CN105356917A - Interference suppression transmission method in large-scale MIMO (Multiple-Input Multiple-Output) heterogeneous network - Google Patents

Abstract

The invention provides an interference suppression transmission method in a massive MIMO heterogeneous network, belonging to the technical field of wireless communication. This method first judges whether the interference intensity of the micro base station by the macro base station is strong interference or weak interference, and then the micro base station performs different processing to further obtain the precoding matrix of the micro base station, and maximize the data rate; when it is weak interference, the macro base station The interference is treated as noise; when it is strong interference, it is processed by interference deletion. The method of the present invention solves the problems of large amount of calculation and low efficiency in the existing precoding design, and the solution of the precoding matrix can be obtained only by one calculation, which improves the calculation efficiency, shortens the system processing time, and greatly contributes to improving the efficiency of the communication system Significant.

Description

An Interference Suppression Transmission Method in Massive MIMO Heterogeneous Networks

technical field

The invention belongs to the technical field of wireless communication, and relates to a transmission method aimed at suppressing inter-user interference and cross-layer interference in a large-scale MIMO (Multiple-Input Multiple-Output) heterogeneous network scenario.

Background technique

MIMO is the abbreviation of MultipleInputMultipleOutput (Multiple Input Multiple Output). Equipped with multiple antennas, MIMO technology can realize transmit/receive diversity and space division multiplexing, thereby providing diversity gain and multiplexing gain, and can double the system capacity without increasing spectrum resources and antenna transmit power. Today, MIMO technology has been widely used in the field of wireless communication. For example, in the current 4G communication, multi-antenna transmit diversity and multi-stream space division multiplexing have been applied. The promotion of MIMO technology makes space a resource that can be used to improve performance and increase the coverage of wireless systems.

With the continuous popularization and development of multimedia terminals, the user's demand for communication efficiency is increasing day by day, and the demand for the number of antennas is also increasing accordingly, resulting in the emergence of massive MIMO. The massive MIMO system has a very large number of array elements, which can provide higher diversity gain and multiplexing gain, and can meet the increasingly higher data rate requirements.

At the same time, in order to ensure the service quality of each user, the cell will also present the characteristics of heterogeneity. In a heterogeneous network, there will be a problem of cell coverage. In addition to inter-user interference, cross-layer interference cannot be ignored. A common means of suppressing interference is to design precoding to meet system requirements. However, the traditional method usually adopts iterative algorithm, which has a large amount of calculation and low efficiency. In terms of massive MIMO, there has been a lot of related work. Literature [1] considered a two-layer precoding system and proposed a method to obtain the precoding matrix through iteration. Literature [2] combines the outer coding with the channel, thereby reducing the dimension of the channel, which can reduce the complexity of channel estimation. At the same time, it is assumed that the range of user activities in the cluster is small, so the inner code is updated at a shorter time interval, and the outer code is updated at a longer time interval. The precoding design of the outer layer is a combination of zero-forcing (ZF) precoding and matched filter, and their respective proportions are controlled by assigning different weights to ZF and matched filtering. Inner coding is used to suppress residual interference. Literature [3] considered a more complex situation, combined MIMO with Orthogonal Frequency Division Multiplexing (OFDM), and performed IFFT (Inverse Fast Fourier Transform) and FFT (Fast Fourier Transform) at the sending end and receiving end respectively. deal with. The outer layer encoding uses a pre-quantized codebook (the codebook is the array vector of the direction angle, and the direction angle is quantized according to the equal angular interval or the equal sine value interval), and the elements of the beam pointing path are selected in the codebook as the outer layer coding. The peak-to-average power ratio and other characteristics of the system are discussed under the condition that the coding is known to have normalization characteristics. Literature [4] considers the two-layer precoding under massive MIMO, the receiving end has a single antenna, and the users are clustered. The outer layer precoding is designed through channel statistical information to suppress inter-cluster interference, and the inner layer precoding suppresses inter-user interference. . Moreover, when the channel satisfies a certain condition, the modulus of each element of the outer layer precoding is equal.

There has also been some work on the suppression of inter-cell interference. Literature [5] uses a "partial frequency reuse" method, that is, users at the center of the cell use the same frequency resources, while users at the edge of the cell use orthogonal frequency resources, so that interference can be suppressed. Literature [6] uses the time-domain orthogonal method, that is, the macro base station sends some blank frames in the time range, so as to reduce the cross-layer interference to the micro cell. Both literature [7] and literature [8] used nonlinear methods to suppress interference. Literature [7] considers the interference suppression method using active interference removal in mixed interference scenarios, and obtains precoding through iteration. Literature [8] divides the interference scenarios in the heterogeneous network into three types according to the strength of the interference (weak and weak, strong and weak, strong and strong), and the user chooses the transmission scheme that treats the interference as noise or deletes the active interference in different scenarios, giving The pre-encoded analytical solutions for each scenario are presented.

At the same time, there are also some related works on the application of large-scale antennas in heterogeneous network scenarios. Literature [9] proposes a scenario in which a macro base station in a heterogeneous network is equipped with a large-scale antenna and coexists with multiple single-antenna micro cells, analyzes the system performance, and gives a curve of micro cell spatial spectrum efficiency versus micro cell density. Literature [10] considers the suppression of cross-layer interference in large-scale antenna heterogeneous networks. First, users are grouped, and different collaborations are adopted between macro cells and micro cells to manage cross-layer interference.

Although existing works have proposed interference management measures for large-scale antennas and heterogeneous networks from different perspectives, they have not given precoding analytical solutions for different interference scenarios such as strong interference or weak interference in massive MIMO heterogeneous networks.

References [1] to [10] are as follows:

[1] AlkhateebA, AyachOE, LeusG, et al. Hybridprecodingformillimeterwavecellularsystemwithpartialchannelknowledge[C]//InformationTheoryandApplicationsWorkshop(ITA).SanDiego, CA:IEEE,2013:1-5;

[2] ChenJunting, and LauVKN.Two-TierPrecodingforFDDMulti-cellMassiveMIMOTime-VaryingInterferenceNetworks[J].IEEEJournalonSelectedAreasinCommunications,2014,32(6):1230-1238;

[3] KimC, SonJS, KimT, et al. On the Hybrid Beamforming with Shared Array Antennaformm Wave MIMO-OFDM Systems [C]//Wireless Communications and Networking Conference (WCNC). Istanbul: IEEE, 2014: 335-340;

[4] AdhikaryA, NamJ, AhnJY, et al. Joint Spatial Division and Multiplexing—The Large-Scale Array Regime [J]. IEEE Transactions on Information Theory, 2013, 59 (10): 6441-6463;

[5] BoudreauG, PanickerJ, GuoN, et al.Interferencecoordinationandcancellationfor4Gnetworks[J].IEEECommunicationsMagazine,2009,47(4):74-81;

[6] Lopez-PerezD, GuvencI, G. DLR, etal. Enhanced Inter-Cell Interference Coordination Challenges in Heterogeneous Networks [J]. IEEE Wireless Communications, 2011, 18 (3): 22-30;

[7] Wang Yunlu, Tian Yafei, Li Yang, et al. Coordinated precoding and proactive interference cancellation in mixed interference scenarios [C]//Wireless Communications and Networking Conference (WCNC). Istanbul: IEEE, 2014: 554-558;

[8] Li Yang, Tian Yafei, and Yang Chenyang. Beamforming Design with Proactive Interference Cancellation in MISO Interference Channels [J]. EURASIP Journalon Advances in Signal Processing, 2015:67 (2Aug.2015);

[9] KountourisM, PappasN. Het Nets and massive MIMO: Modeling, potential gains, and performance analysis [C] // Antennas and Propagation in Wireless Communications (APWC). Torino: IEEE, 2013: 1319-1322;

[10] AdhikaryA, DhillonHS, CaireG. Massive-MIMOMeetsHetNet-InterferenceCoordinationThroughSpatialBlanking[J].IEEEJournalonSelectedAreasinCommunications,2015,33(6):1171-1186.

Contents of the invention

Aiming at the problems of large amount of calculation and low efficiency in the existing precoding design, in order to suppress inter-user interference and cross-layer interference, the present invention provides an interference suppression transmission method in a massive MIMO heterogeneous network. The present invention aims at the situation that the macro cell and the micro cell overlap with each other, and the macro base station is equipped with a large-scale antenna, and the micro user adopts different interference processing methods according to the intensity of the interference generated by the macro base station: active interference deletion for strong interference, and weak interference Interference is treated as noise. Under different processing methods, the micro base station precoding gives analytical solutions.

The interference suppression transmission method in the massive MIMO heterogeneous network of the present invention, the implementation steps are as follows:

Step 1, judging whether the interference intensity of the micro base station by the macro base station is strong interference or weak interference;

Step 2, the micro base station adopts a corresponding processing method according to the interference intensity of the macro base station: (1) when it is weak interference, treat the interference of the macro base station as noise, and perform step 2.1; (2) use interference deletion when it is strong interference For processing, go to step 2.2.

In step 2 shown, let H _mm be the channel vector from the macro base station to the g-th cluster macro user, V _m be the outer precoding matrix or the complete pre-coding matrix of the macro base station to the g-th cluster macro user data, and H _pp be The channel from the gth micro-base station to the g-th micro-user, V _p is the precoding matrix of the g-th micro-base station, H _pm is the channel from the macro-base station to the g-th micro-user; P _m is the transmit power and noise of the macro-base station normalized power of power, P _p is the normalized power of the micro base station transmit power and noise power, P _macro is the transmit power of the macro base station, P _pico is the transmit power of the micro base station, and P _N is the noise power; g is a positive integer.

Step 2.1, under processing method (1): the data rate R _pico of the micro-user is expressed as:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Among them, the superscript H means transpose, I is the identity matrix, is a normal matrix, the normal matrix is decomposed, expressed as: Q ₁ is the middle matrix of decomposing the normal matrix on the left side;

(1.1) When micro-users transmit single-stream data, V _p is The eigenvector x corresponding to the largest eigenvalue λ _max of ;

(1.2) When the micro-user transmits n stream data, n is an integer greater than 1, and the obtained The largest n eigenvalues are arranged in order from large to small as λ ₁ , λ ₂ ,...λ _n , x ₁ , x ₂ ,...x _n are the corresponding features of λ ₁ , λ ₂ ,...λ _n vector to obtain a precoding matrix V _p =[x ₁ x ₂ . . . x _n ].

Step 2.2, under processing mode (2): and the data rate R is expressed as:

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span><span\; class="notranslate">\; ;</span>$

Among them, the superscript H means transpose, I is the identity matrix, and the normal matrix Decomposed into Q ₂ is the intermediate matrix obtained by decomposition;

(2.1) When the micro-user transmits single-stream data, set V _p as The eigenvector x corresponding to the largest eigenvalue λ _max of ;

(2.2) When the micro-user transmits n stream data, n is an integer greater than 1, select Orthogonal eigenvectors x ₁ , x ₂ , ... x _n corresponding to the largest n eigenvalues form a precoding matrix V _p , where V _p =[x ₁ x ₂ ... x _n ].

The advantages and positive effects of the present invention are: compared with the traditional method, the interference suppression transmission method of the present invention does not need iterative calculation, and only needs to perform one calculation to obtain the solution of the precoding matrix, which improves the calculation efficiency and shortens the system processing time. Therefore, the interference suppression transmission method provided by the present invention is very meaningful for improving the efficiency of the communication system.

Description of drawings

FIG. 1 is a schematic diagram of a scenario where a macro cell and a micro cell cover each other;

2 is a schematic diagram of an interference suppression transmission method in a massive MIMO heterogeneous network according to the present invention;

Fig. 3 is a schematic diagram of performance curves of different transmission methods.

detailed description

The present invention will be further described in detail with reference to the accompanying drawings and embodiments.

The interference suppression transmission method in the massive MIMO heterogeneous network of the present invention is aimed at the situation that the macro cell and the micro cell overlap each other, and the macro base station is equipped with a large-scale antenna to cooperate to suppress interference. The macro base station and the micro base station share channel information and precoding information, and the micro base station calculates the precoding at the receiving end according to the channel information and precoding information of the macro base station. The macro base station can adopt any precoding method, such as single-layer precoding, two-layer precoding, and the outer layer of two-layer precoding can be analog precoding or digital precoding. The micro base station adopts different processing methods according to the interference intensity of the macro base station: when the interference is strong, the interference cancellation is used, and when the interference is weak, the interference of the macro base station is regarded as noise.

In a massive MIMO heterogeneous network scenario, consider a situation where a macro cell and a micro cell cover each other, as shown in FIG. 1 .

Let the macro base station be the MBS, the macro user be the Muser, the micro base station be the PBS, and the micro user be the Puser. Macro users are clustered, and there is a micro cell near each cluster of macro users. In the macro cell, the micro cell near the gth cluster of macro users is defined as the gth micro cell, and the corresponding micro base station and micro user in the micro cell are respectively the gth cell. The micro base station and the gth micro user.

The number of macro base station antennas is M _macro , the number of gth micro base station antennas is M _pico-BS-g , and the number of gth micro user antennas is M _pico-User-g . Consider that in a heterogeneous network scenario, the macro cell and the micro cell work in the same frequency band. The macro cell is a massive MIMO system, the macro base station is equipped with large-scale antennas, and the macro users are divided into G clusters, users have multiple antennas, each user receives multiple streams of data, each cluster receives K _g stream data in total, and the macro base station sends a total of S streams Data; each micro cell serves 1 micro user, and the gth micro user receives N _g flow data. The transmit power of the macro base station is P _macro , the transmit power of the micro base station is P _pico , and the noise power is P _N . The normalized power of the macro base station transmit power and noise power is The normalized power of the micro base station transmit power and noise power is

Assuming that the number of antennas of the i-th macro user is M _i , the received signal y _i can be expressed as:

${\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; \&NotEqual;</span>\mathrm{<span\; class="notranslate">\; i</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; m</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; x</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; no</span>}}_{\mathrm{<span\; class="notranslate">\; i</span>}}$

H _i,m is the channel between the macro base station and the i-th macro user, in, Represents M _i ×M _macro -dimensional complex vector space, V _i,m is the precoding matrix of the i-th macro user data, S _i is the number of data streams received by the i-th macro user, Represents M _macro × S-dimensional complex vector space, x _{i, m} represent the sending data of the i-th macro user; is the interference generated by other macro users received by the i-th macro-user, V _k,m is the precoding matrix of the k-th macro-user data, x _k,m represents the transmitted data of the k-th macro-user; H _ip V _p x _p represents the interference from the micro user received by the i-th macro user, H _ip is the interference channel from the micro base station to the i-th macro user, V _p is the precoding matrix of the micro base station, x _p is the transmitted data of the micro base station; n _i represents the noise of the i-th macro user, It means that it obeys the distribution with mean 0 and variance I. Among them, I represents the identity matrix, that is, the autocorrelation value of n _i is 1, and multiple random generation is independent of each other. In the method of the present invention, the channel between the base station and the user refers to the downlink channel.

Since the macro-users of each cluster are geographically far apart, the interference between different micro-cells and the interference caused by non-adjacent clusters to micro-users are negligible. Therefore, the signal y _g received by the gth micro-user can be expressed as:

y _g ＝P _p H _g,p V _g,p x _g,p +P _m H _gm V _m x _m +n _p

H _g,p is the channel between the gth micro user and the micro base station serving it, Represents the M _pico-User-g ×M _pico-BS-g dimensional complex vector space, V _g,p represents the precoding matrix of the gth micro user data, x _g,p is the data sent by the micro base station, Represents an N _g -dimensional complex vector space, is the interference channel from the macro base station to the micro user, Represents M _pico-User-g × M _macro dimensional complex vector space, is the data sent by the macro base station, Represents an S-dimensional complex vector space, P _m H _pm V _m x _m is the interference received by the user from the macro base station, n _p represents the noise of the micro user, and the noise received by the micro user is uniformly represented by n _p in the present invention, for noise. H _pm represents the channel from the micro base station to the macro user, V _m represents the precoding matrix of the macro base station, and x _m represents the transmitted data of the macro base station.

The macro base station can adopt single-layer precoding or two-layer precoding, and the outer layer of the two-layer precoding can be analog or digital precoding. It should be pointed out here that both the macro user and the micro user need to feed back channel information to the base station, and at any time, the micro base station needs to receive the channel information of the macro cell from the macro base station. At the same time, when the macro base station adopts single-layer precoding, the micro base station needs to use all the precoding information of the macro base station. When the macro base station adopts two-layer precoding, the micro base station only needs to use the information of the outer layer precoding of the macro base station. All these information need to be sent by the macro base station to the micro base station.

According to the different interference generated by the macro base station to the micro base station, the micro base station adopts different precoding schemes in different SINR scenarios, namely: Scenario (1) The micro user receives weak interference, and directly decodes the received data using the interference as noise ; Scenario (2) The micro-user is subject to strong interference, and the method of active interference deletion is used to eliminate the impact of interference. In different scenarios, the precoding forms of the micro base station are different.

Since the spatial positions of each cluster of macro users are different, the correlation between the channels of users in different clusters is poor. Therefore, when the micro base station performs cross-layer interference suppression, it is assumed that the g-th micro-user is only affected by the g-th cluster macro-user signal interference.

First, the definition of the characters involved is explained: H _mm is the channel from the macro base station to the g-th cluster macro user, V _m is the outer precoding matrix or the complete precoding matrix of the macro base station for the g-th cluster macro user data, and H _pp is The channel from the gth micro-base station to the g-th micro-user, V _p is the precoding matrix of the g-th micro-base station, H _pm is the channel from the macro-base station to the g-th micro-user, and H _mp is the channel from the g-th micro-base station to the g-th micro-user. The channel of the g-th cluster macro user.

The interference suppression transmission method in the massive MIMO heterogeneous network of the present invention mainly includes two major steps, as shown in Figure 2: the first step is to judge the interference intensity generated by the macro base station; the second step is to judge the interference intensity generated by the macro base station; Different processing methods are adopted corresponding to different interference strengths: when the interference is strong, interference deletion is performed, and when the interference is weak, the interference of the macro base station is regarded as noise.

First of all, the following two methods can be used to judge the interference intensity generated by the macro base station:

(1) Calculate the data rate corresponding to the two processing methods in advance at the micro base station: if the data rate of interference as noise is higher than the data rate of interference deletion, the interference intensity of the macro base station is considered to be weak interference, and the interference is regarded as noise ; On the contrary, it is considered that the interference intensity of the macro base station is strong interference, and interference cancellation will be performed.

(2) Calculate the 2-norm of the channel matrix at the receiving end of the micro-user: if the 2-norm of the channel matrix of the direct transmission link from the micro-base station to the micro-user is greater than the 2-norm of the channel matrix of the cross-link channel from the macro base station to the micro-user, then It is regarded as a weak interference scene, and the interference is directly regarded as noise; otherwise, it is considered a strong interference scene, and the interference is deleted.

When the micro base station adopts different receiving end processing methods, its precoding matrix is also different, and the precoding is calculated according to the precoding information of the macro base station, and the channel information of the macro cell and the micro cell to obtain an analytical solution.

The implementation process of calculating and obtaining the precoding of the micro base station to maximize the data rate in the manner of treating the interference of the macro base station as noise and performing interference cancellation is respectively described below.

In step 2.1, the micro user treats the interference of the macro base station as noise.

First consider the case where a micro-user transmits single-stream data. When the interference generated by the macro base station to the micro user is weak, the micro user can regard the interference generated by the macro base station as noise. At this time, the data rate R _pico of the micro user is:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

which is

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Among them, the superscript H means transpose, I is the identity matrix, is a normal matrix, and the normal matrix can be decomposed, which can be written as:

${<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; ,</span>$

Among them, Q1 is the middle matrix that decomposes the normal matrix _on the left side, is the transpose matrix of matrix Q ₁ .

Therefore, you can get:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Using the matrix properties of |I+BC|=|I+CB|, we can get:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Will Writing the singular value decomposition in the form $I+P_p{Q}_1{h}_{pp}{V}_p{V}_p^h{h}_{pp}^h{Q}_1^h={\mathrm{UAU}}^h,$ where U is the matrix The eigenvector matrix of A is a matrix with diagonal elements A matrix of singular values; then there is

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; pico</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; UAU</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; u</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; |</span>{\mathrm{<span\; class="notranslate">\; u</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; A</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}\left|\begin{array}{cccc}{\mathrm{<span\; class="notranslate">\; K</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}\\ \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 0</span>}& \mathrm{<span\; class="notranslate">\; 1</span>}\end{array}\right|<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; K</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>$

intermediate matrix $K=\left|\right|\sqrt{P_p}{Q}_1{h}_{pp}{V}_p|{|}_2.$

Set _Vp to The eigenvector x corresponding to the largest eigenvalue λ _max of , at this time R _pico obtains the maximum value R _picomax is

R _picomax = log(P _p λ _max +1)

When micro-users transmit n streams of data, n is an integer greater than 1, then select The orthogonal eigenvectors x ₁ , x ₂ , ... x _n corresponding to the largest n eigenvalues λ ₁ , λ ₂ ,…λ _n are used as the beamforming vectors of each stream data, and the obtained precoding matrix is:

V _p ＝[x ₁ x ₂ … x _n ],

At this time, R _pico achieves the maximum value R _picomax is

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span>$

Among them, λ ₁ , λ ₂ ,...λ _n are arranged in descending order, and x ₁ , x ₂ ,...x _n are eigenvectors corresponding to λ ₁ , λ ₂ ,...λ _n respectively.

In step 2.2, the micro-user performs active interference deletion.

The present invention first solves the interference as a signal, and then subtracts the interference from the useful signal. Theoretically speaking, as long as the transmission rate of the interference is lower than the channel capacity when the interference is regarded as the received signal, the interference can be resolved without error. Therefore, it is sufficient not to write the interference item directly in the data rate item representing interference cancellation in the following formula (because the transmission rate of the interference is stipulated in the condition to be smaller than the capacity).

Likewise, first consider the case where a micro-user transmits single-stream data. When the micro-user performs active interference cancellation to suppress the interference of the macro base station, the formula for sum data rate R is expressed as:

R＝R _macro +R _pico ,

R _macro represents the data rate of the macro base station.

Among them, in order to ensure that the data of the macro base station can be resolved on the direct transmission link and the cross-link, the data rate of the macro base station must meet the following conditions:

$\left\{\begin{array}{c}{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>\\ {\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; ma</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>\end{array}\right.$

Therefore, there are

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}\mathrm{<span\; class="notranslate">\; min</span>}<span\; class="notranslate">\; (</span><span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span><span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

Since the micro base station is closer to the micro user than the macro base station, when solving the cross-link interference, the signal transmitted by the micro base station to the micro user is regarded as interference. At this time, the data of the micro base station constitutes strong interference. Therefore, the present invention assumes that the channel environment of the cross-link is worse. Therefore, the macro user data rate formula can be written as

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>\frac{{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}{\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

And because of

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Therefore, there are

$\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; r</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; i</span>}\mathrm{<span\; class="notranslate">\; c</span>}\mathrm{<span\; class="notranslate">\; o</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; ,</span>$

Further transforming this formula, we get

$\begin{array}{c}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>\\ \begin{array}{c}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}\\ <span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}\end{array}\left|\begin{array}{c}\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}\\ \mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}\end{array}\right|\begin{array}{c}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span>\\ <span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span>\end{array}\end{array}<span\; class="notranslate">\; ,</span>$

In the above formula, Can be regarded as a constant.

Similarly, the normal matrix Decomposed, expressed as:

${<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span>}^{<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; ,</span>$

Among them, Q ₂ is the intermediate matrix obtained by decomposing the left normal matrix, is the transpose matrix of matrix _Q2 .

Then transform R into:

$\begin{array}{c}\mathrm{<span\; class="notranslate">\; R</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span>\\ <span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; log</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; p</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; Q</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; |</span>\end{array}<span\; class="notranslate">\; ,</span>$

Set _Vp to The eigenvector x corresponding to the largest eigenvalue λ _max of , at this time R obtains the maximum value R _max , as follows:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

Similarly, if the micro-user transmits n-stream data, select Orthogonal eigenvectors x ₁ , x ₂ , ... x _n corresponding to the largest n eigenvalues are used as the beamforming vectors of each stream data, that is, the precoding matrix V _p =[x ₁ x ₂ ... x _n ], at this time , R obtains the maximum value R _max is:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; a</span>}\mathrm{<span\; class="notranslate">\; x</span>}}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; |</span>\mathrm{<span\; class="notranslate">\; I</span>}<span\; class="notranslate">\; +</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}{\mathrm{<span\; class="notranslate">\; V</span>}}_{\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}{\mathrm{<span\; class="notranslate">\; h</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}\mathrm{<span\; class="notranslate">\; m</span>}}^{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; |</span><span\; class="notranslate">\; +</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{<span\; class="notranslate">\; \&Sigma;</span>}}\mathrm{<span\; class="notranslate">\; l</span>}\mathrm{<span\; class="notranslate">\; o</span>}\mathrm{<span\; class="notranslate">\; g</span>}<span\; class="notranslate">\; (</span>{\mathrm{<span\; class="notranslate">\; P</span>}}_{\mathrm{<span\; class="notranslate">\; p</span>}}{\mathrm{<span\; class="notranslate">\; \&lambda;</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; .</span>$

Among them, λ ₁ , λ ₂ ,...λ _n are arranged in descending order, and x ₁ , x ₂ ,...x _n are eigenvectors corresponding to λ ₁ , λ ₂ ,...λ _n respectively.

Next, the method of the present invention is simulated and tested. The simulation parameters are selected as follows: the number of macro base station antennas is 128, the radius of the macro cell is 500m, and it is divided into 4 clusters, and the total number of user antennas in each cluster is 4; there is a micro cell near each cluster of the macro cell, and its radius is 60m, the number of micro base station antennas is 4, each micro base station serves 1 micro user, and the number of micro user antennas is 4. The transmit power of the macro base station is 46dBm, the transmit power of the micro base station is 30dBm, and the signal-to-noise ratio at the edge of the macro cell is 5dB. The channel of the macro cell adopts the 3GPP channel model, the small-scale fading of the micro-cell channel obeys the Rayleigh distribution, and the large-scale fading meets the definition of the 3GPP channel model. The channel statistical information is obtained through statistics after 1000 channel generation, and the channel instantaneous information is realized through a single generation of the channel under the same conditions. The results were averaged by repeating the experiment 1000 times.

Simulation results are shown in Figure 3. Select MRT, eICIC method and the method of the present invention to compare. Wherein, the MRT method means that the micro base station adopts MRT precoding (Maximum Ratio Transmit Precoding), and treats all interference as noise. The eICIC (Enhanced Inter-Cell Interference Coordination) method makes the macro cell and the micro cell orthogonal in time. The curves eICIC_2, eICIC_3, and eICIC_4 respectively correspond to the data rates when the number of time slots occupied by the macro cell is 2, 3, and 4. Here, the total number of downlink time slots in each data frame is considered to be 6. It should be noted that as the number of time slots allocated by the macro cell increases, the system throughput decreases, because the channel condition of the micro cell is better, and the system throughput is mainly provided by the micro cell. It can be seen from the simulation results that the system and data rate of the scheme proposed in this application are obviously superior to the MRT method and the eICIC method.

Claims (5)

1. An interference suppression transmission method in a large-scale MIMO heterogeneous network is characterized by comprising the following implementation steps:

step one, judging whether the interference intensity of a macro base station on a micro base station is strong interference or weak interference;

step two, the micro base station adopts a corresponding processing mode according to the interference intensity of the macro base station: (1) when the interference is weak interference, the interference of the macro base station is taken as noise processing, (2) when the interference is strong interference, the interference deletion processing is adopted;

let H_mmChannel direction from macro base station to macro user in the g-th clusterAmount, V_mOuter precoding matrix or complete precoding matrix for macro base station to macro user data of the g cluster, H_ppFor the channel from the g-th micro base station to the g-th micro user, V_pPrecoding matrix for the g-th micro base station, H_pmA channel from the macro base station to the g-th micro user; p_mNormalized power for macro base station transmit power and noise power,P_pnormalized power for the micro base station transmit power and noise power,P_macrotransmitting power, P, for macro base stations_picoFor transmitting power, P, to the micro base station_NIs the noise power; g is a positive integer;

step 2.1, the treatment mode (1) is as follows: data rate R of micro-users_picoExpressed as:

$R_{pico}=log|I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H)}^{-1}|,$

wherein the upper corner H represents transpose, I is an identity matrix,to be a regular matrix, the regular matrix is decomposed and represented as:Q₁a middle matrix for decomposing the left normal matrix;

(1.1) when a micro-user transmits a single stream data, V_pIs composed ofMaximum eigenvalue λ of_maxA corresponding feature vector x;

(1.2) when the micro-user transmits n stream data, n is an integer greater than 1, obtainingN maximum eigenvalues and arranged in descending order of λ₁,λ₂,…λ_n，x₁,x₂,…x_nAre each lambda₁,λ₂,…λ_nCorresponding eigenvector to obtain a precoding matrix V_p＝[x₁x₂…x_n]；

Step 2.2, the treatment mode (2) is as follows: and the data rate R is expressed as:

$R=log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+log|I+P_p{(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H)}^{-1}{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H)|;$

wherein, the upper corner mark H represents transposition, I is an identity matrix, and the normal matrix isIs decomposed intoQ₂Is an intermediate matrix obtained by decomposition;

(2.1) when the micro-user transmits single stream data, set V_pIs composed ofMaximum eigenvalue λ of_maxA corresponding feature vector x;

(2.2) when the micro-user transmits n stream data, n is an integer greater than 1, selectingOrthogonal eigenvector x corresponding to the largest n eigenvalues₁,x₂,…x_nForming a precoding matrix V_p，V_p＝[x₁x₂…x_n]。

2. The method according to claim 1, wherein in the first step, the method for determining the interference strength generated by the macro base station comprises: calculating the 2 norm of the channel matrix at the receiving end of the micro-user: if the 2 norm of the channel matrix of the direct transmission link from the micro base station to the micro user is greater than the 2 norm of the channel matrix of the cross link from the macro base station to the micro user, the interference strength of the macro base station is weak interference; otherwise, the interference strength of the macro base station is strong interference.

3. The method according to claim 1, wherein in the first step, the method for determining the interference strength generated by the macro base station comprises: and D, calculating the data rates corresponding to the two processing modes in the step two in advance at the micro base station end: if the data rate of the interference as noise is higher than the data rate of the interference deletion, the interference strength of the macro base station is weak interference; otherwise, the interference strength of the macro base station is strong interference.

4. The method as claimed in claim 1, wherein in step 2.1, the interference generated by the macro base station is regarded as noise by the micro user, and the interference is considered as noise by the micro user

$R_{pico}=log|I+\frac{P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H}{I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H}|,$

Namely, it is

$R_{pico}=log|I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H)}^{-1}|,$

Will be regular matrixIs decomposed into

Thus obtaining $R_{pico}=log|I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{Q}_1^H{Q}_1|,$ Further obtaining:

$R_{pico}=log|I+P_p{Q}_1{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{Q}_1^H|,$

will be provided withWritten form of singular value decomposition $I+P_p{Q}_1{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{Q}_1^H={\mathrm{UAU}}^H,$ Then there is

$R_{pico}=log\left|{\mathrm{UAU}}^H\right|=log\left|U\right|\left|A\right|\left|U^H\right|=log\left|A\right|=log\left|\begin{array}{cccc}{K}^2+1& 0& 0& 0\\ 0& 1& 0& 0\\ 0& 0& 1& 0\\ 0& 0& 0& 1\end{array}\right|=log(K^2+1),$

Wherein U and A are matrices respectivelyThe eigenvector matrix and the matrix with diagonal elements as singular values; intermediate matrix $K=\left|\right|\sqrt{P_p}{Q}_1{H}_{pp}{V}_p|{|}_2;$

If the micro-user transmits single stream data, V_pIs composed ofMaximum eigenvalue λ of_maxCorresponding feature vector x, when R_picoObtaining a maximum value R_picomax，R_picomax＝log(P_pλ_max+1)；

If the micro-user transmits n stream data, selectingMaximum n eigenvalues λ₁,λ₂,…λ_nCorresponding orthogonal eigenvectors x₁,x₂,…x_nObtaining a precoding matrix V as a beamforming vector for each stream data_p＝[x₁x₂…x_n]At this time, R_picoObtaining a maximum value R_picomax， $R_{picomax}=\underset{k=1}{\overset{n}{\&Sigma;}}log(P_p{\mathrm{\&lambda;}}_k+1).$

5. The method according to claim 1, wherein in step 2.2, when the micro user performs active interference cancellation to suppress interference of the macro base station, the sum data rate R is:

R＝R_macro+R_pico,

R_macrodata rate, R, of macro base station_picoRepresenting the data rate of the micro-user;

in order to ensure that the data of the macro base station can be solved in the direct transmission link and the cross link, the data rate of the macro base station meets the following conditions:

$\left\{\begin{array}{c}{R}_{macro}\&le;\log |I+\frac{P_m{H}_{mm}{V}_m{V}_m^H{H}_{mm}^H}{I+P_p{H}_{mp}{V}_p{V}_p^H{H}_{mp}^H}|,\\ R_{macro}\&le;\log |I+\frac{P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H}{I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H}|,\end{array}\right.$

thus, there are

$R_{macro}=\log \min \left(\right|I+\frac{P_m{H}_{mm}{V}_m{V}_m^H{H}_{mm}^H}{I+P_p{H}_{mp}{V}_p{V}_p^H{H}_{mp}^H}|,|I+\frac{P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H}{I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H}\left|\right)$

Because the micro base station is closer to the micro user than the macro base station, when the cross link interference is solved, the signal transmitted to the micro user by the micro base station is taken as the interference, and at the moment, the data of the micro base station forms strong interference; assuming that the channel environment of the cross-link is worse, the macro-user data rate is further expressed as

$R_{macro}=\log |I+\frac{P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H}{I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H}|,$

And due to

$R_{pico}=log|I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H|,$

Thus, the resulting sum data rate R is:

$R=R_{macro}+R_{pico}=log|I+P_p{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|,$

further transformation of the above equation yields:

$\begin{array}{c}R=log\left|(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H)(I+P_p{\left(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H\right)}^{-1}{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H)\right|\\ =log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H||I+P_p{\left(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H\right)}^{-1}{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H)|\\ =log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+\log |I+P_p{\left(I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H\right)}^{-1}{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H)|\end{array},$

will be regular matrixIs decomposed into

Then the sum data rate R is further expressed as:

$\begin{array}{c}R=log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+\log |I+P_p{Q}_2^H{Q}_2{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H)|\\ =log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+\log |I+P_p{Q}_2{H}_{pp}{V}_p{V}_p^H{H}_{pp}^H{Q}_2^H)|\end{array},$

when a micro-user transmits single-stream data, set V_pIs composed ofMaximum eigenvalue λ of_maxCorresponding feature vector x, when R takes the maximum value R_max， $R_{max}=log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+log(P_p{\mathrm{\&lambda;}}_{max}+1);$

When the micro-user transmits n stream data, n is an integer greater than 1, selectingOrthogonal eigenvector x corresponding to the largest n eigenvalues₁,x₂,…x_nForming a precoding matrix V_pWhen R takes the maximum value R_max：

$R_{max}=log|I+P_m{H}_{pm}{V}_m{V}_m^H{H}_{pm}^H|+\underset{k=1}{\overset{n}{\&Sigma;}}log(P_p{\mathrm{\&lambda;}}_k+1)$

Wherein λ is₁,λ₂,…λ_nArranged in descending order, x₁,x₂,…x_nAre each lambda₁,λ₂,…λ_nThe corresponding feature vector.