A multi-cell massive MIMO downlink coordinated beamforming method and system

Abstract

The present invention relates to the field of wireless communication technology, and is a method and system for cooperative beamforming of a multi-cell large-scale MIMO downlink. The method includes: establishing a multi-cell MIMO communication system; constructing an uplink and downlink transmission link model of the communication system; constructing a cooperative beamforming downlink optimization model, and establishing an energy efficiency maximization model under the conditions of transmission power constraints and user service quality constraints; introducing auxiliary variables through fractional programming, and converting the fractional problem of energy efficiency maximization into a coupled non-convex optimization problem without fractions; converting and solving the coupled non-convex optimization problem through convex optimization approximation, and finally obtaining a beam direction vector. The present invention improves the energy efficiency of a multi-cell communication system under the introduction of quantization errors, while ensuring the user's service quality and improving the energy efficiency of a cellular mobile communication system.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technology, and in particular to a multi-cell large-scale MIMO downlink coordinated beamforming method and system. Background Art

[0002] In today's 5G mobile communication technology, cellular mobile communication systems generally use MIMO technology to improve network capacity and data transmission rate. Compared with MIMO, massive MIMO can serve a large number of users at the same time by using large-scale array antennas. Additional antennas also help to concentrate the energy of the transmitting antenna into a smaller spatial area, while improving diversity gain, and the throughput and radiation energy efficiency are further improved.

[0003] However, massive MIMO technology uses a large number of antennas for communication systems, and the energy consumption of the system will increase. In order to reduce the energy consumption of the communication system, the base station antenna end can use low-precision ADC / DAC modules with lower cost and lower power consumption. However, the use of low-precision ADC / DAC modules will lead to quantization effects and quantization errors. At the same time, in order to reduce interference between cells and improve the service quality of users at the edge of the cell, multi-cell mobile communication systems can use cooperative beamforming technology. However, the traditional cooperative beamforming algorithm does not take quantization errors into account, and the algorithm effect will be worse than expected. Summary of the invention

[0004] In order to solve the problems existing in the prior art, the present invention provides a multi-cell large-scale MIMO downlink collaborative beamforming method and system, which can improve the energy efficiency of the cellular mobile communication system while ensuring the minimum service quality of users and the power constraints of the base station, taking into account the quantization error introduced by the low-precision ADC / DAC module used in the base station.

[0005] In a first aspect, an embodiment of the present invention provides a multi-cell massive MIMO downlink coordinated beamforming method, comprising the following steps:

[0006] Establishing a multi-cell MIMO communication system, wherein users of the established communication system include users served by each cell;

[0007] Establishing an uplink model and a downlink model of a communication system; the received signal in the uplink model includes the transmitted signal of the communication system user, the introduced quantization error, noise, and the downlink signal; the received signal in the downlink model includes the transmitted signal of all base stations, the introduced quantization error, and noise;

[0008] Construct a coordinated beamforming downlink optimization model and establish an energy efficiency maximization model under the conditions of transmit power constraint and user service quality constraint; the transmit power constraint means that the total transmit power of the base station shall not exceed a given power threshold; the user service quality constraint means that the user's transmission rate shall not be lower than a given transmission threshold;

[0009] Transform the energy efficiency maximization model, adopt the fractional programming method, transform the fractional problem of the energy efficiency maximization model into an outer iterative coupled non-convex optimization problem without fractions, and introduce auxiliary variables for iterative solution;

[0010] Transform the coupled non-convex optimization problem and establish the inner layer iteration; transform the problem into the geometric programming problem and the second-order cone programming problem of the inner layer iteration through the convex optimization approximation of the inner layer iteration; solve the geometric programming problem using the convex optimization toolkit to obtain the cooperative beamforming power; solve the second-order cone programming problem using the Lagrangian strong duality relationship of the uplink and downlink to obtain the beamforming direction vector;

[0011] According to the convex optimization approximation results, the outer iterative auxiliary variables are updated, and the geometric programming problem and the second-order cone programming problem are repeatedly solved until the change of the auxiliary variables is less than a given threshold;

[0012] A coordinated beamforming direction vector optimization model is established to optimize the coordinated beamforming direction vector in the inner iteration.

[0013] In a second aspect, a multi-cell massive MIMO downlink coordinated beamforming system provided in an embodiment of the present invention includes the following modules;

[0014] A multi-cell communication system establishment module is used to divide the cooperation unit clusters between the cells, divide all the cells into various cooperation clusters, and the cells in the cooperation clusters adopt the coordinated beamforming method; and is used for the cells in the cooperation clusters to transmit channel state parameters;

[0015] A collaborative beamforming optimization model building module is used to build a collaborative beamforming optimization model and establish an energy efficiency maximization model under the conditions of transmission power constraints and user service quality constraints; the user service quality constraint indicates that the user's data transmission rate is not lower than a given transmission rate threshold;

[0016] The outer iteration module is used to update the auxiliary variable θ. When the change of θ is less than a given threshold, the iteration is stopped to obtain the final cooperative beamforming vector.

[0017] The inner iterative convex approximation module is used to transform the coupled non-convex optimization problem into a decoupled convex optimization problem. It is also used to fix the cooperative beamforming direction vector, and adopt an iterative method to solve the power allocation and the signal-to-interference-to-noise ratio of each user through the convex optimization toolkit. The iteration is stopped when the change of the signal-to-interference-to-noise ratio of each user is less than a given threshold.

[0018] The inner iterative cooperative beamforming direction vector module is used to further optimize the cooperative beamforming direction vector under the signal-to-interference-noise ratio constraint of each user obtained in the convex approximation module to minimize the transmission power.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The multi-cell massive MIMO downlink collaborative beamforming method and system of the present invention considers collaborative beamforming that maximizes energy efficiency under quantization errors when base station antennas use low-precision ADC / DAC modules in a multi-cell communication system. Existing collaborative beamforming algorithms and systems do not take the quantization errors of the communication system into account. Therefore, when the quantization errors cannot be ignored, the service quality of users in the cell will be reduced and the expected effect cannot be achieved. Compared with existing collaborative beamforming algorithms and systems, the present invention improves the energy efficiency of the communication system while ensuring the user communication quality when the quantization errors cannot be ignored. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A schematic diagram of a process of a multi-cell massive MIMO downlink coordinated beamforming method provided in an embodiment of the present invention;

[0022] Figure 2 A schematic diagram of the structure of a multi-cell low-precision ADC / DAC large-scale MIMO communication system established in an embodiment of the present invention;

[0023] Figure 3 A schematic diagram of the relationship between the number of antennas and energy efficiency after adopting the downlink coordinated beamforming method provided by an embodiment of the present invention;

[0024] Figure 4 A schematic diagram showing how the total spectrum efficiency varies with the number of antennas after the downlink coordinated beamforming method provided by an embodiment of the present invention is adopted;

[0025] Figure 5 A schematic diagram showing how energy efficiency varies with resolution bits after adopting the downlink coordinated beamforming method provided by an embodiment of the present invention;

[0026] Figure 6 It is a structural block diagram of a multi-cell massive MIMO downlink coordinated beamforming system provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0027] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0028] Example 1

[0029] like Figure 1 As shown, this embodiment provides a multi-cell massive MIMO downlink coordinated beamforming method, which maximizes the energy efficiency of multi-cell downlink coordinated beamforming under the introduction of quantization error. The method specifically includes the following steps:

[0030] S1. Establish a multi-cell massive MIMO communication system; Figure 2 As shown, the cell base station antenna adopts a low-precision ADC / DAC module, and the users of the established multi-cell communication system include the users served by each cell.

[0031] In this embodiment, the number of cells is N. c , the number of users in each cell is N u , the number of cell base station antennas is N t , the number of user antennas in a cell is 1, and each cell shares the channel state parameters.

[0032] In the multi-cell massive MIMO communication system of this step, the channel vector between the transmitting antenna of cell i and the receiving antenna of user u in cell j is denoted as h i,j,u , the channel vector matrix of cell i is recorded as The beamforming vector corresponding to user u in cell i is denoted as w i,u .

[0033] S2. Establish an uplink model and a downlink model of the communication system. The received signal in the uplink model includes the transmitted signal of the communication system user, the introduced quantization error, noise, and the downlink signal. The received signal in the downlink model includes the transmitted signals of all base stations, the introduced quantization error, and noise. When the quantization error is introduced, the uplink and downlink are in a Langrange strong dual relationship.

[0034] For user u in cell i in the downlink model, the received signal y i,u It is expressed as:

[0035]

[0036] The number of cells is Nc , the number of users in each cell is N u , The received symbol for user l in cell i is w m,l is the coordinated beamforming vector corresponding to user l in cell m, h m,i,u is the channel vector from cell m to user u in cell i, Indicates h m,i,u The conjugate transpose of is the quantization noise generated by the downlink of cell j, is the total quantization noise received by user u in cell i, is the channel noise of user u in cell i,

[0037] For the uplink model, the received signal r of the base station in cell i is i It is expressed as:

[0038]

[0039] Among them, λ j,u is the transmission power of user u in cell j, is the quantization noise received by the uplink of the base station in cell i, in

[0040] For user u in cell i in the downlink model, the spectrum efficiency is It is expressed as:

[0041]

[0042] in, is the power of the quantization noise generated by the ADC / DAC module of cell j when it reaches user u in cell i, N t is the number of base station antennas in each cell, W j is the coordinated beamforming matrix of cell j, α is the quantization coefficient; w i,u is the coordinated beamforming vector corresponding to user u in cell i, w j,k is the coordinated beamforming vector corresponding to user k in cell j; h i,i,u is the channel vector from the base station in cell i to user u in cell i, Indicates h i,i,u The conjugate transpose of .

[0043] S3. Construct a collaborative beamforming downlink optimization model and establish an energy efficiency maximization model under the constraints of transmit power and user service quality.

[0044] The transmit power constraint means that the total transmit power of the base station shall not exceed the given power threshold; the user service quality constraint means that the user's transmission rate shall not be lower than the given transmission threshold. When the user service quality constraint and the transmit power constraint cannot be met at the same time, it is considered as a transmission interruption.

[0045] In the coordinated beamforming downlink optimization model constructed in this step, the energy efficiency maximization model uses the ratio function of total spectrum efficiency to total power consumption as the objective function, the coordinated beamforming weights as the optimization variables, and the transmit power inequality and signal-to-interference-to-noise ratio inequality as the constraints for optimization. Among them, the total spectrum efficiency is the sum of the spectrum efficiencies of all users in the cell; and the total power consumption is the sum of the power consumption of the cell base station, including the fixed consumption of the base station, the fixed consumption of the antenna, and the coordinated beamforming consumption.

[0046] The energy efficiency maximization model can be specifically expressed as the following problem

[0047]

[0048]

[0049]

[0050] The meanings of the parameters in formulas (1a) to (1b) are as follows:

[0051] Formula (1a) represents the total power constraint consumed by the beamformer, P is the upper limit of the total power resource, w is the beamformer of the base station user, is the spectrum efficiency of user u in cell i, ξ is the power amplifier efficiency, ξ≥1, P c is the fixed power consumption of the antenna, P 0 The fixed power consumption of the base station; N t is the number of base station antennas, and the number of cells is N c , the number of users in each cell is N u .

[0052] Formula (1b) indicates that the signal-to-interference-to-noise ratio (SINR) at the user receiving end must not be less than a given threshold. i,u is the signal-to-interference-and-noise ratio of user u in cell i, ensuring the communication quality of cell users.

[0053] S4. Transform the energy efficiency maximization model, that is, transform the collaborative beamforming optimization model. Use the fractional programming method to transform the fractional problem of the energy efficiency maximization model into an outer iterative coupled non-convex optimization problem without fractions, and introduce auxiliary variables for iterative solution.

[0054] In this step, when transforming the cooperative beamforming optimization model, fractional programming is used and an auxiliary variable θ is introduced for iterative solution.

[0055] Specifically, the transformation of the coordinated beamforming optimization model in this step is as follows:

[0056] S41. Define auxiliary variables

[0057] S42. Using fractional programming and introducing auxiliary variables θ, the energy efficiency maximization model is transformed into the following problem

[0058]

[0059] st formula (1a)~(1b)

[0060] S43. Solve the above problems Iteratively update auxiliary variables θ max ,θ min ,θ 0 .

[0061] The specific operations are as follows:

[0062] when When θ min =θ 0 , when When θ max =θ 0 , Until θ max With θ min The difference is less than the preset threshold. Among them, λ is updated with the outer layer iteration.

[0063] S5. Transform the coupled non-convex optimization problem and establish an inner iteration. Through the convex optimization approximation of the inner iteration, transform the problem into a geometric programming problem and a second-order cone programming problem of the inner iteration. The geometric programming problem is solved using a convex optimization toolkit to obtain the collaborative beamforming power. The second-order cone programming problem is solved using the Lagrangian strong duality relationship of the uplink and downlink to obtain the beamforming direction vector. According to the convex optimization approximation results, update the outer iteration auxiliary variables, and repeatedly solve the geometric programming problem and the second-order cone programming problem until the auxiliary variable changes are less than a given threshold.

[0064] Through the problem Convex optimization approximation is performed to decouple the coupled non-convex optimization problem into a non-convex optimization problem. For this optimization problem, the coordinated beamforming weights can be optimized in two steps. The first step is to fix the coordinated beamforming direction vector and optimize the power allocation and the signal-to-interference-to-noise ratio of each user. The second step is to optimize the coordinated beamforming direction vector under the signal-to-interference-to-noise ratio constraint of each user after optimization in the first step.

[0065] In this step, the transformation steps of the inner iterative convex optimization approximation are as follows:

[0066] The inequality of the convex approximation transformation adopted is The parameters x 0 Indicates approximate point;

[0067] Transform the coupled non-convex optimization problem into a problem

[0068]

[0069] st formula (1a)~(1b)

[0070] use p j,v is the cooperative beamforming power p j,v =||w j,v || 2 , is the cooperative beamforming direction vector,

[0071] Fixed the cooperative beamforming direction vector, the above problem Transform into a problem

[0072]

[0073] st formula (1a)~(1b)

[0074] The above issues This is a geometric programming problem that can be solved using the CVX optimization toolkit to obtain the power allocation and the signal-to-interference-noise ratio of each user.

[0075] When the change of each user's signal to interference and noise ratio is greater than a given threshold, the user's signal to interference and noise ratio at this time is used as an approximate point to update the parameter Continue to solve the above problems Until the change of each user's signal to interference and noise ratio is less than a given threshold.

[0076] S6. Establish a coordinated beamforming direction vector optimization model to optimize the coordinated beamforming direction vector in the inner iteration. Take the signal to interference plus noise ratio of each user obtained in step S5 as a constraint to minimize the total transmission power.

[0077] The coordinated beamforming direction vector optimization model is as follows:

[0078]

[0079] st formula (1a)

[0080]

[0081] Among them, η i,u is the signal to interference and noise ratio of each user finally obtained in step S5.

[0082] Regarding the above issues It can be solved through the strong duality between uplink and downlink. Assuming that the uplink adopts the MMSE criterion of maximizing the signal-to-interference-noise ratio, the filter vector f of user u in cell i can be obtained. i,u :

[0083]

[0084] Due to the strong duality between uplink and downlink, the coordinated beamforming direction vector is linearly related to the filter vector and can be expressed as The coordinated beamforming direction vector is further obtained by solving the problem.

[0085] definition in

[0086]

[0087] Among them, when i=j,u=v:

[0088]

[0089] otherwise:

[0090]

[0091] Where η j,u is the signal to interference and noise ratio of user u in cell i finally obtained in step S5.

[0092] For λ in the uplink i,u , can be expressed as:

[0093]

[0094] The iteration converges to get;

[0095] In this embodiment, mathematical modeling simulation is performed on the system. The simulation adopts 2.4 GHz carrier frequency, 8.7 dB logarithmic shadow fading variance and 5 dB noise factor, 10 MHz bandwidth, and 4 cells and 4 users. Figure 3-Figure 5 shown.

[0096] from Figure 3 It can be seen from the comparison that when the resolution bit is 4 and the number of antennas is 16, the energy efficiency is the highest. When the number of antennas is too high or too low, the energy efficiency will decrease.

[0097] from Figure 4 It can be seen from the comparison that as the number of antennas increases, the total spectrum efficiency also increases. Figure 3 It can be seen that when the number of antennas is too low, the total spectrum efficiency is low, resulting in a decrease in energy efficiency; when the number of antennas is too high, the total spectrum efficiency increases, but the power consumption caused by the increase in the number of antennas also increases, and the latter has a greater impact on energy efficiency, ultimately leading to a decrease in energy efficiency. In order to maintain higher energy efficiency, the cell base station can use an appropriate number of antennas.

[0098] from Figure 5 It can be seen from the comparison that, without considering the impact of ADC / DAC module accuracy on power consumption, the higher the accuracy, the weaker the quantization effect, the lower the power consumed to achieve the same spectrum efficiency, and the higher the energy efficiency. At the same time, when the resolution bit is low, the energy efficiency that can be achieved is also ideal enough, indicating that the base station uses a low-precision ADC / DAC module, and the energy efficiency of the communication system is also high enough.

[0099] Example 2

[0100] like Figure 6 As shown, this embodiment and Embodiment 1 are based on the same inventive concept, and provide a cell massive MIMO downlink coordinated beamforming system, including:

[0101] A multi-cell communication system establishment module is used to divide the cooperation unit clusters between the cells, divide all the cells into various cooperation clusters, the cells in the cooperation clusters adopt the coordinated beamforming method, and each cell base station adopts a low-precision ADC / DAC module; and is used for the cells in the cooperation cluster to transmit channel state parameters;

[0102] A collaborative beamforming optimization model building module is used to build a collaborative beamforming optimization model and establish an energy efficiency maximization model under the conditions of transmission power constraints and user service quality constraints; the user service quality constraint indicates that the user's data transmission rate is not lower than a given transmission rate threshold;

[0103] The outer iteration module is used to update the auxiliary variable θ according to the cooperative beamforming weight vector obtained after the inner iteration is completed; when the change of θ is less than a given threshold, the iteration is stopped to obtain the final cooperative beamforming vector;

[0104] The inner iterative convex approximation module is used to transform the coupled non-convex optimization problem containing the auxiliary variable θ of the outer iterative layer into a decoupled convex optimization problem; and is used to fix the cooperative beamforming direction vector, adopt an iterative method, and use the CVX convex optimization toolkit to solve the power allocation and the signal-to-interference-to-noise ratio of each user. The iteration is stopped when the change of the signal-to-interference-to-noise ratio of each user is less than a given threshold;

[0105] The inner iterative cooperative beamforming direction vector module is used to further optimize the cooperative beamforming direction vector under the signal-to-interference-noise ratio constraint of each user obtained in the convex approximation module to minimize the transmission power.

[0106] Among them, in the collaborative beamforming optimization model construction module, the energy efficiency maximization model takes the ratio function of total spectral efficiency to total power consumption as the objective function, the collaborative beamforming weights as the optimization variables, and the transmit power inequality and the user receive signal-to-interference-and-noise ratio inequality as the constraints for optimization; the user service quality constraint indicates that the user's transmission rate must not be lower than a given transmission threshold.

[0107] The modules of this embodiment are respectively used to implement the steps of Embodiment 1. Please refer to Embodiment 1 for the detailed implementation process, which will not be described in detail.

[0108] It can be seen from the detailed technical solutions described in the above embodiments that the technical solutions of the present invention have the following beneficial effects:

[0109] The multi-cell massive MIMO downlink cooperative beamforming scheme of the present invention can obtain a cooperative beamforming weight vector that maximizes energy efficiency according to the channel state under the constraints of transmission power and user service quality. The existing co-beamforming scheme does not take into account the quantization effect produced by the low-precision ADC / DAC module used in the cell base station, and cannot achieve the expected effect in this scenario. The scheme of the present invention takes into account the quantization effect and has a good effect in this scenario.

[0110] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A multi-cell massive MIMO downlink coordinated beamforming method, It is characterized in that The following steps are involved: Establishing a multi-cell MIMO communication system, wherein users of the established communication system include users served by each cell; Establishing an uplink model and a downlink model of a communication system; the received signal in the uplink model includes the transmitted signal of the communication system user, the introduced quantization error, noise, and the downlink signal; the received signal in the downlink model includes the transmitted signals of all base stations, the introduced quantization error, and noise; Construct a coordinated beamforming downlink optimization model and establish an energy efficiency maximization model under the conditions of transmit power constraint and user service quality constraint; the transmit power constraint means that the total transmit power of the base station shall not exceed a given power threshold; the user service quality constraint means that the user's transmission rate shall not be lower than a given transmission threshold; Transform the energy efficiency maximization model, adopt the fractional programming method, transform the fractional problem of the energy efficiency maximization model into an outer iterative coupled non-convex optimization problem without fractions, and introduce auxiliary variables for iterative solution; Transform the coupled non-convex optimization problem and establish the inner layer iteration; transform the problem into the geometric programming problem and the second-order cone programming problem of the inner layer iteration through the convex optimization approximation of the inner layer iteration; solve the geometric programming problem using the convex optimization toolkit to obtain the cooperative beamforming power; solve the second-order cone programming problem using the Lagrangian strong duality relationship of the uplink and downlink to obtain the beamforming direction vector; According to the convex optimization approximation results, the outer iterative auxiliary variables are updated, and the geometric programming problem and the second-order cone programming problem are repeatedly solved until the change of the auxiliary variables is less than a given threshold; Establish a coordinated beamforming direction vector optimization model to optimize the coordinated beamforming direction vector in the inner iteration; For user u in cell i in the downlink model, the received signal y i,u It is expressed as: The number of cells is N c , the number of users in each cell is N u , The received symbol for user l in cell m, w m,l is the coordinated beamforming vector corresponding to user l in cell m, h m,i,u is the channel vector from cell m to user u in cell i, Indicates h m,i,u The conjugate transpose of is the quantization noise generated by the downlink of cell j is the total quantization noise received by user u in cell i, is the channel noise of user u in cell i; α is the quantization coefficient; For the uplink model, the received signal r of the base station in cell i is i It is expressed as: Among them, λ j,u is the transmission power of user u in cell j, is the quantization noise received by the uplink of the base station in cell i, and α is the quantization coefficient; is the channel state matrix from cell i to cell j; Send a symbol for user u in cell j; is the channel noise of uplink cell i; For user u in cell i in the downlink model, the spectrum efficiency is It is expressed as: in, is the power of the quantization noise generated by the ADC / DAC module of cell j when it reaches user u in cell i, N t is the number of base station antennas in each cell, W j is the coordinated beamforming matrix of cell j, W j The conjugate transpose of i,u is the coordinated beamforming vector corresponding to user u in cell i, w j,k is the coordinated beamforming vector corresponding to user k in cell j; h i,i,u is the channel vector from the base station in cell i to user u in cell i, Indicates h i,i,u The conjugate transpose of ; In the constructed cooperative beamforming optimization model, the energy efficiency maximization model takes the ratio of total spectrum efficiency to total power consumption as the objective function, the cooperative beamforming weight as the optimization variable, and the transmit power inequality and signal-to-interference-noise ratio inequality as the constraints for optimization; the total spectrum efficiency is the sum of the spectrum efficiencies of all users in the cell, and the total power consumption is the sum of the power consumption of the cell base stations; The steps to transform the energy efficiency maximization model include: Define the energy efficiency maximization optimization function as the problem Formula (1a) represents the total power constraint consumed by the beamformer, P is the upper limit of the total power resource, w is the beamformer of the base station user, is the spectrum efficiency of user u in cell i, ξ is the power amplifier efficiency, ξ≥1, P c is the fixed power consumption of the antenna, P 0 The fixed power consumption of the base station; N t is the number of base station antennas, and the number of cells is N c , the number of users in each cell is N u ; γ i,u is the signal-to-interference-and-noise ratio constraint of user u in cell i; Formula (1b) indicates that the signal-to-interference-to-noise ratio (SINR) at the user receiving end must not be less than a given threshold. i,u is the signal-to-interference-and-noise ratio of user u in cell i; By introducing auxiliary variables θ through fractional programming, the fractional problem of the energy efficiency maximization model is transformed into an outer iterative coupled non-convex optimization problem without fractions, that is, the problem st formula (1a)~(1b) Among them, the auxiliary variable Using convex approximation transformation, the coupled non-convex optimization problem is transformed into a non-coupled convex optimization problem; Using convex approximation inequality in, x 0 Indicates approximate point; Transform the coupled non-convex optimization problem into an uncoupled convex optimization problem, namely problem P3: st formula (1a)~(1b) δ i,u represents the approximate constant of user u in cell i after using the convex approximate inequality; use p j,v is the cooperative beamforming power p j,v =||w j,v || 2 , w j,v is the coordinated beamforming vector corresponding to user v in cell j, is the cooperative beamforming direction vector, α is the quantization coefficient; Fixed cooperative beamforming direction vector, the above problem It is a geometric programming problem, which is solved by using the CVX toolkit to obtain the power allocation p and the signal-to-interference-noise ratio of each user; When optimizing the coordinated beamforming direction vector, the signal-to-interference-noise ratio of each user is used as a constraint to minimize the total transmit power; The coordinated beamforming direction vector optimization model is as follows: st formula (1a) Among them, η i,u is the signal-to-interference-and-noise ratio of each user finally obtained; Regarding the problem The strong duality between the uplink and the downlink is used to solve the problem. The MMSE criterion is used for filtering in the uplink. The filter vector of user u in cell i can be obtained based on the linear relationship between the coordinated beamforming vector in the downlink and the filter vector in the uplink. The coordinated beamforming direction vector is then solved based on the linear relationship between the coordinated beamforming direction vector and the filter vector.

2. A multi-cell massive MIMO downlink cooperative beamforming system, It is characterized in that Includes the following modules; A multi-cell communication system establishment module is used to divide the cooperation unit clusters between the cells, divide all the cells into various cooperation clusters, and the cells in the cooperation clusters adopt the coordinated beamforming method; and is used for the cells in the cooperation clusters to transmit channel state parameters; A collaborative beamforming optimization model building module is used to build a collaborative beamforming optimization model and establish an energy efficiency maximization model under the conditions of transmission power constraints and user service quality constraints; the user service quality constraint indicates that the user's data transmission rate is not lower than a given transmission rate threshold; The outer iteration module is used to update the auxiliary variable θ. When the change of θ is less than a given threshold, the iteration is stopped to obtain the final cooperative beamforming vector. The inner iterative convex approximation module is used to transform the coupled non-convex optimization problem into a decoupled convex optimization problem. It is also used to fix the cooperative beamforming direction vector, and adopt an iterative method to solve the power allocation and the signal-to-interference-to-noise ratio of each user through the convex optimization toolkit. The iteration is stopped when the change of the signal-to-interference-to-noise ratio of each user is less than a given threshold. The inner iterative cooperative beamforming direction vector module is used to further optimize the cooperative beamforming direction vector under the signal-to-interference-noise ratio constraint of each user obtained in the convex approximation module to minimize the transmission power; In the cooperative beamforming optimization model construction module, the energy efficiency maximization model uses the ratio function of total spectrum efficiency to total power consumption as the objective function, the cooperative beamforming weight as the optimization variable, and the transmit power inequality and the user receive signal-to-interference-to-noise ratio inequality as the constraint conditions for optimization; the user service quality constraint indicates that the user's transmission rate must not be lower than a given transmission threshold; The steps for transforming the energy efficiency maximization objective function are as follows: The energy efficiency maximization optimization function is: Where w is the beamformer of the base station user, is the spectrum efficiency of user u in cell i, ξ is the power amplifier efficiency, ξ≥1, w i,u is the coordinated beamforming vector corresponding to user u in cell i, P c is the fixed power consumption of the antenna, P 0 The fixed power consumption of the base station; N t is the number of base station antennas, and the number of cells is N c , the number of users in each cell is N u ; Through fractional programming, the above equation is transformed into a coupled non-convex optimization problem, namely: Among them, the auxiliary variable w i,j is the coordinated beamforming vector corresponding to user j in cell i; The inner iterative convex approximation module transforms the coupled non-convex optimization problem into a decoupled convex optimization problem. The cooperative beamforming direction vector is fixed, and the power allocation and the signal-to-interference-to-noise ratio of each user are solved by an iterative method through the convex problem optimization toolkit. The iteration is stopped when the change of the signal-to-interference-to-noise ratio of each user is less than a given threshold. The inner iterative convex approximation transformation steps are as follows: x 0 Indicates approximate point; Transform the coupled non-convex optimization problem into an uncoupled convex optimization problem, that is: δ i,u represents the approximate constant of user u in cell i after using the convex approximate inequality; w j,k is the coordinated beamforming vector corresponding to user k in cell j; h i,i,u is the channel vector from the base station in cell i to user u in cell i, Indicates h i,i,u The conjugate transpose of j,i,u is the channel vector from the base station in cell j to user u in cell i, Indicates h j,i,u The conjugate transpose of ; is the power of the quantization noise generated by the ADC / DAC module of cell j when it reaches user u in cell i, N t is the number of base station antennas in each cell, W j is the coordinated beamforming matrix of cell j, W j The conjugate transpose of ; use p j,v is the cooperative beamforming power p j,v =||w j,v || 2 , w j,v is the coordinated beamforming vector corresponding to user v in cell j, is the cooperative beamforming direction vector, α is the quantization coefficient; Fixed the cooperative beamforming direction vector, the above problem is a geometric planning problem; The convex optimization toolkit is used to solve the problem and obtain the power allocation p and the signal-to-interference-noise ratio of each user.

Images