CN109194373B - A Massive MIMO Beam Domain Joint Unicast Multicast Transmission Method - Google Patents

Abstract

The invention provides a large-scale MIMO beam domain combined unicast and multicast transmission method, wherein a large-scale antenna array is configured at a base station side of wireless communication, and a large-scale beam set covering the whole cell is generated through beam forming. And the base station communicates with the users in the cell by adopting a mode of combining beam domain unicast and multicast on the same time-frequency resource. And the base station counts the channel state information according to the beam domain of each user in the cell, and performs power distribution on the multicast signals of the beam domain and the unicast signals sent to each user. The beam domain power allocation is based on an MM iterative algorithm and a deterministic equivalence method, a beam domain power allocation matrix is obtained by iteratively solving a convex optimization problem, and dynamic updating is carried out along with the change of statistical channel state information. The invention solves the power distribution optimization problem of the beam domain combined unicast and multicast transmission of which the base station side only knows the statistical channel information, improves the unicast and multicast transmission rate of the system and effectively reduces the complexity of realization.

Description

A Massive MIMO Beam Domain Joint Unicast Multicast Transmission Method

technical field

The invention belongs to the field of communications, and in particular relates to a beam-domain wireless transmission method using a large-scale antenna array to perform joint unicast multicast under the same time-frequency resource.

Background technique

In a massive MIMO system, a base station arranges a massive antenna array to serve multiple users simultaneously. The use of massive MIMO technology can effectively reduce the interference between users and greatly improve the spectrum utilization and power efficiency of wireless communication systems. Beam domain transmission means that the base station side converts the transmitted signal to the beam domain through a unified unitary transformation, and transmits the signal in the beam domain channel, making full use of the spatial angular resolution of the large-scale antenna array and the locality of the user channel in the beam domain. .

In the scenario of joint unicast multicast, the base station simultaneously transmits the multicast signal to all users in the cell and the unicast signal for a single user on the same time-frequency resource. In this scenario, it is often necessary to construct and solve the power allocation problem of the transmitted signal, so as to maximize the weighted average of the unicast rate and multicast rate of the entire system, and the optimization objective function of such problems is often non-convex, which is usually difficult to obtain. global optimal solution. At the same time, in the optimization process, the calculation of the unicast multicast rate needs to be calculated, and the implementation complexity is very high. To this end, the present invention proposes a low-complexity massive MIMO beam-domain joint unicast-multicast transmission method using statistical channel state information.

SUMMARY OF THE INVENTION

Purpose of the invention: The purpose of the present invention is to provide a transmission method for joint unicast-multicast using statistical channel state information in a scenario where a cell base station sends a unicast signal and a multicast signal at the same time.

Technical scheme: In order to realize the above-mentioned purpose of the invention, the technical scheme adopted in the present invention is:

A massive MIMO beam-domain joint unicast-multicast transmission method, comprising the following steps:

(1) In the scenario of joint unicast-multicast communication between the base station and the user group, the base station equipped with a large-scale antenna array generates an a set of beams covering the entire cell;

(2) The base station uses the beam domain statistics of the channel state information of the users in the cell to construct and solve the beam domain joint unicast multicast power allocation optimization problem to allocate power to the transmitted signal; the beam domain joint unicast multicast power allocation optimization problem The optimization goal is to maximize the weighted average of the system unicast rate and multicast rate, and the optimization variable is the covariance matrix of the multicast signal sent by the base station and the unicast signal to each user; the constraint condition is the covariance of the total signal sent by the base station. The matrix satisfies the power constraint;

(3) During the movement of each user, with the change of statistical channel state information between the base station and each user, the base station side dynamically implements beam domain power allocation, and dynamically updates in conjunction with the unicast-multicast process.

In the step (1), the base station generates a large-scale beam set that can cover the entire cell to realize beam domain division of spatial resources, and the base station performs joint unicast-multicast communication on the same time-frequency resource for users in the cell. The process of communication is carried out on the beam domain.

In the step (2), the base station uses the beam domain statistical channel state information of the users in the cell to allocate power to the transmitted signal. The base station estimates the beam domain statistical channel state information required to implement beam domain power allocation according to the sounding signal sent by the cell user in the uplink channel sounding phase. The specific allocation method is based on the MM (Minorize-Maximization) iterative algorithm and the deterministic equivalent method.

The above-mentioned power allocation methods based on the MM iteration algorithm and the deterministic equivalent method include:

(a) The interference rate term in the weighted average expression of unicast rate and multicast rate in the current iteration process is approximated by a first-order Taylor expansion, and the non-convex problem is transformed into a convex optimization problem about power allocation in the beam domain, and then Solve using interior point methods or other optimization methods.

Substitute the solution of the optimization problem in the current iteration process into the optimization objective to generate the optimization problem of the next iteration, and solve it again. This step is repeated until the weighted average difference between the system unicast rate and the multicast rate in two adjacent iterative processes is less than a given threshold, and the solution of the last iterative process is the solution of the optimization problem.

(b) According to the large-dimensional random matrix theory, calculate the deterministic equivalent expression of the total unicast rate term and the total multicast rate term including the interference rate in the weighted average expressions of the system unicast rate and multicast rate respectively, to avoid high The complexity of the expectation operation.

In the step 3), with the dynamic movement of each user, the beam domain statistical channel state information between the base station and each user changes, and the base station re-implements the aforementioned beam domain power allocation according to the changed statistical channel state information, thereby. Implement dynamic updates for joint unicast-multicast procedures. The change of the statistical channel state information in the beam domain is related to the specific application scenario. The typical statistical time window is several times or tens of times the short-term transmission time window, and the acquisition of the relevant statistical channel state information is also in a large time width. conduct.

Beneficial effect: Compared with the prior art, the present invention has the following advantages:

1. The base station and cell users implement joint unicast multicast communication under the same time-frequency resources in the beam domain, which can match the spatial characteristics of their wireless channels, so as to obtain the improvement of power efficiency and spectral efficiency brought by the use of large-scale antenna arrays .

2. Use the beam domain statistical channel state information of the cell users to design the transmitted signal. The required beam domain statistical channel state information of each user can be obtained by sparse sounding signals. The proposed joint unicast multicast transmission method is also applicable. For time division duplex and frequency division duplex systems.

3. Using the MM iteration algorithm and the deterministic equivalent method, the implementation complexity of joint unicast-multicast communication is significantly reduced, and the method can obtain approximately optimal performance.

Description of drawings

FIG. 1 is a flowchart of a massive MIMO beam-domain joint unicast-multicast wireless transmission method using statistical channel state information.

FIG. 2 is a schematic diagram of a massive MIMO joint unicast multicast system.

FIG. 3 is a flowchart of a power allocation method based on the MM iteration algorithm and the deterministic equivalent method.

Detailed ways

In order to make those skilled in the art better understand the solution of the present invention, the accompanying drawings in the embodiments of the present invention will be combined below.

As shown in FIG. 1 , a massive MIMO beam-domain joint unicast-multicast transmission method using statistical channel state information disclosed in an embodiment of the present invention mainly includes the following steps:

1) The base station is configured with a large-scale antenna array, and a large-scale beam set that can cover the entire cell is generated by the beamforming method. In this step, the base station generates a large-scale beam set that can cover the entire cell by means of analog multi-beam forming or digital multi-beam forming, so as to realize beam domain division of spatial resources. The base station performs joint unicast-multicast communication with cell users on the same time-frequency resource, and the process of joint unicast-multicast communication is implemented in the beam domain.

2) The base station uses the beam domain statistics of the channel state information of the users in the cell to construct and solve the beam domain joint unicast-multicast power allocation optimization problem to allocate power to the transmitted signal.

3) During the dynamic movement of each user, with the change of the statistical channel state information in the beam domain between the base station and the cell users, the base station side dynamically implements the beam domain power allocation, and the joint unicast-multicast process is dynamically updated.

为小区用户集合，每个用户配置N根接收天线。基站可以采用模拟多波束赋形或数字多波束赋形或模拟与数字混合波束赋形的方法将发送的空间域信号变换到波束域。The method of the embodiment of the present invention is described in detail below by taking the massive MIMO joint unicast multicast scenario shown in FIG. 2 as an example. Considering a single-cell scenario, M (M is on the order of 10 ² or 10 ³ ) transmit antennas are configured on the base station side, and the antenna spacing is half wavelength.

For a set of cell users, each user is configured with N receiving antennas. The base station can transform the transmitted spatial domain signal into the beam domain by adopting analog multi-beamforming, digital multi-beamforming, or hybrid analog and digital beamforming.

其中H_k为第k个用户的波束域信道矩阵，运算符⊙为矩阵Hadamard乘积，*为矩阵的共轭，

表示期望运算。In the channel sounding phase, the cell user sends an uplink sounding signal, and the base station estimates the beam domain statistical channel state information of the cell user according to the received sounding signal, that is,

where H _k is the beam-domain channel matrix of the kth user, operator ⊙ is the Hadamard product of matrices, * is the conjugate of the matrix,

Indicates the desired operation.

其中x^m为多播信号，

为基站到用户k的单播信号。波束域多播信号的协方差矩阵为

发送给用户k的单播信号的协方差矩阵为

多播速率可以表示为：Assume that the beam-domain joint unicast-multicast signal sent by the base station is

where x ^m is the multicast signal,

is the unicast signal from the base station to user k. The covariance matrix of the beam-domain multicast signal is

The covariance matrix of the unicast signal sent to user k is

The multicast rate can be expressed as:

是多播过程中的干扰速率项，

同样，单播速率可以表示为

is the interference rate term in the multicast process,

Likewise, the unicast rate can be expressed as

是单播过程中基站对用户k的干扰速率项，

is the interference rate term of the base station to user k in the unicast process,

In the above expression, log represents the logarithmic operation, det represents the determinant of the matrix, and H is the conjugate transpose of the matrix.

Weighted average of system unicast rate and multicast rate

η∈[0,1] is the weighting factor of multicast.

均为对角矩阵。注意到在波束域联合单播多播传输过程中，为了获得更高的系统和速率，需要对发送信号的协方差矩阵

进行优化，即在基站侧对发射波束进行功率分配，即解决如下优化问题：Considering the low correlation on the base station side of the beam domain channel, the base station sends independent data streams on each beam, that is, the covariance matrix of multicast signals Λ ^m and the covariance matrix of unicast signals.

Both are diagonal matrices. It is noted that in the process of joint unicast multicast transmission in the beam domain, in order to obtain a higher system and rate, the covariance matrix of the transmitted signal needs to be

To optimize, that is, to allocate the power of the transmit beam on the base station side, that is, to solve the following optimization problems:

Among them, P is the total power constraint of the base station, tr(·) represents the trace of the calculation matrix, and ≥ represents that the matrix is non-negative definite.

The objective function of this problem is non-convex, it is difficult to obtain the global optimal solution, and the implementation complexity is very high. To this end, the embodiment of the present invention adopts the MM iteration algorithm and the deterministic equivalent method to solve the above-mentioned beam-domain multicast power allocation optimization problem.

The above-mentioned power allocation methods based on the MM iteration algorithm and the deterministic equivalent method include:

(a) Perform a first-order Taylor expansion approximation of the interference rate term in the multicast rate term and the interference rate term in the unicast rate term in the current iteration process, and transform the non-convex problem as follows: Convex optimization problem:

分别为

和

的梯度，其对角线上的元素可以表示为in,

respectively

and

The gradient of , whose diagonal elements can be expressed as

Then use the interior point method or other optimization methods to solve the optimization problem in equation (6).

Substitute the solution of the optimization problem in the current iteration process into the optimization objective to generate the optimization problem of the next iteration, and solve it again. This step is repeated until the weighted average difference between the system unicast rate and the multicast rate in two adjacent iterative processes is less than a given threshold, and the solution of the last iterative process is the solution of the optimization problem.

和

的确定性等同表达

和

(b) In order to reduce the computational complexity, according to the large-dimensional random matrix theory, calculate the weighted average terms of the system unicast rate and multicast rate respectively.

and

The deterministic equivalent expression of

and

in,

四个辅助变量通过迭代计算得到：

Four auxiliary variables are calculated iteratively:

和

表示生成对角矩阵，对角线上的元素分别为in,

and

Indicates that a diagonal matrix is generated, and the elements on the diagonal are

From this, the deterministic equivalent expression of the weighted average of the system unicast rate and multicast rate is obtained

Figure 3 shows the implementation process of the MM iterative algorithm and the power allocation method of the deterministic equivalent method. The detailed process of the algorithm is as follows:

设置迭代次数指示i＝0。在初始化发送信号的协方差矩阵

时，可以假设均匀功率分配，即这K+1个协方差矩阵都是

其中I是M×M的单位矩阵。Step 1: Initialize the covariance matrix of the transmitted signal

Set the number of iterations to indicate i=0. Initializing the covariance matrix of the transmitted signal

When , we can assume uniform power distribution, that is, the K+1 covariance matrices are all

where I is an M×M identity matrix.

该步骤首先根据初始化发送信号的协方差矩阵迭代计算确定性等同辅助变量

直至辅助变量收敛，利用辅助变量计算

和

的确定性等同，代入公式(21)进行计算。Step 2: Calculate the deterministic equivalent initial value of the weighted average of the system unicast rate and multicast rate

This step firstly calculates the deterministic equivalent auxiliary variables iteratively according to the covariance matrix of the initialized transmitted signal

Until the auxiliary variables converge, use the auxiliary variables to calculate

and

The certainty of is the same, and it is substituted into formula (21) for calculation.

计算当次迭代梯度项

和

采用MM迭代算法线性化多播干扰速率项

和单播干扰速率项

得到如公式(6)所示本次迭代过程中的优化问题。Step 3: Utilize

Calculate the current iteration gradient term

and

Linearization of Multicast Interference Rate Terms Using MM Iterative Algorithm

and unicast interference rate terms

The optimization problem in this iteration process is obtained as shown in formula (6).

迭代计算确定性等同辅助变量

直至辅助变量收敛。Step 4: Utilize

Iteratively computes deterministic equivalent auxiliary variables

until the auxiliary variables converge.

和

的确定性等同表达

和

并用确定性等同表达替换优化目标中的

和

Step 5: According to formulas (9) and (10), the weighted average term of the system unicast rate and multicast rate is obtained.

and

The deterministic equivalent expression of

and

and replace in the optimization objective with a deterministic equivalent expression

and

Step 6: Use the interior point method or other convex optimization methods to solve the optimization problem to get the solution of the current iteration

Step 7: Using the solution of the current iterative optimization problem, use formula (21) to calculate the deterministic equivalence of the weighted average of the system unicast rate and multicast rate

和

如果两者之间的差值小于预先设定的阈值ε则迭代结束，此时的

即优化问题的解。否则令i＝i+1，返回步骤2。Step 8: Compare

and

If the difference between the two is less than the preset threshold ε, the iteration ends. At this time, the

That is, the solution of the optimization problem. Otherwise, let i=i+1, and return to step 2.

During the movement of each user, as the beam domain statistical channel state information between the base station and the user changes, the base station side repeats the foregoing steps according to the updated statistical channel state information to perform beam domain joint unicast multicast power allocation. Thereby, the dynamic update of the joint unicast-multicast transmission process is realized. The variation of statistical channel state information in the beam domain is related to specific application scenarios. The typical statistical time window is several times or tens of times that of the short-term transmission time window, and the acquisition of relevant statistical channel state information is also carried out in a large time width. .

It should be pointed out that the above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited to this. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention. , all should be covered within the protection scope of the present invention. All components not specified in this embodiment can be implemented by existing technologies.

Claims (5)

1. a massive MIMO beam domain joint unicast multicast transmission method is characterized in that: comprise the steps: (1) In the scenario of joint unicast-multicast communication between the base station and the user group, the base station equipped with a large-scale antenna array generates an a set of beams covering the entire cell; (2) The base station uses the beam domain statistics of the users in the cell to calculate the channel state information, constructs and solves the beam domain joint unicast multicast power allocation optimization problem based on the MM iteration algorithm and the deterministic equivalent method, and allocates power to the transmitted signal; the beam domain The joint unicast-multicast power allocation optimization problem is expressed as:

Among them, η∈[0,1] is the weight factor of multicast,

is the set of users in the cell, K is the total number of users in the cell,

is the base station multicast rate,

is the unicast rate from the base station to user k, Λ ^m is the covariance matrix of the beam domain multicast signal sent by the base station,

is the covariance matrix of the beam domain unicast signal from the base station to the user k, P is the total power constraint of the base station, tr( ) represents the trace of the calculation matrix, ≥ 0 means that the matrix is non-negative definite; in the above optimization objectives:

in,

H _k is the beam domain channel matrix from the base station to user k, I is the identity matrix, the superscript H is the conjugate transpose of the matrix, det is the determinant of the matrix,

for expectation; (3) During the movement of each user, with the change of statistical channel state information between the base station and each user, the base station side dynamically implements beam domain power allocation, and dynamically updates in conjunction with the unicast-multicast process. 2. The massive MIMO beam-domain joint unicast-multicast transmission method according to claim 1, characterized in that: in the step (1), the base station generates a large-scale beam set that can cover the entire cell and realizes the beam domain of spatial resources The base station performs joint unicast-multicast communication with users in the cell on the same time-frequency resource, and the process of joint unicast-multicast communication is implemented in the beam domain. 3. The massive MIMO beam-domain joint unicast-multicast transmission method according to claim 1, wherein the beam-domain statistical channel state information is estimated by the base station according to the uplink sounding signals sent by the users in the cell. . 4. The massive MIMO beam-domain joint unicast multicast transmission method according to claim 1, wherein the MM iteration algorithm and the deterministic equivalent method used to solve the optimization problem in the step (2) include the following two: aspects: (a) In the current iteration process, the interference rate term in the weighted average expression of the system unicast rate and multicast rate is approximated by a first-order Taylor series expansion, and the non-convex problem is transformed into a convex power distribution in the beam domain. Optimization problem, which transforms the optimization problem into solving the following problem:

in,

and

is an M×M diagonal matrix, and the elements on the diagonal are:

Among them, M is the number of base station antennas, N is the number of user antennas,

is the statistical channel state information in the beam domain, the superscript i marks the number of iterations, the subscript t marks the row and column number of the matrix, ⊙ is the Hadamard product of the matrix, and the superscript * is the conjugate of the matrix; Substitute the solution of the optimization problem in the current iteration process into the optimization objective to generate the optimization problem of the next iteration, and solve it again until the weighted average difference between the unicast rate and the multicast rate of the system in the adjacent two iterations is less than the given value. Set the threshold, the solution of the last iteration process is the solution of the optimization problem; (b) According to the large-dimensional random matrix theory, calculate separately

and

The deterministic equivalent expression of

and

Avoid high-complexity expectation operations;

in,

Four auxiliary variables are calculated iteratively:

and

Indicates that a diagonal matrix is generated, and the elements on the diagonal are

5. The massive MIMO beam-domain joint unicast-multicast transmission method according to claim 1, wherein during the dynamic movement of each user, as the statistical channel state information changes between the base station and each user, the base station side The beam domain power allocation is dynamically implemented, and the joint unicast-multicast process is dynamically updated; the changes of the beam domain statistical channel state information are related to the specific application scenarios, and the statistical time window is several times or tens of times the short-term transmission time window.

Images