and indicates the RB resource of the scheduled user equipment. For the scheduled user, the process proceeds to step 4 for the contention access. The user proceeds to step 5;

Step 4: The scheduled user equipment sends and receives data according to the allocated RB resources through the SCMA mode, wherein the scheduled uplink user sends data, and the scheduled downlink user receives the data from the BS, and proceeds to step 6.

Step 5: The contention access user equipment selects the access according to the probability p, and selects no access according to the (1-p) probability. If the access is selected, the SCMA layer consisting of several RBs is randomly selected, and the data is sent by using the SCMA method. Go to step 6; if you choose not to access, discard this transmission opportunity, the process ends;

The probability p is set by the base station through downlink control signaling.

Step 6: The BS separates the information of each uplink user by using the multi-user joint detection method from the superimposed scheduling user and the competing user signal, and the user equipment adopts multiple users from the superimposed scheduling user and the competing user signal. The joint detection method separates the downlink information required by itself, and the process ends.

Concrete embodiment 3

As shown in FIG. 4, Embodiment 3 focuses on implementing the scheduling and contention fine-grained fusion method proposed by the present invention in a cellular network based on Non-orthogonal Multiple Access (NOMA).

Step 1: For the same time slot, frequency, and spatial stream, the BS divides the resource into N power layers (Power Layer) from the power domain for multiple user equipments to send uplink data, and each power layer corresponds to different transmit power. Then go to step 2;

Step 2: The BS directly schedules a part of the power layers to the corresponding uplink user equipment by using the downlink control signaling, and proceeds to step 3.

Step 3: The BS allocates the power layer of the remaining power layer in step 2 to the contention access uplink user equipment by using the downlink control signaling, and proceeds to step 4 for the scheduled user, and proceeds to step 5 for the contention access user;

Step 4: The scheduled user equipment sends uplink data according to the power required by the allocated power layer, and proceeds to step 6;

Step 5: The contention access user equipment selects the access according to the probability p, selects no access according to the (1-p) probability, and if the access is selected, randomly selects a power layer that can be used for the competition, according to the selected power layer. The required power is sent to the uplink data, and the process proceeds to step 6. If the access is not selected, the transmission opportunity is abandoned, and the process ends.

The probability p is set by the base station by using downlink control signaling;

Step 6: The BS separates the information of each user by using the multi-user joint detection method from the superimposed scheduling user and the competing user signal, and the process ends.

Concrete embodiment 4

As shown in FIG. 5, the fourth embodiment is based on the implementation method of the third embodiment. The difference is that the fifth embodiment introduces the uplink and downlink full-duplex technology. Therefore, in the fourth embodiment, the uplink scheduling, the downlink scheduling, and the competing users share the same RB resource. .

Step 1: For the same time slot, frequency and spatial stream, the BS divides the resource into N power layers from the power domain, and proceeds to step 2;

Step 2: The BS uses the downlink control signaling to directly dispatch a part of the power layer to the corresponding uplink user equipment, and directly allocates some of the remaining power layers to the corresponding downlink user equipment, and proceeds to step 3;

Step 3: The downlink control signaling is used by the BS to allocate the subset power layer allocated to the remaining power layer after the uplink and downlink in step 2 to the contention access user equipment, and the scheduled user is transferred to step 4 for the contention access. The user proceeds to step 5;

Step 4: The scheduled uplink user equipment sends uplink data according to the power required by the allocated power layer; the scheduled downlink user equipment waits to receive downlink data, and proceeds to step 6;

Step 5: The contention access user equipment selects the access according to the probability p, selects no access according to the (1-p) probability, and if the access is selected, randomly selects a power layer that can be used for the competition, according to the requirements of the selected power layer. The power is sent to the uplink data, and the process proceeds to step 6. If the access is not accessed, the transmission opportunity is abandoned, and the process ends.

Specific embodiment 5

As shown in FIG. 6, Embodiment 5 focuses on implementing the scheduling and contention fine-grained fusion method proposed by the present invention in an SCMA-based wireless local area network. To ensure compatibility, the radio frame structure of Embodiment 6 follows the IEEE 802.11ax specification. .

Step 1: When the AP sends a Trigger Frame (TF), it first calculates the time-frequency resource unit (RU) position of the scheduling-competitive convergence access mode according to the scheduling algorithm, and is scheduled. The user equipment allocates an uplink RU, and the allocated RU overlaps with the RU occupied by the scheduling-competitive convergence access mode, and proceeds to step 2;

Step 2: The BS sends a TF frame broadcast resource allocation result, indicating the location of the RU occupied by the scheduling-competition and convergence access mode, and indicates the RU resource of the scheduled user equipment. For the scheduled user, the user is transferred to step 3 for the contention. Enter the user and go to step 4;

Step 3: The scheduled user equipment sends data according to the allocated RU resources by SCMA, and proceeds to step 5;

Step 4: The contention access user equipment determines whether the current TF is accessed based on the backoff policy. If the access is selected, the SCMA layer consisting of several RUs is randomly selected, and the data is sent by the SCMA method, and the process proceeds to step 5; If you access, you will give up this transmission opportunity and the process ends.

Step 5: The AP separates the information of each uplink user by using the multi-user joint detection method from the superimposed scheduling user and the competing user signal, and the process ends.

Concrete embodiment 6

As shown in FIG. 7, Embodiment 6 focuses on implementing the scheduling and competition fine-grained in the present invention based on a multi-user shared access (MUSA)-based cellular network. Method.

Step 1: Each user equipment adopting the MUSA access mode has a corresponding multi-domain multi-code sequence corresponding thereto, and proceeds to step 2,

The complex domain multi-code sequence corresponding to the user equipment is obtained in one of two ways:

Manner 1: The BS assigns a complex domain multi-code sequence to the user equipment by using downlink signaling;

Manner 2: The user equipment itself is uniquely bound to a complex domain multi-code sequence;

Step 2: The BS will schedule a part of users to perform uplink transmission by using downlink control signaling, and the BS proceeds to step 3, and the scheduled user equipment proceeds to step 4;

Step 3: The BS triggers the random access user to perform random access by using the downlink control signaling, the BS proceeds to step 6, and the random access user equipment proceeds to step 5;

Step 4: The scheduled user equipment uses its corresponding complex domain multi-code to perform the spread spectrum modulation transmission uplink data, and proceeds to step 6.

Step 5: The contention access user equipment selects the access according to the probability p, and selects no access according to the (1-p) probability. If the access is selected, the complex domain multi-code corresponding to the multi-domain multi-code is used for the spread spectrum modulation to transmit the uplink data. The method is familiar to the industry, and proceeds to step 6. If you choose not to access, the transmission opportunity is abandoned, and the process ends.

Step 6: The BS separates the information of each user by using the serial interference cancellation method from the superimposed scheduling user and the competing user signal, and the process ends.

Concrete embodiment 7

As shown in FIG. 8, Embodiment 7 focuses on implementing the scheduling and contention fine-grained fusion method proposed by the present invention in a Pattern-Division Multiple Access (PDMA)-based cellular network.

Step 1: The base station pre-configures the PDMA multi-user coding matrix H _PDMA by means of downlink signaling, multi-user coding matrix is multiplied by the generalized channel matrix H _{CH to} form an equivalent channel transmission matrix H, and proceeds to step 2;

The PDMA multi-user coding matrix H _PDMA is M rows and N columns, wherein the H _PDMA row set represents a resource dimension, including one or more of a spatial domain dimension, a coding domain dimension, and a power domain dimension, each row pair A resource unit in the resource dimension, the column set of H _PDMA represents the user equipment, and each column corresponds to one user equipment; each element has a value of 1 or 0, wherein 1 represents the corresponding user of the column uses the corresponding row of the row The resource sends data, and 0 means that the user corresponding to the column does not use the resource corresponding to the row to send data;

Step 2: The BS schedules a part of users to perform uplink transmission by using downlink control signaling, and the BS proceeds to step 3, and the scheduled user equipment proceeds to step 4;

Step 4: The scheduled user equipment performs uplink data transmission based on the resource unit whose value is 1 in the column of the PDMA multi-user coding matrix, which is well known in the industry, and proceeds to step 6.

Step 5: The contention access user equipment selects access according to the probability p, and selects no access according to the (1-p) probability. If the access is selected, the user equipment is based on the resource with the value 1 in the corresponding column of the PDMA multi-user coding matrix. The uplink data transmission is performed on the unit, which is well known in the industry, and proceeds to step 6. If the selection is not accessed, the transmission opportunity is abandoned, and the process ends.

Step 6: The BS separates the information of each user by using the front-end detection and the serial interference cancellation from the superimposed scheduling user and the competing user signal, and the process ends.

Embodiment 8

As shown in FIG. 9, Embodiment 8 focuses on implementing the scheduling and contention fine-grained fusion method proposed by the present invention in an Interleave-Division Multiple Access (IDMA)-based cellular network.

Step 1: Each user equipment has a unique corresponding interleaving sequence, and the IDMA system distinguishes different user equipments based on the interleaving sequence, and proceeds to step 2;

The interleaving sequence is generated in the following three ways:

Manner 1: The base station allocates a unique interleaving sequence for each user equipment;

Mode 2: Each user equipment uniquely determines a static interleaving sequence;

Manner 3: The interleaving sequence of each user equipment changes randomly with time, but the interleaving sequence of the base station is consistent with the interleaving sequence of the user equipment;

Step 2: The BS schedules a part of users to perform uplink transmission by using downlink control signaling, and proceeds to step 3;

Step 3: The BS triggers the random access user to perform random access by using the downlink control signaling, and proceeds to step 4.

Step 4: The data of the scheduled user equipment first passes through the universal encoder, and then passes through its own interleaver, and then is modulated and sent to the communication channel, which is well known in the industry, and proceeds to step 6;

The universal encoder has the following three forms:

Form 1: There is only one forward error corrector;

Form 2: only one spreader;

Form 3: first through the forward error correction, after passing through the spreader;

Step 5: The contention access user equipment selects access according to the probability p, and selects no access according to the (1-p) probability. If the access is selected, the data of the user equipment first passes through the universal encoder, and then passes through its own interleaver. Then, the modulation is sent to the communication channel, which is well known in the industry, and proceeds to step 6. If the selection is not accessed, the transmission opportunity is abandoned, and the process ends.

Step 6: The BS separates the information of each user from the superimposed scheduling user and the competing user signal by using a chip-level iterative decoder, and the process ends.

Claims

A non-orthogonal multiple access method based on fine-grained convergence of scheduling and competition, characterized in that the method comprises the following steps:

Step 1: Resource allocation phase

The central node in the wireless network calculates the resource area and the resource location occupied by the scheduling-competitive convergence access mode, and allocates resources for the scheduling user in the resource area, and proceeds to step 2;

The central node refers to a base station in a cellular network, and refers to a wireless access point in a wireless local area network;

The resource refers to a time-frequency two-dimensional resource, the resource size is measured by the number of time-frequency resource blocks, and the resource location is determined by the j-th frequency unit of the ith time slot;

Step 2: Resource indication phase

Step 2.1: The central node sends a resource allocation indication frame to the user equipment of the local cell.

Step 2.2: In the RAIF, indicate the resource location of the scheduling-competition and convergent access mode of the cell, and the specific indication mode is that the jth frequency of the i th slot is used for the scheduling-competition and convergent access mode, and the indicated resource location is scheduled to be used by the user. Share with competing users;

Step 2.3: In the resource location of the scheduling-competition and converged access mode, the RAIF indicates the resource allocation location for the scheduling user, and the specific indication manner is that the jth frequency of the i th slot is allocated to the user equipment u, and the process proceeds to step 3. ;

Step 3: Scheduling-competitive convergence access

Step 3.1: The scheduled user equipment implements uplink access by using a non-orthogonal multiple access technology at the scheduled resource location;

Step 3.2: The contention access user performs random competition in the resource location of the scheduling-competitive convergence access mode. If the competition is successful, the non-orthogonal multiple access technology is used to implement uplink access; if the competition fails, the current access is abandoned;

The manner of random competition includes but is not limited to the backoff mode and the p probability mode;

Step 3.3: The scheduled users and competing users in steps 3.1 and 3.2 will share the same time-frequency resources, and the transmitted information is superimposed on these shared time-frequency resources, and proceeds to step 4;

Step 4: Receiver User Separation Phase

The receiving end parses the information of the multiple users that are superimposed by the multi-user joint detection method, and the process ends.
The non-orthogonal multiple access method based on scheduling and contention fine-grained fusion according to claim 1, wherein the receiver user of the non-orthogonal multiple access method based on scheduling and contention fine-grained fusion is used The SCMA method is used in the separation phase, and the information analysis at the receiving end of the BS includes the following steps:

Step a): The original codebook corresponding to all the scheduled user equipments constitutes a deterministic codebook set, the set is not only the codebook is known, but the codebook must be active, and the active means that the user must use the codebook to send Data, go to step b);

Step b): adding a constellation point (0, 0) corresponding to each competing user, which is called an incremental codebook, and forming an incremental codebook of all competing users into an active to be determined codebook set. The codebook in the set is known, but the activity is unknown, and proceeds to step c);

Step c): combining the deterministic codebook set and the active to-be-determined codebook set to form an SCMA receiving codebook set, and then implementing a multi-user detection by using a message passing algorithm, if the determined codebook corresponding constellation is (0, 0) If it is determined to be inactive, otherwise it is determined to be active and the flow ends.