CN108924934B

CN108924934B - Heterogeneous network interference management method based on multi-dimensional resource allocation

Info

Publication number: CN108924934B
Application number: CN201810738168.0A
Authority: CN
Inventors: 赵林靖; 刘亚楠; 张岗山; 李钊; 张顺
Original assignee: Xidian University
Current assignee: Xidian University
Priority date: 2018-07-06
Filing date: 2018-07-06
Publication date: 2021-08-20
Anticipated expiration: 2038-07-06
Also published as: CN108924934A

Abstract

The invention provides a heterogeneous network interference management method based on multi-dimensional resource allocation, which mainly solves the problem of large cross-layer interference in a heterogeneous network in the prior art. The implementation steps are as follows: 1) initializing a heterogeneous network; 2) a user initializes an associated base station according to Max RSSI (received signal strength indicator) criteria; 3) the macro base station allocates an RB orthogonal to the small base station user for the macro user which is strongly related to the small base station user; 4) the small base station groups and allocates RBs for the associated users, and sends CSI, user grouping and RB allocation results from the small base station users to the macro base station; 5) the macro base station selects macro users orthogonal to the small base station users, and allocates the same RB as the small base station users to the macro users; 6) judging whether the algorithm is finished or not, and if not, executing the step 7); 7) the base station adopts the correlation control to adjust the user grouping and RB allocation; 8) judging whether the algorithm is finished or not, if eta is less than eta_θLet η be η + Δ η, go to step 3). The invention reduces cross-layer interference and interaction overhead.

Description

Heterogeneous network interference management method based on multi-dimensional resource allocation

Technical Field

The invention belongs to the technical field of wireless communication, relates to an interference management method for a heterogeneous network, in particular to an interference management method based on multi-dimensional resource allocation under partial channel state information in an orthogonal frequency division multiple access OFDMA network, and can be used for an uplink of the heterogeneous wireless network with a macro base station and a small base station coexisting.

Background

In recent years, rapid development of wireless communication technology and urgent needs of users for information communication, office work or entertainment with other people anytime and anywhere promote large-scale popularization of intelligent terminal equipment, and meanwhile, the demand of users for data traffic is rapidly increased. In order to improve network throughput, an operator deploys a plurality of small base stations in a coverage area of a macro base station to form a heterogeneous wireless network in which the macro base station and the small base stations coexist.

The introduction of small base stations, while providing benefits to heterogeneous wireless networks, also presents challenges. In order to improve the spectrum utilization rate, the small base station reuses the spectrum resources of the macro base station, and the coverage ranges of the small base station and the macro base station are overlapped, so that serious cross-layer interference exists in the network. Because the coverage range difference between the macro base station and the small base station is large, and the uplink and the downlink of the heterogeneous wireless network have asymmetry, the macro user can only obtain part of channel state information CSI of the adjacent small base station. The traditional heterogeneous network interference management method comprises the following steps: the method comprises a time domain enhanced inter-cell interference coordination technology eICIC, a frequency domain partial frequency reuse technology FFR, a space domain beam forming technology, power control, association control and a multidimensional resource allocation technology. The multidimensional resource allocation technology combines a plurality of single-dimensional interference management technologies, and performs joint resource allocation for users from a plurality of dimensions, and a common multidimensional resource allocation technology includes: the method comprises the steps of combining frequency domain resource allocation with power control, combining frequency domain resource allocation with association control, allocating time-frequency joint resources and allocating space-frequency joint resources.

Under the condition of dense heterogeneous wireless network users, the problems of shortage of spectrum resources, high correlation of partial user channels and the like exist, so that the problem of interference cannot be solved from a single dimension. In the related research of OFDMA, a base station carries out orthogonal grouping on users and reasonably allocates resources of a space frequency domain according to a grouping result, so that the frequency spectrum utilization rate can be effectively improved, the interference among the users can be reduced, and the network throughput can be improved. Pao W C, Lou W T, Chen Y F, et al. resource allocation for multiple input multiple output-orthogonal frequency division multiple access systems [ J ]. Iet Communications,2014,8(18): 3424-. The method comprises the following implementation steps: the base station compares the channel correlation among the users with an orthogonal decision threshold, divides the users with orthogonal channels into a group, allocates different space domain resources of the same RB to different users in a user group, and allocates one RB to only one user group. The method improves the network throughput, but does not consider the interference of macro users to the small base station and the interference of small base station users to the macro base station, can not effectively reduce cross-layer interference, and has limited improved network throughput.

Disclosure of Invention

Aiming at the defects of the prior art, the invention provides a heterogeneous network interference management method based on multi-dimensional resource allocation, which solves the problem of large cross-layer interference in the prior art and improves the network throughput.

The technical idea of the invention is as follows: the macro base station and the small base station cooperate to perform joint optimization on allocation of space domain resources, frequency domain resources and associated base stations, the macro base station and the small base station cooperate to perform orthogonal grouping and RB allocation on users, then associated base stations and user groups of the users which do not meet requirements are changed by using association control, RB allocation is adjusted, and cross-layer interference in the network is reduced from multiple dimensions. The method comprises the following implementation steps:

(1) initializing the heterogeneous network:

the method comprises the steps that a base station set in a heterogeneous network is assumed to be Λ, Λ ═ 1,2, a. Assume that the set of users in the heterogeneous network is Ω, Ω ═ 1,2_KThe set of small base station users is omega_kWherein, U represents the total number of users, U represents the U-th user, each user is configured with an antenna, and the transmitting power of each user is equal; assuming that the RB set of resource blocks in the heterogeneous network is Φ, Φ ═ 1,2,. and N.. and N }, the RB set of the small base station k is Φ, Φ ═ 1,2

The RB set orthogonal to the k frequency spectrum of all small base stations is pi_C，

Wherein N represents the total number of RBs, and N represents the nth RB; an initial value of an orthogonality judgment threshold eta is assumed to be 0.01;

(2) the user initializes the associated base station according to Max RSSI criteria:

each user u in omega calculates the signal strength of the pilot signal received from the base station in lambda, and according to the maximum received signal strength Max RSSI rule, the base station corresponding to the pilot signal with the maximum received signal strength is associated, and the user associated to the macro base station K belongs to the set omega_KUsers associated to small base station k belong to the set omega_k；

(3) Macro base station acquires macro user set omega strongly related to small base station users_I：

(3a) Definition of phi_C＝Π_C；

(3b) A macro base station K receives channel state information CSI from all users to K in a set omega, and calculates channel correlation coefficients rho (u, v) from any two users u and v to K in the set omega according to the CSI to obtain a channel correlation matrix psi_Kρ (u, v) is calculated as:

where ρ (u, v) represents the channel correlation coefficient between user u and users v to K, h_u,KRepresenting the channel vector, h, from user u to macro base station K_v,KRepresenting a channel vector from the user v to the macro base station K;

(3c) macro base station K according to Ψ_KFrom the set Ω_KSelecting channel correlation coefficient greater than 1-eta with small base station user_θThe macro user forms a macro user set omega strongly related to the small base station user_IWherein η_θRepresenting an upper limit of the orthogonal decision threshold;

(4) the macro base station K is a macro user set omega strongly related to small base station users_IThe macro user in the system is allocated with an RB orthogonal to the frequency spectrum of the small base station user:

the macro base station K depends on the channel correlation matrix Ψ_KAnd an orthogonality decision threshold eta, in the macro-user set omega_KMiddle is omega_IEach macro user in the system selects users with orthogonal channels to obtain an orthogonal user group set G_KAnd will aggregate phi_CRB n in (1) is allocated to the ith orthogonal user group G_K,iWherein G is_K,i∈G_KThe method comprises the following implementation steps:

(4a) initializing i to 1;

(4b) macro base station K according to Ψ_KAnd eta, from macro user set omega in priority order of relevance from low to high_KSelecting orthogonal macro user pair (a, b), and putting the macro users a, b into user group G_K,iIn the method, channel correlation coefficients rho (a, b) of macro users a, b to K are less than or equal to eta, the eta is an orthogonal decision threshold, and a belongs to omega_I，b∈Ω_KUser b prefers to select from the set omega_ISelecting;

(4c) macro base station K from macro user set omega_KSelecting macro user w and user group G_K,iAre simultaneously orthogonal and put in G_K,iIn, up to G_K,iThe number of users in (1) is equal to the number of antennas M or the macro user set omega_KWith no macro user and G_K,iWhere macro users w are preferentially selected from the set omega_ISelecting;

(4d) macro base station K will set Φ_CRB n in (1) to a user group G_K,iGroup G of users_K,iFrom the macro user set omega strongly correlated with the small base station users_IMacro user set omega_KDeletion of RB n from set phi_CDeleting;

(4e) letting i ═ i +1, the macro base station K repeats steps (4b) - (4d) to select a new orthogonal user group until the macro user set Ω strongly associated with the small base station users_IAs empty or aggregate Φ_CUntil the collection is empty;

(5) k of small base station is omega_kAllocating RBs by the small base station users:

(5a) k of small base station is omega_kSmall cell user in and omega_IGrouping macro users with interference to the small base station k to obtain an orthogonal user group set G_k；

(5b) Small base station k will aggregate phi_kRB in (1)n is assigned to the jth orthogonal user group G_k,jSmall base station user in middle, and give user group G_k,jThe macro user to k channel in (1) gives way to a spatial dimension, where G_k,j∈G_k；

(6) The small base station K sends CSI from the small base station user to the K and a set G to the macro base station K_kAnd RB allocation results:

small base station k will user group G_k,jPerforming random equivalence on CSI from the small base station user to the macro base station K, namely from the user group G_k,jRandomly selecting a small base station user, and using CSI from the small base station user to a macro base station K to represent G_k,jAll the middle users to the CSI of the macro base station K, and equivalent small base station users to the CSI of the K, a set G_kAnd transmitting the RB allocation result of the small base station user to a macro base station K;

(7) the macro base station K is omega_KThe macro user in the system is allocated with the same RB as the frequency spectrum of the small base station user:

(7a) macro base station K gives user group G_k,jThe macro user in the group is allocated with the same RB as the small base station users in the group;

(7b) macro base station K according to Ψ_KAnd η, in the set Ω_KSelecting orthogonal macro users for each equivalent small base station user to obtain an orthogonal user group set gamma_K；

(7c) The macro base station K gives the s-th orthogonal user group gamma_K,sThe macro user in the group is allocated with the same RB as the equivalent small base station user in the group, and a space domain dimension is given to CSI from the equivalent small base station user to the macro base station, wherein gamma is_K,s∈Γ_K；

(7d) The macro base station K judges whether an unallocated RB exists in the set phi or not, if yes, K judges whether an unallocated RB exists in the set phi or not according to psi_KAnd η to Ω_KIn the method, users without RB allocation are grouped to obtain a user group set T, and the unallocated RB n in the phi is allocated to the z-th orthogonal user group T_zWherein, T_z∈T；

(8) Judging whether the algorithm is finished:

the macro base station K and the small base station K calculate the SINR of the users associated with the macro base station K and the small base station K, and judge whether the SINR of all the users is larger than that of the usersUser' S lowest SINR demand S I N R_TIf yes, finishing the algorithm, otherwise, executing the step (9);

(9) the base station adopts the associated control to adjust RB allocation:

(9a) the macro base station K will satisfy every omega that meets the SINR requirement_KThe macro user a in the base station carries out channel orthogonality matching with user groups of other base stations, and selects the base station corresponding to the user group orthogonal to the macro user a to obtain a base station set lambda_a；

(9b) Macro base station K calculates omega that will not meet SINR requirements_KMacro user a in (A) re-associates to the set lambda from the original associated base station_aAnd associating macro user a to set Λ_aThe base station with the maximum medium gain and the positive gain distributes the RB used by the user group orthogonal to a under the base station for the macro user a;

(9c) the small base station k sets each small base station user omega which does not meet the SINR requirement_kThe small base station user c in the system is subjected to channel orthogonality matching with user groups of other base stations, and a base station corresponding to the user group orthogonal to the small base station user c is selected to obtain a base station set lambda_c；

(9d) The small base station k calculates the set omega of the small base station users_kSmall cell user c in (2) re-associates from the original associated base station to the set Λ_cNetwork throughput gain generated by the base station in (1) and associating small base station user c to set Λ_cThe base station with the maximum middle gain and the positive gain distributes the RB used by the user group orthogonal to the c under the base station to the small base station user c;

(9e) correspondingly updating omega according to the associated control result_KAnd omega_k；

(10) Judging whether the algorithm is finished:

the macro base station K and the small base station K calculate the SINR of the users associated with the macro base station K and the small base station K, and if the SINR of all the users is larger than the lowest SINR requirement S I N R of the users_TIf the algorithm is finished, otherwise, judging eta < eta_θAnd (3) if the judgment result is positive, enabling eta to be eta + delta eta, and executing the step (3), otherwise, ending the algorithm, wherein the delta eta is the judgment threshold change amount.

Compared with the prior art, the invention has the following advantages:

(1) under the conditions of limited frequency spectrum resources and strong correlation of partial user channels, on one hand, a macro base station and a small base station cooperate to carry out space-frequency joint resource allocation on users, and RB is allocated to the users according to orthogonal user groups, so that cross-layer interference is reduced; on the other hand, the base station changes the associated base station of the user which does not meet the requirement, adjusts the RB allocation of the user, and further reduces the cross-layer interference.

(2) In the process of grouping users and allocating resources by the cooperation of the macro base station and the small base station, the small base station performs random equivalence on CSI from the small base station user group to the macro base station, and interaction overhead between the small base station and the macro base station is reduced.

Drawings

FIG. 1 is a diagram of an application scenario of the present invention;

FIG. 2 is a flow chart of an implementation of the present invention;

FIG. 3 is a graph comparing the present invention with the prior art with respect to cumulative distribution of throughput over each RB;

FIG. 4 is a graph comparing throughput as a function of the number of network users in accordance with the present invention and the prior art;

Detailed Description

The invention is described in further detail below with reference to the figures and specific examples:

referring to fig. 1, the heterogeneous network of the present invention includes 1 macro base station and 3 small base stations, and each base station is configured with 4 antennas. 360 users exist in the heterogeneous network, the users are randomly distributed in the coverage area of the macro base station, and the users in the hot spot area where the small base station is located are dense. Each user can only be associated with one base station, and each RB under the base station is allocated to 4 users at most. In order to improve the spectrum utilization rate and avoid the same-layer interference between small base stations, the small base stations partially reuse the spectrum resources of the macro base station, and the adjacent small base stations use orthogonal spectrum, and the small base stations with longer distance use the same spectrum.

Referring to fig. 2, a method for managing interference in a heterogeneous network based on multidimensional resource allocation specifically includes the following steps:

step 1, initializing a heterogeneous network:

assuming that a set of base stations in the heterogeneous network is Λ, Λ ═ 1,2,3,4}, where when k ═ 4, it represents a macro base station, and k ≠ 4, it represents a kth small base station, the number of small base stations is 3, and each base station is configured with 4 antennas; assume that the set of users in the heterogeneous network is Ω, Ω ═ 1,2_KThe set of small base station users is omega_kThe total number of users is 360, u represents the u-th user, each user is configured with an antenna, and the transmitting power of each user is equal; assume that the RB set of resource blocks in the heterogeneous network is Φ, Φ ═ 1,2

Wherein the total number of RBs in Φ is 50, n represents the nth RB, Φ_kTotal RB number of (1) is 10 ^ Π_CThe total number of RBs in (1) is 20; an initial value of an orthogonality judgment threshold eta is assumed to be 0.01;

step 2, the user initializes the associated base station according to Max RSSI standard:

Step 3, the macro base station acquires a macro user set omega strongly related to the small base station users_I：

Step 3a) defining phi_C＝Π_C；

Step 3b), the macro base station K receives channel state information CSI from all users to K in the set omega, and calculates channel correlation coefficients of any two users u, v to K in the set omega according to the CSI

Deriving the channel correlation matrix Ψ_KWhere ρ (u, v) represents the channel correlation coefficient between user u and users v to K, and h_u,KRepresenting the channel vector, h, from user u to macro base station K_v,KRepresenting a channel vector from the user v to the macro base station K; in this example, K ═ 4;

step 3c) macro base station K according to psi_KFrom the set Ω_KSelecting channel correlation coefficient greater than 1-eta with small base station user_θThe macro user forms a macro user set omega strongly related to the small base station user_IWherein η_θRepresenting an upper limit of the orthogonal decision threshold; in this example, η_θ＝0.3；

Step 4), the macro base station K is omega_IThe macro user in the system is allocated with an RB orthogonal to the frequency spectrum of the small base station user:

step 4a) initializing i to 1;

step 4b) macro base station K according to psi_KAnd eta, selecting orthogonal macro user pairs (a, b) according to the priority sequence of the correlation from low to high, and putting the macro users a and b into the ith orthogonal user group G_K,iIn the method, channel correlation coefficients rho (a, b) of macro users a, b to K are less than or equal to eta, the eta is an orthogonal decision threshold, and a belongs to omega_I，b∈Ω_KUser b prefers to select from the set omega_ISelecting;

step 4c) macro base station K slave set omega_KSelecting macro user w and user group G_K,iAre simultaneously orthogonal and put in G_K,iIn, up to G_K,iThe number of users in (1) is equal to the number of antennas M, or omega_KWith no macro user and G_K,iWhere macro users w are preferentially selected from the set omega_ISelecting; in this example, the number of antennas M is 4;

step 4d) the macro base station K will gather phi_CRB n in (1) is allocated to the ith orthogonal user group G_K,iAvoiding the cross-layer interference of macro users to the small base station from the frequency domain, and combining the user group G_K,iFrom the set omega_I、Ω_KDeletion of RB n from set phi_CDeleting;

step 4e) let i ═ i +1, the macro base station K repeats steps 4b) -4d) to select a new orthogonal user group until the set Ω_IAs empty or aggregate Φ_CUntil the collection is empty;

step 4f) defining a set omega_I,kRepresents the set of macro users with strong interference to the small base station k, if Ω_IIf it is empty, then

If omega_IIf not empty, then pair

If the interference of the macro user a to the small base station k is strongest, the macro user a is put into the set omega_I,kPerforming the following steps;

step 5), the k of the small base station is omega_kThe user in (1) allocates the RB:

step 5a) the small base station k is omega_kSmall cell user in and omega_I,kThe macro users in the group are grouped to obtain an orthogonal user group set G_k：

Step 5a1) macro base station K will Ω_I,kCSI from the macro user to the small base station k is sent to the small base station k, and omega is defined_kSmall cell user in and omega_I,kThe set of macro users in (1) is omega_k-KCalculate the set Ω_k-KChannel correlation matrix Ψ between users_k(ii) a In this example, k is 1,2, 3;

step 5a2) Small cell K according to Ψ_kAnd η, set Ω according to a low-to-high priority order of correlation_k-KThe users of (2) are grouped to obtain a set G of orthogonal user groups_k，

Rho (d, e) is less than or equal to eta, wherein G_k,jDenotes the jth orthogonal user group in the small base station k, and G_k,j∈G_k；

Step 5b) the small base station k will gather phi_kRB n in (1) to a user group G_k,jSmall base station user in middle, and give user group G_k,jMacro user to k channel in (1)Giving out a space domain dimension, and eliminating cross-layer interference of macro users to the small base station k from the space domain;

step 6) the small base station K sends CSI from the small base station user to the K and a set G to the macro base station K_kAnd RB allocation results:

step 7) macro base station K is omega_KThe macro user in the system is allocated with the same RB as the frequency spectrum of the small base station user:

step 7a) macro base station K gives user group G_k,jThe macro user in the group is allocated with the same RB as the small base station users in the group;

step 7b) macro base station K according to psi_KAnd η, in the set Ω, in the order of priority of the correlation from low to high_KUntil the number of mutually orthogonal users is equal to the number of antennas M, or omega_KUntil no macro user is orthogonal with equivalent small base station user, obtaining an orthogonal user group set gamma_K，

Rho (u, v) is less than or equal to eta, wherein gamma is_K,sDenotes the s-th orthogonal user group in the macro base station K, and Γ_K,s∈Γ_K；

Step 7c) the macro base station K gives the user group gamma_K,sThe macro user allocates the same RB as the equivalent small base station users in the group, gives out an airspace dimension to CSI from the equivalent small base station users to the macro base station, and eliminates cross-layer interference of the small base station users to the macro base station K from the airspace;

step 7d) the macro base station K judges whether an unallocated RB exists in the set phi or not, if yes, K judges whether an unallocated RB exists in the set phi or not according to psi_KAnd eta to omega in the order of low to high correlation priority_KIn the method, users without RB allocation are grouped to obtain a user group set T, and the users without RB allocation n are allocated to a z-th orthogonal user group T_zWherein, T_z∈T；

Step 8) judging whether the algorithm is finished:

the macro base station K and the small base station K respectively calculate the SINR of users associated with the macro base station K and the small base station K, and judge whether the SINR of all the users is larger than the lowest SINR requirement S I N R of the users_TIf yes, the algorithm is ended, otherwise, step 9 is executed, and the calculation formulas of SINR are respectively as follows:

wherein, the SINR_a,KSINR, SINR of macro user a calculated by macro base station K_c,kSINR, w representing the small base station user c calculated by the small base station k_a ^TAnd w_c ^TRespectively representing the filter vectors of the macro user a and the small base station user c under the Minimum Mean Square Error (MMSE) detection criterion, E_SRepresenting the transmission power, σ, of the user_n ²Representing the power spectral density of gaussian white noise,

and

respectively representing the interference from other co-frequency users suffered by the macro user a and the small base station user c; in this example, E_S＝23dBm，σ_n ²＝-174dBm/Hz，SΙΝR_T＝15；

Step 9), the base station adjusts RB allocation by adopting associated control:

step 9a) macro base station K will each omega not meeting SINR requirements_KThe macro user a in the system communicates with the user group of other base stationsAnd matching the channel orthogonality, and selecting a base station corresponding to a user group orthogonal to the macro user a to obtain a base station set lambda_a；

Step 9b) the macro base station K calculates omega which will not meet the SINR requirement_KMacro user a in (A) re-associates to the set lambda from the original associated base station_aNetwork throughput gain generated at the base station of (1):

macro user a is re-associated to set Λ_aBefore the base station K', the throughput gain deltaC of the macro base station K_a(K)：

Wherein the user group G_K,nIndicating that when macro user a is associated with macro base station K, macro base station K uses RB n's user group, { G_K,nA represents a new user group using RB n in the macro base station K after the macro user a is associated with the base station again;

the macro user a is associated to the set Λ_aAfter the base station k', the throughput gain Δ C of the base station k_a(k′)：

Wherein the user group G_k′,nRepresenting macro user a to be associated to set Λ_aBefore base station k ', the user group using RB n in base station k' { G_k′,nA represents a new user group using RB n in the base station k 'after the macro user a is associated to the base station k';

the macro base station K calculates the network throughput gain Δ C (a, K'):

ΔC(a,K,k′)＝ΔC_a(k′)-ΔC_a(K)；

step 9c) macro base station K associates macro user a to set Λ_aThe base station with the maximum medium gain and the positive gain distributes the RB used by the user group orthogonal to a under the base station for the macro user a;

step 9d) small base station k sends each small base station user which does not meet the SINR requirementSet omega_kThe small base station user c in the system is subjected to channel orthogonality matching with user groups of other base stations, and a base station corresponding to the user group orthogonal to the small base station user c is selected to obtain a base station set lambda_c；

Step 9e) the small base station k calculates the user set omega of the small base station_kSmall cell user c in (2) re-associates from the original associated base station to the set Λ_cNetwork throughput gain generated at the base station of (1):

small base station user c is re-associated to set Λ_cBefore the medium base station k', the throughput gain Δ C of the small base station k_v(k)：

Wherein the user group G_k,nIndicating that when the small cell user c is associated with the small cell k, the user group using RB n in the small cell k, { G_k,nC represents a new user group using RB n in the small cell k after the small cell user c re-associates with the base station.

Small base station user c is associated to set Λ_cAfter base station k', the throughput gain Δ C of base station k_c(k″)：

Wherein the user group G_k″_,nRepresenting association of small base station user c to set Λ_cBefore base station k ' in base station k ', user group using RB n in base station k ' { G_k″_,nC represents a new user group using RB n in the base station k 'after the small base station user c is associated to the base station k';

the small base station k calculates the network throughput gain ac (C, k, k "):

ΔC(c,k,k″)＝ΔC_c(k″)-ΔC_c(k)；

step 9f) associating small base station users c to the set Λ_cThe base station with the maximum medium gain and the positive gain is allocated to the small base station user cThe RB used by the user group orthogonal to the c under the base station;

step 9g) updating the set omega correspondingly according to the associated control result_KAnd omega_k；

Step 10) judging whether the algorithm is finished:

the macro base station K and the small base station K calculate the SINR of the users associated with the macro base station K and the small base station K, and if the SINR of all the users is larger than the lowest SINR requirement S I N R of the users_TIf the algorithm is finished, otherwise, judging eta < eta_θIf yes, let η be η + Δ η, and execute step 3), otherwise, end the algorithm, where Δ η is the decision threshold variation; in this example, Δ η is 0.01.

The effects of the present invention can be further illustrated by the following simulations:

1. simulation parameters:

the application bandwidth of the invention is 10MHz, including 50 RB, the coverage radius of macro base station K is 250 m, the coverage radius of small base station K is 20 m, the number of macro base stations is 1, the number of small base stations is 3, the number of base station antennas is 4, the number of users U in the application network is 360, the transmitting power of users is 23dBm, the user detection mode is MMSE, the noise power spectral density sigma is_n2 was-174 dBm/Hz, and the path loss model was as follows:

macro-to-macro base station: PL (dB) ═ 15.3+37.6log₁₀d

Small base station user to macro base station: PL (dB) ═ 15.3+37.6log₁₀ d+Ls_ow

Macro-user to small base station: PL (dB) max (15.3+37.6 log)₁₀d,38.46+20log₁₀d)+Ls_ow

Small cell user to small cell: PL (dB) ═ 38.46+20log₁₀ d

2. The simulation method comprises the following steps: the invention relates to a space-frequency joint interference management method based on orthogonal user grouping.

3. Simulation content and results:

simulation 1, when the number of users in the network is 360, the throughput on each RB is simulated by using the present invention and the prior art, and a cumulative distribution function diagram of the throughput on each RB is obtained, and the result is shown in fig. 3.

Fig. 3 shows that, compared with the prior art, the interference management method for joint allocation of space domain resources, frequency domain resources and associated base stations through cooperation of the macro base station and the small base station obviously improves the throughput on each RB.

Simulation 2, when the number of users in the network is 40, 80, 120, 160, 200, 240, 280, 320, 360, the small base station throughput, the macro base station throughput, and the network throughput are simulated by using the present invention and the prior art, and the results are shown in fig. 4(a), (b), and (c), respectively.

Fig. 4 shows that, as the number of network users increases, the throughput of the small base station, the throughput of the macro base station and the throughput of the network obtained by the method are all obviously higher than those of the prior art, so that a larger throughput gain is obtained, and cross-layer interference is effectively suppressed.

Claims

1. A heterogeneous network interference management method based on multi-dimensional resource allocation is characterized by comprising the following steps:

(1) initializing the heterogeneous network:

(3a) Definition of phi_C＝Π_C；

(3c) macro base station K according to Ψ_KFrom the macro user set omega_KSelecting channel correlation coefficient greater than 1-eta with small base station user_θThe macro user forms a macro user set omega strongly related to the small base station user_IWherein η_θRepresenting an upper limit of the orthogonal decision threshold;

(4a) initializing i to 1;

(5b) Small base station k will aggregate phi_kRB n in (1) is allocated to the jth orthogonal user group G_k,jSmall base station user in middle, and give user group G_k,jThe macro user to k channel in (1) gives way to a spatial dimension, where G_k,j∈G_k；

(7b) macro base station K according to Ψ_KAnd η at Ω_KSelecting orthogonal macro users for each equivalent small base station user to obtain an orthogonal user group set gamma_K；

(8) Judging whether the algorithm is finished:

the macro base station K and the small base station K calculate the SINR of the users associated with the macro base station K and the small base station K, and judge whether the SINR of all the users is larger than the lowest S of the usersINR demand S I N R_TIf yes, finishing the algorithm, otherwise, executing the step (9);

(9) the base station adopts the associated control to adjust RB allocation:

(10) Judging whether the algorithm is finished:

2. The method as claimed in claim 1, wherein the small cell k in step (5a) is Ω_kSmall cell user in and omega_IGrouping macro users without RB allocation, and the implementation steps are as follows:

(5a1) macro base station K will be Ω_ICSI from a macro user with no RB allocated in the middle to a small base station k is sent to the small base station k, and omega is defined_kSmall cell user in and omega_IThe set of macro users with unallocated RBs is omega_k-KCalculate the set Ω_k-KChannel correlation matrix Ψ between users_k；

(5a2) Small base station k according to Ψ_kAnd η, set Ω according to a low-to-high priority order of correlation_k-KThe users of (2) are grouped to obtain a set G of orthogonal user groups_k，

Rho (d, e) is less than or equal to eta, wherein eta is an orthogonal decision threshold, G_k,jDenotes the jth orthogonal user group in the small base station k, and G_k,j∈G_k。

3. The method for managing interference in a heterogeneous network based on multidimensional resource allocation according to claim 1, wherein the macro base station K and the small base station K in step (8) calculate the SINR of the user associated with them, and their calculation formulas are respectively:

wherein, the SINR_a,KSINR, SINR of macro user a calculated by macro base station K_c,kRepresenting the cell user c calculated by the cell kSINR，w_a ^TAnd w_c ^TRespectively representing the filter vectors of the macro user a and the small base station user c under the Minimum Mean Square Error (MMSE) detection criterion, E_SRepresenting the transmission power, σ, of the user_n ²Representing the power spectral density of gaussian white noise,

and

respectively representing the interference from other co-channel users suffered by the macro user a and the small base station user c.

4. The method as claimed in claim 1, wherein the macro base station K in step (9b) re-associates macro user a that does not satisfy SINR requirement from the original associated base station to the set Λ for the macro base station K calculation_aThe network throughput gain generated by the base station in (1) is calculated by the following formula:

Wherein is made ofHouse group G_k′,nRepresenting macro user a to be associated to set Λ_aBefore base station k ', the user group using RB n in base station k' { G_k′,nA represents a new user group using RB n in the base station k 'after the macro user a is associated to the base station k';

the macro base station K calculates the network throughput gain Δ C (a, K'):

ΔC(a,K,k′)＝ΔC_a(k′)-ΔC_a(K)。