Cell identification allocation method and related equipment for wireless communication system

Abstract

The cell identifier allocation method and related equipment for a wireless communication system proposed in this application embodiment include communication cells with multiple cellular configurations. The method includes: first, obtaining iterative optimization parameters and iterative penalty parameters, as well as obtaining a quotient optimization model and a remainder iterative update model, which are obtained based on the system parameters of the wireless communication system and the conflict reduction objective; then, updating the iterative optimization parameters based on the iterative penalty parameters and the remainder iterative update model, and obtaining the identifier remainder based on the updated iterative optimization parameters; finally, solving the quotient optimization model based on the identifier remainder to obtain the identifier quotient, obtaining the identifier allocation number based on the identifier quotient and the identifier remainder, and allocating an identifier to each communication cell based on the identifier allocation number, thereby effectively improving the identifier allocation efficiency of each physical cell when allocating identifiers for multiple physical cells in the wireless communication system.

Description

Technical Field

[0001] This application relates to the field of industrial Internet of Things (IoT) technology, and in particular to cell identifier allocation methods and related equipment for wireless communication systems. Background Technology

[0002] In 5G networks, the Physical Cell Identifier (PCI), as a unique identifier for a cell, is particularly important for optimizing key network system parameters. It plays a crucial role in helping User Equipment (UE) search for connecting cells, identify neighboring cells for cell reselection, and assist with handover. Incorrect PCI allocation can lead to co-channel interference between cells, thereby affecting communication quality and user experience. Therefore, proper PCI allocation is especially important in next-generation mobile communication technologies such as 5G.

[0003] In related technologies, for the allocation of physical cell identifiers among multiple physical cells in a wireless communication system, the wireless communication system is typically mathematically modeled to obtain an identifier allocation optimization problem. Heuristic algorithms are then used to solve this problem to obtain appropriate physical cell identifier allocations for the multiple physical cells, thereby ensuring the communication quality of user equipment. However, as the number of physical cells in a wireless communication system increases, the efficiency of this method for physical cell identifier allocation decreases significantly. Summary of the Invention

[0004] This application provides a cell identifier allocation method and related equipment for a wireless communication system, which can improve the identifier allocation efficiency while ensuring the communication quality of user equipment when allocating multiple physical cell identifiers in a wireless communication system.

[0005] To achieve the above objectives, a first aspect of this application proposes a cell identifier allocation method for a wireless communication system, the wireless communication system comprising multiple cellular communication cells, the method comprising:

[0006] Obtain iterative optimization parameters and iterative penalty parameters, as well as quotient optimization model and remainder iterative update model, wherein the quotient optimization model and remainder iterative update model are obtained based on the system parameters of the wireless communication system and the conflict reduction objective;

[0007] The iterative optimization parameters are updated based on the iterative penalty parameters and the remainder iterative update model, and the identifier remainder is obtained based on the updated iterative optimization parameters;

[0008] The quotient optimization model is solved based on the identifier remainder to obtain the identifier quotient. The identifier allocation number is obtained based on the identifier quotient and the identifier remainder. The identifier allocation number is then used to allocate an identifier to each of the communication cells.

[0009] In some embodiments, obtaining the quotient optimization model and the remainder iterative update model includes:

[0010] Obtain the co-frequency neighbor cell relationship, second-order co-frequency neighbor cell relationship, co-frequency overlapping coverage neighbor cell relationship, and identification modal interference relationship between every two communication cells, and obtain the interference matrix based on the co-frequency overlapping coverage neighbor cell relationship;

[0011] Obtain the identifier allocation number parameter corresponding to the identifier allocation number, obtain the identifier modulus relationship based on the identifier allocation number parameter, obtain the identifier allocation constraint corresponding to the identifier allocation number parameter, and obtain the identifier allocation model based on the interference matrix, the identifier allocation number parameter, the identifier modulus interference relationship, and the identifier allocation constraint;

[0012] The identifier allocation model is divided into the quotient optimization model and the remainder optimization model, and the remainder iterative update model is obtained based on the remainder optimization model.

[0013] In some embodiments, dividing the identifier allocation model into the quotient optimization model and the remainder optimization model includes:

[0014] Obtain the remainder divisor decomposition function, and based on the remainder divisor decomposition function, decompose the identifier allocation number parameter in the identifier allocation model into an identifier remainder parameter and an identifier quotient parameter, wherein the identifier remainder parameter corresponds to the identifier remainder and the identifier quotient parameter corresponds to the identifier quotient.

[0015] The identifier allocation model is updated based on the identifier remainder parameter and the identifier quotient parameter to obtain the identifier quotient remainder allocation model;

[0016] Based on the identifier remainder parameter and the identifier quotient parameter, the identifier quotient remainder allocation model is split to obtain the quotient optimization model and the remainder optimization model.

[0017] In some embodiments, obtaining the remainder iterative update model based on the remainder optimization model includes:

[0018] Obtain one-hot coding parameters and the coding conversion relationship between the one-hot coding parameters and the identifier remainder parameters, and update the remainder optimization model using the coding conversion relationship and the one-hot coding parameters to obtain a one-hot coding optimization model, wherein the one-hot coding optimization model includes one-hot coding constraints;

[0019] Perform a probabilistic simplex equivalence transformation on the one-hot encoded constraints to obtain simplex equivalence constraints;

[0020] The simplex unrelaxed optimization model is obtained by replacing the one-hot encoding constraints of the simplex equivalent constraint, and the remainder iterative update model is obtained based on the simplex unrelaxed optimization model.

[0021] In some embodiments, the simplex equivalence constraints include a first simplex equivalence constraint and a second simplex equivalence constraint, the simplex unrelaxed optimization model includes a disturbance function, and obtaining the remainder iterative update model based on the simplex unrelaxed optimization model includes:

[0022] Based on the smooth quadratic penalty function and the iterative penalty parameter, the second simplex equivalence constraint is penalized, and then summed with the interference function to obtain the interference penalty function.

[0023] Based on the interference penalty function, the first simplex equivalence constraint, and the one-hot encoding parameters, a remainder interference penalty model is obtained, and the remainder interference penalty model is subjected to mirror gradient descent processing to obtain the remainder iterative update model.

[0024] In some embodiments, performing mirror gradient descent on the remainder interference penalty model to obtain the remainder iterative update model includes:

[0025] Obtain the iteration step size, the strongly convex differentiable function, and the corresponding KL divergence relation;

[0026] The interference penalty function of the remainder interference penalty model is differentiated to obtain the interference penalty derivative function. Then, using mirror gradient descent, based on the interference penalty function, the interference penalty derivative function, the KL divergence relation, and the iteration step size, the remainder mirror gradient descent model of the remainder interference penalty model is obtained.

[0027] Based on the strongly convex differentiable function and the remainder mirror gradient descent model, a remainder iterative update model is obtained, which is the parameter update form of the remainder mirror gradient descent model.

[0028] In some embodiments, obtaining the iterative optimization parameters includes:

[0029] An initial factor matrix is ​​obtained based on a uniform distribution between zero and one, and the initial factor matrix includes multiple initial factors consistent with the number of communication cells;

[0030] Based on the ratio of each initial factor to the first norm of the initial factor matrix, the iterative optimization parameters corresponding to each initial factor are obtained.

[0031] In some embodiments, solving the quotient optimization model based on the identifier remainder to obtain the identifier quotient, and obtaining the identifier allocation number based on the identifier quotient and the identifier remainder, includes:

[0032] Obtain multiple subgraph discrete numbers, and generate multiple subgraphs with the same number of subgraph discrete numbers based on the identifier modulo-remainder interference relationship, the identifier remainder, and the subgraph discrete parameters;

[0033] The quotient optimization model is updated one by one using each subgraph to obtain the updated quotient optimization model, and the updated quotient optimization model is solved to obtain the identifier quotient corresponding to each identifier remainder;

[0034] The identifier allocation number is obtained based on the remainder-divisor decomposition function, the multiple identifier remainders, and the identifier quotient corresponding to each identifier remainder.

[0035] To achieve the above objectives, a second aspect of this application provides a cell identifier allocation device for a wireless communication system, the wireless communication system including multiple cellular communication cells, the device comprising:

[0036] The acquisition module is used to acquire iterative optimization parameters and iterative penalty parameters, as well as to acquire the quotient optimization model and the remainder iterative update model, wherein the quotient optimization model and the remainder iterative update model are obtained based on the system parameters of the wireless communication system;

[0037] An iterative update module is used to update the iterative optimization parameters based on the iterative penalty parameters and the remainder iterative update model, and to obtain the identifier remainder based on the updated iterative optimization parameters;

[0038] The identifier allocation module is used to solve the quotient optimization model based on the identifier remainder to obtain the identifier quotient, obtain the identifier allocation number based on the identifier quotient and the identifier remainder, and allocate an identifier to each of the communication cells based on the identifier allocation number.

[0039] To achieve the above objectives, a third aspect of this application provides an electronic device, which includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the cell identifier allocation method of the wireless communication system as described in the first aspect.

[0040] To achieve the above objectives, a fourth aspect of the present application provides a storage medium, which is a computer-readable storage medium storing a computer program that, when executed by a processor, implements the cell identifier allocation method for the wireless communication system described in the first aspect.

[0041] The present application proposes a cell identifier allocation method and related equipment for a wireless communication system. The wireless communication system includes multiple cellular communication cells. The method includes: first, obtaining iterative optimization parameters and iterative penalty parameters, as well as obtaining a quotient optimization model and a remainder iterative update model, which are obtained based on the system parameters of the wireless communication system and the conflict reduction objective; then, updating the iterative optimization parameters based on the iterative penalty parameters and the remainder iterative update model, and obtaining the identifier remainder based on the updated iterative optimization parameters; finally, solving the quotient optimization model based on the identifier remainder to obtain the identifier quotient, obtaining the identifier allocation number based on the identifier quotient and the identifier remainder, and allocating an identifier to each communication cell based on the identifier allocation number. This application embodiment utilizes a quotient optimization model and a remainder iterative update model corresponding to the system parameters of the wireless communication system and the goal of reducing conflicts. It further solves for the identifier remainder and identifier quotient, and combines the relationship between the physical cell identifier and the quotient and remainder after division with remainder to perform reverse derivation of division with remainder. This allows for the rapid acquisition of the identifier allocation number for each physical cell. Thus, when allocating identifiers for multiple physical cells in a wireless communication system, it minimizes PCI conflicts between communication cells, ensures the communication quality of user equipment in the wireless communication system, and effectively improves the identifier allocation efficiency of each physical cell.

[0042] Other features and advantages of this application will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0043] Figure 1 This is a schematic diagram of the structure of a wireless communication system provided in an embodiment of this application.

[0044] Figure 2 This is a flowchart of a cell identifier allocation method for a wireless communication system provided in an embodiment of this application.

[0045] Figure 3 yes Figure 2 The flowchart for step 201.

[0046] Figure 4 yes Figure 2 Another flowchart for step 201.

[0047] Figure 5 yes Figure 4 The flowchart for step 403.

[0048] Figure 6 yes Figure 4 Another flowchart for step 403.

[0049] Figure 7 yes Figure 6 The flowchart for step 603.

[0050] Figure 8 yes Figure 7 The flowchart for step 702.

[0051] Figure 9 yes Figure 2 The flowchart for step 203.

[0052] Figure 10 This is a flowchart illustrating another cell identifier allocation method provided in another embodiment of this application.

[0053] Figure 11 This is a simulation diagram of the cell identifier allocation performance provided in another embodiment of this application.

[0054] Figure 12 This is a schematic diagram of the structure of a cell identifier allocation device for a wireless communication system provided in an embodiment of this application.

[0055] Figure 13 This is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0057] It should be noted that although functional modules are divided in the device schematic diagram and the logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart.

[0058] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0059] First, let's analyze some of the terms used in this application:

[0060] A Physical Cell Identifier (PCI) is an identifier used to uniquely identify each cell in a wireless communication system. Different wireless communication systems (such as GSM, UMTS, LTE, and 5G) have different cell identifier allocation methods.

[0061] The Channel Quality Indicator (CQI) is a parameter reported by the User Equipment (UE) to the eNodeB to reflect the quality of the current radio channel. The CQI typically ranges from 1 to 15, with each value corresponding to a different modulation and coding scheme (MCS).

[0062] In 5G networks, the Physical Cell Identifier (PCI), as a unique identifier for a cell, is particularly important for optimizing key network system parameters. It plays a crucial role in helping User Equipment (UE) search for connecting cells, identify neighboring cells for cell reselection, and assist with handover. Incorrect PCI allocation can lead to co-channel interference between cells, thereby affecting communication quality and user experience. Therefore, proper PCI allocation is especially important in next-generation mobile communication technologies such as 5G.

[0063] In related technologies, for the allocation of physical cell identifiers among multiple physical cells in a wireless communication system, the wireless communication system is typically mathematically modeled to obtain an identifier allocation optimization problem. Heuristic algorithms are then used to solve this problem to obtain the physical cell identifier allocation for the multiple physical cells. However, as the number of physical cells in a wireless communication system increases, the efficiency of this method for physical cell identifier allocation decreases significantly.

[0064] To improve identifier allocation efficiency while ensuring communication quality for user equipment (UAEs) when allocating identifiers for multiple physical cells in a wireless communication system, this application utilizes a quotient optimization model and a remainder iterative update model corresponding to the system parameters of the wireless communication system and the goal of reducing conflicts. It further solves for the identifier remainder and identifier quotient, and combines this with the relationship between the physical cell identifier and the quotient and remainder after division with remainder to perform a reverse derivation of the remainder with remainder. This allows for the rapid acquisition of the identifier allocation number for each physical cell. Therefore, when allocating identifiers for multiple physical cells in a wireless communication system, PCI conflicts between communication cells are minimized, ensuring communication quality for UAEs in the wireless communication system and effectively improving the identifier allocation efficiency for each physical cell.

[0065] To better describe the cell identifier allocation method for the wireless communication system provided in this application, the wireless communication system to which the cell identifier allocation method is applied is first described below. (Refer to...) Figure 1 This is a schematic diagram of the structure of a wireless communication system provided in an embodiment of this application. Figure 1 As shown, a wireless communication system includes multiple cellular communication cells, each of which needs to be equipped with a cell allocation identifier (PCI).

[0066] In a wireless communication system, if a communication device connected to the master control of the i-th communication cell can simultaneously receive the signal from the j-th communication cell, then the i-th communication cell is called a communication neighbor cell of the j-th communication cell. If, at this time, the i-th communication cell and the j-th communication cell share the same frequency, then the i-th communication cell is a communication co-frequency neighbor cell of the j-th communication cell. If the master control device of the i-th communication cell receives a signal strength p from the i-th communication cell... i The signal strength p of its communication neighbor cell (i.e., the j-th communication cell) j The difference is less than or equal to a given threshold δ, i.e., p i -p j If ≤δ, then the j-th communication cell is called the overlapping coverage neighbor cell of the i-th communication cell.

[0067] When a communication cell and its neighboring communication cells experience a co-channel, co-PCI conflict, a PCI conflict will occur between the two communication cells. For example... Figure 1 As shown, communication cell 11 is a co-frequency neighbor of communication cell 1. Therefore, when communication cell 1 and communication cell 11 share the same PCI, a conflict will occur between the two cells. User equipment that should be connected to communication cell 1 may incorrectly connect to communication cell 11, and this incorrect connection determination will lead to the service terminal and ultimately the incorrect allocation of downlink network resources.

[0068] Furthermore, PCI confusion can occur when two or more co-channel neighboring cells of a master cell have the same PCI. For example... Figure 1 As shown, since cell 3 is a neighbor of cell 6, and cell 7 is also a neighbor of cell 6, confusion will occur when cells 3 and 7 operate on the same frequency and have the same PCI. PCI confusion can lead to signal interruption and incorrect resource allocation for users in the downlink network during service handover.

[0069] Modulo-k interference occurs when the remainders of the PCI allocated to the master cell and its overlapping neighboring cells are the same when modulo k. In 5G communication networks, the case of k=3 is mainly considered. When PCI modulo-3 interference occurs, the signal quality received by user equipment will be significantly degraded due to the superposition of reference signals between communication cells. This phenomenon also causes incorrect CQI assessment and downlink network delays.

[0070] Based on the aforementioned wireless communication system, the cell identifier allocation method and related equipment for the wireless communication system provided in this application embodiment will be further described below. The cell identifier allocation method for the wireless communication system provided in this application embodiment can be applied to the controller in the wireless communication system.

[0071] The cell identifier allocation method of the wireless communication system in the embodiments of this application will be described in detail below. (Refer to...) Figure 2 This is an optional flowchart of a cell identifier allocation method for a wireless communication system provided in an embodiment of this application. Figure 2 The method may include, but is not limited to, steps 201 to 203. It is also understood that this embodiment... Figure 2 The order of steps 201 to 203 is not specifically limited. The order of steps can be adjusted or some steps can be reduced or added according to actual needs.

[0072] Step 201: Obtain the iterative optimization parameters and iterative penalty parameters, as well as the quotient optimization model and the remainder iterative update model.

[0073] Step 201 will be described in detail below.

[0074] In some embodiments, to appropriately allocate cell identifiers among multiple communication cells in a wireless communication system, thereby minimizing PCI conflicts between communication cells and ensuring the communication quality of user equipment in the wireless communication system, while effectively improving the identifier allocation efficiency of each physical cell, in practical applications, when performing actual physical cell identifier allocation in a wireless communication system, it is necessary to first obtain the iterative optimization parameter X. (m) And the iteration penalty parameter ρ, where m represents the m-th iteration. The following describes how to obtain the iterative optimization parameters.

[0075] Reference Figure 3 To obtain the iterative optimization parameters, the steps 301 to 302 are as follows.

[0076] Step 301: Obtain the initial factor matrix based on the uniform distribution between zero and one.

[0077] Step 302: Based on the ratio of each initial factor to the first norm of the initial factor matrix, obtain the iterative optimization parameters corresponding to each initial factor.

[0078] Steps 301 to 302 are described in detail below.

[0079] In some embodiments, to avoid simplex cell assignment before physical cell identifier allocation, For non-differentiable initial points on the boundary, this application proposes a random initialization update strategy to obtain the initial iterative optimization parameters. Specifically, this includes obtaining the initial factor matrix based on a uniform distribution between zero and one, i.e., for the j-th column... Spelling z~U k×1 [0,1]. Next, a normalization operation is performed, that is, based on the ratio of each initial factor to the first norm of the initial factor matrix, the iterative optimization parameters corresponding to each initial factor are obtained as follows:

[0080] Through steps 301 to 302 above, the iterative optimization parameter X is utilized. (m) Proper initial generation can prevent non-differentiability on the simplex boundary during subsequent processing of iterative optimization parameters, thereby improving the reliability of iterative optimization parameter processing.

[0081] In some embodiments, after obtaining the iterative optimization parameter X (m) In addition to the iterative penalty parameter ρ, it is also necessary to obtain the system parameters based on the wireless communication system and the corresponding quotient optimization model and remainder iterative update model with the goal of reducing the conflict between multiple communication cells in the wireless communication system. This will facilitate the subsequent use of the quotient optimization model, remainder iterative update model, and iterative optimization parameter X. (m) The optimal cell identifier for each communication cell in the wireless communication system is obtained by solving for the iterative penalty parameter ρ. The following section will further describe how to obtain the quotient optimization model and the remainder iterative model.

[0082] Reference Figure 4 The process of obtaining the quotient optimization model and the remainder iterative update model includes the following steps 401 to 403.

[0083] Step 401: Obtain the co-frequency neighbor cell relationship, second-order co-frequency neighbor cell relationship, and co-frequency overlapping coverage neighbor cell relationship between every two communication cells, and obtain the interference matrix based on the co-frequency overlapping coverage neighbor cell relationship.

[0084] Step 402: Obtain the identifier allocation number parameter corresponding to the identifier allocation number, obtain the identifier modulus-residual relationship based on the identifier allocation number parameter, and obtain the identifier allocation constraint corresponding to the identifier allocation number parameter. Based on the interference matrix, the identifier allocation number parameter, the identifier modulus-residual interference relationship, and the identifier allocation constraint, obtain the identifier allocation model.

[0085] Step 403: Divide the identifier allocation model into a quotient optimization model and a remainder optimization model, and obtain the remainder iterative update model based on the remainder optimization model.

[0086] Steps 401 to 403 are described in detail below.

[0087] In some embodiments, based on such Figure 1The goal of PCI planning in the illustrated wireless communication system is to allocate PCI to each physical cell to minimize PCI collisions, confusion, and modulo-k interference in the wireless communication system. In practical wireless communication systems, there are various methods to represent the amount of PCI collisions, confusion, and modulo-k interference. These methods include using drive test data points or gridding the entire city and analyzing all grids. Among these, measurement report (MR) data is most commonly used by network optimization departments. Each MR data entry mainly includes the master cell accessed by the UE during communication and its received neighboring cell information, as well as the corresponding signal strength values. Since MR reports are submitted at fixed time intervals, the distribution of MR data can basically reflect the distribution of traffic. Furthermore, the temporal and spatial comprehensiveness of MR data allows it to accurately reflect PCI indicators in the network. Evaluating the effectiveness of network PCI allocation by using the amount of collisions, confusion, and modulo-k interference in MR data is a relatively accurate method.

[0088] MR data is a report that the UE device periodically reports during communication.

[0089] The PCI planning problem based on MR data is as follows: Given a set of N communication cells V = {1, 2, ..., N}, traverse all MR data of these communication cells to obtain the co-frequency neighbor relationship (i.e., conflict relationship) between every two communication cells E = {(i,j)|, cell j is a neighbor of cell i}. If the i-th communication cell and the j-th communication cell are co-frequency, and there exists a cell in the MR data that the i-th communication cell is the controlling cell and the j-th communication cell is a neighbor, then let (i,j) ∈ E; Second-order co-frequency neighbor relationship (confusion relationship), N = {(i,j)|, there exists a cell k such that (k,i), (k,j) ∈ E}. If the i-th communication cell and the j-th communication cell are co-frequency, and there exists a cell in the MR data that the i-th communication cell and the j-th communication cell are co-frequency and are both neighbors of another k-th communication cell, then let (i,j) ∈ N; Based on the co-frequency overlapping coverage neighbor relationship, obtain the interference matrix W = [w ij ] N×N If the i-th communication cell and the j-th communication cell are on the same frequency, then w ij The value is the number of MRs of the overlapping neighboring cells of the i-th communication cell as the controlling cell and the j-th communication cell as the controlling cell of the i-th communication cell; otherwise, w ij The value is 0.

[0090] Obtain the identifier allocation parameter PCI = (PCI1, PCI2, ..., PCI) corresponding to the identifier allocation number of each physical cell in the wireless communication system. N ), obtain the residual interference relationship C of the identifier module k (PCI)={(i,j)∈V×V|PCI i ≡PCIj (mod k)}, when the PCI values ​​of the i-th and j-th communication cells have the same remainder with respect to k, i.e., (i,j)∈C k (PCI) indicates that the PCI of the i-th communication cell and the PCI of the j-th communication cell are the same when modulo k, meaning that a modulo-remainder interference relationship has occurred. For example, if the PCI of cell i is 4 and the PCI of cell j is 7 when k=3, then (i,j)∈C k (PCI), because their PCI values ​​are all equal to 1 modulo 3, and the identifier allocation constraint corresponding to the identifier allocation number parameter is obtained. This identifier allocation constraint characterizes the number of identifiers allocated to each physical cell in the wireless communication system as an integer between 0 and 1007. Next, based on the interference matrix W = [w ij ] N×N The identifier allocation model is obtained by considering the identifier allocation parameter PCI, the identifier modulus interference relationship, and the identifier allocation constraints, as shown in the following formula (1).

[0091]

[0092] The goal of the identifier allocation model is to minimize modulus k interference while eliminating both collisions and interference. Since there are no collisions, confusions, or interferences between cells of different frequencies, we assume that all N cells are co-frequency cells.

[0093] Furthermore, note that the PCI allocation range in 5G is an integer between 0 and 1007, resulting in a decision space of 1008 for this problem. N This is extremely difficult to handle. Therefore, this embodiment divides the identifier allocation model into a quotient optimization model and a remainder optimization model, and then uses the quotient optimization model and the remainder optimization model to solve the original identifier allocation model step by step. The following further describes how to obtain the identifier allocation model divided into a quotient optimization model and a remainder optimization model.

[0094] Reference Figure 5 The identifier allocation model is divided into a quotient optimization model and a remainder optimization model, including the following steps 501 to 503.

[0095] Step 501: Obtain the remainder divisor decomposition function, and based on the remainder divisor decomposition function, decompose the identifier allocation number parameter in the identifier allocation model into the identifier remainder parameter and the identifier quotient parameter.

[0096] Step 502: Update the identifier allocation model based on the identifier remainder parameter and the identifier quotient parameter to obtain the identifier quotient remainder allocation model.

[0097] Step 503: Based on the identifier remainder parameter and the identifier quotient parameter, the identifier quotient remainder allocation model is split to obtain the quotient optimization model and the remainder optimization model.

[0098] Steps 501 to 503 are described in detail below.

[0099] In some embodiments, firstly, a remainder division decomposition function for dividing the identifier allocation number parameter is obtained, and based on the remainder division decomposition function, the identifier allocation number parameter in the identifier allocation model is decomposed into an identifier remainder parameter r and an identifier quotient parameter q as shown in the following formula (2), wherein the identifier remainder parameter corresponds to the identifier remainder, and the identifier quotient parameter corresponds to the identifier quotient.

[0100] PCI = k × q + r (2)

[0101] Next, based on the identifier remainder parameter r and identifier quotient parameter q corresponding to the cell identifiers of all communication cells, the optimization variables in the identifier allocation model (1) are updated to obtain the new identifier quotient remainder allocation model as shown in the following formula (3).

[0102]

[0103] Where q = (q1, ..., q) N ) represents the identifier quotient parameter after taking the PCI (cell identifier) ​​modulo k, r = (r1, ..., r2) N This represents the identifier remainder parameter after the PCI modulo operation on k. This indicates neighboring cell pairs with the same identifier remainder parameter. Furthermore, it is noted that the objective function in the new identifier quotient remainder allocation model (3) is only related to the identifier remainder parameter r. Therefore, in this embodiment, the identifier remainder parameter r is calculated first, followed by the identifier quotient parameter q, to ​​obtain the final PCI allocation scheme.

[0104] Therefore, based on the identifier remainder parameter r and the identifier quotient parameter q, the identifier quotient remainder allocation model (3) is split to obtain the remainder optimization model as shown in the following formula (4) and the quotient optimization model as shown in the following formula (5).

[0105]

[0106] In the remainder optimization model (4), the first and fourth constraints on the remainder parameter r of the original identifier quotient remainder allocation model (3) are omitted. This is because the decision space of the identifier quotient parameter q is much larger than that of the identifier remainder parameter r, so these missing constraints can be easily satisfied by allocating different quotients.

[0107] The solution identifying the remainder parameter r is obtained by solving the remainder optimization model (4). Next, the quotient optimization model is solved further. The following describes how to solve the remainder optimization model (4). First, the remainder optimization model (4) is transformed into a remainder iterative update model, and then the solution identifying the remainder parameter r is obtained using the remainder iterative update model. The following section will further describe how to obtain the remainder for iterative model updates.

[0108] Reference Figure 6 The remainder optimization model is used to obtain the remainder iterative update model, which includes the following steps 601 to 603.

[0109] Step 601: Obtain the one-hot coding parameters and the coding conversion relationship between the one-hot coding parameters and the identifier remainder parameters. Then, use the coding conversion relationship and the one-hot coding parameters to update the remainder optimization model and obtain the one-hot coding optimization model.

[0110] Step 602: Perform a probabilistic simplex equivalent transformation on the one-hot coding constraints to obtain simplex equivalent constraints.

[0111] Step 603: Replace the one-hot encoding constraints of the simplex equivalence optimization model with the one-hot encoding constraints to obtain the simplex unrelaxed optimization model, and obtain the remainder iterative update model based on the simplex unrelaxed optimization model.

[0112] Steps 601 to 603 are described in detail below.

[0113] In some embodiments, in order to solve the remainder optimization model (4), one-hot encoding is first used to represent each identifier remainder parameter r. i Instead of using label encoding, the variable is represented using one-hot encoding. One-hot encoding eliminates the order inherent in label encoding. Specifically, the one-hot encoding parameter x is first obtained. i The encoding conversion relationship between the unique hot coding parameters and the identifier remainder parameters is shown in the following formula (6).

[0114]

[0115] in It means only the r-th i A k-dimensional unit vector with one element being 1 and the rest being 0.

[0116] Next, using the encoding transformation relation (6) and the one-hot encoding parameter x i The remainder optimization model (5) is updated to obtain the one-hot encoding optimization model as shown in the following formula (7).

[0117]

[0118] Among them, the one-hot encoding optimization model (7) includes one-hot encoding constraints Since the one-hot encoding constraints of the one-hot encoding optimization model (7) are discrete, to solve this thorny problem, in this embodiment, the one-hot encoding constraints are subjected to a probability simplex equivalent transformation. That is, first, the one-hot encoding constraints are convex hull relaxed, that is, the constraints on x i are relaxed to the k-dimensional probability simplex of its convex hull to obtain the first simplex equivalent constraint Obviously, this relaxation will cause the original one-hot encoding constraints to lose the characteristic that each x i has only one non-zero element. To solve this problem, note that a non-zero vector x has and only has one non-zero element if and only if ‖x‖ p = ‖x‖ q , where 1 ≤ p < q, that is, this is used as the second simplex equivalent constraint. The joint action of these two constraints makes the simplex equivalent constraint and the one-hot encoding constraint a non-relaxed equivalent transformation

[0119] After replacing the one-hot encoding constraints of the one-hot encoding optimization model (6) with the first simplex equivalent constraint and the second simplex equivalent constraint , the simplex non-relaxed optimization model is obtained as shown in the following formula (8).

[0120]

[0121] Next, based on the simplex non-relaxed optimization model, a remainder iterative update model is obtained, so as to obtain the solution of the identification remainder parameter r by using the remainder iterative update model Next, how to obtain the remainder iterative update model based on the simplex non-relaxed optimization model is further described

[0122] Referring to Figure 7 , based on the simplex non-relaxed optimization model, a remainder iterative update model is obtained, including the following steps 701 to step 702

[0123] Step 701: Based on the smooth quadratic penalty function and the iterative penalty parameter, the second simplex equivalent constraint is penalized to obtain a penalty function, and the penalty function is added to the interference function to obtain an interference penalty function

[0124] Step 702: Based on the interference penalty function, the first simplex equivalent constraint, and the one-hot encoding parameter, a remainder interference penalty model is obtained, and the remainder interference penalty model is subjected to mirror gradient descent processing to obtain a remainder iterative update model

[0125] Next, steps 701 to step 702 are described in detail

[0126] In some embodiments, after obtaining the simplex unrelaxed optimization model (8), in order to solve the simplex unrelaxed optimization model (8), this application proposes to construct a series of problems based on the penalty function and solve them using the mirror gradient descent algorithm. Specifically, the second simplex equivalent constraint is first penalized based on the smooth quadratic penalty function and the iterative penalty parameter, and then summed with the disturbance function to obtain the disturbance penalty function as shown in the following formula (9).

[0127]

[0128] Where Tr(·) represents the matrix trace operation, W represents the interference matrix, I represents the identity matrix, and ρ is the iteration penalty parameter.

[0129] Next, based on the interference penalty function (9), the first simplex equivalence constraint and the one-hot encoding parameters, the remainder interference penalty model is obtained as shown in the following formula (10).

[0130]

[0131] Next, after completing the construction of the remainder interference penalty model (10), the simplex equivalent optimization model (8) is solved using the mirror gradient descent algorithm. For a fixed iteration penalty parameter ρ, the interference penalty function (9) is a continuous, smooth, non-convex problem defined on the probabilistic simplex. Therefore, it can be solved using the mirror gradient descent algorithm. That is, the remainder interference penalty model needs to be processed by mirror gradient descent to obtain the formula for each iteration as shown in the following formula (11).

[0132]

[0133] in Let t denote the derivative of function F with respect to X, m denote the number of iterations, and t m B represents the step size chosen in the m-th iteration. ω (X,X (m) Let be the Bregman divergence. The definition of the Bregman divergence is shown in the following formula (12).

[0134]

[0135] Based on this iterative formula (11), the remainder iterative update model can be further obtained. The following will further describe how to obtain the remainder iterative update model.

[0136] Reference Figure 8 The remainder interference penalty model is subjected to mirror gradient descent processing to obtain the remainder iterative update model, including the following steps 801 to 803.

[0137] Step 801: Obtain the iteration step size, the strongly convex differentiable function, and the corresponding KL divergence relation.

[0138] Step 802: Differentiate the interference penalty function of the remainder interference penalty model to obtain the interference penalty derivative function, and use mirror gradient descent to obtain the remainder mirror gradient descent model of the remainder interference penalty model based on the interference penalty function, the interference penalty derivative function, the KL divergence relation, and the iteration step size.

[0139] Step 803: Based on the strongly convex differentiable function and the remainder mirror gradient descent model, the remainder iterative update model is obtained.

[0140] Steps 801 to 803 are described in detail below.

[0141] In some embodiments, after determining the iterative formula (11), it is necessary to first determine the iteration step size t. m Since solving the problem using the mirror gradient descent algorithm requires full utilization of the set properties of the probability simplex itself, and noting that the Kullback-Leibler divergence (KL divergence) is a good measure of the distance between probability distributions and the variance within a distribution, this application uses KL divergence as a specific Bregman divergence to design the algorithm. Therefore, the strongly convex differentiable function in the Bregman divergence is determined. X,Y∈C, as shown in the following formula (13).

[0142]

[0143] The KL divergence relation, which is the Bregman divergence, is shown in the following formula (14).

[0144]

[0145] Differentiating the interference penalty function of the remainder interference penalty model yields the interference penalty derivative function. And using mirror gradient descent, based on the interference penalty function, the interference penalty derivative function, the KL divergence relation, and the iteration step size, the remainder mirror gradient descent model of the remainder interference penalty model is obtained as shown in the following formula (15).

[0146]

[0147] By combining the strongly convex differentiable function (13) with the remainder mirror gradient descent model (15), it can be proved that the m-th iteration of the mirror gradient descent algorithm has a closed update form, namely the remainder iteration update model, as shown in the following formula (16).

[0148]

[0149] It is understandable that the remainder iterative update model (16) is the parameter update form of the remainder mirror gradient descent model (15).

[0150] Through steps 401 to 403, 501 to 503, 601 to 603, 701 to 702, and 801 to 803, an identifier allocation model based on relevant system parameters in a wireless communication system is proposed. The model is decomposed into a remainder and quotient modulo k using the remainder division method, thus transforming the identifier allocation model into a modulo optimization problem and a graph coloring problem. This effectively reduces the complexity of identifier allocation processing while ensuring user communication quality. For the difficult-to-handle modulo optimization problem, a probabilistic simplex equivalent transformation is performed on the one-hot coding constraint to obtain a simplex unrelaxed optimization model. Compared to traditional relaxation methods, this scheme does not require relaxation of the original problem, thus avoiding relaxation and rounding errors. Furthermore, to address the newly proposed no-relaxation problem, it transforms the problem into a smooth non-convex optimization task by appropriately penalizing the norm condition (Exact Penalty). By utilizing KL divergence as the Bregman divergence in mirror descent (MD), it effectively solves this series of penalty problems, thereby obtaining a more suitable remainder iterative update model. This enables cell identifier allocation in practical wireless communication systems to effectively ensure user communication quality while reducing the complexity of identifier allocation processing, thereby improving the efficiency of cell identifier allocation.

[0151] Step 202: Update the iterative optimization parameters based on the iterative penalty parameters and remainders, and obtain the identifier remainder based on the updated iterative optimization parameters.

[0152] Step 202 will be described in detail below.

[0153] In some embodiments, after obtaining the iterative penalty parameter ρ and the remainder, the iterative update model (16) and the iterative optimization parameter X are obtained. (m) Then, the iterative optimization parameter X is updated in each iteration using the iterative penalty parameter ρ and the remainder (16). (m) Until convergence, and based on the converged iterative optimization parameters and formula (6), the identifier remainder is obtained.

[0154] Step 203: Solve the quotient optimization model based on the identifier remainder to obtain the identifier quotient. Obtain the identifier allocation number based on the identifier quotient and the identifier remainder, and allocate an identifier to each communication cell based on the identifier allocation number.

[0155] Step 203 will be described in detail below.

[0156] In some embodiments, after obtaining the identifier remainder Next, the quotient optimization model (5) is solved based on the identifier remainder to obtain a suitable identifier quotient. The following further describes how to solve the quotient optimization model (5) based on the identifier remainder.

[0157] Reference Figure 9 The identification quotient is obtained by solving the quotient optimization model based on the identification remainder. The identification allocation number is obtained based on the identification quotient and the identification remainder, including the following steps 901 to 903.

[0158] Step 901: Obtain multiple subgraph discrete numbers, and generate multiple subgraphs with the same number of subgraph discrete numbers based on the identifier modulus interference relationship, identifier remainder, and subgraph discrete parameters.

[0159] Step 902: Update the quotient optimization model one by one using each subgraph to obtain the updated quotient optimization model, and solve the updated quotient optimization model to obtain the identifier quotient corresponding to each identifier remainder.

[0160] Step 903: Based on the remainder divisor decomposition function, multiple identifier remainders, and the identifier quotient corresponding to each identifier remainder, obtain the identifier allocation number.

[0161] Steps 901 to 903 are described in detail below.

[0162] In some embodiments, the solution to the remainder optimization model (4) is obtained (i.e., the remainder is identified). Next, the discrete numbers k of multiple subgraphs are first obtained, and then the interference relationship between the identifier modulus and the identifier remainder are determined. The subgraph discrete parameter k generates multiple subgraphs with the same number of discrete elements as the subgraph, denoted as k. Where r′=0,1,…,k-1,

[0163]

[0164] Next, an update quotient optimization model is established for each subgraph, resulting in the update quotient optimization model. Then, based on the existing graph coloring algorithm based on a greedy algorithm, the update quotient optimization model is solved to obtain the identifier quotient corresponding to each identifier remainder. After obtaining the identifier quotient and identifier remainder Then, the identifier allocation number for each physical cell in the wireless communication system is obtained again using the remainder divisor decomposition function (2). The identifier allocation number for each physical cell is the identifier allocated to each communication cell.

[0165] In some embodiments, before the algorithm begins, to avoid [further issues] in the simplex [process]... For non-differentiable initial points on the boundary, a stochastic initialization update strategy is proposed to obtain the initial iterative optimization parameters. Specifically, for the j-th column Spelling z~U k×1 [0,1], then perform a normalization operation, let

[0166] In the post-processing stage, a local search algorithm is proposed to further optimize the quality of the solutions generated by the PMD algorithm, and to obtain solutions based on the PMD algorithm. Then, the values ​​of PCI modulo k for any two cells are enumerated, and the selection that maximizes the decrease in the objective function is calculated. These two cells are then updated. This process is repeated until all cell pairs have been enumerated, thus obtaining a better solution. The complexity of this local search algorithm

[0167] In some embodiments, the PMD algorithm for solving the remainder optimization model and the greedy graph coloring algorithm for solving the quotient optimization model are combined to obtain the final PCI allocation algorithm. The entire process of the algorithm includes: firstly, optimizing the remainder r of PCI modulo k. Specifically, for each optimization problem corresponding to a fixed ρ, the stable point of the problem is found by the iterative algorithm proposed in the above formula (16), and the penalty coefficient ρ is continuously increased by the penalty function product factor γ until the algorithm satisfies the convergence condition. After the algorithm converges and obtains the solution, the local search algorithm as shown above is used on the solution until the value of the objective function no longer decreases, thereby obtaining the final allocation of the remainder r obtained by PCI modulo k. Then, construct k subgraphs, denoted as... Then, based on a greedy coloring algorithm, these subgraphs are colored to obtain the PCI quotient allocation with respect to k. Finally, the complete PCI allocation can be reconstructed based on the relationship in equation (1) above, that is...

[0168] Reference Figure 10 This is a flowchart illustrating another cell identifier allocation method provided in an embodiment of this application. Figure 10 As shown, first, the engineering parameters (including the base station frequency and cell number of each communication cell) and MR data (including the base station numbers of each main communication cell and its neighboring communication cells, as well as the received signal strength) are input. Then, the engineering parameters and MR data are processed using the cell identifier allocation method described above to output the PCI allocation result.

[0169] To further verify the reliability of the cell identifier allocation method for the wireless communication system proposed in this application, this embodiment performs performance simulations on the proposed cell identifier allocation method (i.e., the GGC algorithm) and three other existing algorithms (including the genetic algorithm, the BQP algorithm, and the PMD algorithm), referring to... Figure 11 This is a simulation diagram illustrating the cell identifier allocation performance provided in an embodiment of this application. Figure 11 As shown, the cell identifier allocation method (i.e., GGC algorithm) proposed in this application has a collision number and confusion number that can guarantee the quality of user communication while significantly reducing its running time.

[0170] The cell identifier allocation method and related equipment for a wireless communication system proposed in this application embodiment include a wireless communication system comprising multiple cellular communication cells. The method includes: first, obtaining iterative penalty parameters; second, obtaining an initial factor matrix based on a uniform distribution between zero and one; third, obtaining iterative optimization parameters corresponding to each initial factor based on the ratio of each initial factor to the first norm of the initial factor matrix.Furthermore, the co-frequency neighbor relationships, second-order co-frequency neighbor relationships, co-frequency overlapping coverage neighbor relationships, and identifier modulus-remainder interference relationships between every two communication cells are obtained. Based on the co-frequency overlapping coverage neighbor relationships, the interference matrix is ​​obtained, and the identifier allocation number parameters corresponding to the identifier allocation number are obtained. Based on the identifier allocation number parameters, the identifier modulus-remainder relationship and the identifier allocation constraints corresponding to the identifier allocation number parameters are obtained. Based on the interference matrix, identifier allocation number parameters, identifier modulus-remainder interference relationships, and identifier allocation constraints, the identifier allocation model is obtained. The identifier allocation model is divided into a quotient optimization model and a remainder optimization model. Based on the remainder optimization model, the remainder iterative update model is obtained, and the remainder-divisor decomposition function is obtained. Based on the remainder-divisor decomposition function, the identifier allocation parameter in the identifier allocation model is decomposed into identifier remainder parameter and identifier quotient parameter. The identifier remainder parameter corresponds to the identifier remainder, and the identifier quotient parameter corresponds to the identifier quotient. The identifier allocation model is updated based on the identifier remainder parameter and the identifier quotient parameter to obtain the identifier quotient remainder allocation model. Based on the identifier remainder parameter and the identifier quotient parameter, the identifier quotient remainder allocation model is further decomposed to obtain the quotient optimization model and the remainder optimization model. The one-hot encoding parameters are obtained, as well as the encoding conversion relationship between the one-hot encoding parameters and the identifier remainder parameter. The remainder optimization model is updated using the encoding conversion relationship and the one-hot encoding parameters to obtain the one-hot encoding optimization model. The one-hot encoding optimization model includes one-hot encoding constraints. A probabilistic simplex equivalence transformation is performed on these constraints to obtain simplex equivalence constraints. These simplex equivalence constraints are then used to replace the one-hot encoding constraints in the optimization model, resulting in a simplex unrelaxed optimization model. A remainder iterative update model is then derived based on this model. A second simplex equivalence constraint is penalized using a smooth quadratic penalty function and iteration penalty parameters, and this penalty is summed with an interference function to obtain an interference penalty function. Based on this interference penalty function, the first simplex equivalence constraint, and the one-hot encoding parameters, a remainder interference penalty model is obtained. This model is then subjected to mirror gradient descent to obtain the remainder iteration. The model is updated by obtaining the iteration step size, strongly convex differentiable function, and corresponding KL divergence relation. The interference penalty function of the remainder interference penalty model is differentiated to obtain the interference penalty derivative function. Using mirror gradient descent, based on the interference penalty function, interference penalty derivative function, KL divergence relation, and iteration step size, the remainder mirror gradient descent model of the remainder interference penalty model is obtained. Based on the strongly convex differentiable function and the remainder mirror gradient descent model, the remainder iterative update model is obtained, which is the parameter update form of the remainder mirror gradient descent model. The iterative optimization parameters are updated based on the iterative penalty parameters and the remainder iterative update model, and the identifier remainder is obtained based on the updated iterative optimization parameters.Finally, multiple subgraph discrete numbers are obtained. Based on the identifier modulus-remainder interference relationship, identifier remainder, and subgraph discrete parameters, multiple subgraphs with the same number of discrete numbers as the subgraphs are generated. The update quotient optimization model is then used for each subgraph to obtain the update quotient optimization model. Solving the update quotient optimization model yields the identifier quotient corresponding to each identifier remainder. Based on the remainder-divisor decomposition function, multiple identifier remainders, and the identifier quotient corresponding to each identifier remainder, the identifier allocation number is obtained. Based on the identifier quotient and identifier remainder, the identifier allocation number is obtained, and identifiers are allocated to each communication cell based on the identifier allocation number.

[0171] This application utilizes a quotient optimization model and a remainder iterative update model corresponding to the system parameters of the wireless communication system and the goal of reducing conflicts. It further solves for the identifier remainder and identifier quotient, and combines this with the relationship between the physical cell identifier and the quotient and remainder after division with remainder, to perform a reverse derivation of division with remainder, quickly obtaining the identifier allocation number for each physical cell. This minimizes PCI conflicts between communication cells when allocating identifiers for multiple physical cells in a wireless communication system. Furthermore, it proposes an identifier allocation model generated based on relevant system parameters in the wireless communication system, and uses division with remainder to decompose the identifier allocation model into a decomposition of the remainder and quotient modulo k, thus transforming the identifier allocation model into a modulo optimization problem and a graph coloring problem. This effectively reduces the complexity of identifier allocation processing while ensuring user communication quality. For the difficult-to-handle modulo optimization problem, a probabilistic simplex equivalent transformation is performed on the one-hot coding constraint to obtain a probabilistic simplex unrelaxed modulo optimization model. Compared with traditional relaxation methods, this scheme does not require relaxation of the original problem, thus avoiding relaxation and rounding errors. In addition, to solve the newly proposed no-relaxation problem, the problem is transformed into a smooth non-convex optimization task by appropriately penalizing the norm condition (Exact Penalty). The KL divergence is used as the Bregman divergence in mirror descent (MD), which effectively solves this series of penalty problems, thereby obtaining a more suitable remainder iterative update model. This allows for effective reduction of the complexity of the identifier allocation process while ensuring user communication quality in actual wireless communication systems, thereby improving the efficiency of cell identifier allocation.

[0172] This application also provides a cell identifier allocation apparatus for a wireless communication system, which can implement the cell identifier allocation method of the above-mentioned wireless communication system, referring to... Figure 12 The device 1200 includes:

[0173] The acquisition module 1210 is used to acquire iterative optimization parameters and iterative penalty parameters, as well as quotient optimization model and remainder iterative update model. The quotient optimization model and remainder iterative update model are obtained based on the system parameters of the wireless communication system.

[0174] The iterative update module 1220 is used to update the iterative optimization parameters based on the iterative penalty parameters and the remainder of the iterative update model, and to obtain the identifier remainder based on the updated iterative optimization parameters;

[0175] The identifier allocation module 1230 is used to solve the quotient optimization model based on the identifier remainder to obtain the identifier quotient, obtain the identifier allocation number based on the identifier quotient and the identifier remainder, and allocate an identifier to each communication cell based on the identifier allocation number.

[0176] In some embodiments, the acquisition module 1210 is further configured to:

[0177] Obtain the co-frequency neighbor cell relationship, second-order co-frequency neighbor cell relationship, co-frequency overlapping coverage neighbor cell relationship, and identification modal interference relationship between every two communication cells, and obtain the interference matrix based on the co-frequency overlapping coverage neighbor cell relationship;

[0178] Obtain the identifier allocation number parameter corresponding to the identifier allocation number, obtain the identifier modulus-remainder relationship based on the identifier allocation number parameter, obtain the identifier allocation constraint corresponding to the identifier allocation number parameter, and obtain the identifier allocation model based on the interference matrix, identifier allocation number parameter, identifier modulus-remainder interference relationship and identifier allocation constraint;

[0179] The identifier allocation model is divided into a quotient optimization model and a remainder optimization model, and a remainder iterative update model is obtained based on the remainder optimization model.

[0180] In some embodiments, the acquisition module 1210 is further configured to:

[0181] Obtain the remainder-divisor decomposition function, and based on the remainder-divisor decomposition function, decompose the identifier allocation number parameter in the identifier allocation model into the identifier remainder parameter and the identifier quotient parameter. The identifier remainder parameter corresponds to the identifier remainder, and the identifier quotient parameter corresponds to the identifier quotient.

[0182] The identifier allocation model is updated based on the identifier remainder parameter and the identifier quotient parameter to obtain the identifier quotient remainder allocation model;

[0183] Based on the identifier remainder parameter and the identifier quotient parameter, the identifier quotient remainder allocation model is split to obtain the quotient optimization model and the remainder optimization model.

[0184] In some embodiments, the acquisition module 1210 is further configured to:

[0185] Obtain the one-hot coding parameters and the coding conversion relationship between the one-hot coding parameters and the identifier remainder parameters. Then, use the coding conversion relationship and the one-hot coding parameters to update the remainder optimization model to obtain the one-hot coding optimization model, which includes one-hot coding constraints.

[0186] By performing a probabilistic simplex equivalent transformation on the one-hot encoded constraints, we obtain simplex equivalent constraints.

[0187] By replacing the one-hot encoding constraints of the simplex equivalence constraint with those of the one-hot encoding optimization model, a simplex unrelaxed optimization model is obtained, and a remainder iterative update model is derived based on the simplex unrelaxed optimization model.

[0188] In some embodiments, the acquisition module 1210 is further configured to:

[0189] Based on the smooth quadratic penalty function and the iterative penalty parameter, the second simplex equivalent constraint is penalized to obtain the penalty function, which is then summed with the disturbance function to obtain the disturbance penalty function.

[0190] Based on the interference penalty function, the first simplex equivalence constraint, and the one-hot encoding parameters, the remainder interference penalty model is obtained. Then, the remainder interference penalty model is processed by mirror gradient descent to obtain the remainder iterative update model.

[0191] In some embodiments, the acquisition module 1210 is further configured to:

[0192] Obtain the KL divergence relation of the iteration step size and the iteration penalty parameter, as well as the strongly convex differentiable function of the iteration penalty parameter;

[0193] The interference penalty function of the remainder interference penalty model is differentiated to obtain the interference penalty derivative function. Then, using mirror gradient descent, based on the interference penalty function, the interference penalty derivative function, the KL divergence relation, and the iteration step size, the remainder mirror gradient descent model of the remainder interference penalty model is obtained.

[0194] Based on strongly convex differentiable functions and the remainder mirror gradient descent model, a remainder iterative update model is obtained, which is the parameter update form of the remainder mirror gradient descent model.

[0195] In some embodiments, the acquisition module 1210 is further configured to:

[0196] Based on a uniform distribution between zero and one, an initial factor matrix is ​​obtained, which includes multiple initial factors consistent with the number of communication cells.

[0197] Based on the ratio of each initial factor to the first norm of the initial factor matrix, the iterative optimization parameters corresponding to each initial factor are obtained.

[0198] In some embodiments, the identifier allocation module 1230 is further configured to:

[0199] Obtain the discrete numbers of multiple subgraphs, and generate multiple subgraphs with the same number of discrete numbers as the subgraphs based on the identification modulus interference relationship, the identification remainder, and the subgraph discrete parameters;

[0200] The quotient optimization model is updated one by one using each subgraph to obtain the updated quotient optimization model, and the updated quotient optimization model is solved to obtain the identifier quotient corresponding to each identifier remainder;

[0201] Based on the remainder divisor decomposition function, multiple identifier remainders, and the identifier quotient corresponding to each identifier remainder, the identifier allocation number is obtained.

[0202] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, the specific implementation of the cell identifier allocation device of the wireless communication system is basically the same as the specific implementation of the cell identifier allocation method of the wireless communication system described above, and will not be repeated here.

[0203] In this embodiment, the cell identifier allocation device of the wireless communication system utilizes the quotient optimization model and remainder iterative update model corresponding to the system parameters of the wireless communication system and the conflict reduction objective, and further solves for the identifier remainder and identifier quotient. It also combines the relationship between the physical cell identifier and the quotient and remainder after band remainder division to perform reverse derivation of band remainder division, quickly obtaining the identifier allocation number for each physical cell. This minimizes PCI conflicts between communication cells when allocating identifiers for multiple physical cells in the wireless communication system. Furthermore, it proposes an identifier allocation model generated based on relevant system parameters in the wireless communication system, and uses band remainder division to decompose the identifier allocation model into a decomposition of the remainder and quotient modulo k, thereby transforming the identifier allocation model into a modulo optimization problem and a graph coloring problem. This effectively reduces the complexity of identifier allocation processing while ensuring user communication quality. For the difficult-to-handle modulo optimization problem, probabilistic simplex relaxation is applied to the one-hot coding constraint to obtain a modulo optimization model without probabilistic simplex relaxation. Compared with traditional relaxation methods, this scheme does not require relaxation of the original problem, thus avoiding relaxation and rounding errors. In addition, to solve the newly proposed no-relaxation problem, the problem is transformed into a smooth non-convex optimization task by appropriately penalizing the norm condition (ExactPenalty). By using KL divergence as the Bregman divergence in mirror descent (MD), this series of penalty problems is effectively solved, resulting in a more suitable remainder iterative update model. This allows for effective reduction of the complexity of cell identifier allocation processing while ensuring user communication quality in actual wireless communication systems, thereby improving the efficiency of cell identifier allocation.

[0204] This application also provides an electronic device, including:

[0205] At least one memory;

[0206] At least one processor;

[0207] At least one program;

[0208] The program is stored in a memory, and the processor executes the at least one program to implement the cell identifier allocation method of the wireless communication system described above in this application. The electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), in-vehicle computers, etc.

[0209] Please see Figure 13 , Figure 13 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0210] The processor 1301 can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this application.

[0211] The memory 1302 can be implemented in the form of ROM (Read-Only Memory), static storage device, dynamic storage device, or RAM (Random Access Memory). The memory 1302 can store the operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1302 and is called and executed by the processor 1301 to execute the cell identifier allocation method of the wireless communication system of the embodiments of this application.

[0212] The input / output interface 1303 is used to implement information input and output;

[0213] The communication interface 1304 is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0214] Bus 1305 transmits information between various components of the device (e.g., processor 1301, memory 1302, input / output interface 1303, and communication interface 1304);

[0215] The processor 1301, memory 1302, input / output interface 1303 and communication interface 1304 are connected to each other within the device via bus 1305.

[0216] This application embodiment also provides a storage medium, which is a computer-readable storage medium, storing a computer program that, when executed by a processor, implements the cell identifier allocation method of the above-described wireless communication system.

[0217] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0218] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0219] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0220] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0221] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0222] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0223] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0224] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of the units described above is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. The coupling or direct coupling or communication connection between the shown or discussed units may be through some interfaces, or indirect coupling or communication connection between the apparatus or units, and may be electrical, mechanical, or other forms.

[0225] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0226] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0227] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0228] The preferred embodiments of the present application have been described above with reference to the accompanying drawings, but this does not limit the scope of the claims of the present application. Any modifications, equivalent substitutions, and improvements made by those skilled in the art without departing from the scope and substance of the embodiments of the present application shall be within the scope of the claims of the present application.

Claims

1. A cell identifier allocation method for a wireless communication system, characterized in that, The wireless communication system includes multiple communication cells configured with cellular arrays, and the method includes: Obtain iterative optimization parameters and iterative penalty parameters, as well as quotient optimization model and remainder iterative update model, wherein the quotient optimization model and remainder iterative update model are obtained based on the system parameters of the wireless communication system and the conflict reduction objective; The iterative optimization parameters are updated based on the iterative penalty parameters and the remainder iterative update model, and the identifier remainder is obtained based on the updated iterative optimization parameters; The quotient optimization model is solved based on the identifier remainder to obtain the identifier quotient. The identifier allocation number is obtained based on the identifier quotient and the identifier remainder. An identifier is then allocated to each of the communication cells based on the identifier allocation number. The acquisition of the quotient optimization model and the remainder iterative update model includes: acquiring the co-frequency neighbor cell relationship, second-order co-frequency neighbor cell relationship, co-frequency overlapping coverage neighbor cell relationship, and identifier modulus-remainder interference relationship between every two communication cells; obtaining an interference matrix based on the co-frequency overlapping coverage neighbor cell relationship; acquiring identifier allocation number parameters corresponding to the identifier allocation number; acquiring identifier modulus-remainder relationship based on the identifier allocation number parameters; acquiring identifier allocation constraints corresponding to the identifier allocation number parameters; and obtaining an identifier allocation model based on the interference matrix, the identifier allocation number parameters, the identifier modulus-remainder interference relationship, and the identifier allocation constraints; dividing the identifier allocation model into the quotient optimization model and the remainder optimization model; and obtaining a remainder iterative update model based on the remainder optimization model. The step of dividing the identifier allocation model into the quotient optimization model and the remainder optimization model includes: obtaining a remainder-divisor decomposition function, and based on the remainder-divisor decomposition function, decomposing the identifier allocation parameter in the identifier allocation model into an identifier remainder parameter and an identifier quotient parameter, wherein the identifier remainder parameter corresponds to the identifier remainder, and the identifier quotient parameter corresponds to the identifier quotient; updating the identifier allocation model based on the identifier remainder parameter and the identifier quotient parameter to obtain an identifier quotient remainder allocation model; and splitting the identifier quotient remainder allocation model based on the identifier remainder parameter and the identifier quotient parameter to obtain the quotient optimization model and the remainder optimization model. The process of obtaining the remainder iterative update model based on the remainder optimization model includes: acquiring one-hot encoding parameters, acquiring the encoding transformation relationship between the one-hot encoding parameters and the identifier remainder parameters, and updating the remainder optimization model using the encoding transformation relationship and the one-hot encoding parameters to obtain a one-hot encoding optimization model, wherein the one-hot encoding optimization model includes one-hot encoding constraints; performing a probabilistic simplex equivalence transformation on the one-hot encoding constraints to obtain simplex equivalence constraints; replacing the one-hot encoding constraints of the one-hot encoding optimization model based on the simplex equivalence constraints to obtain a simplex unrelaxed optimization model, and obtaining the remainder iterative update model based on the simplex unrelaxed optimization model. The simplex equivalence constraints include a first simplex equivalence constraint and a second simplex equivalence constraint. The simplex no-relaxation optimization model includes an interference function and the simplex equivalence constraints. Obtaining the remainder iterative update model based on the simplex no-relaxation optimization model includes: penalizing the second simplex equivalence constraint based on a smooth quadratic penalty function and the iterative penalty parameter to obtain a penalty function, and summing it with the interference function to obtain an interference penalty function; obtaining a remainder interference penalty model based on the interference penalty function and the first simplex equivalence constraint, and performing mirror gradient descent on the remainder interference penalty model to obtain the remainder iterative update model. The step of solving the quotient optimization model based on the identifier remainder to obtain the identifier quotient, and obtaining the identifier allocation number based on the identifier quotient and the identifier remainder, includes: obtaining multiple subgraph discrete numbers; generating multiple subgraphs with the same number of subgraph discrete numbers based on the identifier modulus-remainder interference relationship, the identifier remainder, and the subgraph discrete parameters; updating the quotient optimization model one by one using each subgraph to obtain an updated quotient optimization model, and solving the updated quotient optimization model to obtain the identifier quotient corresponding to each identifier remainder; and obtaining the identifier allocation number based on the remainder-based divisor decomposition function, the multiple identifier remainders, and the identifier quotient corresponding to each identifier remainder.

2. The cell identifier allocation method for a wireless communication system according to claim 1, characterized in that, The step of performing mirror gradient descent processing on the remainder interference penalty model to obtain the remainder iterative update model includes: Obtain the iteration step size, the strongly convex differentiable function, and the corresponding KL divergence relation; The interference penalty function of the remainder interference penalty model is differentiated to obtain the interference penalty derivative function. Then, using mirror gradient descent, based on the interference penalty function, the interference penalty derivative function, the KL divergence relation, and the iteration step size, the remainder mirror gradient descent model of the remainder interference penalty model is obtained. Based on the strongly convex differentiable function and the remainder mirror gradient descent model, a remainder iterative update model is obtained, which is the parameter update form of the remainder mirror gradient descent model.

3. The cell identifier allocation method for a wireless communication system according to claim 1, characterized in that, The process of obtaining iterative optimization parameters includes: An initial factor matrix is ​​obtained based on a uniform distribution between zero and one, and the initial factor matrix includes multiple initial factors consistent with the number of communication cells; Based on the ratio of each initial factor to the first norm of the initial factor matrix, the iterative optimization parameters corresponding to each initial factor are obtained.

4. An electronic device, characterized in that, The system includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the cell identifier allocation method of the wireless communication system according to any one of claims 1 to 3.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the cell identifier allocation method of the wireless communication system according to any one of claims 1 to 3.

Images