阅读说明：本技术 一种基于非正交多址系统的子信道调度与功率分配联合优化方法 (Sub-channel scheduling and power distribution joint optimization method based on non-orthogonal multiple access system ) 是由 吴江 黄勇 徐伟强 朱升宏 于 2019-09-23 设计创作，主要内容包括：本发明涉及一种基于非正交多址系统的子信道调度与功率分配联合优化方法,包括以下步骤：(1)初始化参数,包括：协作多点聚类内基站集B,每个小区的移动用户集M,联合子信道集K,参考信道增益阈值V,信源功率上限P<Sub>s</Sub>,信宿处噪声功率<Image he="71" wi="89" file="DDA0002210814550000011.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>(2)利用用户选择和偏好集排序算法得到信道增益矩阵、等效信道增益矩阵、参考数据速率,分别记为D<Sub>b</Sub>、<Image he="71" wi="99" file="DDA0002210814550000012.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>R<Sub>sum</Sub>；采用二进制元素<Image he="78" wi="86" file="DDA0002210814550000015.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>表示小区b的联合子信道k是否分配给用户M<Sub>j</Sub>,<Image he="86" wi="90" file="DDA0002210814550000013.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>表示小区b的联合子信道k分配给用户j的功率；(3)利用联合子信道-用户匹配算法和注水功率法得到步骤(2)中<Image he="90" wi="235" file="DDA0002210814550000014.GIF" imgContent="drawing" imgFormat="GIF" orientation="portrait" inline="no"></Image>的最优解。本发明对子信道调度和功率分配联合考虑,最大化总和速率的同时保证了用户公平性,改善小区边缘用户性能,提高用户使用体验。(The invention relates to a sub-channel scheduling and power allocation joint optimization method based on a non-orthogonal multiple access system, which comprises the following steps: (1) initializing parameters, including: a base station set B in the coordinated multi-point clustering, a mobile user set M of each cell, a combined sub-channel set K, a reference channel gain threshold V and an information source power upper limit P s Noise power at sink (2) Obtaining a channel gain matrix, an equivalent channel gain matrix and a reference data rate by utilizing a user selection and preference set ordering algorithm, and respectively recording the channel gain matrix, the equivalent channel gain matrix and the reference data rate as D b 、 R sum (ii) a Using binary elements Indicating whether the joint subchannel k of cell b is allocated to user M j ， Represents the power allocated to user j by the joint subchannel k of cell b; (3) obtaining the power of step (2) by using a joint subchannel-user matching algorithm and a water filling power method The optimal solution of (1). The invention considers the sub-channel scheduling and the power allocation jointly, ensures the user fairness while maximizing the sum rate, improves the performance of the users at the edge of the cell and improves the user experience.)

1. A sub-channel scheduling and power allocation joint optimization method based on a non-orthogonal multiple access system is characterized by comprising the following steps:

(1) initializing parameters, including: a base station set B in the coordinated multi-point clustering, a mobile user set M of each cell, a combined sub-channel set K, a reference channel gain threshold V and an information source power upper limit P_sNoise power at sink

(2) Obtaining channel gain matrix and equivalent channel by using user selection and preference set ordering algorithmThe gain matrix and the reference data rate are respectively marked as D_b、

(3) obtaining the power of step (2) by using a joint subchannel-user matching algorithm and a water filling power method

2. The method of claim 1, wherein the step (1) further comprises:

dividing users of each cell into a center user and an edge user according to the equivalent channel gain, and respectively representing the users by a CCU (central channel unit) and a CEU (central channel unit); the center user is a non-CoMP user, and the edge user is a CoMP user;

assuming that B base stations are a CoMP cluster, the CoMP user set actually scheduled by all joint base stations in the CoMP cluster is: CEU ═ CEU₁,CEU₂,...,CEU_B]；

The user set actually scheduled by each CoMP base station is: u shape_b＝[CCU_b,CEU],(b∈B)；

The number of users actually scheduled by each CoMP base station is: u shape_b＝card(U_b)；

The total users of the base station with the largest number of scheduling users in the CoMP cluster are: max (U)_b)。

3. The method of claim 2, wherein the step (2) comprises:

suppose that

wherein, phi_jIndicating the out-of-cell interference experienced by user j,

when card (S)_j) When 1, the user j of the cell b is a non-CoMP user, and the user j can eliminate

When card (S)_j)>1, user j of cell b is CoMP user, then user j can eliminate joint sub-channelInternal interference caused by other users having upper channel gains smaller than their own channel gains, i.e.

Therefore, when M is_j∈CCU_bWhen the temperature of the water is higher than the set temperature,upper M_jThe sum rate of (c) is:

wherein the content of the first and second substances,

when M is_jWhen the element belongs to the CEU,

wherein the content of the first and second substances,

of cell b

introducing a K multiplied by U joint sub-channel distribution matrix, and evaluating the system performance by the sum rate of all users in CoMP cluster:

4. the method as claimed in claim 3, wherein the sub-channel scheduling and power allocation joint optimization method based on the non-orthogonal multiple access system is characterized in that

is provided with

wherein, the objective function is formula (9a), and the system sum rate of CoMP clustering is determined by the subchannel and the power; equation (9b) ensures that each subchannel is superimposed by q at most_uA user; equation (9c) ensures that each user consists of q at most_lScheduling the sub-channels; equation (9d) is the interference term of the objective function, and the optimization problem is a non-convex optimization problem; each user power coefficient satisfies equations (9e) and (9 f).

5. The method of claim 4, wherein the user selection and preference set ordering algorithm of step (2) comprises the following steps:

(2.1) the base station broadcasts the acquired reference channel gain set as:

(2.2) CoMP user partitioning: setting a reference channel gain threshold V of an algorithm, and dividing a user set according to channel gains between sub-channels and users in a cell;

if max (D)_b,j) If V is less than or equal to V, then CEU_bJ, j; otherwise, CCU_bJ, j; the base station sends signals with the same reference power, and then the users U of the cell_b,jThe channel gain of (2) can be equivalent channel gain

and (3) calculating data rate sets of the users of the cells in different sub-channels when the same reference power is distributed according to the following formulas (2), (5) and (11):

6. the method of claim 5, wherein the step (3) comprises the steps of:

(3.1) establishing a set { K }_bmatch }, recording users matched with each sub-channel in the cell b at present;

(3.2) preparation of

(3.3) according to { P (U)_b) And { P (K) } and { P_b) Judging the result of each round of mutual selection, and updatingAnd { P (U)_b,j)}；

(3.4) the power allocation uses a water-filling power algorithm as follows:

7. the method of claim 6, wherein the step (3.3) specifically comprises:

(a) input { P (U)_b)}，{P(K_b)}；

(b) Building a setRecording users matched with each sub-channel in the cell b at present;

(c) sub-channel matching process: each U_b,j∈U_bSelf-referral to preference set { P (U)_b,j) The sub-channel with the highest satisfaction:

if it is not

(d) Judging whether to schedule the self-recommended edge users: if it is not

(e) updating the preference set of the sub-channels and the preference set of the user:

(f) Judging whether the loop condition of the algorithm is met: if it is not

8. The method as claimed in claim 1, wherein the method for joint optimization of sub-channel scheduling and power allocation based on non-orthogonal multiple access system is characterized in that

9. The method of claim 1, wherein each mobile subscriber in the mobile subscriber set and each base station in the base station set are both single antenna.

Technical Field

The invention belongs to the technical field of wireless communication, and particularly relates to a sub-channel scheduling and power allocation joint optimization method based on a non-orthogonal multiple access system.

Background

The rapid development of mobile communication technology has made the demand for data transmission rate and communication service quality higher and higher. In a Non-Orthogonal Multiple Access (NOMA) system, on one hand, the Non-Orthogonal Multiple Access technology can improve the spectrum efficiency and the network throughput, and thus becomes one of the key technologies for the next generation of mobile communication; on the other hand, the traditional non-orthogonal multiple access system superimposes user transmission on the same resource block, which results in increased interference to edge users using the same spectrum resource, and reduces the service quality and user fairness of cell edge users.

The Coordinated Multi-Point (CoMP) technology has the characteristics of reducing inter-cell interference and improving cell throughput and cell edge user performance. Meanwhile, a coordinated multi-point-based non-orthogonal multiple access (NOMA-CoMP) technology has significant theoretical research and application values, and the technology can improve the spectrum efficiency and simultaneously reduce the inter-cell interference, so that the overall throughput of a cell is improved. In the prior art, research on a NOMA-CoMP system mainly focuses on optimizing multi-cell user power allocation or only considering a single-cell subchannel scheduling problem, and the multi-cell subchannel scheduling and power allocation joint optimization problem is not considered in the technologies.

Disclosure of Invention

Based on the above disadvantages in the prior art, the present invention provides a method for joint optimization of sub-channel scheduling and power allocation based on a non-orthogonal multiple access system.

In order to achieve the purpose, the invention adopts the following technical scheme:

a sub-channel scheduling and power allocation joint optimization method based on a non-orthogonal multiple access system comprises the following steps:

(1) initializing parameters, including: a base station set B in the coordinated multi-point clustering, a mobile user set M of each cell, a combined sub-channel set K, a reference channel gain threshold V and an information source power upper limit P_sNoise power at sink

(2) Obtaining a channel gain matrix, an equivalent channel gain matrix and a reference data rate by utilizing a user selection and preference set ordering algorithm, and respectively recording the channel gain matrix, the equivalent channel gain matrix and the reference data rate as D_b、

R_sum(ii) a Using binary elements

Indicating whether the joint subchannel k of cell b is allocated to user M_j，

Represents the power allocated to user j by the joint subchannel k of cell b;

(3) obtaining the power of step (2) by using a joint subchannel-user matching algorithm and a water filling power method

The optimal solution of (1).

Preferably, the step (1) further comprises:

dividing users of each cell into a center user and an edge user according to the equivalent channel gain, and respectively representing the users by a CCU (central channel unit) and a CEU (central channel unit); the center user is a non-CoMP user, and the edge user is a CoMP user;

assuming that B base stations are a CoMP cluster, the CoMP user set actually scheduled by all joint base stations in the CoMP cluster is: CEU ═ CEU₁,CEU₂,...,CEU_B]；

The user set actually scheduled by each CoMP base station is: u shape_b＝[CCU_b,CEU],(b∈B)；

The number of users actually scheduled by each CoMP base station is: u shape_b＝card(U_b)；

The total users of the base station with the largest number of scheduling users in the CoMP cluster are: max (U)_b)。

Preferably, the step (2) comprises:

suppose that

The transmission signal, S, on subchannel k representing cell b_jRepresents the set of base stations that scheduled user j,

representing the channel coefficient of user j on sub-channel k of cell, the transmission signal on joint sub-channel k of cell b at the receiving end of user j is represented as:

wherein, phi_jIndicating the out-of-cell interference experienced by user j,

representing the superposition of white gaussian noise,

is a noise variable;

when card (S)_j) When 1, user j of cell b is notCoMP user, then user j can cancel

Internal interference caused by other users having upper channel gains smaller than their own channel gains, i.e.

When card (S)_j)>1, user j of cell b is CoMP user, then user j can eliminate joint sub-channel

Internal interference caused by other users having upper channel gains smaller than their own channel gains, i.e.

And external interference

Therefore, when M is_j∈CCU_bWhen the temperature of the water is higher than the set temperature,upper M_jThe sum rate of (c) is:

wherein the content of the first and second substances,

represents the CCU_bM of (A)_jIn that

The internal interference suffered by the above-mentioned method,

represents the CCU_bM of (A)_jThe external interference suffered by the system is reduced,

when M is_jWhen the element belongs to the CEU,

upper M_jThe sum rate of (c) is:

wherein the content of the first and second substances,

m representing CEU_jIn that

The internal interference suffered by the above-mentioned method,

of cell b

The sum rate of (c) is:

introducing a K multiplied by U joint sub-channel distribution matrix, and evaluating the system performance by the sum rate of all users in CoMP cluster:

preferably, the above-mentionedThe optimal solution of (a) is:

is provided withThe overall and speed of the system are maximized, and the optimization problem is converted into:

wherein, the objective function is formula (9a), and the system sum rate of CoMP clustering is determined by the subchannel and the power; equation (9b) ensures that each subchannel is superimposed by q at most_uA user; equation (9c) ensures that each user consists of q at most_lScheduling the sub-channels; equation (9d) is the interference term of the objective function, and the optimization problem is a non-convex optimization problem; each user power coefficient satisfies equations (9e) and (9 f).

Preferably, the user selection and preference set ordering algorithm of step (2) comprises the following steps:

(2.1) the base station broadcasts the acquired reference channel gain set as:

(2.2) CoMP user partitioning: setting a reference channel gain threshold V of an algorithm, and dividing a user set according to channel gains between sub-channels and users in a cell;

if max (D)_b,j) If V is less than or equal to V, then CEU_bJ, j; otherwise, CCU_bJ, j; the base station sends signals with the same reference power, and then the users U of the cell_b,jThe channel gain of (2) can be equivalent channel gain

Represents; the equivalent set of reference gains is expressed as:

and (3) calculating data rate sets of the users of the cells in different sub-channels when the same reference power is distributed according to the following formulas (2), (5) and (11):

preferably, the step (3) comprises the following steps:

(3.1) establishing a set { K }_bmatch }, recording users matched with each sub-channel in the cell b at present;

(3.2) preparation of

And

solving a set of user preferences { P (U)_b) And a joint subchannel preference set { P (K) }_b) And i.e.:

(3.3) according to { P (U)_b) And { P (K) } and { P_b) Judging the result of each round of mutual selection, and updating

And { P (U)_b,j)}；

(3.4) the power allocation uses a water-filling power algorithm as follows:

preferably, the step (3.3) specifically comprises:

(a) input { P (U)_b)}，{P(K_b)}；

(b) Building a set

Recording users matched with each sub-channel in the cell b at present;

(c) sub-channel matching process: each U_b,j∈U_bSelf-referral to preference set { P (U)_b,j) The sub-channel with the highest satisfaction:

if it is not

Then select

The user of (1) is reserved; otherwise, from selection

Is selected from among the users of (1)_uThe user with the highest satisfaction degree updates

(d) Judging whether to schedule the self-recommended edge users: if it is not

Select CEU_bIn (1) U_b,jIf the joint scheduling set S is present_jBase station in (1) selects scheduling U at the same time_b,jThen, U is reserved_b,j(ii) a If not, then,

updating

Otherwise, the next step;

(e) updating the preference set of the sub-channels and the preference set of the user:

from

In deleting the selected U_b,jUpdateIf it is not

Then is in { P (U)_b,j) In (1) } deletion

Update { P (U)_b,j) }; otherwise, in { P (U)_b,j) Delete the selected U_b,jCorresponding preference set sequence, updating { P (U)_b,j)}；

(f) Determining if algorithm cycles are satisfiedRing conditions: if it is not

Or

Returning to the step (c); otherwise, the algorithm is ended.

Preferably, the above-mentioned

The power of the joint subchannel k of the representation cell b to be distributed to the user j satisfies

And

wherein, P_sThe total transmit power of each base station is equal for the total transmit power of each base station.

As a preferred scheme, each mobile user in the mobile user set and each base station in the base station set are both a single antenna.

The invention has the beneficial effects that: the invention considers the sub-channel scheduling problem and the power distribution problem jointly, can ensure the user fairness while maximizing the sum rate, improves the performance of users at the edge of the cell, and improves the use experience of wireless network users. In addition, the invention simplifies the complex non-convex model into a many-to-many bilateral matching problem, thereby greatly saving complexity.

Drawings

Fig. 1 is a diagram of a system model for a method of joint optimization of subchannel scheduling and power allocation based on a non-orthogonal multiple access system according to an embodiment of the present invention.

Fig. 2 is a specific flowchart of a sub-channel scheduling and power allocation joint optimization method based on a non-orthogonal multiple access system according to an embodiment of the present invention.

Fig. 3 is a diagram of average sum rate of cells versus the number of users in the cell according to an embodiment of the present invention.

Fig. 4 is a diagram of the average sum rate of edge users of a cell according to an embodiment of the present invention.

Detailed Description

In order to more clearly illustrate the embodiments of the present invention, the following description will explain the embodiments of the present invention with reference to the accompanying drawings. It is obvious that the drawings in the following description are only examples of the invention, and that for a person skilled in the art, other drawings and embodiments can be derived from them without inventive effort. The following describes the sub-channel scheduling and power allocation joint optimization method based on the non-orthogonal multiple access system in detail.

As shown in fig. 1, the present invention considers a dual-cell NOMA-CoMP system model, where B denotes a base station set, and M denotes a mobile user set of each cell, assuming that both the user and the base station are single antennas. The base station divides the available bandwidth into a set of subchannels, K. Assuming that the BS can know complete Channel State Information (CSI), the BS can perform joint subchannel scheduling and power allocation for users according to the complete CSI.

As shown in fig. 2, in the sub-channel scheduling and power allocation joint optimization method based on the non-orthogonal multiple access system according to the embodiment of the present invention, a channel gain matrix, an equivalent channel gain matrix, and a reference data rate are obtained by using the proposed user selection and preference set ordering algorithm, so as to further obtain a user preference set { P (U) } in the embodiment of the present invention_b) And a joint subchannel preference set { P (K) }_b)}. Then, converting the optimization problem into a many-to-many bilateral matching problem by using a matching theory; secondly, the proposed joint sub-channel-user matching algorithm is used for solving the optimal joint sub-channel-user matching

And finally, solving a user distribution power matrix of each cell by using a water injection power algorithm. The invention considers the sub-channel scheduling problem and the power distribution problem jointly, can ensure the user fairness while maximizing the sum rate, improves the performance of users at the edge of the cell, and improves the use experience of wireless network users. Specifically, the method comprises the following steps:

(1) initializing parameters: a base station set B in the CoMP cluster, a mobile user set M of each cell, a combined sub-channel set K, a reference channel gain threshold V and an information source power upper limit P_sNoise power at sink

(2) Obtaining a channel gain matrix, an equivalent channel gain matrix and a reference data rate by using the proposed user selection and preference set ordering algorithm, and respectively recording the channel gain matrix, the equivalent channel gain matrix and the reference data rate as D_b、

R_sum(ii) a Using binary elementsIndicating whether the joint subchannel k of cell b is allocated to user M_j，Represents the power allocated to user j by the joint subchannel k of cell b;

(3) obtaining the optimal solution of the problem in the step (2) by using the proposed joint sub-channel-user matching algorithm and the water filling power method

Specifically, the present invention divides users of each cell into center users and edge users, denoted as CCU and CEU, respectively, according to the equivalent channel gain.

The center user is a non-CoMP user, and the edge user is a CoMP user. Assuming that B base stations are a CoMP cluster, the CoMP user set actually scheduled by all joint base stations in the CoMP cluster is: CEU ═ CEU₁,CEU₂,...,CEU_B]. The user set actually scheduled by each CoMP base station (including the user scheduling the local cell as the main base station and the CoMP user scheduling the cooperative cell as the cooperative base station) is: u shape_b＝[CCU_b,CEU]And (B. epsilon. B). The number of users actually scheduled by each CoMP base station is: u shape_b＝card(U_b). The total users of the base station with the largest number of scheduling users in the CoMP cluster are: max (U)_b)。

The power of the subchannel k of the expression cell b to be distributed to the user j satisfies

And

wherein P is_sFor the total transmit power of each base station, it is assumed that the total transmit power of the respective base stations is equal.

The embodiment of the invention considers that the transmission channel is a block fading channel, and assumes that

The transmission signal, S, on subchannel k representing cell b_jRepresents the set of base stations that scheduled user j,

representing the channel coefficient of user j on sub-channel k of cell, the transmission signal on sub-channel k of cell b at the receiving end of user j is represented as:

wherein, phi_jIndicating the out-of-cell interference experienced by user j,

representing superimposed white gaussian noise (AWGN),

is a noise variable.

When card (S)_j) When 1, the user j of the cell b is a non-CoMP user, and the user j can eliminate

Internal interference caused by other users having upper channel gains smaller than their own channel gains, i.e.

When card (S)_j)>1, user j of cell b is CoMP user, then user j can eliminate joint sub-channelInternal interference caused by other users having upper channel gains smaller than their own channel gains, i.e.

And external interference

Therefore, when M is_j∈CCU_bWhen the temperature of the water is higher than the set temperature,

upper M_jThe sum rate of (c) is:

wherein the content of the first and second substances,

represents the CCU_bM of (A)_jIn that

The internal interference suffered by the above-mentioned method,

represents the CCU_bM of (A)_jThe external interference suffered by the system is reduced,

when M is_jWhen the element belongs to the CEU,upper M_jThe sum rate of (c) is:

wherein

M representing CEU_jIn that

The internal interference suffered by the above-mentioned method,

of cell b

The sum rate of (c) is:

introducing a K multiplied by U joint sub-channel distribution matrix, and evaluating the system performance by the sum rate of all users in CoMP cluster:

the purpose of the invention is to provide

The overall and speed of the system are maximized, and the optimization problem is converted into:

the constraint (9b) ensures that each subchannel is superimposed by q at most_uA user (9c) ensures that each user consists of q at most_lAnd scheduling the sub-channels. Due to the base station transmit power limitation, each user power coefficient must satisfy conditions (9e) and (9 f).

Since the constraint (9d) is also the interference term of the objective function, it can be seen that the above optimization problem is a non-convex optimization problem. The invention solves the problems of sub-channel allocation and power allocation of each CoMP cell respectively.

As can be seen from the objective function (9a), the system sum rate of CoMP clustering is determined by both the subchannel and the power. Considering the system computation complexity, the method firstly allocates the combined sub-channel of the CoMP cell, and the combined sub-channel-user many-to-many bilateral matching strategy is implemented by the following steps:

in the first step, the user selection and preference set ordering algorithm is implemented by the following steps:

1) the base station broadcast acquires a set of reference channel gains denoted as

2) CoMP user division: and setting a reference channel gain threshold V of the algorithm, and dividing the user set according to the channel gain between each subchannel and the user in the cell. If max (D)_b,j) If V is less than or equal to V, then CEU_bJ, j; otherwise, CCU_bJ. Since the base stations transmit signals with the same reference power, the users U of the cell_b,jThe channel gain of (a) may be equivalent channel gain:

and (4) showing.

The equivalent reference gain set is represented as

The data rate sets of the users in different sub-channels of each cell when the same reference power is allocated can be obtained by the above equations (2), (5) and (11):

wherein the content of the first and second substances,

in the second step, the joint subchannel-user matching algorithm is implemented by the following steps:

1) and (3) conversion of many-to-many bilateral matching problems: the subchannel set and the actually scheduled user set of each cell are taken as a group of two non-cooperative sets, and players in the two sets in each group are selfish and rational and are targeted to maximize benefits of the players. If the subchannel of cell b

Is allocated to a scheduling user U_b,jThen call

And U_b,jAre paired with each other and form a matched pair

Where Θ represents the matching mapping.

2) Assume that each player in the same set within a group has a complete set of preferences for other players in another set within the group.

The set of users of group b centralizes the player's preference set as:

the set of players' preferences in the set of subchannels of group b is represented as:

the core idea of the algorithm is that each user of each cell self-recommends a respective preference set P (U)_b) The joint sub-channel with the highest satisfaction, for example: suppose that each user sends respective resume to the joint sub-channel (the non-CoMP users only send the resume to the sub-channel with the highest satisfaction degree of the cell where the non-CoMP users are located, and the CoMP users send the resume to the sub-channel with the highest satisfaction degree of each cell in the CoMP cluster where the non-CoMP users are located), and then each sub-channel of each cell sends the resume to the sub-channel with the highest satisfaction degree of each cell according to the preference set of each user

And the user can be refused or accepted, and once all users submit resumes to the sub-channel with the highest degree of satisfaction, the round of mutual selection is called to be finished.

The specific process is as follows:

(a) input { P (U)_b)}，{P(K_b)}；

(b) Building a setRecording users matched with each sub-channel in the cell b at present;

(c) sub-channel matching process: each U_b,j∈U_bSelf-referral to preference set { P (U)_b,j) The sub-channel with the highest satisfaction:

if it is not

Then select

The user of (1) is reserved; otherwise, from selection

Is selected from among the users of (1)_uThe user with the highest satisfaction degree updates

(d) Judging whether to schedule the self-recommended edge users: if it is not

Select CEU_bIn (1) U_b,jIf the joint scheduling set S is present_jBase station in (1) selects scheduling U at the same time_b,jThen, U is reserved_b,j(ii) a If not, then,updatingOtherwise, the next step;

(e) updating the preference set of the sub-channels and the preference set of the user:

from

In deleting the selected U_b,jUpdateIf it is not

Then is in { P (U)_b,j) In (1) } deletion

Update { P (U)_b,j)}: otherwise, in { P (U)_b,j) Delete the selected U_b,jCorresponding preference set sequence, updating { P (U)_b,j)}；

(f) Judging whether the loop condition of the algorithm is met: if it is not

OrReturning to the step (c); otherwise, the algorithm is ended.

The strategy for combining the subchannel-user many-to-many bilateral matching in the step (3) specifically comprises the following steps:

(3.1) establishing a set { K }_bmatch records the users matched with each sub-channel in the cell b at present;

(3.2) preparation ofAnd

further solving a user preference set { P (U)_b) And a joint subchannel preference set { P (K) }_b) And i.e.:

(3.3) according to { P (U)_b) And { P (K) } and { P_b) Judging the result of each round of mutual selection, and updating

And { P (U)_b,j)}。

(3.4) implementing a water filling power algorithm:

fig. 3-4 are simulation verifications of a designed solution by an embodiment of the present invention through Mtalab. The parameters are specifically designed as follows: the peak power of the base station is set to P_s46dBm, noise variance of

And it is assumed that users are randomly distributed in the respective cells at each time. The simulation results were averaged over 1000 time slots.

Figure 3 shows the relationship between the average sum rate of users per CoMP cell and the number of users per cell, where each cell has 6 joint subchannels. As can be seen from fig. 3, the performance of the proposed NOMA-CoMP system based on the joint subchannel-user matching algorithm is 83.39% higher than that of the orthogonal multiple access algorithm based on the coordinated multiple points, because each subchannel of the orthogonal multiple access system can only schedule one user in the same time slot, and the base station does not fully utilize the spectrum resources. In the maximum throughput algorithm, assuming no difference among the sub-channels, the power distribution and sub-distribution of the sub-channels are firstly carried out on each user, and each sub-channel is sequentially distributed to a CoMP user or a non-CoMP user according to the algorithm provided by the literature. It is assumed herein that each subchannel is differentiated, and the proposed joint subchannel-user matching algorithm performs power allocation after subchannel allocation. When the difference between sub-channels of the cells is assumed, the performance of the algorithm is 12.4% higher than that of the maximum throughput algorithm.

Figure 4 shows the relationship between the average sum rate of edge users per CoMP cell and the number of users per cell, where each cell has 6 subchannels. As can be seen from fig. 4, the algorithm proposed herein can well protect the probability of edge user selection even when the number of users increases, and improve user fairness, so that the average sum rate of edge users of each CoMP cell of the joint subchannel-user matching algorithm is better than that of other algorithms.

Under the condition of considering the differentiation of the combined sub-channels, the invention provides a user selection and preference set ordering algorithm based on a CoMP user selection mode and a combined sub-channel-user matching algorithm based on an expanded Galer Shapril version, and develops a combined sub-channel-user many-to-many bilateral matching strategy of a non-orthogonal multi-access wireless network based on coordinated multiple points on the basis, and meanwhile, a water injection power method is adopted to distribute power, so that the total rate of all coordinated multiple point cells is maximized and the fairness of users is ensured; in addition, in the invention, the complicated non-convex model is simplified into a many-to-many bilateral matching problem, so that the complexity can be greatly saved.

The foregoing has outlined rather broadly the preferred embodiments and principles of the present invention and it will be appreciated that those skilled in the art may devise variations of the present invention that are within the spirit and scope of the appended claims.