CN107888367B - Dynamic pilot frequency distribution method based on user classification in large-scale MIMO system - Google Patents

Abstract

The invention relates to the field of mobile communication, in particular to a dynamic pilot frequency allocation method based on user classification in a large-scale MIMO system, which comprises the following steps: the method comprises the steps of averagely dividing an orthogonal pilot frequency set provided by a system into three groups; respectively calculating the signal intensity of each user in three adjacent cells; dividing the users of the three adjacent cells into edge users and center users according to the signal intensity of each user; distributing mutually orthogonal pilot frequency groups to the edge users of the three adjacent cells respectively, and distributing another two pilot frequency groups which are not distributed in the edge users of the cell to the center users of the three adjacent cells respectively; the pilot frequency distribution scheme designed by the invention reduces the orthogonal pilot frequency overhead by improving the traditional soft pilot frequency multiplexing scheme, simultaneously reduces the inter-cell interference generated by the central users of the adjacent cells to the edge users of the target cell, and improves the uplink and downlink SINR of the edge users and the channel capacity of the system.

Description

Dynamic pilot frequency distribution method based on user classification in large-scale MIMO system

Technical Field

The invention relates to the field of mobile communication, in particular to a dynamic pilot frequency distribution method based on user classification in a large-scale Multiple-input Multiple-output (MIMO) system.

Background

The pilot pollution problem is not only present in Massive MIMO (Massive MIMO) systems, but is a common phenomenon in communication systems, both in Code Division Multiple Access (CDMA) systems and L TE-Advanced coordinated multipoint (COMP), in conventional MIMO systems, the pilot pollution problem is also present, but its impact on Massive MIMO system communication is much greater than that of conventional MIMO communication, because in conventional MIMO systems, the communication process is usually interfered by other users inside the cell and interference between cells (pilot pollution), so in conventional MIMO systems, reducing intra-cell interference is an effective way to improve system capacity and spectral efficiency, but in Massive MIMO, due to the large number of antennas of the base station, the channels of different users in the same cell of the system tend to be orthogonal, so the interference between different users in the same cell can be ignored, so that the main factors limiting the system capacity spectral efficiency in Massive MIMO systems are different small-cell interference factors, and the pilot pollution problem has been thoroughly investigated for several years.

The pilot frequency distribution strategy is to maintain the existing pilot frequency sequence group of the system unchanged and redistribute the pilot frequency distribution situation in the system, thereby reducing the degree of pilot frequency pollution in the system and the influence of the pilot frequency pollution on the communication process of the system as much as possible. The pilot frequency distribution scheme capable of utilizing the pilot frequency time shift to reduce the pilot frequency pollution effectively restrains the pilot frequency pollution under different cell synchronous transmission schemes by aligning the pilot frequency time slot in a certain cell to the data time slot of the adjacent cell. Another method for reducing pilot pollution is a pilot distribution scheme based on sector classification, and user pilots in different sectors are mutually independent in physical space, so that the pilot pollution is effectively reduced, and when the number of base station antennas is large enough, the system throughput is greatly improved. It can also be considered that the same pilot frequency is distributed to the users in the center of the cell, and the orthogonal pilot frequency is distributed to the users at the edge of the cell, thereby reducing the pilot frequency pollution and improving the capacity of the system.

The pilot allocation strategy is mainly to design a reasonable allocation scheme according to the signal strength of the users and the inter-cell interference strength (pilot pollution) to allocate the pilots to the users. The main problems in the current pilot allocation strategy are as follows:

(1) current pilot allocation strategies based on user classification generally take the following three approaches when classifying users:

a) dividing users according to the distance from the users to the base station;

b) setting a fixed value according to the signal intensity of the users to classify the users in the cell;

c) classifying users according to the signal intensity difference of the useful signals and the interference signals of the users;

although the former two ways are simple and easy to implement, the classification standard cannot be dynamically adjusted according to the distribution condition of the user in the cell, so that the classification accuracy is not high; the third classification method has high accuracy, but needs inter-cell cooperation, which increases the system overhead.

(2) The pilot allocation algorithm adopted by the literature in the aspect of the current pilot allocation strategy has the following two disadvantages:

a) based on the algorithm of system capacity maximization, the whole system capacity is improved, but the communication performance of edge users is still poor.

b) The algorithm based on the user fairness criterion improves the communication performance of the edge users but cannot effectively improve the overall system capacity.

(3) Orthogonal pilot overhead problem for partial pilot multiplexing schemes

In the partial pilot multiplexing scheme (the pilot multiplexing factor of the central user is 1, and the pilot multiplexing factor of the edge user is 3), although the overall performance of the system is improved, the number of required orthogonal pilots is greatly increased. Due to the system coherence time limitation, the number of orthogonal pilots provided by the system is limited.

Disclosure of Invention

In view of this, the present invention is directed to solve the problem of ensuring better communication performance of edge users while the overall system capacity is larger in the process of pilot allocation in a massive MIMO system. The invention provides a dynamic pilot frequency distribution method based on user classification in a large-scale MIMO system, which effectively improves the system capacity on the basis of ensuring the communication performance of edge users.

The invention aims to realize the following technical scheme, and the dynamic pilot frequency allocation method based on user classification in the large-scale MIMO system comprises the following steps:

s1: the orthogonal pilot frequency set psi provided by the system is divided into three groups phi₁、Φ₂、Φ₃；

S2: respectively calculating the signal intensity of each user in three adjacent cells; sorting the three adjacent cells according to the descending order of the mean values of the central user signal intensities of the three adjacent cells; the mean values of the signal strengths of the central users of the first cell are respectively

l＝1,2,3，

S3: according to the descending order of the signal intensity of the users, dividing the users of the ith cell into edge users and center users respectively;

s4: respectively allocating a pilot frequency group phi to the edge users of the ith cell₁、Φ₂、Φ₃And respectively allocating { phi ] to users in the center of the ith cell₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂I.e. the edge user of the 1 st cell is allocated with a pilot group phi₁Assigning a pilot group phi to edge users of the 2 nd cell₂Assigning a pilot group phi to edge users of the 3 rd cell₃Then, the central user of the 1 st cell is allocated with a pilot frequency group { phi₂，Φ₃Center user of 2 nd cell allocates pilot frequency group { phi }₁，Φ₃Allocating pilot frequency group { phi ] to central user of 3 rd cell₁，Φ₂}。

Further, the orthogonal pilot frequency set psi provided by the system is averagely divided into three groups phi₁、Φ₂、Φ₃The method comprises the following steps: will be provided withOrthogonal pilot sequence composition set psi ═ { psi ═ provided by system₁,ψ₂,......,ψ_NThe method is divided into three groups, which specifically comprise the following steps:

Φ₁＝{ψ₁,ψ₂,...,ψ[_N/3]}

Φ₂＝{ψ_[N/3]+1,ψ_[N/3]+2,...,ψ_2×[N/3]}

Φ₃＝{ψ_2×[N/3]+1,ψ_2×[N/3]+2,...,ψ_3×[N/3]}

wherein psi_i(i-1, 2 … … N) represents an orthogonal pilot sequence, N is the number of orthogonal pilots that the system can provide, [ X []Represents the rounding function of X.

Further, the calculating the signal strength of each user in the three adjacent cells respectively comprises β according to the large-scale fading factor from the kth user in the ith cell to the base station of the ith cell_l,l,kCalculating the signal strength of the kth user:

calculating the average value of the signal intensity of the central user in the ith cell according to the signal intensity of each central user in the ith cell respectively:

wherein, l ═ 1,2,3}, k ═ 1,2,... M }, S ═ 1,2_l,iRepresents the signal strength of the ith central user in the cell, M3. N/3]M is the number of users in each cell, N is the number of orthogonal pilots that the system can provide, [ X ]]Represents the rounding function of X.

Further, the dividing users of the ith cell into edge users and center users according to the decreasing order of the signal strength of the users specifically includes: for each cell user, according to the user signal strength S_l,kThe users are sorted in descending order, where l is {1,2,3}, k is {1,2_l＝{u_l,1,u_l,2......u_l,MGet the user set U_lFront of2M/3 item composition center user set

The last M/3 items form an edge user set

M is the number of users per cell.

Further, distributing pilot frequency group phi to edge users of three adjacent cells₁、Φ₂、Φ₃The method specifically comprises the following steps:

distributing pilot frequency in pilot frequency group to the edge user by random distribution mode, and distributing the pilot frequency group phi₁、Φ₂、Φ₃Edge user set sequentially allocated to the l cell

Namely: pilot group phi₁Edge user set assigned to 1 st cell

Pilot group phi₂Edge user set assigned to cell 2

Pilot group phi₃Edge user set assigned to 3 rd cell

Preferably, the central users of the three neighboring cells are respectively assigned { Φ }₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂The method comprises the following steps: according to the mean value of the signal intensity of the central users of the three adjacent cells

In descending order, allocating pilot frequency to the central users of the l cell in sequence;

concrete bagComprises the following steps: preferentially distributing pilot frequency groups to the central users of the 1 st cell, and respectively calculating the average value of the signal intensity of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell

Wherein S is_l,iRepresents the signal strength of the ith central user in the ith cell, i ═ 1, 2. The signal strength of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell are respectively

If it is

The pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 3 rd cell₃Additional central users { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 2 nd cell₂Center user { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}；

Further, the pilot group allocated to the center user in the 2 nd cell is: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 2 u_2,1,u_2,2......u_2,M/3And then, allocating the edge users of the 3 rd cellPilot group phi of₃Additional central users { u } assigned to cell 2_2,M/3+1,u_2,M/3+2......u_2,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Pilot frequency is allocated to center user { u } of 2 nd cell_2,1,u_2,2......u_2,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 1 st cell₁Center user allocated to cell 2 u_2,M/3+1,u_2,M/3+2......u_2,2M/3}；

Further, the pilot group allocated to the center user in the 3 rd cell is: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 3 u_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 2 nd cell₂Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Pilot frequency is allocated to center user { u } of 2 nd cell_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 1 st cell₁Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}。

Further, the allocating pilot frequencies to the central users of the 1 st cell, the 2 nd cell, and the 3 rd cell in sequence specifically includes:

according to the pilot pollution intensity formula among users:

obtaining a pilot pollution intensity matrix, and selecting the element with the maximum value in each column of the matrix in turn

I.e. the kth cell in the L th cell_eFor at one edgePilot frequency distributed by user is distributed to kth cell_cA central user;

wherein the content of the first and second substances,

denotes the kth cell_cThe central user and the kth L th cell_eThe strength of the pilot pollution between the individual edge users,

denotes the kth cell_cThe large scale fading factor of the center user to the base station of the l-th cell,

denotes the kth cell_cThe large scale fading factor of the individual center users to the L th cell base station,

denotes the kth cell in the L th cell_eThe large scale fading factor of the individual edge users to the L th cell base station,<l,k_e>denotes the kth cell_cThe number of the central users is such that,<L,k_c>denotes the kth cell in the L th cell_eAn edge user, 1,2,3, L, 1,2,3, l ≠ L.

Further, the pilot pollution intensity matrix includes:

wherein H_lc,LeA pilot pollution strength matrix representing the center user of the ith cell and the edge users of the L th cell.

The invention has the beneficial effects that: compared with the traditional partial pilot frequency multiplexing scheme (the pilot frequency multiplexing factor of a central user is 1, and the pilot frequency multiplexing factor of an edge user is 3), the dynamic pilot frequency allocation scheme adopted by the invention has the advantages that the pilot frequency multiplexing factor among the cells is 1, and the orthogonal pilot frequency overhead is effectively reduced; according to the scheme, the users are classified according to the signal intensity of the users in each cell, so that the calculation complexity is reduced, and meanwhile, the accuracy of user classification is improved; optimizing the pilot frequency distribution sequence according to the signal intensity of the central user and the signal intensity of the edge user of each cell, namely firstly distributing the pilot frequency to the central user of the cell with the larger average signal intensity of the central user; meanwhile, the distributed pilot frequency groups in the cell with smaller average signal intensity of the edge users are distributed to the central users of the adjacent cells, and the pilot frequency distribution sequence is optimized, so that the inter-cell interference of the central users of the adjacent cells to the edge users of the target cell is effectively reduced, the uplink and downlink SINR of the edge users is improved, and the channel capacity of the system is also effectively improved.

Drawings

Fig. 1 is a flowchart illustrating a dynamic pilot allocation method based on user classification according to the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more clearly and completely apparent, the technical solutions in the embodiments of the present invention are described below with reference to the accompanying drawings, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

A dynamic pilot allocation method based on user classification in a massive MIMO system, as shown in fig. 1, includes:

s1: the orthogonal pilot frequency set psi provided by the system is divided into three groups phi₁、Φ₂、Φ₃；

S2: respectively calculating the signal intensity of each user in three adjacent cells; sorting the three adjacent cells according to the descending order of the mean values of the central user signal intensities of the three adjacent cells; the mean values of the signal strengths of the central users of the first cell are respectively

S3: according to the descending order of the signal intensity of the users, dividing the users of the ith cell into edge users and center users respectively;

s4: respectively allocating a pilot frequency group phi to the edge users of the ith cell₁、Φ₂、Φ₃And respectively allocating { phi ] to users in the center of the ith cell₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂I.e. the edge user of the 1 st cell is allocated with a pilot group phi₁Assigning a pilot group phi to edge users of the 2 nd cell₂Assigning a pilot group phi to edge users of the 3 rd cell₃Then, the central user of the 1 st cell is allocated with a pilot frequency group { phi₂，Φ₃Center user of 2 nd cell allocates pilot frequency group { phi }₁，Φ₃Allocating pilot frequency group { phi ] to central user of 3 rd cell₁，Φ₂}。

The invention optimizes the pilot frequency distribution sequence according to the signal intensity of the central users and the signal intensity of the edge users of each cell, namely, the pilot frequency is distributed to the central users of the cell with the larger average signal intensity of the central users; meanwhile, the distributed pilot frequency groups in the cell with smaller average signal intensity of the edge users are distributed to the central users of the adjacent cells, and the pilot frequency distribution sequence is optimized, so that the inter-cell interference of the central users of the adjacent cells to the edge users of the target cell is effectively reduced, the uplink and downlink SINR of the edge users is improved, and the channel capacity of the system is also effectively improved.

Further, the orthogonal pilot frequency set psi provided by the system is averagely divided into three groups phi₁、Φ₂、Φ₃The method comprises the following steps: the orthogonal pilot sequences provided by the system are combined into a set psi ═ { psi ═ phi₁,ψ₂,......,ψ_NThe method is divided into three groups, which specifically comprise the following steps:

Φ₁＝{ψ₁,ψ₂,...,ψ_[N/3]}

Φ₂＝{ψ_[N/3]+1,ψ_[N/3]+2,...,ψ_2×[N/3]}

Φ₃＝{ψ_2×[N/3]+1,ψ_2×[N/3]+2,...,ψ_3×[N/3]}

wherein psi_i(i-1, 2 … … N) represents an orthogonal pilot sequence, N is the number of orthogonal pilots that the system can provide, [ X []Represents the rounding function of X.

Further, the calculating the signal strength of each user in the three adjacent cells respectively comprises β according to the large-scale fading factor from the kth user in the ith cell to the base station of the ith cell_l,l,kCalculating the signal strength of the kth user:

calculating the average value of the signal intensity of the central user in the ith cell according to the signal intensity of each central user in the ith cell respectively:

wherein, l ═ 1,2,3}, k ═ 1,2,... M }, S ═ 1,2_l,iRepresents the signal strength of the ith central user in the cell, M3. N/3]M is the number of users in each cell, N is the number of orthogonal pilots that the system can provide, [ X ]]Represents the rounding function of X.

Further, dividing the users in the ith cell into edge users and center users according to the decreasing order of the signal strength of the users specifically includes: for each cell user, according to the user signal strength S_l,kThe users are sorted in descending order, where l is {1,2,3}, k is {1,2_l＝{u_l,1,u_l,2......u_l,MGet the user set U_lThe first 2M/3 items of (A) constitute a central user set

The last M/3 items form an edge user set

M is the number of users per cell.

Further, distributing pilot frequency group phi to edge users of three adjacent cells₁、Φ₂、Φ₃The method specifically comprises the following steps:

distributing pilot frequency in pilot frequency group to the edge user by random distribution mode, and distributing the pilot frequency group phi₁、Φ₂、Φ₃Edge user set sequentially allocated to the l cell

I.e. the pilot group phi₁Edge user set assigned to 1 st cell

Pilot group phi₂Edge user set assigned to cell 2

Pilot group phi₃Edge user set assigned to 3 rd cell

Preferably, the central users of the three neighboring cells are respectively assigned { Φ }₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂The method comprises the following steps: according to the mean value of the signal intensity of the central users of the three adjacent cells

In descending order, allocating pilot frequency to the central users of the l cell in sequence;

the method specifically comprises the following steps: preferentially distributing pilot frequency groups to the central users of the 1 st cell, and respectively calculating the average value of the signal intensity of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell

Wherein S is_l,iSignals representing the ith central user in the ith cellIntensity, i ═ 1, 2.... 2M/3 }; the signal strength of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell are respectively

If it is

The pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 3 rd cell₃Additional central users { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 2 nd cell₂Center user { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}；

Further, the pilot group allocated to the center user in the 2 nd cell is: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 2 u_2,1,u_2,2......u_2,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 3 rd cell₃Additional central users { u } assigned to cell 2_2,M/3+1,u_2,M/3+2......u_2,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Pilot frequency is allocated to center user { u } of 2 nd cell_2,1,u_2,2......u_2,M/3And then the edge users of the 1 st cell are allocatedPilot group phi₁Center user allocated to cell 2 u_2,M/3+1,u_2,M/3+2......u_2,2M/3}；

Further, the pilot group allocated to the center user in the 3 rd cell is: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 3 u_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 2 nd cell₂Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Pilot frequency is allocated to center user { u } of 2 nd cell_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 1 st cell₁Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}。

Further, the allocating pilot frequencies to the central users of the 1 st cell, the 2 nd cell, and the 3 rd cell in sequence specifically includes:

according to the pilot pollution intensity formula among users:

obtaining a pilot pollution intensity matrix, and selecting the element with the maximum value in each column of the matrix in turn

I.e. the kth cell in the L th cell_eThe pilot frequency distributed by each edge user is distributed to the kth cell_cA central user, wherein,

denotes the kth cell_cThe central user and the kth L th cell_eThe strength of the pilot pollution between the individual edge users,

denotes the kth cell_cThe large scale fading factor of the center user to the base station of the l-th cell,

denotes the kth cell_cThe large scale fading factor of the individual center users to the L th cell base station,

denotes the kth cell in the L th cell_eThe large scale fading factor of the individual edge users to the L th cell base station,<l,k_e>denotes the kth cell_cThe number of the central users is such that,<L,k_c>denotes the kth cell in the L th cell_eAn edge user, 1,2,3, L, 1,2,3, l ≠ L.

Further, the pilot pollution intensity matrix includes:

wherein H_lc,LeA pilot pollution strength matrix representing the center user of the ith cell and the edge users of the L th cell.

Preferably, if

Respectively calculating the kth cell in the 2 nd cell_eLarge scale fading factor of edge user to base station of said 2 nd cell

And kth cell in 2 nd cell_cLarge scale fading factor of individual central users to base station of 1 st cell

Kth in 1 st cell_cThe size of the central user to the base station of the 1 st cellDegree fading factor

And kth cell in 1 st cell_cLarge scale fading factor of individual central users to base station of 2 nd cell

Respectively calculating the pilot pollution intensity between every two users, and finally calculating to obtain a pilot pollution intensity matrix between the edge users of the 2 nd cell and the center users of the 1 st cell:

wherein, the column of the matrix represents the edge user, the row of the matrix represents the center user, and the column number of the matrix is according to the signal intensity of the edge user in the 2 nd cell

The increasing order increasing in turn, i.e.

Selecting the element with the largest column in turn

2 nd cell_eThe pilot frequency distributed by each edge user is distributed to the kth cell in the 1 st cell_cA central user until there is { u }_1,1,u_1,2......u_1,M/3Central users are allocated to obtain pilot frequency; according to the pilot frequency distribution situation of 3 rd cell edge users to another { u } in 1 st cell_1,M/3+1,u_1,M/3+2......u_1,2M/3Pilot frequencies are allocated to the central users.

The pilot frequency distribution scheme of the invention can fully consider the benefits of cooperation among base stations on pilot frequency distribution, and effectively solves the problem of high pilot frequency overhead in the pilot frequency distribution scheme. By improving the traditional soft pilot frequency multiplexing scheme, the orthogonal pilot frequency overhead is reduced, the inter-cell interference generated by the central user of the adjacent cell to the edge user of the target cell is reduced, and the uplink and downlink SINR of the edge user and the channel capacity of the system are improved.

The method is simple and practical, has low algorithm complexity, and has good application prospect and economic benefit.

The above-mentioned embodiments, which further illustrate the objects, technical solutions and advantages of the present invention, should be understood that the above-mentioned embodiments are only preferred embodiments of the present invention, and should not be construed as limiting the present invention, and any modifications, equivalents, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (8)

1. A dynamic pilot frequency distribution method based on user classification in a large-scale MIMO system is characterized by comprising the following steps:

s1: the orthogonal pilot frequency set psi provided by the system is divided into three groups phi₁、Φ₂、Φ₃；

S2: respectively calculating the signal intensity of each user in three adjacent cells; sorting the three adjacent cells according to the descending order of the mean values of the central user signal intensities of the three adjacent cells; wherein, the average values of the signal intensity of the central users in the first cell are respectively

S3: according to the descending order of the signal intensity of the users, the user at the front 2/3 in the descending order of the signal intensity in the ith cell is taken as a center user, and the user at the back 1/3 is taken as an edge user;

s4: respectively allocating a pilot frequency group phi to the edge users of the ith cell₁、Φ₂、Φ₃And respectively allocating { phi ] to users in the center of the ith cell₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂Calculating the signal intensity of edge users of three adjacent cells respectively; distributing the pilot frequency groups distributed by the edge users in two adjacent cells to the central user of the first cell; i.e. the edge users of cell 1 are assigned a pilot group phi₁Assigning a pilot group phi to edge users of the 2 nd cell₂Assigning a pilot group phi to edge users of the 3 rd cell₃Then, the pilot frequency group allocated by the edge users of the 2 nd cell and the 3 rd cell is allocated to the center user of the 1 st cell { phi₂，Φ₃Allocating pilot frequency groups (phi) allocated by edge users of the 1 st cell and the 3 rd cell to the center user of the 2 nd cell₁，Φ₃Allocating pilot frequency groups (phi) allocated by edge users of the 1 st cell and the 2 nd cell to the center user of the 3 rd cell₁，Φ₂}; the order of allocating pilot frequency to the central user of the l cell is to judge the signal intensity of the edge users in two adjacent cells, and allocate the pilot frequency groups allocated to the edge users in the two adjacent cells to the central user of the l cell in turn according to the ascending order of the edge signal intensity.

2. The method as claimed in claim 1, wherein the orthogonal pilot set Ψ provided by the system is divided into three groups Φ₁、Φ₂、Φ₃The method comprises the following steps: the orthogonal pilot sequences provided by the system are combined into a set psi ═ { psi ═ phi₁,ψ₂,......,ψ_NThe method is divided into three groups, which specifically comprise the following steps:

Φ₁＝{ψ₁,ψ₂,...,ψ_[N/3]}

Φ₂＝{ψ_[N/3]+1,ψ_[N/3]+2,...,ψ_2×[N/3]}

Φ₃＝{ψ_2×[N/3]+1,ψ_2×[N/3]+2,...,ψ_3×[N/3]}

wherein psi_i(i-1, 2 … … N) represents an orthogonal pilot sequence, where N is a number that can be extracted by the systemSupply of orthogonal pilot number, [ X ]]Represents the rounding function of X.

3. The method of claim 1, wherein the step of calculating the signal strength of each user in three neighboring cells comprises β according to the large-scale fading factor from the kth user in the ith cell to the base station of the ith cell_l,l,kCalculating the signal strength of the kth user in the ith cell:

calculating the average value of the signal intensity of the central user in the ith cell according to the signal intensity of each central user in the ith cell respectively:

where l ═ 1,2,3, i ═ 1,2,.. 2M/3, S_l,iRepresents the signal strength of the ith central user in the ith cell, M3. N/3]M is the number of users in each cell, N is the number of orthogonal pilots that the system can provide, [ X ]]Represents the rounding function of X.

4. The method of claim 1, wherein the step of using the front 2/3 user as the center user and the rear 1/3 user as the edge user in the descending order of signal strength in the ith cell according to the descending order of signal strength of users specifically comprises: for each cell user, according to the user signal strength S_l,kSequencing the users in descending order, and obtaining M user sets of the first cell as U_l＝{u_l,1,u_l,2...u_l,k...u_l,M}；

Wherein u is_l,kDenotes the kth user of the l cell, and S_l,k≥S_l,k+1，S_l,kIndicating the kth central user in the l cellNumber intensity, S_l,k+1Represents the signal strength of the (k + 1) th central user in the l cell; taking the user set U as {1,2,3}, and k as {1,2, … …, M }, respectively_lThe first 2M/3 items of (A) constitute a central user set

The last M/3 items form an edge user set

M is the number of users per cell.

5. The dynamic pilot allocation method based on user classification in massive MIMO system as claimed in claim 1 wherein, the pilot group Φ is allocated to the edge users of three adjacent cells respectively₁、Φ₂、Φ₃The method comprises the following steps: distributing pilot frequency in pilot frequency group to the edge user by random distribution mode, and distributing the pilot frequency group phi₁、Φ₂、Φ₃Edge user set sequentially allocated to the l cell

Namely: pilot group phi₁Edge user set assigned to 1 st cell

Pilot group phi₂Edge user set assigned to cell 2

Pilot group phi₃Edge user set assigned to 3 rd cell

6. The dynamic pilot based on user classification in massive MIMO system as claimed in claim 1An allocation method, characterized in that the central users of said three neighbouring cells are allocated { Φ } respectively₂，Φ₃}、{Φ₁，Φ₃And phi₁，Φ₂The method comprises the following steps: according to the mean value of the signal intensity of the central users of the three adjacent cells

In descending order, allocating pilot frequency to the central users of the l cell in sequence;

the method specifically comprises the following steps: preferentially distributing pilot frequency groups to the central users of the 1 st cell, and respectively calculating the average value of the signal intensity of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell

Wherein S is_l,iRepresents the signal strength of the ith central user in the ith cell, i ═ 1, 2. The signal strength of the edge users of the 1 st cell, the 2 nd cell and the 3 rd cell are respectively

If it is

The pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 3 rd cell₃Additional central users { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Center user { u } assigned to cell 1_1,1,u_1,2......u_1,M/3And then dividing the edge users of the 2 nd cell intoMatched pilot group phi₂Center user { u } assigned to cell 1_1,M/3+1,u_1,M/3+2......u_1,2M/3}；

The pilot group is allocated to the center user of the 2 nd cell as follows: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 2 u_2,1,u_2,2......u_2,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 3 rd cell₃Additional central users { u } assigned to cell 2_2,M/3+1,u_2,M/3+2......u_2,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge user of the 3 rd cell₃Pilot frequency is allocated to center user { u } of 2 nd cell_2,1,u_2,2......u_2,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 1 st cell₁Center user allocated to cell 2 u_2,M/3+1,u_2,M/3+2......u_2,2M/3}；

The pilot group is allocated to the center user of the 3 rd cell as follows: if it is

The pilot group phi preferentially allocated to the edge users of the 1 st cell₁Center user allocated to cell 3 u_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 2 nd cell₂Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}; otherwise, the pilot group phi preferentially allocated to the edge users of the 2 nd cell₂Pilot frequency is allocated to center user { u } of 2 nd cell_3,1,u_3,2......u_3,M/3Fourthly, the pilot frequency group phi distributed to the edge user of the 1 st cell₁Center user allocated to cell 3 u_3,M/3+1,u_3,M/3+2......u_3,2M/3}。

7. The dynamic pilot allocation method based on user classification in the massive MIMO system according to claim 6, wherein the allocating pilots to the central users of the 1 st cell, the 2 nd cell, and the 3 rd cell in sequence specifically comprises:

according to the pilot pollution intensity formula among users:

obtaining a pilot pollution intensity matrix, and selecting the element with the maximum value in each column of the matrix in turn, namely the kth element in the L th cell_eThe pilot frequency distributed by each edge user is distributed to the kth cell_cA central user;

wherein the content of the first and second substances,

denotes the kth cell_cThe central user and the kth L th cell_eThe strength of the pilot pollution between the individual edge users,

denotes the kth cell_cThe large scale fading factor of the center user to the base station of the l-th cell,

denotes the kth cell_cThe large scale fading factor of the individual center users to the L th cell base station,

denotes the kth cell in the L th cell_eThe large scale fading factor of the individual edge users to the L th cell base station,<l,k_e>denotes the kth cell_cThe number of the central users is such that,<L,k_c>denotes the kth cell in the L th cell_eAn edge user, 1,2,3, L, 1,2,3, l ≠ L.

8. The method of claim 6, wherein the pilot pollution strength matrix comprises:

wherein H_lc,LeA pilot pollution strength matrix representing the center user of the ith cell and the edge users of the L th cell.

Images