CN108834155B - Method for optimizing spectrum efficiency based on multiple parameters of large-scale antenna system - Google Patents

Abstract

The present invention provides a method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system, which can improve the spectrum efficiency of the system. The method includes: establishing a multi-parameter optimization spectrum efficiency model, wherein the multi-parameters include: antenna selection matrix, beamforming vector, transmit power and pilot sequence length; Selection matrix; optimize the beamforming vector and transmit power according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix; optimize the established multi-parameter spectral efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power power, and optimize the length of the pilot sequence; bring the optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency. The present invention is suitable for spectral efficiency optimization operations.

Description

A method for optimizing spectral efficiency based on multi-parameter of large-scale antenna system

technical field

The present invention relates to the technical field of wireless communication, in particular to a method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system.

Background technique

In recent years, with the rapid development of global wireless communication technologies, the demand for wireless communication data services has grown exponentially. In addition, the current wireless communication system adopts a static (fixed) spectrum allocation strategy, and spectrum resources are increasingly scarce. Massive Multiple-Input Multiple-Output (Massive MIMO) technology does not require large-scale updating of user's terminal equipment, and only needs to transform the base station, which can improve the system's spectrum utilization and system capacity, and solve the problem of spectrum shortage. .

In large-scale antenna systems, multiple solutions to pilot pollution, effective channel estimation, power consumption, and hardware equipment costs need to be studied urgently. Currently commonly used methods are:

Based on the antenna selection technology, the problem of hardware equipment cost can be effectively solved; based on the beamforming technology, a large array gain can be obtained, the channel capacity can be significantly improved, and the problem of power consumption can be solved; based on the channel estimation and pilot scheduling, the pilot frequency can be reduced or eliminated. At present, these methods are effective in solving their respective problems. However, these methods cannot comprehensively solve many problems such as pilot pollution, power consumption, high system complexity and cost in large-scale antenna systems.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is to provide a method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system, so as to solve the problems of pilot frequency pollution, high power consumption, high system complexity and high cost existing in the prior art .

To solve the above technical problems, an embodiment of the present invention provides a method for optimizing spectral efficiency based on multiple parameters of a large-scale antenna system, including:

establishing a model for optimizing spectral efficiency with multiple parameters, wherein the multiple parameters include: antenna selection matrix, beamforming vector, transmit power and pilot sequence length;

According to the established multi-parameter optimization spectrum efficiency model, the antenna selection matrix is optimized;

Optimize the beamforming vector and transmit power according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix;

Optimize the length of the pilot sequence according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power;

The optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length are brought into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency.

Further, the established multi-parameter optimized spectral efficiency model is:

arg max R

variables w, S, p _s , τ;

where variables represent the optimization variables of the model, R represents the spectral efficiency, w is the beamforming vector, S is the antenna selection matrix, _ps is the transmit power, and τ is the length of the pilot sequence sent by the user.

Further, the spectral efficiency R is expressed as:

Among them, K is the number of users in the large-scale antenna system, I _K is the K × K dimensional identity matrix, and the non-superscript H is the M × K dimensional channel matrix between the M antennas of the base station BS and the K users in the system , the superscript H represents the conjugate of the matrix, and || represents the determinant of the matrix.

Further, according to the established multi-parameter optimization spectrum efficiency model, optimizing the antenna selection matrix includes:

According to the step-by-step optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, and the model is transformed from a non-convex objective function to a convex objective function;

In the obtained convex objective function, after fixing the beamforming vector, the transmit power, and the length of the pilot sequence, calculate the derivative g _i of the convex objective function to the antenna selection matrix;

According to the derivative g _i of the convex objective function to the antenna selection matrix, find the maximum value S _i+1 of the lower boundary of the convex objective function;

According to the maximum value S _i+1 of the lower boundary of the convex objective function, the optimized antenna selection matrix is obtained.

Further, the convex objective function obtained after transformation is expressed as:

where α is the convex function coefficient.

Further, according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, the optimized beamforming vector and transmit power include:

According to the step-by-step optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, and the model is transformed from a non-convex objective function to a convex objective function:

arg maxf(w)=f _cave (w)+f _vex (w)

Among them, f _cave (w) represents the concave function in the convex objective function f(w), and f _vex (w) represents the convex function in the convex objective function f(w);

其中，

表示对凹函数的变量求偏导数；According to the obtained convex objective function, after fixing the length of the pilot sequence, the partial derivative of the concave function in the convex objective function to the beamforming vector is calculated by using the antenna selection matrix obtained by optimization.

in,

Represents the partial derivative of the variable of the concave function;

求凸目标函数下边界的最大值w_i+1；Partial Derivatives of Beamforming Vectors According to Concave Functions in Convex Objective Functions

Find the maximum value w _i+1 of the lower boundary of the convex objective function;

According to the maximum value w _i+1 of the lower boundary of the convex objective function, the optimized beamforming vector and transmit power are obtained.

Further, according to the established multi-parameter optimized spectrum efficiency model and the antenna selection matrix, beamforming vector and transmit power obtained by optimization, the optimized pilot sequence length includes:

According to the principle of maximizing energy efficiency, construct a pilot sequence length optimization function;

According to the established multi-parameter optimized spectral efficiency model, after fixing the spectral efficiency to a preset constant, the optimized antenna selection matrix, beamforming vector, and transmit power are used to search for the length of the pilot sequence that maximizes the energy efficiency. the length of the subsequent pilot sequence.

Further, the constructed pilot sequence length optimization function is expressed as:

Further, the channel refers to the uplink channel of the large-scale antenna system.

The beneficial effects of the above-mentioned technical solutions of the present invention are as follows:

In the above scheme, a multi-parameter optimized spectral efficiency model is established; according to the established multi-parameter optimized spectral efficiency model, the antenna selection matrix is optimized; according to the established multi-parameter optimized spectral efficiency model and the optimized antenna selection matrix, the beamforming vector and Transmit power; optimize the pilot sequence length according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power; optimize the antenna selection matrix, beamforming vector sum, transmit power and optimize The length of the pilot sequence is brought into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency. In this way, by optimizing the four parameters of the antenna selection matrix, beamforming vector, transmit power and pilot sequence length in the established spectrum efficiency model, the pilot pollution and large power consumption in large-scale antenna systems can be comprehensively solved. , high complexity and high cost, thereby improving the spectral efficiency of the system.

Description of drawings

1 is a schematic flowchart of a method for optimizing spectral efficiency based on multiple parameters of a large-scale antenna system according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a Massive MIMO communication system provided by an embodiment of the present invention;

FIG. 3 is a schematic diagram illustrating a comparison between a method for optimizing spectral efficiency with multiple parameters of a large-scale antenna system and a GA effect provided by an embodiment of the present invention.

Detailed ways

In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

Aiming at the existing problems of pilot frequency pollution, high power consumption, high system complexity and high cost, the present invention provides a method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system.

As shown in FIG. 1 , a method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system provided by an embodiment of the present invention includes:

S101, establishing a model for optimizing spectrum efficiency with multiple parameters, wherein the multiple parameters include: an antenna selection matrix, a beamforming vector, a transmit power, and a pilot sequence length;

S102, optimize the antenna selection matrix according to the established multi-parameter optimization spectrum efficiency model;

S103, optimize the beamforming vector and transmit power according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix;

S104, according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power, optimize the length of the pilot sequence;

S105: Bring the optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length into the established multi-parameter optimized spectral efficiency model to obtain optimized spectral efficiency.

In the method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system according to the embodiment of the present invention, a model for optimizing spectrum efficiency with multiple parameters is established; according to the established model for optimizing spectrum efficiency with multiple parameters, an antenna selection matrix is optimized; The spectral efficiency model and the optimized antenna selection matrix are used to optimize the beamforming vector and transmit power; according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power, the pilot sequence length is optimized; The optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length are brought into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency. In this way, by optimizing the four parameters of the antenna selection matrix, beamforming vector, transmit power and pilot sequence length in the established spectrum efficiency model, the pilot pollution and large power consumption in large-scale antenna systems can be comprehensively solved. , high complexity and high cost, thereby improving the spectral efficiency of the system.

In the foregoing specific implementation of the method for optimizing spectral efficiency based on multi-parameters of a large-scale antenna system, further, the established multi-parameter optimized spectral efficiency model is:

arg max R

variables w, S, p _s , τ;

where variables represent the optimization variables of the model, R represents the spectral efficiency, w is the beamforming vector, S is the antenna selection matrix, _ps is the transmit power, and τ is the length of the pilot sequence sent by the user.

Further, the spectral efficiency R is expressed as:

Among them, K is the number of users in the large-scale antenna system, I _K is the K × K dimensional identity matrix, and the non-superscript H is the M × K dimensional channel matrix between the M antennas of the base station BS and the K users in the system , the superscript H represents the conjugate of the matrix, and || represents the determinant of the matrix.

In the embodiment of the present invention, as shown in FIG. 2 is a schematic structural diagram of a large-scale antenna system, it is assumed that a base station (BS) is equipped with M antennas, K single-antenna users, and an uplink channel conforming to flat fading. All users are movable within the designated range with the base station as the center point, and of course they may not move.

In this embodiment, it is assumed that a large-scale antenna system includes: a base station (BS) with M transmitting antennas and K single-antenna users, where M≥K, and K users occupy the same time-frequency resources to transmit data, then the base station BS The received signal y is expressed as:

y=HSwx+n

Among them, y is the received signal of the base station (BS), an M×1-dimensional vector; H is the M×K-dimensional channel matrix H=[h ₁ , h ₂ between the M antennas of the base station BS and the K users in the system ,...,h _m ], the channel between the mth antenna of the BS and the kth user is h _k , the K×1 dimension h _k corresponds to the channel vector of the mth transmission antenna; w is the M transmission The beamforming weight value of the antenna, representing the beamforming vector, is an M×1-dimensional matrix, w=[w ₁ ,w ₂ ,...,w _m ] ^T ∈R ⁺ , R ⁺ represents a positive real number, and || w||≤1, w _m is the beamforming weight value of the mth transmit antenna, || || represents the norm of the matrix, ∈ represents belonging to the set; x is the transmitted signal vector, x=[x ₁ , x ₂ ,...,x _m ] ^T , where T represents the transpose of the matrix, x _m is the symbol transmitted from the mth transmit antenna; n is a zero-mean additive white Gaussian noise vector, an M×1-dimensional vector; S is The antenna selection matrix is a diagonal matrix of M×M dimension, trace(S)=K, trace represents the trace of the matrix, so that:

After the received signal y of the base station BS passes through the linear detector, the received signal r is obtained, which can be expressed as:

r=A ^Hy

Among them, A is an M×K-dimensional linear detection matrix. When a linear zero-forcing detector is used, A=H, and the non-superscript H is the M×K-dimensional channel between the M antennas of the base station (BS) and the K users. Matrix; superscript H denotes the conjugate of the matrix.

According to the obtained received signal r, the multi-parameter optimized spectral efficiency model of the uplink system is established as:

arg max R

variables w, S, p _s , τ;

where variables represent the optimization variables of the model, R represents the spectral efficiency, w is the beamforming vector, S is the antenna selection matrix, _ps is the transmit power, and τ is the length of the pilot sequence sent by the user.

The spectral efficiency R is expressed as:

Among them, K is the number of users in the large-scale antenna system, I _K is the K × K dimensional identity matrix, and the non-superscript H is the M × K dimensional channel matrix between the M antennas of the base station BS and the K users in the system , the superscript H represents the conjugate of the matrix, and || represents the determinant of the matrix.

In the embodiment of the present invention, in S102, S103, and S104, the multi-parameter optimization spectrum efficiency model established in S101 can be used, and the step-by-step optimization equivalent to the multi-optimization can be used to optimize the antenna selection matrix S, the beamforming vector w, and the transmit power step by step. p _S and the pilot sequence length τ.

In this embodiment, the optimized antenna selection matrix according to the established multi-parameter optimized spectrum efficiency model may specifically include:

According to the step-by-step optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, the antenna selection matrix is optimized step-by-step, and the model is transformed from a non-convex objective function to a convex objective function, and the convex objective function represents for:

Among them, a is the convex function coefficient, α is a constant, a>0;

In the obtained convex objective function, the beamforming vector, transmit power and pilot sequence length are fixed first, and then the derivative g _i of the convex objective function to the antenna selection matrix is calculated;

According to the derivative g _i of the calculated convex objective function to the antenna selection matrix, find the maximum value S _i+1 of the lower boundary of the convex objective function:

表示定义；in,

express definition;

According to the maximum value S _i+1 of the lower boundary of the convex objective function, the optimized antenna selection matrix is obtained.

In this embodiment, the optimized beamforming vector and transmit power may specifically include:

According to the stepwise optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, and the beamforming vector w and transmit power p _S are optimized step by step, and the model is transformed from a non-convex objective function to a convex objective function:

arg max f(w)=f _cave (w)+f _vex (w)

Among them, f _cave (w) represents the concave function in the convex objective function f(w), and f _vex (w) represents the convex function in the convex objective function f(w);

其中，

表示对凹函数的变量求偏导数；In the obtained convex objective function, the length of the pilot sequence is fixed first, and the partial derivative of the concave function in the objective function to the beamforming vector is calculated by using the antenna selection matrix obtained by optimization.

in,

Represents the partial derivative of the variable of the concave function;

求解凸目标函数下边界的最大值w_i+1：The partial derivative of the beamforming vector according to the concave function in the objective function

Find the maximum value w _i+1 of the lower bound of the convex objective function:

According to the maximum value w _i+1 of the lower boundary of the convex objective function, the optimized beamforming vector and transmit power are obtained.

In this embodiment, the optimized pilot sequence length may specifically include:

According to the principle of maximizing energy efficiency, construct a pilot sequence length optimization function;

According to the established multi-parameter optimized spectral efficiency model, after fixing the spectral efficiency to a preset constant, the optimized antenna selection matrix, beamforming vector, and transmit power are used to search for the length of the pilot sequence that maximizes the energy efficiency. the length of the subsequent pilot sequence.

In this embodiment, S102-S104 are repeatedly executed until the preset iterative termination condition is satisfied, that is, the iterative increase value of the spectral efficiency of the system is less than or equal to 0.01, and the antenna selection matrix, beamforming vector sum, and transmit power sum that are finally optimized are optimized. The length of the pilot sequence is brought into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency.

The channel involved in this embodiment refers to the uplink channel of the large-scale antenna system.

In order to better understand the method for optimizing spectral efficiency based on multi-parameters of a large-scale antenna system according to the embodiment of the present invention, a specific embodiment is used to describe it in detail:

In this embodiment, it is assumed that the base station (BS) is equipped with 100 antennas and 55 single-antenna users, which conform to the uplink channel of flat fading. All users are movable within the designated range with the base station as the center point, and of course they may not move.

In the embodiment of the present invention, it is assumed that a large-scale antenna array is configured at the base station in the system to provide services for multiple user terminals at the same time, and the number of antennas configured at the base station is much larger than the number of users at the receiving end; for flat fading channels, due to spatial fading and shadow fading are non-sequential and can be assumed to be constant, so only fast fading is considered; the coherence time is assumed to be 196ms, then the model for establishing multi-parameter optimization of spectral efficiency can include:

A11, the received signal y of the base station BS is represented as follows:

y=HSwx+n

Among them, n is a zero-mean additive white Gaussian noise vector, a 100 × 1-dimensional vector;

A12, using A11 to obtain the received signal of the base station, through the linear zero-forcing detector, the received signal of the base station BS is expressed as follows:

r=A ^Hy

Among them, A is a 100×55-dimensional linear detection matrix; y is the received signal of the base station (BS), a 100×1-dimensional vector;

A13, the received signal r of the base station BS is known, and the multi-parameter optimization spectrum efficiency model of the uplink system is established as follows:

variables w, S, p _s , τ;

Among them, K indicates that the number of users in the system is 55; I _K is a 55 × 55-dimensional identity matrix; τ is the length of the pilot sequence of the pilot transmitted by the user, 55≤τ≤196; p _s is the transmit power; non-superscript H is the 100×55-dimensional channel matrix between the 100 antennas of the base station (BS) and 55 users, H=[h ₁ , h ₂ ,...,h _m ], the mth antenna of the BS and the kth antenna The channel between users is h _k , the 55×1-dimensional h _k corresponds to the channel vector of the mth transmission antenna; S is the antenna selection matrix, a 100×100-dimensional diagonal matrix, trace(S)=55, so that :

where x is the transmitted signal vector, x=[x ₁ ,x ₂ ,...,x _m ] ^T (T represents the transpose of the matrix), x _m is the symbol transmitted from the mth transmit antenna; w is the M root The beamforming weight of the transmit antenna, a 100×1-dimensional matrix, w=[w ₁ ,w ₂ ,…,w _m ] ^T ∈R ⁺ (R ⁺ represents a positive real number) and ||w||≤1,w _m is the beamforming weight of the mth transmit antenna; T is the coherence time; the superscript H denotes the conjugate of the matrix; trace denotes the trace of the matrix; || || denotes the norm of the matrix; | | denotes the determinant of the matrix .

In this embodiment, it is assumed that a 100×1-dimensional signal vector x is randomly generated, the optimized beamforming vector w is a 100×1-dimensional vector randomly generated with ||w||≤1, and the transmit power p _S =w ^H wE{| x| ² }, the length of the pilot sequence τ=K=55; to optimize the antenna selection matrix S, the specific steps are as follows:

A21, according to the stepwise optimization equivalent to the joint optimization, the antenna selection matrix is optimized step by step, and the parameter model is transformed from a non-convex objective function to a convex objective function;

A22, in the obtained convex objective function, firstly fix the beamforming vector as 100×1-dimensional elements of 0.1, transmit power w ^H wE{|x| ² }, random signal generation, and pilot sequence length of 55;

A23, calculate the derivative of the convex objective function to the antenna selection matrix;

A24, according to the derivative of the convex objective function to the antenna selection matrix, find the maximum value S _i+1 of the lower boundary of the convex objective function;

A25, according to the maximum value S _i+1 of the lower boundary of the convex objective function, an optimized antenna selection matrix is obtained.

In this embodiment, the antenna selection matrix is obtained, the pilot sequence length is fixed, and the beamforming vector w and the transmit power p _S are optimized. The specific steps may include:

A31, according to the stepwise optimization equivalent to the joint optimization, the beamforming vector w and the transmit power p _S are optimized step by step, and the parameter model is transformed from a non-convex objective function to a convex objective function:

arg max f(w)=f _cave (w)+f _vex (w)

Among them, f _cave (w) represents the concave function in the convex objective function f(w), and f _vex (w) represents the convex function in the convex objective function f(w);

其中，

表示对凹函数的变量求偏导数；A32, in the obtained convex objective function, first fix the pilot sequence length of 55, and use the optimized antenna selection matrix to calculate the partial derivative of the concave function in the objective function to the beamforming vector

in,

Represents the partial derivative of the variable of the concave function;

构造波束成形矢量优化函数w_i+1：A33, the partial derivative of the beamforming vector according to the concave function in the objective function

Construct the beamforming vector optimization function w _i+1 :

A34, the beamforming vector and the transmit power can be obtained by solving the formula in A33.

In this embodiment, according to the antenna selection matrix, beamforming vector, and transmit power obtained by the above solution, the pilot sequence length τ is optimized step by step, and the specific steps are as follows:

A41, according to the principle of maximizing energy efficiency, construct a pilot sequence length optimization function;

A42. According to the established multi-parameter optimized spectral efficiency model, after fixing the spectral efficiency to a preset constant, use the optimized antenna selection matrix, beamforming vector, and transmit power to search for a pilot sequence length that maximizes energy efficiency, The optimized pilot sequence length is obtained.

Finally, the optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length are brought into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency.

In this embodiment, FIG. 3 is a schematic diagram illustrating the comparison of the spectral efficiency effect of the method for optimizing the spectral efficiency of a large-scale antenna system with multiple parameters and a traditional genetic (GA) algorithm provided in the embodiment. The abscissa in Figure 3 is the signal-to-noise ratio (dB), and the ordinate is the spectral efficiency. The result is that when the number of transmitting antennas is 100, the number of serving users is 55, the coherence time is 196ms, and the signal-to-noise ratio is equally increased under the same conditions, the experiment was repeated under the same conditions. out. It can be seen from FIG. 3 that, compared with the traditional genetic (GA) algorithm, the method for improving spectral efficiency provided by the embodiment of the present invention not only reduces the pilot frequency pollution existing in the system, but also reduces power consumption, system complexity and cost. , and got better performance. This shows that the method proposed by the present invention is comprehensive and effective. At the same time, the comparison results also show that the method for optimizing the spectral efficiency based on multi-parameters of a large-scale antenna system provided by the embodiment of the present invention is superior to the performance of the spectral efficiency of the traditional genetic (GA) algorithm.

To sum up, the method for optimizing spectrum efficiency based on multi-parameters of a large-scale antenna system provided by the embodiments of the present invention mainly has the following advantages:

1). Effectively reduce pilot pollution. The multi-parameter optimization spectrum efficiency method for a large-scale antenna system provided by the embodiment of the present invention can estimate the channel based on the pilot, obtain complete channel state information (CSI) of the channel, and The pilot frequency is scheduled, and the length of the pilot frequency sequence is optimized, which effectively reduces the pilot frequency pollution of the system;

2). The complexity and hardware cost of the system are reduced. The multi-parameter optimization spectrum efficiency method of a large-scale antenna system provided by the embodiment of the present invention can select some antennas with the best performance from many antennas of the base station for signal transmission , improving the system spectrum efficiency without affecting the quality of service to users;

3). The power consumption of the system is reduced. The multi-parameter optimization spectrum efficiency method for a large-scale antenna system provided by the embodiment of the present invention uses the beamforming technology to optimize the beamforming vector to obtain a larger array gain. Correspondingly, the transmit power is reduced, and the same quality of service is achieved;

4). The spectral efficiency is improved. Under the same conditions, the spectral efficiency of this method is better than that of the traditional classical genetic algorithm (GA).

It should be noted that, in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any relationship between these entities or operations. any such actual relationship or sequence exists.

The above are the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. These improvements and modifications It should also be regarded as the protection scope of the present invention.

Claims (3)

1. a method for optimizing spectral efficiency based on multiple parameters of a large-scale antenna system, is characterized in that, comprising: establishing a model for optimizing spectral efficiency with multiple parameters, wherein the multiple parameters include: antenna selection matrix, beamforming vector, transmit power and pilot sequence length; According to the established multi-parameter optimization spectrum efficiency model, the antenna selection matrix is optimized; Optimize the beamforming vector and transmit power according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix; Optimize the length of the pilot sequence according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix, beamforming vector and transmit power; Bring the optimized antenna selection matrix, beamforming vector sum, transmit power and optimized pilot sequence length into the established multi-parameter optimized spectral efficiency model to obtain the optimized spectral efficiency; Among them, the established multi-parameter optimized spectral efficiency model is: argmaxR variables w, S, p _s , τ; where variables represent the optimization variables of the model, R represents the spectral efficiency, w is the beamforming vector, S is the antenna selection matrix, _ps is the transmit power, and τ is the length of the pilot sequence sent by the user; Among them, the spectral efficiency R is expressed as:

Among them, K is the number of users in the large-scale antenna system, I _K is the K × K dimensional identity matrix, and the non-superscript H is the M × K dimensional channel matrix between the M antennas of the base station BS and the K users in the system , the superscript H represents the conjugate of the matrix, and || represents the determinant of the matrix; Wherein, the optimized antenna selection matrix according to the established multi-parameter optimized spectrum efficiency model includes: According to the step-by-step optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, and the model is transformed from a non-convex objective function to a convex objective function; In the obtained convex objective function, after fixing the beamforming vector, the transmit power, and the length of the pilot sequence, calculate the derivative g _i of the convex objective function to the antenna selection matrix; According to the derivative g _i of the convex objective function to the antenna selection matrix, find the maximum value S _i+1 of the lower boundary of the convex objective function; According to the maximum value S _i+1 of the lower boundary of the convex objective function, the optimized antenna selection matrix is obtained; Wherein, the optimized beamforming vector and transmit power according to the established multi-parameter optimized spectrum efficiency model and the optimized antenna selection matrix include: According to the step-by-step optimization equivalent to joint optimization, the established model is subjected to convex optimization processing, and the model is transformed from a non-convex objective function to a convex objective function: arg maxf(w)=f _cave (w)+f _vex (w) Among them, f _cave (w) represents the concave function in the convex objective function f(w), and f _vex (w) represents the convex function in the convex objective function f(w); According to the obtained convex objective function, after fixing the length of the pilot sequence, the partial derivative of the concave function in the convex objective function to the beamforming vector is calculated by using the antenna selection matrix obtained by optimization.

in,

Represents the partial derivative of the variable of the concave function; Partial Derivatives of Beamforming Vectors According to Concave Functions in Convex Objective Functions

Find the maximum value w _i+1 of the lower boundary of the convex objective function; According to the maximum value w _i+1 of the lower boundary of the convex objective function, the optimized beamforming vector and transmit power are obtained; According to the established multi-parameter optimized spectrum efficiency model and the antenna selection matrix, beamforming vector and transmit power obtained by optimization, the optimized pilot sequence length includes: According to the principle of maximizing energy efficiency, construct a pilot sequence length optimization function; According to the established multi-parameter optimized spectral efficiency model, after fixing the spectral efficiency to a preset constant, the optimized antenna selection matrix, beamforming vector, and transmit power are used to search for the length of the pilot sequence that maximizes the energy efficiency. the length of the subsequent pilot sequence. 2. the method for optimizing spectral efficiency based on large-scale antenna system multi-parameter according to claim 1, is characterized in that, the convex objective function that obtains after conversion is expressed as:

where a is the convex function coefficient. 3. the method for optimizing spectral efficiency based on large-scale antenna system multi-parameter according to claim 1, is characterized in that, the pilot sequence length optimization function of structure is expressed as:

Images