CN109327894A - Optimal power allocation method for multi-cell MIMO-NOMA based on interference suppression - Google Patents

Abstract

The invention belongs to wireless communication technology fields, disclose a kind of multiple cell MIMO-NOMA optimal power allocation method based on AF panel；The multiple cell MIMO-NOMA optimal power allocation method based on AF panel constructs multiple cell MIMO-NOMA system model, subscriber channel fading model；Inter-cell interference is eliminated by interference alignment techniques, obtains the mathematical model of power distribution problems；Tight lower bound coefficient and corresponding replacement are constructed, converts convex optimization problem for former power distribution problems；Iterative algorithm finds out optimal power allocation scheme.The invention proposes multiple cell MIMO-NOMA to interfere dispelling tactics, and optimal multi-plot joint power distribution problems are derived based on the inhibition to interference, make system and rate reach maximization under the premise of ensureing user fairness, obtains the optimal transmission scheme under optimal multiple cell NIMO-NOMA scene.

Description

Optimal power allocation method for multi-cell MIMO-NOMA based on interference suppression

technical field

The invention belongs to the technical field of wireless communication, and in particular relates to a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression.

Background technique

At present, the existing technologies commonly used in the industry are as follows: Since the Non-orthogonal Multiple Access (NOMA) technology can effectively cope with the challenges of high throughput, low latency, and large connections in future wireless networks, it has been subject to It has received extensive attention and research, and has become one of the alternatives for 5G networks. NOMA technology allows multiple users to share the same power domain channel resources and schedule them differently. Therefore, compared with orthogonal multiple access (orthogonal multiple access, OMA), the NOMA technology can serve more users. In addition, combining NOMA technology and MIMO technology can further improve system performance. In recent years, single-cell NOMA systems for system and rate, user fairness, and energy efficiency have been extensively studied by scholars. With the demand for higher spectrum utilization, the multi-cell NOMA network has received more and more attention and attention. In the scenario of a multi-cell NOMA network, the interference between cells will be very serious, resulting in serious degradation of the performance of cell edge users. According to the power allocation principle of NOMA technology, more power will be occupied by cell-edge users; therefore, the inter-cell interference in multi-cell NOMA network will be more serious than that in multi-cell OMA network, making the performance of cell-edge users worse. Furthermore, joint power allocation in a multi-cell NOMA network is more complex than in a single-cell NOMA network. Several existing NOMA networks are studied as follows: Prior art 1 proposes a cooperative beamforming scheme for a two-cell MIMO-NOMA system, and obtains an effective precoding matrix and decoding matrix. The second prior art proposes a precoding matrix design principle at the base station from the perspective of maximizing throughput for a multi-cell MIMO-NOMA system. In the third prior art, for a multi-cell MIMO-NOMA system, an optimal power allocation strategy is sought for the optimization goals of minimizing the total transmit power and maximizing the system sum rate. In the prior art, the precoding matrix and decoding matrix in a multi-cell NOMA network avoid inter-cell interference, but an efficient power allocation strategy is not considered. The second prior art does not consider inter-cell interference cancellation and power allocation. The power allocation problem in the prior art III does not consider inter-cell interference. Since it is difficult to comprehensively consider a variety of problems, in the above methods, a specific problem in a multi-cell MIMO-NOMA scenario is solved under specific conditions. However, in the actual multi-cell MIMO-NOMA scenario, it is not only necessary to consider the optimal power allocation strategy to improve network capacity, but also to consider the impact of severe interval interference on edge users in multiple cells. Therefore, it is necessary to seek a joint cell. The optimal transmission strategy for inter-interference suppression and power allocation.

To sum up, the problems existing in the prior art are:

(1) Prior art - Precoding matrix and decoding matrix in multi-cell NOMA network to avoid inter-cell interference, but without considering efficient power allocation strategy, the allocation efficiency is not high, it is difficult to serve a large number of users, and the improvement of network capacity is limited , not suitable for actual deployment.

(2) The second prior art does not consider inter-cell interference cancellation and power allocation, which is an optimal power allocation strategy in an ideal situation and cannot be directly used in practical scenarios.

(3) The power allocation problem in the prior art No. 3 does not consider the inter-cell interference, which leads to a serious degradation of the performance of users at the edge of the cell.

The difficulty and significance of solving the above technical problems:

The difficulty of the optimal power allocation method for multi-cell MIMO-NOMA based on interference suppression lies in the comprehensive consideration of the optimal power allocation problem and the problem of inter-multi-cell interference in the actual multi-cell NOMA scenario. While obtaining the optimal power allocation strategy to improve network capacity, it reduces the impact of multi-cell interference on edge users and ensures the performance of all users. Designing a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression is an optimal transmission strategy that combines inter-cell interference suppression and power allocation. It effectively improves resource utilization, ensures user performance and fairness, and provides solutions for the deployment of multi-cell MIMO-NOMA networks in the future.

SUMMARY OF THE INVENTION

In view of the problems existing in the prior art, the present invention provides a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression.

The present invention is implemented in this way, a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression, the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression constructs a multi-cell MIMO-NOMA system model, and the user Channel fading model; eliminate inter-cell interference through interference alignment technology, and obtain the mathematical model of power allocation problem; construct tight lower bound coefficients and corresponding substitutions to transform the original power allocation problem into a convex optimization problem; iterative algorithm to find the optimal power distribution plan.

Further, the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression includes the following steps:

Step 1, based on the multi-cell MIMO-NOMA system model, obtain the user channel fading matrix;

In step 2, the user received signal model is obtained, and the optimal power allocation optimization problem is constructed;

Step 3: Convert the original power allocation problem into a convex optimization problem by constructing a tight lower bound of the original objective function;

Step 4, according to the initial tight lower bound coefficient, solve to obtain a set of power distribution schemes;

In step 5, the optimal power distribution scheme is obtained through an iterative algorithm.

Further, the step 1 specifically includes: there are 1 base stations with Nt antennas, and each base station has J users with Nr antennas; using U[i,1] and U[i,2] to represent the same frequency band of the cell respectively central users and edge users within; matrix is the channel fading coefficient from the i-th BS to the j-th user (m∈I) of the m-th cell; α represents the large-scale fading factor; all elements of the matrix G are independent and identically distributed and follow CN(0 , 1) distribution.

Further, the step 2 specifically includes: using the IA method for cell edge users according to the channel fading matrix, then the received signals of U[i,1] and U[i,2] can be expressed as: in, for the received signal, Represents the interference of other base stations to U[i,1], n _i,1 (n=1,2) is Gaussian white noise; the rates of U[i,1] and U[i,2] are expressed as: in, The three terms are the channel gains; the system sum rate is: where p=[p _1,1 ,p _1,2 ,...,p _I,2 ], the optimal power allocation is modeled as:

Problem1: max _p R(p)

s.t.

C2: p _i,1 -p _i,2 <0,i=1,2...I

C3:R _i,1 ≥C _i,1 ,R _i,2 ≥C _i,2 ,i=1,2...I;

Among them, C1 is the total power constraint condition, C2 is the power constraint condition required by the NOMA network, and C3 is the minimum rate requirement of the multi-cell MIMO-NOMA network user.

Further, the step 3 specifically includes: constructing the tight lower bound of the original objective function, Then the lower bound compact coefficients α _i,j and β _i,j are respectively and is any non-negative number; when time, lower bound Will Bring in the original formula to get:

The optimal power distribution problem translates to:

s.t.

Further, the step 4 specifically includes: transforming the optimal power allocation problem into a dual problem, and transforming the problem into:

Among them, μ _i , υ _i , ω _i and ξ are Lagrange multipliers, take the initial tight lower bound coefficients α _i,j =0 and β _i,j =0, solve L(P2) according to the KKT condition, and get _A set of power allocation schemes p=[p _1,1 , p _1,2 , . . . , p 1,2 ].

Further, the step 5 specifically includes: judging whether the obtained power allocation scheme is converged, if not, according to the obtained power allocation scheme p=[p _1,1 ,p _1,2 ,...,p _{I, 2} ] Calculate the new tight lower bound coefficients α _i,j and β _i,j , and repeatedly solve according to the initial tight lower bound coefficients to obtain a set of power allocation schemes until iteratively converges, and obtain the optimal value of each base station in the multi-cell MIMO-NOMA system. power distribution scheme.

Another object of the present invention is to provide a wireless communication system applying the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression.

Another object of the present invention is to provide a non-orthogonal multiple access system applying the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression.

Another object of the present invention is to provide an information data processing terminal applying the interference suppression-based multi-cell MIMO-NOMA optimal power allocation method.

To sum up, the advantages and positive effects of the present invention are: based on the multi-cell MIMO-NOMA system model, the present invention takes into account the interference between cells, and introduces a new optimal power allocation calculation method; compared with the prior art First, the optimal power allocation calculation method is considered. Compared with the existing technology 2 and technology 3, the interference between multiple cells is considered, and the fairness of users is guaranteed. The invention proposes a multi-cell MIMO-NOMA interference elimination strategy, and derives the optimal multi-cell joint power allocation problem based on the interference suppression, maximizes the system and rate under the premise of ensuring user fairness, and obtains the optimal The optimal transmission scheme in the multi-cell NIMO-NOMA scenario provides a solution for the deployment of the multi-cell MIMO-NOMA scenario in the future.

Description of drawings

FIG. 1 is a flowchart of an optimal power allocation method for multi-cell MIMO-NOMA based on interference suppression provided by an embodiment of the present invention.

FIG. 2 is a schematic diagram of an application scenario of a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression provided by an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

For the prior art, an efficient power allocation strategy is not considered; inter-cell interference cancellation and power allocation are not considered; the problem of inter-cell interference is not considered. The present invention specifically relates to a multi-cell transmission method for joint interference suppression and power allocation, which is applied to a multi-cell MIMO-NOMA wireless network to eliminate inter-cell interference, improve cell edge user performance, and improve overall system performance.

The application principle of the present invention will be described in detail below with reference to the accompanying drawings.

As shown in FIG. 1 , the optimal power allocation method for multi-cell MIMO-NOMA based on interference suppression provided by an embodiment of the present invention includes the following steps:

S101: Obtain a user channel fading matrix based on a multi-cell MIMO-NOMA system model;

S102: Obtain a user received signal model, and construct an optimal power allocation optimization problem;

S103: Convert the original power allocation problem into a convex optimization problem by constructing a tight lower bound of the original objective function;

S104: According to the initial tight lower bound coefficient, solve to obtain a set of power distribution schemes;

S105: Obtain an optimal power distribution scheme according to an iterative algorithm.

The power allocation method for a multi-cell MIMO-NOMA system provided by the embodiment of the present invention specifically includes the following steps:

(1) In a multi-cell MIMO-NOMA system, there is one base station with Nt antennas, and each base station has J users with Nr antennas. U[i,1] and U[i,2] are used to represent the center users and edge users in the same frequency band of the cell, respectively. matrix is the channel fading coefficient from the ith BS to the jth user (m∈I) of the mth cell. Among them, α represents the large-scale fading factor; all elements of the matrix G are independent and identically distributed and follow the CN(0,1) distribution.

(2) According to the channel fading matrix, using the IA method for cell edge users, the received signals of U[i,1] and U[i,2] can be expressed as: in, for the received signal, Represents the interference of other base stations to U[i,1], n _i,1 (n=1,2) is Gaussian white noise. The rates of U[i,1] and U[i,2] can be expressed as: in, The three terms are channel gain. From this, the system sum rate can be obtained as: where p=[p _1,1 ,p _1,2 ,...,p _I,2 ], based on this, the optimal power allocation can be modeled as the following optimization problem:

Problem1: max _p R(p)

s.t.

C2: p _i,1 -p _i,2 <0,i=1,2...I

C3:R _i,1 ≥C _i,1 ,R _i,2 ≥C _i,2 ,i=1,2...I;

Among them, C1 is the total power constraint condition, C2 is the power constraint condition required by the NOMA network, and C3 is the minimum rate requirement of the multi-cell MIMO-NOMA network user.

(3) Construct the tight lower bound of the original objective function, Then the lower bound compact coefficients α _i,j and β _i,j are respectively and is any non-negative number. when time, lower bound Will Bring in the original formula to get:

The optimal power allocation problem can be transformed into:

s.t.

(4) Convert the above optimal power allocation problem into its dual problem, and convert the problem into:

Among them, μ _i , υ _i , ω _i and ξ are Lagrange multipliers, take the initial tight lower bound coefficients α _i,j =0 and β _i,j =0, solve L(P2) according to the KKT condition, and get _A set of power allocation schemes p=[p _1,1 , p _1,2 , . . . , p 1,2 ].

(5) _Judging whether the obtained power distribution scheme is _converged , if not _, calculate a new Tighten the lower bound coefficients α _i,j and β _i,j , repeat the fourth step until iterative convergence, and obtain the optimal power allocation scheme for each base station of the multi-cell MIMO-NOMA system.

The application principle of the present invention will be further described below in conjunction with the accompanying drawings

As shown in Figure 2, there are three cells BS1, BS2, and BS3 in the scene, each cell includes a BS, a cell center user and a cell edge user, and the two users use NOMA for downlink transmission.

Step 1: Obtain the user channel fading matrix based on the multi-cell MIMO-NOMA system model.

According to the scenario in FIG. 2 , there are 3 base stations with Nt antennas, and there are J users with Nr antennas in each base station. U[i,1] and U[i,2] are used to represent the center users and edge users (i∈{1,2,3}) in the same frequency band of the cell, respectively. Allocate power to center users and edge users in the same frequency band. matrix Represents the channel fading coefficients from the i-th BS to the j-th user (m∈I={1, 2, 3}) of the m-th cell. Among them, α represents the large-scale fading factor; all elements of the matrix G are independent and identically distributed and follow the CN(0,1) distribution.

Step 2: Obtain the user received signal model, and construct the optimal power allocation optimization problem;

(2a) The received signals of U[i,1] and U[i,2] can be expressed as:

in, Represents all base stations except the ith base station; the second item on the right side of Y _i,1 and Y _i,2 respectively represents the inter-cell interference to U[i,1] and U[i,2]; is additive white Gaussian noise (AWGN). Furthermore, x ⁱ is the superposition signal of N _t × 1, expressed as: in, is the precoding vector of the i-th BS side, Yes the required number of symbols, Represents the power distribution coefficient of the superimposed signal.

(2b) In order to eliminate severe inter-cell interference, use the IA method for cell-edge users, and use ^vi to concentrate all inter-cell interference into the same signal space, so that other signals are desired signals. Correspondingly, the feasible conditions for IA are:

where d _i,2 is defined as the degree of freedom (DoF), is the unit interference rejection vector at the receiver.

(2c) According to (2a), (2b), convert the received signals of U[i,1] and U[i,2] into: Among them, y _i,1 =Y _i,1 , Then the rates of U[i,1] and U[i,2] can be expressed as: in, From this, the total rate of the system can be obtained as: where p=[p _1,1 ,p _1,2 ,...,p _I,2 ], the optimal power allocation problem can be described as:

Problem1: max _p R(p)

s.t.

C2: p _i,1 -p _i,2 <0,i=1,2...I

C3:R _i,1 ≥C _i,1 ,R _i,2 ≥C _i,2 ,i=1,2...I;

Among them, C1 is the total power constraint condition, C2 is the power constraint condition required by the NOMA network, and C3 is the current condition that the users of the multi-cell MIMO-NOMA network require the lowest rate.

Step 3: Construct the original objective function tight lower bound and logarithmic variation method, and convert the original power allocation problem into a convex optimization problem;

Construct the lower bound compact coefficients α _i,j and β _i,j such that Then the lower bound compact coefficients α _i,j and β _i,j are respectively and is any non-negative number. when time, lower bound Will Bring in the original formula to get:

The original question can be transformed into:

s.t.

Step 4: According to the initial tight lower bound coefficient, solve to obtain a set of power distribution schemes.

(4a) Transform the above optimal power into its dual problem:

min _μ,υ,ω,ξ Dual(P2)

s.t.μ＞0,υ＞0,ω＞0,ξ＞0;

Convert the question to:

Among them, μ _i , υ _i , ω _i and ξ are Lagrange multipliers.

(4b) Initialize relaxation coefficients so that α _i,j =0 and β _i,j =0;

(4c) Solve L(P2) according to the KKT condition, and obtain a set of power distribution schemes p=[p _1,1 ,p _1,2 ,...,p _I,2 ]

Step 5: Find the optimal power distribution scheme according to the iterative algorithm.

Determine whether the obtained power allocation scheme is convergent, if not, calculate a new tight lower bound coefficient according to the obtained power allocation scheme p=[p _1,1 ,p _1,2 ,...,p _I,2 ] α _i,j and β _i,j , the fourth step is repeated until the iterative algorithm converges, and the optimal power allocation scheme of each base station of the multi-cell MIMO-NOMA system is obtained.

The above descriptions are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention shall be included in the protection of the present invention. within the range.

Claims (10)

1. a multi-cell MIMO-NOMA optimal power allocation method based on interference suppression is characterized in that, the described multi-cell MIMO-NOMA optimal power allocation method based on interference suppression builds a multi-cell MIMO-NOMA system model, and the user channel Fading model; Inter-cell interference is eliminated by interference alignment technology, and the mathematical model of power allocation problem is obtained; the tight lower bound coefficient and corresponding substitution are constructed to transform the original power allocation problem into a convex optimization problem; iterative algorithm is used to find the optimal power allocation Program. 2. The multi-cell MIMO-NOMA optimal power allocation method based on interference suppression as claimed in claim 1, wherein the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression comprises the following steps: Step 1, based on the multi-cell MIMO-NOMA system model, obtain the user channel fading matrix; In step 2, the user received signal model is obtained, and the optimal power allocation optimization problem is constructed; Step 3: Convert the original power allocation problem into a convex optimization problem by constructing a tight lower bound of the original objective function; Step 4, according to the initial tight lower bound coefficient, solve to obtain a set of power distribution schemes; In step 5, the optimal power distribution scheme is obtained through an iterative algorithm. 3. the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression as claimed in claim 1, is characterized in that, described step 1 specifically comprises: have 1 base station with Nt antenna, each base station has J users with Nr antennas; use U[i,1] and U[i,2] to represent the center users and edge users in the same frequency band of the cell, respectively; matrix is the channel fading coefficient from the i-th BS to the j-th user (m∈I) of the m-th cell; α represents the large-scale fading factor; all elements of the matrix G are independent and identically distributed and follow CN(0 , 1) distribution. 4. The multi-cell MIMO-NOMA optimal power allocation method based on interference suppression according to claim 1, wherein the step 2 specifically comprises: according to the channel fading matrix, using the IA method for cell edge users, then U The received signals of [i,1] and U[i,2] can be expressed as: in, for the received signal, Represents the interference of other base stations to U[i,1], n _i,1 (n=1,2) is Gaussian white noise; the rates of U[i,1] and U[i,2] are expressed as: in, The three terms are the channel gains; the system sum rate is: where p=[p _1,1 ,p _1,2 ,...,p _I,2 ], the optimal power allocation is modeled as: Problem1: max _p R(p) s.t. C1: C2: p _i,1 -p _i,2 <0,i=1,2...I C3:R _i,1 ≥C _i,1 ,R _i,2 ≥C _i,2 ,i=1,2...I; Among them, C1 is the total power constraint condition, C2 is the power constraint condition required by the NOMA network, and C3 is the minimum rate requirement of the multi-cell MIMO-NOMA network user. 5. the multi-cell MIMO-NOMA optimal power allocation method based on interference suppression as claimed in claim 1, is characterized in that, described step 3 specifically comprises: construct the tight lower bound of original objective function, Then the lower bound compact coefficients α _i,j and β _i,j are respectively and is any non-negative number; when time, lower bound Will Bring in the original formula to get: The optimal power distribution problem translates to: s.t. C4: C5: C6: 6. The multi-cell MIMO-NOMA optimal power allocation method based on interference suppression according to claim 1, wherein the step 4 specifically comprises: converting the optimal power allocation problem into a dual problem, and converting the problem into : Among them, μ _i , υ _i , ω _i and ξ are Lagrange multipliers, take the initial tight lower bound coefficients α _i,j =0 and β _i,j =0, solve L(P2) according to the KKT condition, and get _A set of power allocation schemes p=[p _1,1 , p _1,2 , . . . , p 1,2 ]. 7. The multi-cell MIMO-NOMA optimal power allocation method based on interference suppression as claimed in claim 1, wherein the step 5 specifically comprises: judging whether the obtained power allocation scheme converges, if not, according to The obtained power distribution scheme p=[p _1,1 ,p _1,2 ,...,p _I,2 ] calculates new tight lower bound coefficients α _i,j and β _i,j , and repeats according to the initial tight lower bound coefficients, a set of power allocation schemes are obtained by solving until the iteration converges, and the optimal power allocation scheme for each base station of the multi-cell MIMO-NOMA system is obtained. 8. A wireless communication system applying the interference suppression-based multi-cell MIMO-NOMA optimal power allocation method according to any one of claims 1 to 7. 9. A non-orthogonal multiple access system applying the interference suppression-based multi-cell MIMO-NOMA optimal power allocation method according to any one of claims 1 to 7. 10 . An information data processing terminal to which the interference suppression-based multi-cell MIMO-NOMA optimal power allocation method according to any one of claims 1 to 7 is applied. 11 .