CN106788937A - A kind of pilot distribution method for taking into account fairness between Massive mimo systems overall performance and user - Google Patents

Abstract

The present invention relates to a pilot allocation method that takes into account both the overall performance of the Massive MIMO system and the fairness among users. The optimization goal is to maximize the system rate product. When solving the optimization problem in the present invention, the pilot based on the greedy strategy is adopted. Allocation method, the greedy pilot allocation method has lower complexity and is more conducive to implementation. Therefore, the pilot allocation method proposed by the present invention can realize fairness among users while maximizing the overall system and rate, thereby achieving an effective compromise between overall system performance and fairness among users.

Description

A Pilot that Considers the Overall Performance of Massive MIMO System and the Fairness among Users distribution method

technical field

The invention relates to a pilot allocation method that takes into account both the overall performance of a Massive MIMO system and the fairness among users, and belongs to the technical field of communication systems.

Background technique

As one of the key technologies of 5G wireless communication system, massive antenna technology (Massive MIMO technology) has attracted more and more researchers' attention in recent years because of its ability to provide faster, stronger and better services. The performance of the Massive MIMO system will increase with the increase of the number of base station antennas, but this performance improvement is not unlimited. In TDD mode, when the number of antennas at the base station reaches a certain number, the system performance tends to be saturated. At this time, pilot pollution becomes the main problem that limits system performance. The problem of pilot pollution occurs because there is interference between users sharing the same pilot, which cannot be eliminated by simply increasing the number of antennas at the base station. Studies have shown that allocating pilots reasonably can effectively improve the performance of the system.

In the existing pilot allocation schemes, most of the work is to maximize the overall sum rate of the system by designing the pilot allocation strategy, that is, max-sumrate, and some will focus on the improvement of the minimum user rate in the system, that is, max-min . The pilot design scheme that focuses on the overall system and rate improvement usually only focuses on the optimization of the overall system performance, which will allow users with good channel conditions to have a higher rate, while users with poor channel conditions will have a lower rate, thus increasing the overall performance of the system. However, the pilot allocation scheme that focuses on the minimum user rate increase in the system is usually dedicated to improving the rate of those users with the worst channel conditions in the system, thus greatly reducing the system overall fitness. So far, there is almost no pilot allocation scheme that can achieve fairness among users while improving the overall performance of the system.

Chinese patent document CN 105827273A discloses a dual-antenna pilot interference elimination method for users of a multi-cell massive MIMO system. The uplink pilot pollution and the interference in the downlink data transmission process are considered by considering spatial correlation and controlling user Beamforming weight vector at the end to alleviate. At the same time, the pilot allocation is introduced into the method mentioned above, and a pilot allocation strategy based on the greedy algorithm is proposed. Combined with the user's dual antenna device, the influence of pilot pollution is further reduced, so as to reduce the channel estimation error. , to increase the system speed. However, the optimization goal in this patent is to maximize the system and rate, and the focus is on improving the overall performance of the system. This method of maximizing the system and rate will seriously damage the fairness among users in the system. This patent adds an antenna to the user end, which can effectively improve the accuracy of uplink channel estimation by utilizing the spatial multiplexing gain, but it is usually not suitable to place multiple antennas on the user end equipment.

Contents of the invention

Aiming at the pilot pollution problem in the Massive MIMO system, the present invention proposes a pilot allocation method that takes into account both the overall performance of the Massive MIMO system and the fairness among users.

The present invention proposes a pilot allocation scheme based on a greedy strategy by constructing an optimization problem about pilot allocation, which can realize overall system performance and user-to-user interaction while maximizing the overall sum rate of the system and reducing pilot pollution. Effective compromise of fairness.

Summary of the invention:

The basic principle of pilot allocation in the present invention is to use mutually orthogonal pilots in the same cell and multiplex pilots between different cells. The present invention considers the scenario where the number of antennas at the base station tends to be infinite, and in this scenario, the pilot pollution problem is only related to the large-scale fading factor from the user to the base station.

The basic idea of the present invention is to go through all the possibilities of selecting a user in each cell in each iteration, select the combination that makes the user rate product the largest in the user combination as the optimal solution for this iteration, and then obtain the optimal solution from the system Remove the selected users, and continue to iterate until all users are assigned pilots.

Technical scheme of the present invention is:

A pilot allocation method that takes into account the overall performance of a Massive MIMO system and fairness among users. The Massive MIMO system includes L equivalent hexagonal cells, each cell includes a central base station and K users, and a central The base station serves K users at the same time, and the number of pilots available for allocation in the Massive MIMO system is K, and the specific steps include:

(1) According to the distance from the user to the central base station, obtain the large-scale fading information from each user in each cell in the Massive MIMO system to each central base station refers to the large-scale fading factor from user k in cell i to the central base station in cell j;

(2) Construct an objective function for the pilot allocation problem, and the objective function is to maximize the product of all user rates in the Massive MIMO system;

(3) According to the objective function constructed in the step (2), select a user from each sub-district at a time to form a user combination;

(4) from the various user combinations that step (3) obtains, select the user combination that makes the objective function of step (2) maximum as a pilot group;

(5) Remove the user combinations selected in step (4) from the set of unassigned pilots, and return to step (3), until all user combinations are assigned pilots.

Preferably according to the present invention, described step (2), specific steps include:

Construct the objective function, the objective function is shown in formula (I):

In formula (I), the set of users using the same pilot k is a pilot group, denoted as P _k , then there are k pilot groups in the Massive MIMO system, that is, P ₁ , P ₂ ,..., P _K , There are L users in each pilot group;

U _k (P _k ) is the rate product of the pilot group P _k , that is, the product of the rates of all users in the pilot group P _k ,

refers to the large-scale fading factor from user k in cell l to the central base station in cell l; It refers to the large-scale fading factor of user k in cell l to the central base station in cell j.

In order to take into account the overall performance of the system and the fairness among users while reducing pilot pollution, an objective function is constructed to maximize the rate product of the entire system. The optimization objective function can maximize the system and rate while taking into account the fairness among users.

Preferably according to the present invention, described step (3), specific steps include:

A. Set all users participating in pilot allocation in the Massive MIMO system to form a set S;

B. Obtain k! Kinds of user combinations: each user combination includes L users and the L users all come from different communities, and k is the remaining number of users in each community in the set S;

C. Calculate the product of the rates of the users in the k! kinds of user combinations according to formula (I).

Preferably according to the present invention, described step (4), specific steps include:

From the k! user combinations obtained in step (3), the greedy strategy is used to select the user combination that maximizes the product of user rates in the user combination as a pilot group P _{k that shares pilot k} .

Preferably according to the present invention, described step (5), specific steps include:

Remove the users in the pilot group P _k from the set S, update the users in the set S, and return to step (3) until all users are assigned pilots.

The beneficial effects of the present invention are:

The present invention realizes a pilot allocation scheme that takes into account both system effectiveness and fairness by maximizing the product of system user rates. Compared with the existing pilot allocation method, the pilot allocation scheme considered in the present invention not only maximizes the overall performance of the system but also takes into account the fairness among users. At the same time, when solving the optimization problem in the present invention, a pilot allocation method based on a greedy strategy is adopted. The greedy pilot allocation method has lower complexity and is more conducive to realization.

Description of drawings

Fig. 1 is the schematic diagram of embodiment Massive MIMO system model;

Fig. 2 is a flow diagram of a pilot allocation method that takes into account the overall performance of the Massive MIMO system and the fairness among users of the present invention;

FIG. 3 is a simulation effect diagram of an embodiment on fairness;

Fig. 4 is a simulation effect diagram of the embodiment about the overall performance of the system.

detailed description

The present invention will be further limited below in conjunction with the accompanying drawings and embodiments, but not limited thereto.

Example

A pilot allocation method that takes into account the overall performance of the Massive MIMO system and the fairness between users, as shown in Figure 2, the Massive MIMO system, as shown in Figure 1, includes 3 equivalent hexagonal cells, each cell includes 1 central base station and 5 users, and 1 central base station serves 5 users at the same time. The system model diagram briefly describes the configuration of cell users and central base stations in the system. It can be seen from the model diagram that due to the sharing of the same pilot, the base station will receive mutually interfering pilot information at the same time during uplink pilot estimation, and this mutual interference will cause pilot pollution. The specific steps include:

(1) According to the distance from the user to the central base station, obtain the large-scale fading information from each user in each cell in the Massive MIMO system to each central base station refers to the large-scale fading factor from user k in cell i to the central base station in cell j;

(2) Build an objective function for the pilot assignment problem, and the objective function is to maximize the product of all user rates in the Massive MIMO system; the specific steps include:

Construct the objective function, the objective function is shown in formula (I):

In formula (I), the set of users using the same pilot k is a pilot group, denoted as P _k , then there are k pilot groups in the Massive MIMO system, that is, P ₁ , P ₂ ,..., P _K , There are L users in each pilot group;

U _k (P _k ) is the rate product of the pilot group P _k , that is, the product of the rates of all users in the pilot group P _k ,

refers to the large-scale fading factor from user k in cell l to the central base station in cell l; It refers to the large-scale fading factor of user k in cell l to the central base station in cell j.

In order to take into account the overall performance of the system and the fairness among users while reducing pilot pollution, an objective function is constructed to maximize the rate product of the entire system. The optimization objective function can maximize the system and rate while taking into account the fairness among users.

(3) According to the objective function constructed in step (2), a user is selected from each community at a time to form a user group; the specific steps include:

A. Set all users participating in pilot allocation in the Massive MIMO system to form a set S;

B. Get k! A user combination: each user combination includes L users and the L users are from different cells, and k is the number of users remaining in each cell in the set S;

C. Calculate the product of the rates of the users in the k! kinds of user combinations according to formula (I).

(4) From the multiple user combinations obtained in step (3), select the user combination that maximizes the objective function of step (2) as a pilot group; use the greedy strategy to obtain the k! The user combination that maximizes the product of the user rates in the user combination is selected from among the user combinations as a pilot group P _{k that shares the pilot k} .

(5) Remove the selected user combination in step (4) from the set of unassigned pilots, return to step (3), until all user combinations are assigned pilots: the pilot group P _k The user is removed from the set S, the users in the set S are updated, and the step (3) is returned until all users are assigned pilots.

Figures 3 and 4 show the effect diagrams of the present embodiment that take into account the overall performance of the system and the fairness among users. From the results shown in Figure 3 and Figure 4, when designing the pilot allocation scheme, when the optimization goal is to maximize the sum rate of the system, the system has a better sum rate but the maximum user rate and the minimum user rate in the system The difference between the user rates is very large, and the fairness of the system is poor; when the optimization goal is to maximize the minimum rate, the difference between the maximum user rate and the minimum user rate in the system is effectively reduced, but the system and The rate is greatly reduced; when the optimization goal is to maximize the user product, the system not only has better overall performance but also takes into account the effectiveness. Therefore, the pilot allocation scheme proposed by the present invention can effectively take into account the overall performance of the system and the fairness among users.

Claims (5)

1. a kind of pilot distribution method for taking into account fairness between Massive mimo systems overall performance and user, described Massive mimo systems include L equivalent hexagonal cell, and each cell includes 1 center base station and K user, 1 Center base station services K user simultaneously, and distributable pilot number is K in the Massive mimo systems, and its feature exists In specific steps include：

(1) according to user to center base station distance, each user in Massive mimo systems in each cell is asked in each The large scale fading information of heart base station Refer to user k to the j in i cells The large scale fading factor of the center base station in cell；

(2) object function is built to pilot tone assignment problem, object function is all users in maximization Massive mimo systems The product of speed；

(3) according to the object function built in step (2), a user is selected from each cell every time, constitutes a user Combination；

(4) in the various users combination for being obtained from step (3), the user's combination for causing the object function of step (2) maximum is selected As a pilot group；

(5) will remove in the never set of distribution pilot tone of the user for the selecting combination in step (4), return to step (3), Zhi Daosuo Some user's combinations are all allocated pilot tone.

2. a kind of pilot tone for taking into account fairness between Massive mimo systems overall performance and user according to claim 1 Distribution method, it is characterised in that the step (2), specific steps include：

Object function is built, shown in object function such as formula (I)：

$\max U\left(P\right)=\max {\Π}_{k=1}^K{U}_k\left(P_k\right)---\left(I\right)$

In formula (I), a pilot group is combined into using the collection of the user of same pilot k, is denoted as P_k, then in Massive mimo systems Total k pilot group, i.e. P₁, P₂..., P_K, there is L user in each pilot group；

U_k(P_k) it is pilot group P_kSpeed product, i.e. pilot group P_kThe product of the speed of interior all users,

It refer to the large scale fading factor of the center base station in user's k to the l cells in l cells；In referring to l cells User's k to j cells in center base station large scale fading factor.

3. a kind of pilot tone for taking into account fairness between Massive mimo systems overall performance and user according to claim 2 Distribution method, it is characterised in that the step (3), specific steps include：

All users for participating in pilot tone distribution constitute a set S in A, setting Massive mimo systems；

B, acquisition k！Plant user's combination：Every kind of user's combination includes L user and the L user is all from different cells, and k is Remaining number of users in each cell in set S；

C, according to formula (I) calculate k!The product of the speed of the user in kind user's combination.

4. a kind of pilot tone for taking into account fairness between Massive mimo systems overall performance and user according to claim 3 Distribution method, it is characterised in that the step (4), specific steps include：

The k obtained from step (3) using greedy strategy！Plant and the user rate caused in user's combination is selected during user combines The maximum user's group cooperation of product is a pilot group P of common pilot k_k。

5. a kind of pilot tone for taking into account fairness between Massive mimo systems overall performance and user according to claim 4 Distribution method, it is characterised in that the step (5), specific steps include：

By pilot group P_kIn user removed from set S, update the user in set S, return to step (3) is until all of use Untill family distributes pilot tone.