User selection and power allocation method and system based on space division multiplexing

Abstract

The application provides a user selection and power distribution method and system based on space division multiplexing, comprising: based on a multi-user MIMO scene, acquiring system parameters, including a channel matrix, total transmission power, noise power, the number of base station antennas, the total number of users and the number of users to be selected; calculating the equivalent power gain of each user under the multi-user scene based on the channel matrix; constructing a corresponding search tree based on the single-user equivalent power gain under the single-user scene as a priority reference; constructing an optimization problem of joint user selection and power distribution with a user subset and power distribution as joint optimization variables and the sum of the achievable rates of all served users as an objective function; solving the optimization problem of joint user selection and power distribution and outputting the optimal user subset and the optimal power distribution scheme. The application can dynamically select an optimal user set and distribute power under the constraint of limited transmission power, and realizes the maximization of system capacity.

Description

Technical Field

[0001] This invention belongs to the field of wireless communication technology, and particularly relates to a user selection and power allocation method and system based on spatial multiplexing. Background Technology

[0002] Multiple-input multiple-output (MIMO) technology is the core of modern wireless communication systems. By deploying multiple antennas at the base station and user end, it achieves spatial signal multiplexing and diversity gain. MIMO systems can significantly improve spectrum utilization and system capacity, especially in multi-user scenarios, greatly enhancing the network's concurrency capabilities by spatially separating and serving multiple users simultaneously.

[0003] Spatial Division Multiplexing (SDM) is a key technology in MIMO systems. SDM allows base stations to utilize spatial resources to allocate different data streams to different users or multiple data streams from the same user, achieving "spatial multiplexing" and thus transmitting more information within the same frequency band and time slot. SDM performance is highly dependent on the channel characteristics between users, especially the orthogonality and independence of the channels. If user selection is inappropriate, and the selected user channels are highly correlated, the spatial multiplexing gain will decrease significantly, potentially even limiting system capacity. Therefore, designing an efficient user selection algorithm is crucial for SDM scenarios. This algorithm needs to comprehensively consider user channel gain, spatial correlation, and system resource constraints to dynamically select the optimal user set, providing a solid foundation for subsequent power allocation and precoding, thereby maximizing the capacity and improving the quality of service in multi-user MIMO systems. Traditional exhaustive search methods have extremely high computational complexity and are difficult to apply to large-scale systems; therefore, an efficient joint user selection and power allocation method is urgently needed. Summary of the Invention

[0004] This invention is applicable to 5G and future multi-antenna base station scenarios. Under limited transmit power constraints, it can dynamically select the optimal user set and allocate power to maximize system capacity. Typical applications include cellular base station downlink scheduling and IoT access optimization.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution: A user selection and power allocation method based on space division multiplexing includes the following steps: Based on the multi-user MIMO scenario, obtain system parameters, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Calculate the equivalent power gain for each user in a multi-user scenario based on the channel matrix; Based on the single-user scenario, the equivalent power gain of a single user is used as a priority reference to construct a corresponding search tree; based on the equivalent power gain of each user in the multi-user scenario, a joint user selection and power allocation optimization problem is constructed, with user subsets and power allocation as joint optimization variables and the objective function being to maximize the sum of the reachable rates of all served users. The optimization problem of joint user selection and power allocation is solved, and the optimal user subset and optimal power allocation scheme are output.

[0006] Furthermore, the equivalent power gain for each user in a multi-user scenario is:

[0007] in, For users The equivalent power gain, The first unnormalized ZF precoding matrix List; The channel matrix corresponding to the selected user.

[0008] Furthermore, the reachable rate for each served user is:

[0009] in, For users The achievable rate; For users Distributed power; This represents noise power.

[0010] Furthermore, the optimization problem of joint user selection and power allocation is as follows:

[0011] in, This represents an optimization problem involving joint user selection and power allocation, with user subsets and power allocation as joint optimization variables and maximizing the sum of reachable rates of all served users as the objective function. Indicates constraints; This indicates the total power constraint of the base station; This represents the selected subset of users; Indicates selection Multiple users can be served simultaneously; This represents the total user set.

[0012] Furthermore, solving the optimization problem of the joint user selection and power allocation includes: Based on the equivalent power gain of a single user in a single-user scenario, generate an order or priority to guide the search of user subsets; Based on the order or priority, a combinatorial tree for searching user subsets is constructed. The combinatorial tree is traversed using a depth-first search based on a branch-and-bound strategy. During the traversal, an upper bound of the sum of reachable rates for the partially selected user subsets is calculated and compared with the current global optimal solution to perform pruning operations. For each complete subset of candidate users obtained during the traversal, calculate the equivalent power gain of the users in the subset under zero-forcing precoding, and solve the corresponding optimal power allocation problem to obtain the sum of achievable rates for the subset. After traversal, output the optimal subset of users with the largest sum of reachable rates for all served users and their corresponding optimal power allocation scheme.

[0013] Furthermore, generating the order or priority for guiding the search of user subsets includes: calculating the power gain of each user when served individually, and sorting all users according to the gain value.

[0014] Furthermore, performing pruning operations includes: calculating an upper bound on the reachable rate for nodes representing a subset of users in the combinatorial tree; if the upper bound is lower than the sum of the currently obtained optimal reachable rates, then stopping the search of all subsequent branches of that node.

[0015] Furthermore, the method for the upper bound of the achievable rate includes: assuming that the subsequent candidate users are those with the best channel conditions among the currently unselected users, and that there is no interference between users, and calculating the sum of the theoretically achievable maximum achievable rates under this condition.

[0016] Furthermore, solving the optimal power allocation subproblem involves: for a given subset of users, using a water-filling algorithm to allocate power to maximize the sum of achievable rates for that subset.

[0017] On the other hand, the present invention provides a user selection and power allocation system based on spatial division multiplexing in a multi-user MIMO scenario, comprising: System parameter acquisition module: It is used to acquire system parameters based on multi-user MIMO scenarios, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Equivalent power gain calculation module: It is used to calculate the single-user equivalent power gain for each user based on the channel matrix; The optimization problem construction module is used to construct a corresponding search tree based on the equivalent power gain of a single user in a single-user scenario as a priority reference; and to construct a joint user selection and power allocation optimization problem based on the equivalent power gain of each user in a multi-user scenario, with user subsets and power allocation as joint optimization variables and the objective function being to maximize the sum of the reachable rates of all served users. The solution module is used to solve the optimization problem of joint user selection and power allocation and output the optimal user subset and the optimal power allocation scheme.

[0018] Compared with the prior art, the present invention has the following beneficial effects: This invention, based on reasonable user selection, utilizes ZF precoding to eliminate interference between multiple users and simplifies the rate expression. During the gradual expansion of the selected user set, it efficiently updates the data using the block matrix inverse formula, avoiding redundant calculations. By assuming that subsequent users can achieve optimal results, it calculates the performance upper bound achievable by each node, thus significantly reducing the complexity of combinatorial search using a branch-and-bound strategy. Based on a determined user subset, it uses a water-filling method for power allocation, achieving optimal power distribution. This invention can efficiently implement joint user selection and power allocation in large-scale multi-user MIMO systems, significantly improving system capacity and reducing computational complexity, and is suitable for practical base station scheduling, IoT access, and other scenarios. Its mathematical derivation provides a solid foundation for subsequent algorithm optimization and theoretical analysis. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the process of an embodiment of the present invention; Figure 2 This is a schematic diagram of the search process in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In specific implementation, the method proposed in the technical solution of this invention can be automatically executed by those skilled in the art using computer software technology. System devices for implementing the method, such as computer-readable storage media storing the corresponding computer program of the technical solution of this invention and computer equipment including the computer program running the corresponding computer program, should also be within the protection scope of this invention.

[0023] like Figure 1As shown, this embodiment provides a user selection and power allocation method based on spatial division multiplexing in a multi-user MIMO scenario, including the following steps: Step 1. Based on the multi-user MIMO scenario, obtain system parameters, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Step 2. Calculate the single-user equivalent power gain for each user in a multi-user scenario based on the channel matrix; Step 3. Based on the single-user equivalent power gain meter as the priority reference in the single-user scenario, construct the corresponding search tree; based on the equivalent power gain of each user in the multi-user scenario, construct a joint user selection and power allocation optimization problem with user subset and power allocation as joint optimization variables and the objective function of maximizing the sum of the reachable rates of all served users. Step 4. Solve the optimization problem of joint user selection and power allocation, and output the optimal user subset and the optimal power allocation scheme.

[0024] In the above embodiments, the multi-user MIMO scenario in step 1 is: equipped with The base station with the root transmitting antenna has coverage area Let there be a single-antenna user, and let the total user set be denoted as . The goal is to start from the total user set. Select A subset of users are served simultaneously, and the selected user subset is: Let the power allocated to each user in each user subset be denoted as . The total power constraint of the base station is The overall channel matrix between base stations and total users It is known that the channel matrix corresponding to the selected user is... .

[0025] System parameters include: channel matrix ,in This refers to the number of base station antennas. Number of users; Total transmit power Number of base station antennas Number of users Select the number of users Noise power .

[0026] In step 2, consider the signal received by the user.

[0027] in, In order to receive signals, For the precoding matrix, In order to send a signal, For noise. Zero-forcing (ZF) precoding is used, i.e.

[0028] in, The normalized ZF precoding matrix, The precoding matrix of the unnormalized ZF is... This is the normalized matrix corresponding to the unnormalized ZF precoding matrix. Specifically,

[0029] in, These represent the precoding matrices of the unnormalized ZF. The The reciprocal of the second norm of the column. for The Columns, which means that each column of the precoding matrix is ​​normalized to .in, For normalized That is, the normalized ZF precoding matrix. The corresponding number in List.

[0030] At this point, the equivalent channel power gain for each user in a multi-user scenario can be expressed as:

[0031] in, For the first The channel vector corresponding to each user, i.e., the channel matrix The Okay. Considering The first in The term can be represented as

[0032] Therefore, in multi-user scenarios, users The equivalent power gain can be expressed as

[0033] That is, the equivalent power gain of each user is determined by the diagonal elements of the inverse of the Gram matrix of the channel matrix.

[0034] In step 3 of this embodiment, the optimization problem of joint user selection and power allocation is:

[0035] in, This represents an optimization problem involving joint user selection and power allocation, with user subsets and power allocation as joint optimization variables and maximizing the sum of reachable rates of all served users as the objective function. Indicates constraints; This indicates the total power constraint of the base station; This represents the selected subset of users; Indicates selection Multiple users can be served simultaneously; This represents the total user set.

[0036] The achievable speed for each user is:

[0037] like Figure 2 As shown, step 4 of this embodiment specifically includes: generating an order or priority for guiding the search of user subsets based on the equivalent power gain of a single user in a single-user scenario; Based on the order or priority, a combinatorial tree for searching user subsets is constructed. The combinatorial tree is traversed using a depth-first search based on a branch-and-bound strategy. During the traversal, an upper bound of the sum of reachable rates for the partially selected user subsets is calculated and compared with the current global optimal solution to perform pruning operations. For each complete subset of candidate users obtained during the traversal, calculate the equivalent power gain of the users in the subset under zero-forcing precoding, and solve the corresponding optimal power allocation problem to obtain the sum of achievable rates for the subset. After traversal, output the optimal subset of users with the largest sum of reachable rates for all served users and their corresponding optimal power allocation scheme.

[0038] In this embodiment, the derivation has been selected. In the case of one user, what is the impact of adding another user on the equivalent channel power gain? The channel matrix for the case of one user is: ,definition

[0039] but (Right now The Gram matrix of the channel matrix in the case of each user is a Hermitian matrix and is positive definite. Its inverse matrix is:

[0040] The diagonal matrix formed by its diagonal elements is

[0041] At this point, a new user and its channel vector are added. Then there is

[0042]

[0043] in, , As an intermediate variable; The Gram matrix of the channel matrix for the case of n+1 users; The channel matrix for the case of n+1 users.

[0044] Applying the formula for finding the inverse of a block matrix, we can obtain...

[0045] in,

[0046] in, for An identity matrix of dimension 1, and

[0047] It is projected onto The orthogonal projection matrix of the row space.

[0048] definition

[0049] but

[0050] Based on the above derivation, it can be concluded that... The geometric meaning is that row vectors exist The energy of the projected component on the orthogonal complement of the row space. The geometric meaning is that row vectors exist The projection components in the line space are decomposed into The coordinates on each row vector.

[0051] Therefore, the ideal scenario for adding a new user can be obtained, namely its corresponding channel vector. and The row spaces are orthogonal. At this point, a newly added user will not affect the power gain of previous users, and its own power gain will reach its maximum.

[0052] Based on the above conclusions, we can use a branch-and-bound strategy to search for the optimal subset of users. Specifically: Using the channel matrix input in step 1 Each line in the file represents each user. Calculate the single-user power gain in a single-user scenario:

[0053] in It is the channel matrix The Columns represent users The channel vector.

[0054] After the calculation is completed, all users will be classified according to... Arrange in descending order to obtain the permutation. satisfy:

[0055] After the above steps, the sorted user index sequence can be obtained. and the single-user power gain for each user. .

[0056] Constructing a combinatorial tree and initializing the search state specifically includes: based on the processing results in steps 1 and 2, constructing a combinatorial tree of depth... The tree structure has a root node with a depth of 0, corresponding to an empty user set; intermediate nodes (with a depth of...) () represents a partial set of users ,Include Selected users; Leaf nodes (depth) ): Represents the complete set of users ,Include There are [number] users. The specific node expansion rules during the construction of the composite tree are as follows: each node's child nodes are selected from the remaining users who have not yet been selected; the index order constraint must be satisfied, meaning the index of the newly added user must be greater than the largest index in the current set. Specifically, if the current node set is [number] users... Then its child nodes can add users. ,in .

[0057] After constructing the combinatorial tree, the search state is further initialized. Specifically, this includes initializing the optimal sum and rate: Optimal user set: And optimal power distribution: .

[0058] Depth-First Search (DFS) traversals of a tree structure specifically include the following cases: Case 1: Visiting internal nodes (depth...) Case 2: Accessing the leaf node (depth) ).

[0059] For case 1, i.e., for the set representing a subset of users For internal nodes, first calculate the power gain of the current subset: Then, the upper bound of the performance that can be achieved under the selected subset conditions is calculated. The specific process is as follows: Select from unused users The set of users with the maximum single-user gain is denoted as set A. Assuming the total power Assigned to Users within the group, and no interference between users (optimistic estimate); upper bound of calculation:

[0060] The constraints are

[0061] in,

[0062] After calculating the upper bound, a pruning decision is made: if If the branch is not found, then prune the branch (stop searching downwards); otherwise, continue traversing the child nodes of the node.

[0063] For case 2, i.e., for the representation of the complete user set First, solve the optimal power allocation problem for the leaf nodes.

[0064] in,

[0065] During the solution process, the optimal power allocation will be obtained through the water injection algorithm. Then calculate the corresponding sum rate. .

[0066] After solving the optimal power allocation problem, the global optimal solution is updated, specifically according to the following rule: if Then update

[0067] Otherwise, leave the current optimal solution unchanged.

[0068] The output optimization results include the following process: After the DFS traverses all unpruned nodes, output the globally optimal solution that is retained at this point, i.e., the optimal user set: Optimal power allocation: And the maximum achievable system speed: .

[0069] Example 2 This embodiment describes a user selection and power allocation system based on spatial division multiplexing in a multi-user MIMO scenario, comprising: System parameter acquisition module: It is used to acquire system parameters based on multi-user MIMO scenarios, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Equivalent power gain calculation module: It is used to calculate the single-user equivalent power gain for each user based on the channel matrix; The optimization problem construction module is used to construct a corresponding search tree based on the equivalent power gain of a single user in a single-user scenario as a priority reference; and to construct a joint user selection and power allocation optimization problem based on the equivalent power gain of each user in a multi-user scenario, with user subsets and power allocation as joint optimization variables and the objective function being to maximize the sum of the reachable rates of all served users. The solution module is used to solve the optimization problem of joint user selection and power allocation and output the optimal user subset and the optimal power allocation scheme.

[0070] It should be understood that any parts not described in detail in this specification belong to the prior art.

[0071] It should be understood that the above description of the preferred embodiments is quite detailed, but this should not be construed as limiting the scope of protection of this invention. It is neither necessary nor possible to exhaustively describe all possible implementations. Those skilled in the art, guided by this invention, can make substitutions or modifications without departing from the scope of the claims, all of which fall within the scope of protection of this invention. The scope of protection of this invention should be determined by the appended claims.

Claims

1. A user selection and power allocation method based on spatial division multiplexing, characterized in that, Includes the following steps: Based on the multi-user MIMO scenario, obtain system parameters, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Calculate the equivalent power gain for each user in a multi-user scenario based on the channel matrix; Based on the equivalent power gain of a single user in a single-user scenario as a priority reference, a corresponding search tree is constructed. Based on the equivalent power gain of each user in a multi-user scenario, an optimization problem for joint user selection and power allocation is constructed, with user subsets and power allocation as joint optimization variables, and the objective function being to maximize the sum of the reachable rates of all served users. The optimization problem for joint user selection and power allocation is as follows: in, This represents an optimization problem involving joint user selection and power allocation, with user subsets and power allocation as joint optimization variables and maximizing the sum of reachable rates of all served users as the objective function. Indicates constraints; This indicates the total power constraint of the base station; This represents the selected subset of users; Indicates selection Multiple users can be served simultaneously; Represents the total user set; For users Distributed power; Solve the optimization problem of joint user selection and power allocation, and output the optimal user subset and optimal power allocation scheme; including: Based on the equivalent power gain of a single user in a single-user scenario, generate an order or priority to guide the search of user subsets; Based on the order or priority, a combinatorial tree for searching user subsets is constructed. The combinatorial tree is traversed using a depth-first search based on a branch-and-bound strategy. During the traversal, an upper bound of the sum of reachable rates for the partially selected user subsets is calculated and compared with the current global optimal solution to perform pruning operations. For each complete subset of candidate users obtained during the traversal, calculate the equivalent power gain of the users in the subset under zero-forcing precoding, and solve the corresponding optimal power allocation problem to obtain the sum of achievable rates for the subset. After traversal, output the optimal subset of users with the largest sum of reachable rates for all served users and their corresponding optimal power allocation scheme; Performing pruning operations includes: calculating the upper bound of reachability for nodes representing a subset of users in the composite tree; if the upper bound is lower than the sum of the currently obtained best reachability, then stopping the search of all subsequent branches of that node.

2. The user selection and power allocation method based on spatial division multiplexing according to claim 1, characterized in that, In a multi-user scenario, the equivalent power gain for each user is: in, For users The equivalent power gain, The first unnormalized ZF precoding matrix List; The channel matrix corresponding to the selected user.

3. The user selection and power allocation method based on spatial division multiplexing according to claim 2, characterized in that, The achievable speed for each served user is: in, For users The achievable rate; This represents noise power.

4. The user selection and power allocation method based on spatial division multiplexing according to claim 1, characterized in that, Generating the order or priority for guiding the search of a subset of users includes: calculating the power gain of each user when served individually, and sorting all users according to that gain value.

5. The user selection and power allocation method based on spatial division multiplexing according to claim 1, characterized in that, The method for the upper bound of the achievable rate includes: assuming that the subsequent candidate users are those with the best channel conditions among the currently unselected users, and that there is no interference between users, and calculating the sum of the theoretically achievable maximum achievable rates under this condition.

6. The user selection and power allocation method based on spatial division multiplexing according to claim 1, characterized in that, Solving the optimal power allocation subproblem involves: for a given subset of users, using a water-filling algorithm to allocate power to maximize the sum of achievable rates for that subset.

7. A user selection and power allocation system based on space division multiplexing, characterized in that, include: System parameter acquisition module: It is used to acquire system parameters based on multi-user MIMO scenarios, including channel matrix, total transmit power, noise power, number of base station antennas, total number of users, and number of candidate users; Equivalent power gain calculation module: It is used to calculate the single-user equivalent power gain of each user in a multi-user scenario based on the channel matrix; The optimization problem construction module is used to construct a corresponding search tree based on the single-user equivalent power gain in a single-user scenario as a priority reference. Based on the equivalent power gain of each user in a multi-user scenario, we construct a joint user selection and power allocation optimization problem with user subsets and power allocation as joint optimization variables and the objective function being to maximize the sum of the reachable rates of all served users. The solution module is used to solve the optimization problem of joint user selection and power allocation and output the optimal user subset and the optimal power allocation scheme. The user selection and power allocation system based on spatial division multiplexing is used to perform the steps in the user selection and power allocation method based on spatial division multiplexing as described in any one of claims 1-6.

Images