Power domain interference pattern estimation method under cellular network multi-cell scene

Abstract

The invention discloses a power domain interference pattern estimation method in a cellular network multi-cell scene, and the method comprises the steps: constructing a transmission power matrix based on the condition of power allocated to a plurality of resource blocks by each base station in a preset time period; based on the maximum energy efficiency and the minimum condition number of the sending power matrix, creating an objective function, and constructing an optimization model by taking the allocation condition of the resource blocks, the sending power range and the full rank of the sending power matrix as constraint conditions of the objective function; decomposing the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solving the first sub-optimization model, and solving the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal sending power matrix; based on the optimal sending power matrix and the receiving power, the equivalent channel vector of the user is calculated, the original time domain / frequency domain is expanded to the power domain, and the problems that a traditional method is large in measurement overhead and high in time synchronization precision requirement are fundamentally solved.

Description

Technical Field

[0001] The present invention relates to the field of network interference technology, and in particular to a method for estimating a power domain interference graph in a multi-cell scenario of a cellular network. Background Art

[0002] A multi-cell cellular network is an organizational form of mobile communication network. It achieves wide-range mobile communication coverage by dividing the entire service area into multiple cells and setting up a base station (or a subsystem of a base station) in each cell. With the surge in the number of mobile device users, cellular networks tend to be densely deployed to meet users' needs for network traffic and speed. However, dense base station deployment will inevitably bring about inter-cell interference, which has a significant negative impact on network capacity. Therefore, the communication system must effectively manage these interferences and optimize resource allocation based on the actual system interference situation to maintain network performance.

[0003] Traditional interference management and resource scheduling methods usually assume that inter-cell interference is known, while ignoring the effective and accurate measurement of interference. However, multi-cell cellular networks are dynamic, large-scale and complex, which makes existing interference measurement methods face the problems of high measurement overhead, insufficient accuracy and insufficient flexibility in practical applications.

[0004] The Channel State Information Interference Measurement (CSI-IM) method introduced in the 4G Long Term Evolution (LTE) standard is an interference measurement technology widely used in multi-cell cellular networks. When a base station (BS) to be measured sends information to a UE (User Equipment) on a certain RE (Resource Element), assuming that all base stations have achieved ideal time synchronization, other base stations should send zero information on the corresponding time-frequency resource element. In this way, the signal strength received by the UE on this specific RE will be equal to the inter-cell interference of the base station to be measured on the UE. The signal carried by the base station to be measured on the RE is called CSI-RS, and the mechanism by which the base stations of adjacent cells send zero information for collaborative measurement is called CSI-IM. In a multi-cell scenario, the base stations in each cell take turns to send CSI-RS through scheduling, while ensuring that the base stations in the remaining cells send zero information on the corresponding RE, so as to measure the inter-cell interference of all base stations on the UE.

[0005] The measurement method based on CSI-IM relies on the scheduling of all neighboring base stations to ensure that during the measurement phase, when the base station to be measured sends CSI-RS, all other remaining base stations send zero information on the corresponding RE, resulting in large measurement overhead. In addition, strict symbol-level time synchronization is required between base stations. However, in 5G systems, symbol time is often measured in nanoseconds (ns), which makes the accuracy of time synchronization a difficult challenge for this method to overcome. Summary of the invention

[0006] In view of this, the present invention provides a power domain interference graph estimation method in a multi-cell scenario of a cellular network, the main purpose of which is to solve the problems of high time synchronization accuracy requirement and large measurement overhead in the existing interference measurement method of the cell cellular network.

[0007] According to one aspect of the present application, a method for estimating a power domain interference graph in a multi-cell scenario of a cellular network is provided, the method comprising:

[0008] Acquire the power conditions allocated by each base station to the multiple resource blocks within a preset period, and construct a transmit power matrix based on the power conditions allocated by each base station to the multiple resource blocks within the preset period;

[0009] Based on the maximization of energy efficiency and the minimization of condition number of the transmit power matrix, an objective function is created, and the allocation of the resource blocks, the transmit power range, the signal-to-noise ratio range, and the full rank of the transmit power matrix are used as constraints of the objective function to construct an optimization model;

[0010] Decomposing the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solving the first sub-optimization model, and solving the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix;

[0011] The receiving power corresponding to the power allocated to the user and the multiple resource blocks is obtained, and based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector, an equivalent channel vector of the user is calculated.

[0012] Optionally, constructing a transmit power matrix based on power conditions allocated by each base station to a plurality of resource blocks within a preset time period includes:

[0013] Obtaining the power allocated by each base station to multiple resource blocks of different frequencies within a preset period of time, and dividing a preset number of adjacent resource elements in the same frequency into a resource sub-block according to the time sequence;

[0014] Obtaining the transmit power of each resource element in each resource sub-block, and calculating the average transmit power corresponding to each resource sub-block based on the transmit power of the resource elements in each resource sub-block;

[0015] A transmit power matrix is ​​generated based on the frequency of each resource block, the time corresponding to the resource sub-blocks included therein, and the average transmit power.

[0016] Optionally, the optimization model is:

[0017]

[0018] Where E(P) represents the energy efficiency of the transmit power matrix P, k(P k [d]) represents the condition number of the transmit power matrix formed by the d-th resource block of the k-th base station and its adjacent resource blocks, δ k,z [d][l] indicates whether the kth base station allocates the lth resource subblock of the dth resource block to the zth user in the cell where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,z represents the lower bound of the SINR of the zth user in the cell where the kth base station is located, represents the transmission power of the zth user in the cell where the kth base station is located, represents the upper bound of the total power of the kth base station, P k [d] represents the transmission power of the dth resource block of the kth base station, N RB Indicates the number of RBs causing interference, u k Indicates the number of users in the cell where k base stations are located, RB k represents the RB set of the kth base station, and K represents the total number of base stations.

[0019] Optionally, the first sub-optimization model is:

[0020]

[0021]

[0022] The second sub-optimization model is:

[0023]

[0024] Where α represents the decomposition coefficient.

[0025] Optionally, solving the first sub-optimization model includes:

[0026] Based on the fractional form of energy efficiency, converting the objective function in the first sub-optimization model into a first intermediate objective function;

[0027] Performing relaxation conversion and convexity conversion on the first intermediate objective function in sequence to obtain a second intermediate objective function;

[0028] Performing linear decomposition on the signal-to-noise ratio range constraint in the first sub-optimization model to obtain a first signal-to-noise ratio linear constraint and a second signal-to-noise ratio linear constraint, and performing convexity transformation on the second signal-to-noise ratio linear constraint to obtain a third signal-to-noise ratio linear constraint;

[0029] The second intermediate objective function, the allocation of the resource blocks, the transmission power range, the first signal-to-noise ratio linear constraint and the third signal-to-noise ratio linear constraint are used as a conversion model of the first sub-optimization model, and the conversion model of the first sub-optimization model is solved.

[0030] Optionally, solving the second sub-optimization model based on the solution of the first sub-optimization model includes:

[0031] Decomposing the second sub-optimization model to obtain a first initial optimization model and a second initial optimization model;

[0032] Based on the solution of the first sub-optimization model, converting the first initial optimization model into a first intermediate optimization model;

[0033] Converting the full rank constraint of the transmit power matrix in the first intermediate optimization model into a positive definite constraint to obtain a third sub-optimization model, and solving the third sub-optimization model;

[0034] Based on the solution of the third sub-optimization model, adding a ratio constraint condition of the maximum eigenvalue and the minimum eigenvalue of the transmit power matrix to the second initial optimization model to obtain a second intermediate optimization model;

[0035] The constraint conditions in the second intermediate optimization model are scaled to obtain a fourth sub-optimization model, and the fourth sub-optimization model is solved.

[0036] Optionally, the equivalent channel vector of the user is calculated by the following method:

[0037]

[0038] in, is the received power of the user corresponding to the transmit power of the i-th resource block, v i is the Gaussian white noise power vector, P is the transmit power matrix, s i is the equivalent channel vector on the i-th resource block.

[0039] According to another aspect of the present application, a power domain-based interference measurement device is provided, including:

[0040] A transmission power matrix construction module, used to obtain the power conditions allocated by each base station to multiple resource blocks within a preset period of time, and to construct a transmission power matrix based on the power conditions allocated by each base station to multiple resource blocks within the preset period of time;

[0041] An optimization model building module is used to create an objective function based on maximizing energy efficiency and minimizing condition number of the transmit power matrix, and to construct an optimization model by taking the allocation of the resource blocks, the transmit power range, the signal-to-noise ratio range, and the full rank of the transmit power matrix as constraints of the objective function;

[0042] an optimization model solving module, configured to decompose the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solve the first sub-optimization model, and solve the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix;

[0043] The equivalent channel vector acquisition module is used to obtain the receiving power corresponding to the power allocated to the user and multiple resource blocks, and calculate the equivalent channel vector of the user based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector.

[0044] Optionally, the transmit power matrix construction module is further used for:

[0045] Obtaining the power allocated by each base station to multiple resource blocks of different frequencies within a preset period of time, and dividing a preset number of adjacent resource elements in the same frequency into a resource sub-block according to the time sequence;

[0046] Obtaining the transmit power of each resource element in each resource sub-block, and calculating the average transmit power corresponding to each resource sub-block based on the transmit power of the resource elements in each resource sub-block;

[0047] A transmit power matrix is ​​generated based on the frequency of each resource block, the time corresponding to the resource sub-blocks included therein, and the average transmit power.

[0048] Optionally, the optimization model is:

[0049]

[0050] Where E(P) represents the energy efficiency of the transmit power matrix P, k(P k [d]) represents the condition number of the transmit power matrix formed by the d-th resource block of the k-th base station and its adjacent resource blocks, δ k,z [d][l] indicates whether the kth base station allocates the lth resource subblock of the dth resource block to the zth user in the cell where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,zrepresents the lower bound of the SINR of the zth user in the cell where the kth base station is located, represents the transmission power of the zth user in the cell where the kth base station is located, represents the upper bound of the total power of the kth base station, P k [d] represents the transmission power of the dth resource block of the kth base station, N RB Indicates the number of RBs causing interference, u k Indicates the number of users in the cell where k base stations are located, RB k represents the RB set of the kth base station, and K represents the total number of base stations.

[0051] Optionally, the first sub-optimization model is:

[0052]

[0053] The second sub-optimization model is:

[0054]

[0055] Where α represents the decomposition coefficient.

[0056] Optionally, the optimization model solving module is further used for:

[0057] Based on the fractional form of energy efficiency, converting the objective function in the first sub-optimization model into a first intermediate objective function;

[0058] Performing relaxation conversion and convexity conversion on the first intermediate objective function in sequence to obtain a second intermediate objective function;

[0059] Performing linear decomposition on the signal-to-noise ratio range constraint in the first sub-optimization model to obtain a first signal-to-noise ratio linear constraint and a second signal-to-noise ratio linear constraint, and performing convexity transformation on the second signal-to-noise ratio linear constraint to obtain a third signal-to-noise ratio linear constraint;

[0060] The second intermediate objective function, the allocation of the resource blocks, the transmission power range, the first signal-to-noise ratio linear constraint and the third signal-to-noise ratio linear constraint are used as a conversion model of the first sub-optimization model, and the conversion model of the first sub-optimization model is solved.

[0061] Optionally, the optimization model solving module is further used for:

[0062] Decomposing the second sub-optimization model to obtain a first initial optimization model and a second initial optimization model;

[0063] Based on the solution of the first sub-optimization model, converting the first initial optimization model into a first intermediate optimization model;

[0064] Converting the full rank constraint of the transmit power matrix in the first intermediate optimization model into a positive definite constraint to obtain a third sub-optimization model, and solving the third sub-optimization model;

[0065] Based on the solution of the third sub-optimization model, adding a ratio constraint condition of the maximum eigenvalue and the minimum eigenvalue of the transmit power matrix to the second initial optimization model to obtain a second intermediate optimization model;

[0066] The constraint conditions in the second intermediate optimization model are scaled to obtain a fourth sub-optimization model, and the fourth sub-optimization model is solved.

[0067] Optionally, the equivalent channel vector of the user is calculated by the following method:

[0068]

[0069] in, is the received power of the user corresponding to the transmit power of the i-th resource block, v i is the Gaussian white noise power vector, P is the transmit power matrix, s i is the equivalent channel vector on the i-th resource block.

[0070] According to another aspect of the present application, a storage medium is provided, in which at least one executable instruction is stored, and the executable instruction enables a processor to perform operations corresponding to the power domain interference graph estimation method in the above-mentioned cellular network multi-cell scenario.

[0071] According to another aspect of the present application, a computer device is provided, comprising: a processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus;

[0072] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the power domain interference graph estimation method in the above-mentioned cellular network multi-cell scenario.

[0073] By means of the above technical solution, the technical solution provided by the embodiment of the present invention has at least the following advantages:

[0074] The present application provides a method for estimating a power domain interference graph in a multi-cell scenario of a cellular network. By measuring the transmit power of each base station and the receive power of the user at different frequencies over a period of time, that is, constructing a transmit power matrix, the physical characteristics and requirements of multi-cell cellular network communications are combined to design the transmit power matrix, and the maximum energy efficiency of the transmit power matrix and the minimum condition number of the matrix are used as objective functions. The allocation of resource blocks, the transmit power range, the signal-to-noise ratio range and the full rank of the transmit power matrix are used as constraints to construct an optimization model, solve the optimization model, and obtain the optimal transmit power matrix. The interference is inferred based on the relationship between the transmit power matrix and the receive power vector, thereby expanding the original time domain / frequency domain method to the new dimension of the power domain, fundamentally solving the problems of high measurement overhead and high requirements for time synchronization accuracy of traditional methods.

[0075] The above description is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0077] Figure 1 A flowchart of a method for estimating a power domain interference graph in a cellular network multi-cell scenario provided by an embodiment of the present application is shown;

[0078] Figure 2 A schematic diagram of constructing a transmit power matrix of a power domain interference graph estimation method in a cellular network multi-cell scenario provided by an embodiment of the present application is shown;

[0079] Figure 3 A decomposition diagram of an optimization model of a power domain interference graph estimation method in a cellular network multi-cell scenario provided by an embodiment of the present application is shown;

[0080] Figure 4 A structural block diagram of a power domain-based interference measurement device provided in an embodiment of the present application is shown;

[0081] Figure 5 A schematic structural diagram of a computer device provided by an embodiment of the present invention is shown.

[0082] in,

[0083] Figure 4In: 402-transmitting power matrix building module; 404-optimization model building module; 406-optimization model solving module;

[0084] Figure 5 In it: 502 - processor; 504 - communication interface; 506 - memory; 508 - communication bus; 510 - program. DETAILED DESCRIPTION

[0085] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present invention can be combined with each other without conflict.

[0086] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention application are described in detail below in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0087] In view of the problems that the interference measurement method for the existing cell cellular network has high time synchronization accuracy requirements and large measurement overhead, the embodiment of the present application provides a power domain interference graph estimation method in a cellular network multi-cell scenario, such as Figure 1 As shown, the method includes:

[0088] 102: Obtain power conditions allocated by each base station to multiple resource blocks within a preset period, and construct a transmit power matrix based on the power conditions allocated by each base station to multiple resource blocks within the preset period;

[0089] 104: Based on maximizing energy efficiency and minimizing condition number of the transmit power matrix, an objective function is created, resource block allocation, transmit power range, signal-to-noise ratio range, and full rank of the transmit power matrix are used as constraints of the objective function, and an optimization model is constructed;

[0090] 106: decomposing the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solving the first sub-optimization model, and solving the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix;

[0091] 108: Obtain the receiving power corresponding to the power allocated to the user and the multiple resource blocks, and calculate the equivalent channel vector of the user based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector.

[0092] Specifically, the power allocation of resource blocks (RB) at different frequencies of each base station in a multi-cell cellular network within a period of time is obtained, and then a transmit power matrix is ​​constructed according to these power allocation conditions. Since interference measurement and resource scheduling are coupled, there are two goals that need to be optimized simultaneously. The first is to maximize the energy efficiency (EE) of the transmit power matrix, and the second is to minimize the condition number of the transmit power matrix. Energy efficiency is defined as the ratio between communication rate and transmit power in a communication system. The larger the ratio, the more efficient the communication system is in utilizing energy. The condition number is used to indicate whether a matrix has good computational properties. Specifically, the smaller the condition number of the constructed transmit power matrix, the more accurate the channel size obtained using it. In order to achieve these two goals, the present invention takes the basic requirements of transmit power range, transmit rate size, resource block allocation mode and transmit power full rank in an actual multi-cell system as constraints, and constructs an optimization model with the maximization of energy efficiency and minimization of condition number of the transmit power matrix as the objective function. However, this optimization model has a typical mixed integer nonlinear problem (MINLP), which has NP-hard properties, and the solution process is very complicated and cumbersome. Therefore, the optimization model is decomposed to obtain a first sub-optimization model and a second sub-optimization model. The first sub-optimization model is first solved, and based on the solution of the first sub-optimization model, the second sub-optimization model is solved to obtain the optimal transmit power matrix. The interference is inferred based on the relationship between the transmit power matrix and the receive power vector, that is, based on the optimal transmit power matrix, the receive power and the Gaussian white noise power vector, the user's equivalent channel vector is calculated.

[0093] The present application provides a power domain interference graph estimation method in a multi-cell scenario of a cellular network. Compared with the prior art, by measuring the transmission power of each base station and the receiving power of the user at different frequencies over a period of time, that is, constructing a transmission power matrix, combined with the physical characteristics and requirements of multi-cell cellular network communications, the maximum energy efficiency of the transmission power matrix and the minimum condition number of the matrix are used as objective functions, the allocation of resource blocks, the transmission power range, the signal-to-noise ratio range and the full rank of the transmission power matrix are used as constraints, an optimization model is constructed, the optimization model is solved, and the optimal transmission power matrix is ​​obtained. The interference is obtained by inferring based on the relationship between the transmission power matrix and the receiving power vector, thereby expanding the original time domain / frequency domain method to the new dimension of the power domain, fundamentally solving the problems of high measurement overhead and high requirements for time synchronization accuracy of traditional methods.

[0094] In one embodiment, based on the power allocated by each base station to multiple resource blocks within a preset period, a transmit power matrix is ​​constructed, including:

[0095] Obtaining the power allocated by each base station to multiple resource blocks of different frequencies within a preset period of time, and dividing a preset number of adjacent resource elements in the same frequency into a resource sub-block according to the time sequence;

[0096] Obtaining the transmit power of each resource element in each resource sub-block, and calculating the average transmit power corresponding to each resource sub-block based on the transmit power of the resource elements in each resource sub-block;

[0097] A transmit power matrix is ​​generated based on the frequency of each resource block, the time corresponding to the resource sub-blocks included therein, and the average transmit power.

[0098] Specifically, the execution subject of the present application is a central control unit, which constructs a transmission power matrix according to the resource allocation of each base station. Each base station in a multi-cell cellular network sends the power allocation of resource blocks (RBs) on different frequencies over a period of time to the central control unit, which then constructs the transmission matrix.

[0099] Figure 2 A scenario with three RBs is shown, where each resource block can be considered as a collection of several adjacent REs (resource elements) in the same time unit (symbol). These RBs can come from the same base station or from different base stations, so that the interference between base stations in different cells or between RBs within the same base station in the same cell can be flexibly characterized. Figure 2 It can be seen that in order to measure the interference between these RBs and a certain user, several RBs on the same frequency in continuous time are combined into a "Block", and then the average power of this Block is considered. It is worth noting that on the base station side, based on the power allocated to the RB by the scheduling strategy and the modulation method used, the ideal power value on each block can be obtained by calculating the mathematical expectation, but in the actual system, due to the limited number of signal sampling points used to form the block, there is a deviation between the estimated power value and the ideal value. Based on rigorous mathematical derivation, it is proved that the expected power value can be made close to the actual power by reasonably selecting the block length, thereby ignoring this estimation error. The resource allocation strategy will cause the power of different blocks to change, so its average power is calculated on each adjacent block, and then it is obtained. Figure 1 The transmit power matrix on the right side of the figure. As shown in Figure 1, each row in the 3*3 matrix represents a different RB, and each column represents a block at a different time.

[0100] The transmit power matrix is ​​constructed by calculating the average power, without distinguishing whether the signal is a data signal or a reference signal, thereby fundamentally avoiding the occupation of time-frequency resources by the reference signal.

[0101] The construction of the transmit power matrix can also average out the inter-subcarrier interference from a larger scale. Specifically, inter-subcarrier interference originally exists between REs, and tends to decrease as the interval frequency between REs increases. The present invention follows the principle of using RB as the minimum resource allocation unit in the cellular network standard, and combines multiple REs into RBs for consideration. This method averages the differences in the size of inter-subcarrier interference caused by different interval frequencies between REs. As a result, the order of magnitude of inter-subcarrier interference borne by each RB is less different, which has good robustness for CFO and improves the measurement accuracy as a whole.

[0102] In addition, the construction of the transmit power matrix can also cope with a certain degree of time asynchrony. The traditional CSI-IM method requires strict time synchronization so that the user can only receive signals from a single base station on a specific RE, so as to calculate the inter-cell interference.

[0103] In one embodiment of the present invention, the optimization model is:

[0104]

[0105] Where E(P) represents the energy efficiency of the transmit power matrix P, k(P k [d]) represents the condition number of the transmit power matrix formed by the d-th resource block of the k-th base station and its adjacent resource blocks, δ k,z [d][l] indicates whether the kth base station allocates the lth resource subblock of the dth resource block to the zth user in the cell where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,z is the lower bound of the SINR of the zth user in the cell where the kth base station is located, is the transmission power of the zth user in the cell where the kth base station is located, is the upper bound of the total power of the kth base station, P k [d] represents the transmission power of the dth resource block of the kth base station, N RR is the number of RBs causing interference, u k Indicates the number of users in the cell where k base stations are located, RB k represents the RB set of the kth base station, and K represents the total number of base stations.

[0106] Specifically, resources are allocated according to the needs and physical characteristics of the multi-cell cellular network. This application couples interference measurement and resource scheduling together, so there are two goals that need to be optimized simultaneously. The first is to maximize energy efficiency (EE), and the second is to minimize the condition number of the transmit power matrix. Energy efficiency is defined as the ratio between communication rate and transmit power in a communication system. The larger the ratio, the more efficient the communication system is in utilizing energy. The condition number is used to indicate whether a matrix has good computational properties. Specifically, the smaller the condition number of the constructed transmit power matrix, the more accurate the channel size calculated using it. In order to achieve these two goals, this application uses the basic requirements for transmit power range, resource block allocation mode, and full rank of the transmit power matrix in an actual multi-cell system as constraints, and constructs an optimization model with maximizing energy efficiency and minimizing the condition number as optimization goals. The optimization model is shown below:

[0107]

[0108] Where E(P) represents the EE of the power matrix P, k(P k [d]) represents the power matrix condition number formed by the dth RB of the kth base station and its adjacent RBs, δ k,z [d][l] indicates whether the k-th base station allocates the d-th RB on its l-th block to the z-th user in the cell where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,z represents the lower bound of the SINR of the zth user in the cell where the kth base station is located, represents the upper bound of the total power of the kth base station, N RB Indicates the number of RBs that cause interference.

[0109] In one embodiment, the first sub-optimization model is:

[0110]

[0111]

[0112] The second sub-optimization model is:

[0113]

[0114] Where α represents the decomposition coefficient.

[0115] Specifically, the optimization model is a model with a typical mixed integer nonlinear problem (MINLP), which has NP-hard properties and the solution process is complicated and cumbersome. Therefore, the present application decomposes, transforms and then solves the optimization model, and the solution process is simple and easy. The solution process is mainly divided into three stages. In the first stage, the original optimization problem is decomposed into two sub-problems, namely maximizing EE (energy efficiency) and minimizing the condition number, that is, the optimization model is decomposed into a first sub-optimization model and a second sub-optimization model. In the second stage, the original problem in the first sub-optimization model is converted into a quadratic cone problem for solution by relaxing the nonlinear SINR constraint. In the third stage, the allocation method of the resource blocks obtained in the second stage is used as input, and the condition number and EE (energy efficiency) are optimized as objective functions. The full rank constraint is relaxed by an iterative method to obtain the final resource allocation result.

[0116] The first sub-optimization model is:

[0117]

[0118] The second sub-optimization model is:

[0119]

[0120] Among them, the first sub-optimization model removes the full rank constraint and only considers how to maximize EE (energy efficiency); the second sub-optimization model obtains the first sub-optimization model for δ k,z After obtaining the optimization results, EE and condition number are used as two variables of the objective function. At this time, the second sub-optimization model is a multi-objective optimization problem.

[0121] In one embodiment, solving the first sub-optimization model includes:

[0122] Based on the fractional form of energy efficiency, the objective function in the first sub-optimization model is transformed to obtain a first intermediate objective function;

[0123] Perform relaxation transformation and convexity-concavity transformation on the first intermediate objective function in sequence to obtain a second intermediate objective function;

[0124] Linearly decompose the signal-to-noise ratio range constraint in the first sub-optimization model to obtain a first signal-to-noise ratio linear constraint and a second signal-to-noise ratio linear constraint, and perform convexity transformation on the second signal-to-noise ratio linear constraint to obtain a third signal-to-noise ratio linear constraint;

[0125] The second intermediate objective function, resource block allocation, transmit power range, first signal-to-noise ratio linear constraint and third signal-to-noise ratio linear constraint are used as a conversion model of the first sub-optimization model, and the conversion model of the first sub-optimization model is solved.

[0126] Specifically, at this stage: solving the first sub-optimization model is divided into three sub-steps:

[0127] Step 1: Since EE (energy efficiency) is a fractional objective function, convert it into a subtraction objective function as the first intermediate objective function:

[0128]

[0129] Where f and g are equal to the numerator and denominator in the original EE, respectively. λ is given at the time of initialization and is continuously updated in the subsequent optimization process. The solution is performed according to the Dinkelbach algorithm. Since the algorithm requires the convexity of f and g and the feasible set, the first intermediate objective function is relaxed.

[0130] Step 2: Continuous relaxation

[0131] Let the quadratic integer variable δ k,z Relax to any integer value between 0 and 1. In order to eliminate the influence of this relaxation, a penalty term for the variable is added to the first intermediate objective function to make the value of the variable approach 0 or 1. The specific form is as follows:

[0132]

[0133] Step 3: Convex approximation: Since f and g contain nonlinear log functions, rewrite f into the following form:

[0134]

[0135] And define the following f1, f2 functions:

[0136]

[0137] After that, the first intermediate objective function can be expressed by the following formula after transformation:

[0138]

[0139] In order to make this objective function concave, a first-order Taylor expansion is performed on f2 to obtain the second intermediate objective function. The first-order Taylor expansion of f2 is:

[0140]

[0141] At this time, there is still a nonlinear constraint C4 in the first sub-optimization model. C4 can be decomposed into C 4.1 and C 4.2

[0142]

[0143] Using successive convex approximation (SCA), C 4.2 Convert to C 4.3

[0144]

[0145] in Combining all the above operations, the first sub-optimization model can be written as the transformation model of the first sub-optimization model:

[0146]

[0147] The transformation model of the first sub-optimization model is a second-order cone programming problem, which can be solved iteratively until convergence.

[0148] In one embodiment, solving the second sub-optimization model based on the solution of the first sub-optimization model includes:

[0149] Decomposing the second sub-optimization model to obtain a first initial optimization model and a second initial optimization model;

[0150] Based on the solution of the first sub-optimization model, converting the first initial optimization model into a first intermediate optimization model;

[0151] Converting the full rank constraint of the transmit power matrix in the first intermediate optimization model into a positive definite constraint to obtain a third sub-optimization model, and solving the third sub-optimization model;

[0152] Based on the solution of the third sub-optimization model, a ratio constraint condition of the maximum eigenvalue and the minimum eigenvalue of the transmit power matrix is ​​added to the second initial optimization model to obtain a second intermediate optimization model;

[0153] The constraint conditions in the second intermediate optimization model are scaled to obtain a fourth sub-optimization model, and the fourth sub-optimization model is solved.

[0154] Specifically, Figure 3 As shown, in this stage, the second sub-optimization model is decomposed to obtain the first initial optimization model and the second initial optimization model to optimize EE (energy efficiency) and condition number respectively. In the previous stage, by solving the conversion model of the first sub-optimization model, δ k,z The value of , at this time, the first initial optimization model is degenerated into a multi-objective optimization problem with full rank constraints, which can be rewritten as the first intermediate optimization model:

[0155]

[0156] The full rank constraint can be approximated as a series of semidefinite programming (SDP) constraints. Assume that Z k [d] is a n b *n b A symmetric matrix, Inb is the identity matrix, Where nb is equal to the number of resource blocks, U k [d] is the constraint matrix of the eigenvalues ​​of the transmit power matrix. The constraint conditions of the eigenvalues ​​of the transmit power matrix are added to the first intermediate optimization model to obtain the third sub-optimization model. It is only necessary to iteratively solve the following third sub-optimization model:

[0157]

[0158] Among them, P (t) is the transmit power matrix at the tth iteration, is the convergence number, Includes the corresponding nb The eigenvector of the smallest eigenvalue. (t) Represents the weight that grows with the number of iterations t.

[0159] After obtaining the solution of the third optimization model, its EE (energy efficiency) solution is used as the lower bound to solve the second initial optimization model. Since the condition number is the ratio of the maximum eigenvalue to the minimum eigenvalue of the power matrix P, minimizing the condition number can be converted into constraining the size of this ratio, and the constraint term C8 is introduced to describe it:

[0160]

[0161] And use C9 to ensure that EE (energy efficiency) is not less than the constraint C 3.1 The returned EE (energy efficiency) size:

[0162] (C9)f1(P)-f2(P)≥λ * q(P).

[0163] A second intermediate optimization model is obtained.

[0164] because is not linear, C8 still needs further transformation. When the optimization process converges, Converges to Z k [d]. Therefore, Z k [d] Divide by μ and substitute into the original C8 constraint to obtain

[0165]

[0166] The remaining constraints should also be scaled in the same way to obtain the fourth optimized sub-model. Finally, only the fourth optimized sub-model needs to be solved iteratively to obtain the solution to the entire problem:

[0167]

[0168] In one embodiment, the equivalent channel vector of the user is calculated using the following method:

[0169]

[0170] in, is the received power of the user corresponding to the transmit power of the i-th resource block, v i is the Gaussian white noise power vector, P is the transmit power matrix, s i is the equivalent channel vector on the i-th resource block.

[0171] Specifically, based on the transmit power matrix, the received power of a user on different frequency resource blocks (resource blocks in the transmit power matrix) is obtained. Within several block times corresponding to the transmit power matrix, the received power of a user on the i-th resource block is calculated. This received power can be expressed as the vector Then, based on the characteristic that the received power in wireless communication is equal to the transmitted power * channel response, the following equation is obtained:

[0172]

[0173] Among them, v i represents the Gaussian white noise power vector, s i represents the equivalent channel vector on the i-th resource block. By sending power P with full rank, it is easy to obtain s i Since the optimal transmit power matrix is ​​obtained under the full rank constraint, the equivalent channel vector on the resource block can be obtained by substituting the optimal transmit power matrix into the above formula.

[0174] Furthermore, as a response to the above Figure 1 The implementation of the method shown in the figure, the embodiment of the present invention provides an interference measurement device based on the power domain, such as Figure 4 As shown, the device comprises:

[0175] The transmission power matrix construction module 402 is used to obtain the power allocated by each base station to multiple resource blocks within a preset period of time, and construct a transmission power matrix based on the power allocated by each base station to multiple resource blocks within the preset period of time;

[0176] The optimization model building module 404 is used to create an objective function based on maximizing energy efficiency and minimizing condition number of the transmit power matrix, and to build an optimization model by taking resource block allocation, transmit power range, signal-to-noise ratio range, and full rank of the transmit power matrix as constraints of the objective function;

[0177] The optimization model solving module 406 is used to decompose the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solve the first sub-optimization model, and solve the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix;

[0178] The equivalent channel vector acquisition module 408 is used to obtain the receiving power corresponding to the power allocated to the user and multiple resource blocks, and calculate the equivalent channel vector of the user based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector.

[0179] The present application provides an interference measurement device based on power domain. Compared with the prior art, by measuring the transmission power of each base station and the receiving power of the user at different frequencies within a period of time, that is, constructing a transmission power matrix, combined with the physical characteristics and requirements of multi-cell cellular network communications, the maximum energy efficiency of the transmission power matrix and the minimum condition number of the matrix are used as objective functions, the allocation of resource blocks, the transmission power range, the signal-to-noise ratio range and the full rank of the transmission power matrix are used as constraints, an optimization model is constructed, the optimization model is solved, and the optimal transmission power matrix is ​​obtained. The interference is obtained by inferring based on the relationship between the transmission power matrix and the receiving power vector, thereby expanding the original time domain / frequency domain method to the new dimension of the power domain, fundamentally solving the problems of high measurement overhead and high requirements for time synchronization accuracy of traditional methods.

[0180] In one embodiment, the transmit power matrix building module is further configured to:

[0181] Obtaining the power allocated by each base station to multiple resource blocks of different frequencies within a preset period of time, and dividing a preset number of adjacent resource elements in the same frequency into a resource sub-block according to the time sequence;

[0182] Obtaining the transmit power of each resource element in each resource sub-block, and calculating the average transmit power corresponding to each resource sub-block based on the transmit power of the resource elements in each resource sub-block;

[0183] A transmit power matrix is ​​generated based on the frequency of each resource block, the time corresponding to the resource sub-blocks included therein, and the average transmit power.

[0184] In one embodiment, the optimization model is:

[0185]

[0186] Where E(P) represents the energy efficiency of the transmit power matrix P, k(P k [d]) represents the condition number of the transmit power matrix formed by the d-th resource block of the k-th base station and its adjacent resource blocks, δ k,z [d][l] indicates whether the kth base station allocates the lth resource subblock of the dth resource block to the zth user in the cell where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,z represents the lower bound of the SINR of the zth user in the cell where the kth base station is located, represents the transmission power of the zth user in the cell where the kth base station is located, represents the upper bound of the total power of the kth base station, P k [d] represents the transmission power of the dth resource block of the kth base station, N RB is the number of RBs causing interference, u k Indicates the number of users in the cell where k base stations are located, RB k represents the RB set of the kth base station, and K represents the total number of base stations.

[0187] In one embodiment, the first sub-optimization model is:

[0188]

[0189] The second sub-optimization model is:

[0190]

[0191] Where α represents the decomposition coefficient.

[0192] In one embodiment, the optimization model solving module is further used to:

[0193] Based on the fractional form of energy efficiency, the objective function in the first sub-optimization model is transformed to obtain a first intermediate objective function;

[0194] Perform relaxation transformation and convexity-concavity transformation on the first intermediate objective function in sequence to obtain a second intermediate objective function;

[0195] Linearly decompose the signal-to-noise ratio range constraint in the first sub-optimization model to obtain a first signal-to-noise ratio linear constraint and a second signal-to-noise ratio linear constraint, and perform convexity transformation on the second signal-to-noise ratio linear constraint to obtain a third signal-to-noise ratio linear constraint;

[0196] The second intermediate objective function, resource block allocation, transmit power range, first signal-to-noise ratio linear constraint and third signal-to-noise ratio linear constraint are used as a conversion model of the first sub-optimization model, and the conversion model of the first sub-optimization model is solved.

[0197] In one embodiment, the optimization model solving module is further used to:

[0198] Decomposing the second sub-optimization model to obtain a first initial optimization model and a second initial optimization model;

[0199] Based on the solution of the first sub-optimization model, converting the first initial optimization model into a first intermediate optimization model;

[0200] Converting the full rank constraint of the transmit power matrix in the first intermediate optimization model into a positive definite constraint to obtain a third sub-optimization model, and solving the third sub-optimization model;

[0201] Based on the solution of the third sub-optimization model, a ratio constraint condition of the maximum eigenvalue and the minimum eigenvalue of the transmit power matrix is ​​added to the second initial optimization model to obtain a second intermediate optimization model;

[0202] The constraint conditions in the second intermediate optimization model are scaled to obtain a fourth sub-optimization model, and the fourth sub-optimization model is solved.

[0203] In one embodiment, the equivalent channel vector of the user is calculated using the following method:

[0204]

[0205] in, is the received power of the user corresponding to the transmit power of the i-th resource block, v i is the Gaussian white noise power vector, P is the transmit power matrix, s i is the equivalent channel vector on the i-th resource block.

[0206] According to an embodiment of the present invention, a storage medium is provided, which stores at least one executable instruction. The computer executable instruction can execute the power domain interference graph estimation method in a cellular network multi-cell scenario in any of the above method embodiments.

[0207] Figure 5 A schematic diagram of the structure of a computer device provided according to an embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the computer device.

[0208] like Figure 5 As shown, the computer device may include: a processor (processor) 502 , a communication interface (Communications Interface) 504 , a memory (memory) 506 , and a communication bus 508 .

[0209] The processor 502 , the communication interface 504 , and the memory 506 communicate with each other via a communication bus 508 .

[0210] The communication interface 504 is used to communicate with other devices such as clients or other servers.

[0211] The processor 502 is used to execute the program 510, and specifically can execute the relevant steps in the embodiment of the power domain interference graph estimation method in the multi-cell scenario of the cellular network.

[0212] Specifically, the program 510 may include program codes, which include computer operation instructions.

[0213] The processor 502 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the computer device may be processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0214] The memory 506 is used to store the program 510. The memory 506 may include a high-speed RAM memory, and may also include a non-volatile memory (non-volatile memory), such as at least one disk memory.

[0215] The program 510 may be specifically configured to enable the processor 502 to perform the following operations:

[0216] Acquire the power situation of each base station allocated to multiple resource blocks within a preset period, and construct a transmission power matrix based on the power situation of each base station allocated to multiple resource blocks within the preset period;

[0217] Based on maximizing energy efficiency and minimizing condition number of transmit power matrix, an objective function is created, resource block allocation, transmit power range, signal-to-noise ratio range and full rank of transmit power matrix are used as constraints of the objective function, and an optimization model is constructed.

[0218] Decomposing the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solving the first sub-optimization model, and solving the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix;

[0219] The receiving power corresponding to the power allocated to the user and multiple resource blocks is obtained, and the equivalent channel vector of the user is calculated based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector.

[0220] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices. In one embodiment, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0221] The above embodiments are only exemplary embodiments of the present application and are not intended to limit the present application. The protection scope of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and protection scope of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the protection scope of the present application.

Claims

1. A method for estimating a power domain interference graph in a multi-cell scenario of a cellular network, characterized in that: include: Acquire the power conditions allocated by each base station to the multiple resource blocks within a preset period, and construct a transmit power matrix based on the power conditions allocated by each base station to the multiple resource blocks within the preset period; Based on the maximization of energy efficiency and the minimization of condition number of the transmit power matrix, an objective function is created, and the allocation of the resource blocks, the transmit power range, the signal-to-noise ratio range, and the full rank of the transmit power matrix are used as constraints of the objective function to construct an optimization model; Decomposing the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solving the first sub-optimization model, and solving the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix; The receiving power corresponding to the power allocated to the user and the multiple resource blocks is obtained, and based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector, an equivalent channel vector of the user is calculated.

2. The power domain interference graph estimation method in a cellular network multi-cell scenario as claimed in claim 1, characterized in that: The constructing a transmission power matrix based on the power allocated by each base station to a plurality of resource blocks within a preset time period includes: Obtaining the power allocated by each base station to multiple resource blocks of different frequencies within a preset period of time, and dividing a preset number of adjacent resource elements in the same frequency into a resource sub-block according to the time sequence; Obtaining the transmit power of each resource element in each resource sub-block, and calculating the average transmit power corresponding to each resource sub-block based on the transmit power of the resource elements in each resource sub-block; A transmit power matrix is ​​generated based on the frequency of each resource block, the time corresponding to the resource sub-blocks included therein, and the average transmit power.

3. The power domain interference graph estimation method in a cellular network multi-cell scenario as claimed in claim 1, characterized in that: The optimization model is: Where E(P) represents the energy efficiency of the transmit power matrix P, k(P k [d]) represents the condition number of the transmit power matrix formed by the d-th resource block of the k-th base station and its adjacent resource blocks, δ k,z [d][l] indicates whether the kth base station allocates the lth resource subblock of the dth resource block to the zth user in cell 5 where the base station is located. SINR indicates the signal-to-noise ratio of the user, γ k,z represents the lower bound of the SINR of the zth user in the cell where the kth base station is located, represents the transmission power of the zth user in the cell where the kth base station is located, represents the upper bound of the total power of the kth base station, P k [d] represents the transmission power of the dth resource block of the kth base station, N RB Indicates the number of RBs causing interference, u k Indicates the number of users in the cell where k base stations are located, RB k represents the RB set of the kth base station, and K represents the total number of base stations.

4. The method for estimating a power domain interference graph in a cellular network multi-cell scenario according to claim 3, characterized in that: The first sub-optimization model is: The second sub-optimization model is: Where α represents the decomposition coefficient.

5. The power domain interference graph estimation method in a cellular network multi-cell scenario as claimed in claim 4, characterized in that: Solving the first sub-optimization model includes: Based on the fractional form of energy efficiency, converting the objective function in the first sub-optimization model into a first intermediate objective function; Performing relaxation conversion and convexity conversion on the first intermediate objective function in sequence to obtain a second intermediate objective function; Performing linear decomposition on the signal-to-noise ratio range constraint in the first sub-optimization model to obtain a first signal-to-noise ratio linear constraint and a second signal-to-noise ratio linear constraint, and performing convexity transformation on the second signal-to-noise ratio linear constraint to obtain a third signal-to-noise ratio linear constraint; The second intermediate objective function, the allocation of the resource blocks, the transmission power range, the first signal-to-noise ratio linear constraint and the third signal-to-noise ratio linear constraint are used as a conversion model of the first sub-optimization model, and the conversion model of the first sub-optimization model is solved.

6. The power domain interference graph estimation method in a cellular network multi-cell scenario as claimed in claim 4, characterized in that: The step of solving the second sub-optimization model based on the solution of the first sub-optimization model comprises: Decomposing the second sub-optimization model to obtain a first initial optimization model and a second initial optimization model; Based on the solution of the first sub-optimization model, converting the first initial optimization model into a first intermediate optimization model; Converting the full rank constraint of the transmit power matrix in the first intermediate optimization model into a positive definite constraint to obtain a third sub-optimization model, and solving the third sub-optimization model; Based on the solution of the third sub-optimization model, adding a ratio constraint condition of the maximum eigenvalue and the minimum eigenvalue of the transmit power matrix to the second initial optimization model to obtain a second intermediate optimization model; The constraint conditions in the second intermediate optimization model are scaled to obtain a fourth sub-optimization model, and the fourth sub-optimization model is solved.

7. The method for estimating a power domain interference graph in a cellular network multi-cell scenario according to any one of claims 1 to 5, characterized in that: The equivalent channel vector of the user is calculated using the following method: in, is the received power of the user corresponding to the transmit power of the i-th resource block, v i is the Gaussian white noise power vector, P is the transmit power matrix, s i is the equivalent channel vector on the i-th resource block.

8. An interference measurement device based on power domain, characterized in that: include: A transmission power matrix construction module, used to obtain the power conditions allocated by each base station to multiple resource blocks within a preset period of time, and to construct a transmission power matrix based on the power conditions allocated by each base station to multiple resource blocks within the preset period of time; An optimization model building module is used to create an objective function based on maximizing energy efficiency and minimizing condition number of the transmit power matrix, and to construct an optimization model by taking the allocation of the resource blocks, the transmit power range, the signal-to-noise ratio range, and the full rank of the transmit power matrix as constraints of the objective function; an optimization model solving module, configured to decompose the optimization model to obtain a first sub-optimization model and a second sub-optimization model, solve the first sub-optimization model, and solve the second sub-optimization model based on the solution of the first sub-optimization model to obtain an optimal transmission power matrix; The equivalent channel vector acquisition module is used to obtain the receiving power corresponding to the power allocated to the user and multiple resource blocks, and calculate the equivalent channel vector of the user based on the optimal transmission power matrix, the receiving power and the Gaussian white noise power vector.

9. A storage medium storing at least one executable instruction, wherein the executable instruction enables a processor to perform operations corresponding to the power domain interference graph estimation method in a cellular network multi-cell scenario as described in any one of claims 1-7.

10. A computer device comprising: A processor, a memory, a communication interface and a communication bus, wherein the processor, the memory and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and the executable instruction enables the processor to perform operations corresponding to the power domain interference graph estimation method in a cellular network multi-cell scenario as described in any one of claims 1-7.

Images