CN114785469B

CN114785469B - Pilot frequency determining method, device, electronic equipment and storage medium

Info

Publication number: CN114785469B
Application number: CN202210375188.2A
Authority: CN
Inventors: 李立华; 周茅玲; 孙轩轩
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2022-04-11
Filing date: 2022-04-11
Publication date: 2023-11-14
Anticipated expiration: 2042-04-11
Also published as: CN114785469A

Abstract

The invention provides a pilot frequency determining method, a device, electronic equipment and a storage medium, wherein the method comprises the following steps: determining one or more pilot allocation configurations corresponding to the dense region based on a first set of APs serving the dense region, and determining one or more pilot allocation configurations corresponding to the sparse region based on a second set of APs serving the sparse region; and determining target pilot frequency allocation configuration in one or more pilot frequency allocation configurations corresponding to the dense area and one or more pilot frequency allocation configurations corresponding to the sparse area so as to maximize the system downlink total rate corresponding to the target pilot frequency allocation configuration. According to the embodiment of the invention, the dense area and the sparse area are divided, one or more pilot frequency distribution configurations can be respectively determined for the dense area and the sparse area, and then pilot frequency pollution of the dense area can be relieved by determining the target pilot frequency distribution configuration, so that good communication quality can be obtained for the UE in different geographic positions in the service area.

Description

Pilot frequency determining method, device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of wireless communications technologies, and in particular, to a method and apparatus for determining a pilot frequency, an electronic device, and a storage medium.

Background

In a Cell-free massive multiple-input multiple-output (Cell-free massive Multiple input Multiple output, CF mimo) system is a network of distributed antennas designed to achieve nearly uniform high communication quality over a given geographic area.

The CF mimo system in the related art mainly performs pilot allocation for a User Equipment (UE) in a service area in view of a scenario in which the UE is uniformly distributed in the service area. Since the UEs are freely movable in the service area, the distribution of UEs within a piece of area is not always uniformly distributed. In the region where the UE is denser, the pilot frequency multiplexing condition is more serious, and the generated pilot frequency pollution is more, which is unfavorable for the data transmission quality; in a sparse area of the UE, the pilot frequency multiplexing degree among multiple UE is light, pilot frequency pollution is less, and better data transmission quality can be obtained, so that the UE in different geographic positions in the service area cannot obtain good communication quality.

Disclosure of Invention

The invention provides a pilot frequency determining method, a device, electronic equipment and a storage medium, which are used for solving the defect that in the prior art, UE (user equipment) in different geographic positions in a service area cannot obtain good communication quality, and realizing that the UE in different geographic positions in the service area can obtain good communication quality.

In a first aspect, the present invention provides a method for determining pilot frequency in a honeycomb-free large-scale mimo system, including:

determining one or more pilot allocation configurations corresponding to a dense region based on a first AP set serving the dense region, and determining one or more pilot allocation configurations corresponding to a sparse region based on a second AP set serving the sparse region;

determining a target pilot frequency allocation configuration in one or more pilot frequency allocation configurations corresponding to the dense area and one or more pilot frequency allocation configurations corresponding to the sparse area so as to maximize the system downlink total rate corresponding to the target pilot frequency allocation configuration;

wherein the dense region and the sparse region are determined based on historical user equipment distribution data in a service region, the service region comprising the dense region and the sparse region, the first set of APs having no intersection with the second set of APs.

Optionally, according to the method for determining pilots in a honeycomb-free large-scale multiple input multiple output system provided by the present invention, the determining one or more pilot allocation configurations corresponding to the dense area based on the first AP set of the service dense area includes:

Determining a large-scale fading coefficient between each AP in the first set of APs and the user equipment in the dense area, in the case that the number of the plurality of orthogonal pilot sequences is smaller than the number of the user equipment in the dense area;

and acquiring one or more pilot allocation configurations corresponding to the dense area based on a large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and a distance between each AP in the first AP set and the user equipment in the dense area.

Optionally, according to the method for determining pilots in a honeycomb-free large-scale multiple input multiple output system provided by the present invention, the obtaining one or more pilot allocation configurations corresponding to the dense area based on a large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and a distance between each AP in the first AP set and the user equipment in the dense area includes:

determining channel similarity between each AP in the first set of APs and user equipment in the dense area based on a large scale fading coefficient between each AP in the first set of APs and user equipment in the dense area;

Determining a first joint matrix based on channel similarity between each AP in the first AP set and the user equipment in the dense area and distance between each AP in the first AP set and the user equipment in the dense area, wherein each row in the first joint matrix is used for representing interference degree between the user equipment served by the same AP;

sorting elements corresponding to each row in the first joint matrix from small to large to obtain a second joint matrix corresponding to the first joint matrix;

acquiring a group corresponding to each row of the second joint matrix based on the number of the plurality of orthogonal pilot sequences;

and carrying out pilot frequency distribution on the groups corresponding to each row of the second joint matrix, and obtaining one or more pilot frequency distribution configurations corresponding to the dense area.

Optionally, according to the method for determining pilots in a cellular-free massive multiple-input multiple-output system provided by the present invention, the determining one or more pilot allocation configurations corresponding to the sparse area based on the second AP set serving the sparse area includes:

determining a service relationship between each AP of the second set of APs and the user equipment in the sparse region based on a large-scale fading coefficient between each AP of the second set of APs and the user equipment in the sparse region, the service relationship being used to characterize a situation in which the AP provides service for the user equipment;

Determining a third combining matrix based on a service relationship between each AP of the second set of APs and user equipment in the sparse region and channel estimation between each AP of the second set of APs and user equipment in the sparse region, the third combining matrix being used to characterize the degree of interference between all user equipment in the sparse region;

acquiring a fourth joint matrix based on a target interference threshold and the third joint matrix, wherein the number of rows and columns of the fourth joint matrix is the same as that of rows and columns of the third joint matrix, the value of a second element is 1 when a first element is larger than or equal to the target interference threshold, the value of the second element is 0 when the first element is smaller than the target interference threshold, the first element is any element in the third joint matrix, and the second element is an element with the same row and column number as that of the first element in the fourth joint matrix;

determining a structure of a target graph based on the fourth joint matrix, and determining an information amount corresponding to each edge between all vertexes of the target graph based on the third joint matrix, wherein each vertex of the target graph has a unique corresponding relation with each user equipment in the sparse area, and the number of vertexes of the target graph is the same as the number of the user equipment in the sparse area;

Performing coloring operation on each vertex of the target graph based on the number of the plurality of orthogonal pilot sequences, and determining one or more coloring configurations, wherein any one of the one or more coloring configurations comprises color information corresponding to all vertices of the target graph;

one or more pilot allocation configurations corresponding to the sparse region are determined based on the one or more coloring configurations.

Optionally, according to the method for determining a pilot frequency in a honeycomb-free large-scale multiple input multiple output system provided by the present invention, the first coloring operation includes:

determining one vertex with the largest interference value sum as a starting vertex in all vertexes of the target graph based on the third combining matrix, wherein the interference value sum corresponding to any one target vertex of the target graph is the sum of elements corresponding to a target row in the third combining matrix, and the target vertex corresponds to the target row;

and selecting a first color from a color list, and coloring the initial vertex, wherein the number of the colors of the color list is equal to the number of the plurality of orthogonal pilot sequences.

Optionally, according to the method for determining a pilot frequency in a honeycomb-free large-scale multiple input multiple output system provided by the present invention, the nth coloring operation includes:

Determining a third vertex in one or more second vertexes adjacent to the first vertex based on the information quantity of each side connected with the first vertex, wherein user equipment corresponding to the third vertex has the greatest interference on the user equipment corresponding to the first vertex;

selecting a second color from the color list, and coloring the third vertex so that the color of the third vertex is different from the color corresponding to any vertex adjacent to the third vertex;

configuring an information amount corresponding to an edge between the first vertex and the second vertex to be 0;

wherein the first vertex is a vertex colored in the (N-1) -th coloring operation, N is an integer, and N is greater than or equal to 2.

Optionally, according to the method for determining pilots in a cellular-free massive multiple input multiple output system provided by the present invention, the determining one or more pilot allocation configurations corresponding to the sparse area based on the one or more coloring configurations includes:

screening the one or more coloring configurations based on a color usage frequency threshold and a color usage frequency corresponding to each coloring configuration to obtain one or more target coloring configurations, so that the color usage frequency corresponding to each target coloring configuration is smaller than or equal to the color usage frequency threshold;

Determining one or more pilot allocation configurations corresponding to the sparse region based on the one or more target coloring configurations;

wherein the color usage number threshold is determined based on the number of user devices in the sparse region and the number of the plurality of orthogonal pilot sequences.

Optionally, according to the method for determining pilots in a honeycomb-free large-scale multiple input multiple output system provided by the present invention, before the first AP set based on a service dense area determines one or more pilot allocation configurations corresponding to the dense area, and the second AP set based on a service sparse area determines one or more pilot allocation configurations corresponding to the sparse area, the method further includes:

determining the dense region and the sparse region based on historical user equipment distribution data in the service region;

determining the first set of APs based on a distance threshold and distances between all APs of the service area and a center location of the dense area;

and determining the AP except the first AP set in all the APs as the second AP set.

In a second aspect, the present invention also provides a pilot frequency determining device in a honeycomb-free large-scale mimo system, including:

A first determining module, configured to determine, based on a first AP set serving a dense area, one or more pilot allocation configurations corresponding to the dense area, and determine, based on a second AP set serving a sparse area, one or more pilot allocation configurations corresponding to the sparse area;

a second determining module, configured to determine a target pilot allocation configuration in one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area, so as to maximize a system downlink total rate corresponding to the target pilot allocation configuration;

In a third aspect, the present invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing a method for pilot determination in a non-cellular massive multiple-input multiple-output system as described in any one of the above when executing the program.

In a fourth aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a method of pilot determination in a non-cellular massive multiple-input multiple-output system as described in any of the above.

In a fifth aspect, the present invention also provides a computer program product comprising a computer program which when executed by a processor implements a method of pilot determination in a non-cellular massive multiple-input multiple-output system as described in any one of the above.

According to the pilot frequency determining method, the device, the electronic equipment and the storage medium, the service area is divided into the dense area and the sparse area, one or more pilot frequency distribution configurations corresponding to the dense area can be determined based on the first AP set, one or more pilot frequency distribution configurations corresponding to the sparse area can be determined based on the second AP set, the target pilot frequency distribution configuration is determined so that the total downlink speed of the system corresponding to the target pilot frequency distribution configuration is maximum, one or more pilot frequency distribution configurations corresponding to the dense area and one or more pilot frequency distribution configurations corresponding to the sparse area can be screened, pilot frequency pollution of the dense area can be relieved by the screened pilot frequency distribution configurations, and good communication quality of the UE in different geographic positions in the service area can be achieved.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is one of the schematic diagrams of a CF mMIMO system provided by the related art;

FIG. 2 is a schematic diagram of a second related art CF mMIMO system;

fig. 3 is a schematic flow chart of a pilot frequency determining method in a honeycomb-free large-scale mimo system according to the present invention;

FIG. 4 is a schematic diagram of a CF mMIMO system provided by the present invention;

fig. 5 is a second flow chart of a pilot determining method in a honeycomb-free large-scale mimo system according to the present invention;

fig. 6 is a third flow chart of a pilot determining method in a honeycomb-free large-scale mimo system according to the present invention;

FIG. 7 is one of the experimental simulation schematics provided by the present invention;

FIG. 8 is a second experimental simulation diagram provided by the present invention;

FIG. 9 is a third experimental simulation diagram provided by the present invention;

FIG. 10 is a fourth experimental simulation schematic provided by the present invention;

FIG. 11 is a fifth experimental simulation schematic provided by the present invention;

FIG. 12 is a sixth experimental simulation schematic provided by the present invention;

fig. 13 is a schematic structural diagram of a pilot determining device in a honeycomb-free large-scale mimo system according to the present invention;

fig. 14 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

In order to facilitate a clearer understanding of various embodiments of the present invention, some relevant background knowledge is first presented as follows.

Fig. 1 is one of schematic diagrams of a CF mimo system provided in the related art, as shown in fig. 1, in the CF mimo system, M Access Points (APs) may all be equipped with multiple antennas, and K UEs with single antennas are served simultaneously on the same time-band resource, and M is far greater than K. The link from the UE to the AP is called an uplink, the transmission link from the AP to the UE is called a downlink, and each AP is connected to a central processing unit (Central Processing Unit, CPU) of the CF mimo system through a backhaul link to perform information transmission. The system adopts a time division duplex (Time Division Duplexing, TDD) working mode, and each coherence interval can be divided into 3 phases:

In the uplink training stage of the first stage, the UE transmits a pilot sequence allocated to the UE to the AP through an uplink, and the AP performs channel estimation by using the received pilot signal at a receiving end to obtain channel state information (Channel State Information, CSI);

in the uplink data transmission stage of the second stage, the UE transmits data to the AP, the AP firstly carries out local signal detection and then transmits the data to a CPU of the CF mMIMO system, and the CPU of the CF mMIMO system carries out centralized detection on the UE data according to the received data and the estimated value of the statistical channel;

in the downlink data transmission stage of the third stage, the AP performs power control and precoding on data to be transmitted to the UE through a power coefficient allocated by the CPU of the CF mimo system and a locally estimated channel, and transmits the data to the UE.

The advantages offered by CF mimo over cellular networks are mainly manifested in 3 aspects: (1) The signal-to-noise ratio is higher and more uniform, and the variation of the signal-to-noise ratio is smaller; (2) has stronger anti-interference capability; (3) coherent transmission may increase signal-to-noise ratio.

When a large number of APs are widely distributed in the CF mimo system, in the conventional full connection (as shown in fig. 1) mode, if the current UE is far away from some APs, the APs to serve the UE will cause stronger interference to UEs around themselves, which affects the overall performance of the system. To overcome this drawback, CF mimo with User-center (UC) has been proposed in the related art.

Fig. 2 is a schematic diagram of a second CF mimo system provided by the related art, as shown in fig. 2, in which each UE is served by only a part of APs, compared to a conventional CF mimo system, less backhaul overhead is required, and for most UEs in the network, the UE availability rate is superior to that of the conventional CF mimo system, and the energy efficiency is also higher than that of the conventional CF mimo system.

In practical application scenarios, the UE distribution in a service area is not always uniform, for example, for a city area, there is a small area (such as tourist attractions) where the user equipment is dense on holidays or on active days. In a scenario where UE distribution in a service area is not uniform, the CF mimo system in the related art has the following two drawbacks:

(1) When the number of the pilots is insufficient, the UEs in the same area have more numbers and more pilot multiplexing, and when the channel estimation is performed, the UEs are affected by the interference of other UEs nearby the UEs, so that the pilot pollution becomes more serious, and meanwhile, for the APs in the dense area, the limited power resources are required to be distributed to the signals of a plurality of transmitted users, so that the receiving rate of the UEs is also lost;

(2) If the AP still serves all the UEs, not only is the coverage capability of the AP higher required, but also the CPU of the CF mMIMO system is required to have more computing power; and the interference to UE equipment nearby the AP is stronger when the remote UE is served, so that the cost and the interference are larger.

In order to overcome the defects, the invention provides a pilot frequency determining method, a device, electronic equipment and a storage medium, and the method, the device, the electronic equipment and the storage medium can realize that the UE in different geographic positions in a service area can obtain good communication quality by respectively determining pilot frequency allocation configuration for a dense area and a sparse area.

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Fig. 3 is a schematic flow chart of a pilot frequency determining method in a non-cellular massive multiple input multiple output system according to the present invention, and as shown in fig. 3, an execution subject of the pilot frequency determining method in the non-cellular massive multiple input multiple output system may be an electronic device or a module in the electronic device, for example, a CPU in a CF mimo system, etc. The method comprises the following steps:

Step 301, determining one or more pilot allocation configurations corresponding to a dense area based on a first AP set serving the dense area, and determining one or more pilot allocation configurations corresponding to a sparse area based on a second AP set serving the sparse area;

wherein the dense region and the sparse region are determined based on historical user equipment distribution data in a service region, the service region comprising the dense region and the sparse region, the first set of APs having no intersection with the second set of APs;

specifically, in the pilot allocation process, pilot allocation configurations may be determined for dense areas and sparse areas in the service area, one or more pilot allocation configurations may be determined for UEs in the dense areas based on a first AP set serving the dense areas, and one or more pilot allocation configurations may be determined for UEs in the sparse areas based on a second AP set serving the sparse areas.

Alternatively, the serving cell may be a service area providing communication services for the user equipment.

Alternatively, the one or more pilot allocation configurations corresponding to the dense region and the one or more pilot allocation configurations corresponding to the sparse region may be determined simultaneously by parallel processing.

Alternatively, one or more pilot allocation configurations corresponding to the sparse region may be determined after one or more pilot allocation configurations corresponding to the dense region are determined.

Alternatively, one or more pilot allocation configurations corresponding to dense regions may be determined after one or more pilot allocation configurations corresponding to sparse regions are determined.

Alternatively, one or more pilot allocation configurations may be determined for UEs in dense areas based on the first set of APs by one or more pilot allocation algorithms.

Alternatively, one or more pilot allocation configurations may be determined for the sparse region of UEs based on the second set of APs by one or more pilot allocation algorithms.

Accordingly, pilot allocation configurations may be determined for dense and sparse regions, respectively, and one or more pilot allocation configurations corresponding to the dense region and one or more pilot allocation configurations corresponding to the sparse region may be used to determine a target pilot allocation configuration.

Step 302, determining a target pilot allocation configuration in one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area, so as to maximize a system downlink total rate corresponding to the target pilot allocation configuration;

Specifically, after obtaining one or more pilot allocation configurations corresponding to the dense region and one or more pilot allocation configurations corresponding to the sparse region, the optimization target of the maximum system downlink total rate corresponding to the target pilot allocation configuration may be selected from the pilot allocation configurations corresponding to the dense region and the pilot allocation configurations corresponding to the sparse region, and one pilot allocation configuration corresponding to the dense region and one pilot allocation configuration corresponding to the sparse region may be selected, so that the target pilot allocation configuration may be determined.

It will be appreciated that after determining the target pilot allocation configuration, the CPU in the CF mimo system may issue the target pilot allocation configuration to UEs in the service area through APs in the service area.

Optionally, fig. 4 is a schematic diagram of a CF mimo system provided by the present invention, where, as shown in fig. 4, the CF mimo system may include a CPU, multiple APs and multiple UEs, where the CPU may divide a dense area and a sparse area in a service area based on historical user equipment distribution data in the service area, may divide one or more dense areas in the service area, and may also divide one or more sparse areas in the service area.

According to the pilot frequency determining method in the honeycomb-free large-scale multi-input multi-output system, the service area is divided into the dense area and the sparse area, one or more pilot frequency distribution configurations corresponding to the dense area can be determined based on the first AP set, one or more pilot frequency distribution configurations corresponding to the sparse area can be determined based on the second AP set, the target pilot frequency distribution configuration is determined so that the system downlink total rate corresponding to the target pilot frequency distribution configuration is maximum, one or more pilot frequency distribution configurations corresponding to the dense area and one or more pilot frequency distribution configurations corresponding to the sparse area can be screened, pilot frequency pollution of the dense area can be relieved by the screened pilot frequency distribution configurations, UE (user equipment) in different geographic positions in the service area can be realized, and good communication quality can be obtained.

Optionally, the determining, based on the first AP set serving the dense area, one or more pilot allocation configurations corresponding to the dense area includes:

Specifically, in the pilot allocation process, pilot allocation configurations may be determined for a dense area and a sparse area in a service area, and in the case that the number of multiple orthogonal pilot sequences is smaller than the number of user devices in the dense area, one or more pilot allocation configurations corresponding to the dense area may be obtained based on a large-scale fading coefficient between each AP in the first AP set and the user devices in the dense area, and a distance between each AP in the first AP set and the user devices in the dense area;

specifically, based on the second AP set serving the sparse area, one or more pilot allocation configurations may be determined for the UE in the sparse area, and then, based on an optimization target that a total downlink rate of the system corresponding to the target pilot allocation configuration is maximum, screening may be performed between the pilot allocation configuration corresponding to the dense area and the pilot allocation configuration corresponding to the sparse area, and one pilot allocation configuration corresponding to the dense area and one pilot allocation configuration corresponding to the sparse area may be screened, and then, the target pilot allocation configuration may be determined.

Optionally, the number of pilot sequences (τ _p ) Greater than or equal to the number of UEs in the dense area (|u) ^dense I), orthogonal pilot sequences can be allocated to UEs in the dense area, and UEs in the dense area can be configured to be served by the same AP set in a fully connected manner, and one pilot allocation configuration corresponding to the dense area can be obtained. In this case, interference of UEs in a dense area upon data reception is mainly due to the influence of UEs in a sparse area, while there is no interference at all between UEs in a dense area.

It can be appreciated that the embodiments of the present invention provide a geographic and channel based pilot allocation (Location Hata Pilot Assignment, LHPA) algorithm that reduces pilot multiplexing for nearby UEs in the vicinity of the same region based on geographic location and channel characteristics, thereby reducing pilot pollution in that region.

Optionally, the number of pilot sequences (τ _p ) Less than the number of UEs in dense areas (|u) ^dense I) is providedThe main difference between the UEs in the dense area in this case is geographic location and channel condition, and the LHPA algorithm can be used for pilot allocation;

specifically, the procedure for pilot allocation using LHPA algorithm may include: determining a large-scale fading coefficient between each AP in the first set of APs and the user equipment in the dense area; and further, based on the large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and the distance between each AP in the first AP set and the user equipment in the dense area, one or more pilot frequency allocation configurations corresponding to the dense area can be obtained.

It can be appreciated that, for pilot allocation in dense areas, when the number of pilot resources provided by the system is greater than the number of UEs in the dense areas, no pilot pollution exists between the UEs in the dense areas; even if there are too many UEs in the dense area to multiplex, the effect of pilot multiplexing is reduced compared to an allocation algorithm that does not consider this scenario (does not distinguish between dense and sparse areas). Moreover, pilot multiplexing may be reduced for nearby UEs in the vicinity of the same region based on the geographic location and characteristics of the channel, thereby reducing pilot pollution for that region.

It can be appreciated that for the pilot allocation of sparse areas, the pollution caused by pilot multiplexing is also reduced compared to allocation algorithms that would not consider this scenario (without distinguishing dense and sparse areas) due to the effect of geographical location.

Therefore, based on the large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and the distance between each AP in the first AP set and the user equipment in the dense area, one or more pilot allocation configurations corresponding to the dense area can be determined, by determining the target pilot allocation configuration, one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area can be screened, the screened pilot allocation configurations can be enabled to alleviate pilot pollution of the dense area, and UEs in different geographic positions in the service area can be realized, so that good communication quality can be obtained.

Optionally, the obtaining one or more pilot allocation configurations corresponding to the dense area based on a large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and a distance between each AP in the first AP set and the user equipment in the dense area includes:

Specifically, in the pilot allocation process, pilot allocation configuration can be determined for a dense area and a sparse area in a service area, and under the condition that the number of a plurality of orthogonal pilot sequences is smaller than the number of user equipment in the dense area, based on a large-scale fading coefficient between each AP in a first AP set and the user equipment in the dense area, channel similarity between each AP in the first AP set and the user equipment in the dense area can be determined, and then a first joint matrix can be determined in combination with the distance between each AP in the first AP set and the user equipment in the dense area;

specifically, after the first joint matrix is determined, the elements corresponding to each row in the first joint matrix can be ordered from small to large (ascending order), so as to obtain a second joint matrix, further, based on the number of the plurality of orthogonal pilot sequences, the group corresponding to each row of the second joint matrix can be obtained, further, pilot frequency distribution is performed on the group corresponding to each row of the second joint matrix, and one or more pilot frequency distribution configurations corresponding to the dense area can be obtained;

Alternatively, based on the large scale fading coefficients between each AP in the first set of APs and the user equipment in the dense area, a large scale fading matrix β for all UEs centered on AP service may be determined _m ：

β _m ＝[β _m1 ,β _m2 ,...,β _mK ]；

Wherein,k represents the total number of UEs in the dense area, beta _m1 Representing the large-scale fading coefficient, beta, between the mth AP in the dense area and the 1 st UE in the dense area _m2 Representing the large scale fading coefficient between the mth AP in the dense area and the 2 nd UE in the dense area, and so on, beta _mK Representing a large-scale fading coefficient between an mth AP in the dense area and a kth UE in the dense area;

further, the mean value β of the large-scale fading coefficients of the UE served by each AP can be calculated by the following formula _mCenter ：

Wherein beta is _mk Representing a large-scale fading coefficient between an mth AP in the dense area and a kth UE in the dense area;

further, the channel similarity between an AP and its served UEs in a dense area can be calculated by the following channel similarity metric function:

further, for the user equipment j served by the ith AP, the channel similarity F may be calculated by the following formula _ij ：

The channel similarity matrix F formed by all UEs served by all APs can be obtained by the channel similarity metric function:

wherein,m represents the total number of all APs in the dense area and K represents the total number of UEs in the dense area.

It will be appreciated that the channel similarity matrix F described above may be determined based on the large scale fading coefficients between each AP in the first set of APs and the user equipment in the dense area, the channel similarity matrix F comprising the channel similarities between each AP in the first set of APs and the UEs in the dense area.

Alternatively, the distance between each AP in the first set of APs and the user equipment in the dense area may be determined by the following formula:

wherein p is _j Representing the coordinate position of the jth UE in the dense region, p _i Representing the coordinate position of the ith AP in the dense area, P _ij Representing the distance of the ith AP from the jth UE it serves;

further, based on the distance of the APs in the dense area from all UEs in the dense area, a distance matrix P can be constructed:

It will be appreciated that the distance matrix P includes the distance between each AP in the first set of APs and the user devices in the dense area.

Alternatively, based on the channel similarity matrix F and the distance matrix P described above, a first joint matrix Q may be determined:

wherein the first joint matrixThe elements in Q may represent the corresponding joint values (joint channel similarity and distance) of the UEs in the dense region.

Optionally, the elements corresponding to each row in the first joint matrix Q are sorted from small to large (ascending process), and the second joint matrix S corresponding to the first joint matrix may be obtained:

wherein,matrix element S _ij Representing the joint value of the UE ordered as j on the i-th AP.

Alternatively, the number τ of orthogonal pilot sequences may be multiple after the second combining matrix S is acquired _p Grouping one row corresponding to the ith AP in the second joint matrix S, and dividing the row into N _g Group:

where K represents the total number of UEs in the dense area;

Further, packet case G of the ith AP _i Can be expressed as:

alternatively, in acquiring packet case G of the ith AP _i Thereafter, orthogonal pilot assignments may be made for each group of ith AP, i.e., for τ in each group _p The UE sequentially selects orthogonal pilot sequences to obtain pilot index sets distributed by all the UEs of the ith APThe pilot index set Ps _i Corresponding to a pilot frequency allocation configuration; further, by determining the pilot index set corresponding to each AP in the dense area, one or more pilot allocation configurations corresponding to the dense area may be obtained.

It can be understood that by sorting the elements corresponding to each row in the first joint matrix from small to large, the second joint matrix corresponding to the first joint matrix is obtained, so that UEs with larger interference to other UEs can be located at the position of being sorted forward as much as possible, and further, in the process of pilot frequency distribution, pilot frequencies can be distributed to UEs sorted forward preferentially (i.e. pilot frequency selection priority can be supported), and pilot frequency pollution can be reduced indirectly.

Alternatively, the service mode corresponding to the dense area may be a fully connected service mode, and specifically, each AP in the first AP set may maintain full connection with all UEs in the dense area, and provide services for UEs in the dense area in the fully connected mode.

Optionally, the service mode corresponding to the sparse area may be a service mode supporting AP selection, and specifically, for each AP in the second AP set, the connection may be maintained with all or part of UEs in the sparse area based on the service relationship, so as to provide services for UEs that remain connected.

It can be understood that the service area is divided into a dense area and a sparse area, and different service modes and pilot selection priorities can be adopted for the UEs in the dense area and the UEs in the sparse area to relieve pilot pollution, so that the UEs in the service area can be uniformly and well served.

Optionally, the determining, based on the second AP set serving the sparse region, one or more pilot allocation configurations corresponding to the sparse region includes:

Specifically, in the pilot allocation process, pilot allocation configurations may be determined for dense areas and sparse areas in the service area, and one or more pilot allocation configurations may be determined for UEs in the dense areas based on the first AP set serving the dense areas;

specifically, based on the large-scale fading coefficient between each AP of the second AP set and the user equipment in the sparse area, a service relationship between each AP of the second AP set and the user equipment in the sparse area can be determined, and then, in combination with channel estimation between each AP of the second AP set and the user equipment in the sparse area, a third combining matrix can be determined, and further, based on the target interference threshold and the third combining matrix, a fourth combining matrix for constructing the target graph can be obtained;

Specifically, after the structure of the target graph and the information amount corresponding to each edge in the target graph are determined, each vertex of the target graph can be traversed, coloring operation is performed on each vertex of the target graph in the traversing process, one or more coloring configurations can be determined, and one or more pilot frequency allocation configurations corresponding to the sparse region can be obtained;

It can be appreciated that the embodiment of the present invention provides a joint AP selection and pilot allocation (Graph Color based AP selection, GCAPS) algorithm based on graph coloring, specifically, for pilot allocation in a sparse area, UEs in the sparse area are more dispersed in geographic location, the distance between UEs is farther, meanwhile, the number of UEs in the sparse area is smaller and the pilot resources are smaller, and a UE target graph in the sparse area can be constructed based on a service relationship (a service relationship between each AP in the second AP set and user equipment in the sparse area) to perform a coloring operation, so that a part of UEs with poor remote channel quality are avoided from being connected to the APs.

Optionally, for the AP serving the sparse region UE (the AP in the second AP set), the CPU side of the CF mimo system may construct the following matrix based on the large scale fading coefficients between the sparse region UE and the AP

Wherein M is ^sparse For the number of APs serving sparse areas, |u ^sparse The i represents the number of UEs in the sparse region;

furthermore, for each AP of the sparse region, the UE of the sparse region can be accumulated from high to low to obtainThe accumulated value +.A corresponding to the ith AP can be calculated by the following formula>

Accumulated value at ith APNot less than the current AP total accumulated value +.>When delta%, the current AP finishes the selection of the UE to be served, and stops accumulation, namely the AP only serves the UE participating in accumulation in the downlink data transmission stage, wherein：/>

And then each AP is operated in this way until all APs complete UE selection, so that the service relationship between all APs and all UEs can be determined, and the following service matrix S can be used _a The representation is:

wherein element a _ij Element a, indicating whether the ith AP serves the jth UE _ij A value of 1 represents a service, element a _ij A value of 0 represents no service.

Optionally, based on the service relationship between each AP of the second set of APs and the user equipment in the sparse region, a degree of similarity α between the ith UE in the sparse region and the jth UE in the sparse region served by the APs in the second set of APs may be determined _i ′ _j Alpha 'can be determined specifically by the following formula' _ij ：

Wherein alpha is _i And alpha _j Through the service matrix S _a Determining;

based on channel estimation between each AP of the second set of APs and the user equipment in the sparse region, a degree of channel similarity gamma 'between an ith UE in the sparse region and a jth UE in the sparse region may be determined' _ij Gamma 'can be determined specifically by the following formula' _ij ：

Wherein, gamma _i For the variance of the channel estimation corresponding to the ith UE in the sparse region, the channel estimation corresponding to the ith UE can be determined by the channel estimation information reported by the AP; gamma ray _j For the variance of the channel estimation corresponding to the jth UE in the sparse region, the channel estimation corresponding to the jth UE can be determined through the channel estimation information reported by the AP;

further, based on the degree of similarity α 'of the services' _ij And the channel similarity degree gamma' _ij Can determine the joint service channel similarity value theta' _ij Specifically, θ 'can be determined by the following formula' _ij ：

θ′ _ij ＝α′ _ij *γ′ _ij ；

Further, based on the joint service channel similarity values between all UEs in the sparse region, the following third joint matrix θ' may be determined:

/>

wherein, the dimension of θ' is K.K, K is the number of UEs in the sparse region.

Alternatively, to eliminate interference caused by multiplexing pilots by partial UEs with similar channels and similar services, the average value of joint service channel similarity values among all UEs in the sparse region can be used as the threshold lambda _threshold Filtering the third combined matrix to obtain a fourth combined matrix theta, wherein lambda _threshold ＝sum(θ′)/(|U ^sparse |·|U ^sparse |-|U ^sparse I), the set of all UEs of the sparse region may be U ^sparse All the UE numbers of the sparse region can be determined by U ^sparse Mode length |U ^sparse The i represents, the fourth joint matrix θ may be represented by the following matrix:

wherein, element theta in fourth joint matrix theta _ij Can be expressed as:

It can be understood that the pilot frequency determining method in the non-cellular large-scale mimo system provided by the embodiment of the present invention may be a resource allocation (Hybrid Service Resource Allocation, HSRA) method of a hybrid service mode, which can ensure that UEs in different geographic locations in a service area can obtain good communication quality.

Therefore, one or more pilot allocation configurations corresponding to the dense area are determined based on the first AP set, one or more pilot allocation configurations corresponding to the sparse area can be determined by coloring each vertex of the target graph, and then one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area can be screened by determining the target pilot allocation configuration so as to maximize the total downlink rate of the system corresponding to the target pilot allocation configuration, so that pilot pollution of the dense area can be relieved by the screened pilot allocation configurations, and good communication quality can be obtained for the UE in different geographic positions in the service area.

Optionally, the first coloring operation includes:

Specifically, after determining the structure of the target graph and the information amount corresponding to each edge in the target graph, each vertex of the target graph may be traversed, and in the traversing process, coloring operation may be performed on each vertex of the target graph, where in the coloring operation for the first time, one vertex with the largest sum of interference values may be determined as a starting vertex from all vertices of the target graph based on a third combining matrix, and then coloring may be performed on the starting vertex;

specifically, after coloring the initial vertex, the remaining uncolored vertices in the target graph may be colored, and after coloring all vertices in the target graph, one or more coloring configurations may be determined, and further one or more pilot allocation configurations corresponding to the sparse region may be obtained;

Alternatively, the first color may be one of a list of colors, with "first" of the first colors not being used to describe a particular order or precedence.

It can be appreciated that, by determining the vertex with the largest sum of interference values as the initial vertex, in the pilot allocation process, pilot frequencies can be preferentially allocated to UEs corresponding to the vertex with the largest sum of interference values (i.e. pilot frequency selection priority can be supported), so that pilot pollution can be indirectly reduced.

Optionally, the nth coloring operation includes:

Specifically, after determining the structure of the target graph and the information amount corresponding to each side in the target graph, each vertex of the target graph may be traversed, in which coloring operation is performed on each vertex of the target graph in the traversing process, in the nth coloring operation, a third vertex may be determined from one or more second vertices adjacent to the first vertex based on the information amount of each side connected to the first vertex, and then a second color may be selected in the color list, and the third vertex may be colored, so that the color of the third vertex is different from the color corresponding to any vertex adjacent to the third vertex, and then the information amount corresponding to the side between the first vertex and the second vertex may be configured to be 0;

Specifically, after coloring all vertices in the target graph, one or more coloring configurations may be determined, and then one or more pilot allocation configurations corresponding to the sparse region may be obtained, and after obtaining one or more pilot allocation configurations corresponding to the dense region and one or more pilot allocation configurations corresponding to the sparse region, the optimization target of the maximum system downlink total rate corresponding to the target pilot allocation configuration may be based on screening is performed between the pilot allocation configuration corresponding to the dense region and the pilot allocation configuration corresponding to the sparse region, and then one pilot allocation configuration corresponding to the dense region and one pilot allocation configuration corresponding to the sparse region may be screened, and then the target pilot allocation configuration may be determined.

Alternatively, the second color may be one color in a color list, and "second" in the second color is not used to describe a specific order or sequence, and the second color may be the same color as the first color or may be a different color.

Alternatively, during the nth coloring operation, the probability P of transitioning from vertex i (first vertex) to vertex j (second vertex) may be determined as follows _ij The formula determines the third vertex:

Wherein, θ' _ij And θ' _is Can be determined by a third combining matrix theta ', theta' _is Representing similarity values between other vertices (one or more second vertices) connected to vertex i, select P _ij The vertex corresponding to the maximum value in (a) is used as the next traversal vertex (third vertex) of the vertex i.

Alternatively, during the nth coloring operation, colors whose neighbors have been used may be collected at each vertex (which may be referred to as a dye bucket): traversing all adjacent vertexes, if the adjacent vertexes have colors, putting the colors into the dyeing barrel, selecting a second color which is not in the dyeing barrel for the current node, and assigning the second color to the current vertex. After the current shading operation is completed, the bucket may be emptied and transferred to the next vertex. After the current completion of the dyeing of the vertex (third vertex), the information amount corresponding to the edge between the first vertex and the second vertex can be configured to be 0, so that the cyclic traversal of the immersion map traversal is prevented, and the UE representing the two vertices has completed the allocation, and the repeated allocation is not required.

It can be understood that, after dyeing of one vertex is completed, the target graph structure is updated once, until the information quantity between all vertices in the target graph is 0, which means that dyeing of all vertices is completed, and all dyeing configurations meeting that adjacent vertices are not dyed into the same color can be obtained based on the topological relation of the graph structure.

It can be appreciated that, by determining the third vertex from the one or more second vertices adjacent to the first vertex, the UE corresponding to the third vertex has the greatest interference to the UE corresponding to the first vertex, so that in the pilot allocation process, pilot can be preferentially allocated to the UE corresponding to the third vertex (i.e. pilot selection priority can be supported), and pilot pollution can be indirectly reduced.

Optionally, the determining, based on the one or more coloring configurations, one or more pilot allocation configurations corresponding to the sparse region includes:

Specifically, after the structure of the target graph and the information amount corresponding to each edge in the target graph are determined, each vertex of the target graph can be traversed, coloring operation is performed on each vertex of the target graph in the traversing process, and one or more coloring configurations can be determined;

specifically, after determining one or more coloring configurations, the one or more coloring configurations may be screened based on the color usage frequency threshold and the color usage frequency corresponding to each coloring configuration, to obtain one or more target coloring configurations, and further one or more pilot allocation configurations corresponding to the sparse region may be determined;

Alternatively, in the case where the number Nc of coloring configurations > 0, the threshold value [ |u of the number of color uses may be based on ^sparse |/τ _p ]Screening the plurality of first coloring configurations to obtain one or more second coloring configurations according to the color use times t corresponding to each first coloring configuration, so that the color use times corresponding to each second coloring configuration are smaller than or equal to a color use times threshold, wherein the set of all UE in the sparse area can be U ^sparse All the UE numbers of the sparse region can be determined by U ^sparse Mode length |U ^sparse I represents that the number of orthogonal pilot sequences is τ _p 。

Alternatively, in the case where the number nc=0 of coloring configurations, the update target interference threshold (λ) may be adjusted by the following formula _threshold ) Furthermore, based on the updated target interference threshold value, the joint filtering interference matrix (fourth joint matrix) can be obtained again, the target graph is constructed, and coloring operation is performed on each vertex of the target graph, so that coloring configuration can be determined:

where θ' represents the third combining matrix, the set of all UEs in the sparse region may beU ^sparse Tt represents how many times lambda is adjusted _threshold All the UE numbers of the sparse region can be determined by U ^sparse Mode length |U ^sparse I represents, each time lambda is adjusted _threshold The value of tt is then increased by 1.

Therefore, one or more pilot allocation configurations corresponding to the dense area are determined based on the first AP set, coloring operation is performed on each vertex of the target graph, the obtained coloring configurations can be screened to obtain one or more target coloring configurations, one or more pilot allocation configurations corresponding to the sparse area can be determined, the target pilot allocation configuration is determined to enable the system downlink total rate corresponding to the target pilot allocation configuration to be maximum, one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area can be screened, pilot pollution of the dense area can be relieved by the screened pilot allocation configurations, and good communication quality can be obtained for the UE in different geographic positions in the service area.

Optionally, before the determining, based on the first AP set of the service dense area, one or more pilot allocation configurations corresponding to the dense area, and the determining, based on the second AP set of the service sparse area, one or more pilot allocation configurations corresponding to the sparse area, the method further includes:

Specifically, based on the historical user equipment distribution data in the service area, a dense area and a sparse area can be determined, and then a first AP set can be determined based on a distance threshold and the distance between all APs in the service area and the central position of the dense area, and then the APs except the first AP set in all APs can be determined to be second AP sets, so that the first AP set and the second AP set have no intersection;

specifically, in the pilot allocation process, pilot allocation configurations can be determined for a dense area and a sparse area in a service area respectively, one or more pilot allocation configurations can be determined for UEs in the dense area based on a first AP set serving the dense area, and one or more pilot allocation configurations can be determined for UEs in the sparse area based on a second AP set serving the sparse area;

Alternatively, statistics may be performed on all UE service conditions (historical UE distribution data) in the service area, and a UE service profile of the entire service area may be drawn based on the statistics, and the dense area T may be determined by determining that the number of UEs exceeds a certain threshold ^dense And sparse region T ^sparse . And then N can be collected on the boundary of the dense area ^dense Coordinate position of individual points:the center coordinates C of the dense areas can be determined by calculation according to the following formula ^center ：

Alternatively, the process of determining the AP (first AP set) of the service dense area UE may include: traversing all APs in the service area, and calculating the distance between each AP and the dense area Heart C ^center Is the m-th AP from the center C of the dense area ^center May be of the distance ofWhen->Less than a given distance->(distance threshold), the current AP is selected as the service dense area UE, and it may be determined that all APs except the first AP set are the second AP set.

Alternatively, in determining the AP (first AP set) of the UE in the service dense area, the service matrix S may be used _a Recording the service relationship between the AP and the UE, wherein the service relationship can be determined by the following expression:

wherein the mth AP and UE (user equipment k in dense area) ^dense Sparse region user equipment k ^sparse ) When there is a service relationship between them (i.e. the mth AP provides service to the UE), then S _a A of (a) _mk Set to 1, otherwise 0.

Therefore, the first AP set and the second AP set can be determined based on the distance threshold and the distance between the AP of the service area and the central position of the dense area, one or more pilot allocation configurations corresponding to the dense area are determined based on the first AP set, one or more pilot allocation configurations corresponding to the sparse area are determined based on the second AP set, and then one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area can be screened by taking the maximum system downlink total rate corresponding to the target pilot allocation configuration as an optimization target, so that the screened pilot allocation configurations can alleviate pilot pollution of the dense area, and UEs in different geographic positions in the service area can be realized, and good communication quality can be obtained.

Optionally, fig. 5 is a second schematic flow chart of a pilot frequency determining method in the honeycomb-free large-scale mimo system, as shown in fig. 5, a CF mimo system may include M APs and K UEs, where all APs may be connected to a CPU through a backhaul link, each AP may be equipped with N antennas, and each UE may be a single antenna.

Alternatively, as shown in fig. 5, during the uplink training phase, the UE may send its own assigned pilot sequence to the AP, e.g., UE, via the uplink ₁ Can transmit pilot frequencyUE _k Can transmit pilot +.>UE _K Can transmit pilot +.>

Alternatively, as shown in fig. 5, after receiving the pilot sequence sent by the UE, the AP may perform calculation to obtain channel estimation information and pilot pollution information, and may further send the channel estimation information and the pilot pollution information to the CPU.

Optionally, as shown in fig. 5, the CPU may divide the dense area and the sparse area in the service area based on the historical user equipment distribution data in the service area, and may further determine a first AP set of the dense area and a second AP set of the sparse area, and may further determine one or more pilot allocation configurations corresponding to the dense area based on the first AP set of the dense area, and determine one or more pilot allocation configurations corresponding to the sparse area based on the second AP set of the sparse area, and may further determine a target pilot allocation configuration, and may further calculate the UE downlink reachable rate according to the channel estimation information.

Optionally, aAs shown in fig. 5, the CPU may configure the target pilot allocation (as in fig. 5) The power control coefficients and the AP selection scheme are transmitted to the AP, wherein the AP selection scheme may include information of the first AP set and a service relationship between the second AP set and the UEs in the sparse region.

Alternatively, as shown in fig. 5, during the downlink data transmission phase, the AP may perform power control and precoding on data to be transmitted to the UE through the power coefficient allocated by the CPU of the CF mimo system and the locally estimated channel, and may further perform power control and precoding based on the target pilot allocation configuration (as in fig. 5) Transmitting the data to the UE, so that the UE can receive the downlink signal transmitted by the AP, e.g. the UE ₁ Can receive signal S ₁ ，UE _k Can receive signal S _k ，UE _K Can receive signal S _K 。

It can be appreciated that, depending on whether the location of the UE belongs to a dense area, the UEs in the service area can be classified into UEs in the dense area and UEs in the sparse area, where the set of all UEs in the dense area can be U ^dense The set of all UEs in the sparse region may be U ^sparse All the UE numbers in the dense area may be determined by U ^dense Mode length |U ^dense The i indicates that all UE numbers of sparse regions may pass through U ^sparse Mode length |U ^sparse The I represents, and can be further represented by a service area UE total number formula I U ^dense |+|U ^sparse The total number K of service areas UE is calculated. The kth UE of the dense area may be denoted as k ^dense The set of APs (first set of APs) serving the UEs of the dense area may be denoted as M ^dense The kth UE of the sparse region may be denoted as k ^sparse The set of APs (second set of APs) of the UE serving the sparse region may be denoted as M ^sparse The AP set of the kth UE serving the sparse region may be expressed as

Alternatively, as shown in fig. 5, for the uplink training phase, at a coherence time interval τ _c In this, the duration for uplink pilot training may be τ _p (τ _p Positive integer) satisfies τ _c ＞τ _p And there is tau _p Orthogonal pilot sequences(/>Representing a real set), the pilot sequence transmitted by the kth UE in the service region (including dense and sparse regions) can be expressed as +.> Can represent the power allocated by each pilot frequency, each AP can receive the pilot frequency sequence transmitted by all UE, the mth AP receives the pilot frequency sequence signal transmitted by all UE +.>Specifically, the method can be obtained by the following "pilot sequence signal formula received by the AP":

wherein p is _k Indicating the transmit power allocated by the kth UE in the service area,(/>n in (a) represents the number of antennas equipped by the AP) represents the garment Additive noise from complex gaussian distribution, +.>Representing the channel between the mth AP and the kth UE in the service area.

Channel h between mth AP and kth UE in service area _mk The determination can be made by the following formula:

wherein g _mk Is the small scale fading coefficient g _mk Can be complex Gaussian random variable obeying independent same distributionβ _mk May be a large scale fading coefficient between the mth AP and the kth UE in the service area, which is related to path loss and shadowed fading channels.

Large-scale fading coefficient beta between mth AP and kth UE in service area _mk The determination can be made by the following "large scale fading coefficient formula" formula:

wherein,represents shadow fading, with standard deviation sigma _sh ，z _mk Represents the shading coefficient (shadowing coefficients), and +.>PL _mk Representing the path loss.

Path loss PL _mk The three-slope model formula can be obtained by the following formula:

wherein d _mk Represents the distance between the mth AP and the kth UE in the service area, d ₀ And d ₁ Is a distance parameter of the triclinic model.

L in the above triclinic model formula may be determined by the following formula:

wherein f represents carrier frequency in MHZ; h is a _AP The antenna height of the AP is expressed in meters (m); h is a _u The antenna height in meters (m) is the UE.

In the pilot sequence signal formula received by the APIs the pilot sequence transmitted by the kth UE in the service area,/or->Constraints that satisfy the following expression are required: />

Wherein,represents a set of UEs multiplexing pilots with k UEs, k' being the set of UEs +.>Is a component of the group.

And can then pass throughMultiplied by y _m Acquisition of y _mk To estimate the mth AP and the kth U in the service areaE channels, particularly calculating y _mk The formula of (2) is as follows:

wherein,the above calculation y _mk Second term in the formula (2)>The pollution caused by multiplexing pilot frequency by different UE when pilot frequency resources are limited, and the k-th UE pilot frequency pollution in the service area can be obtained by calculating the following UE pilot frequency pollution formula:

channel estimation can be performed by a minimum mean square error estimator (minimum mean squareerror estimation, MMSE) at the AP end, and specifically, the channel estimation can be performed by the following stepsAcquiring a channel estimate between the mth AP and the kth UE in the service area:

wherein,is the additive Gaussian noise variance of the downlink channel, and can be obtained by the following variance formula of channel estimation>Is the variance of:

thus, for the uplink training phase, y is calculated by the pilot sequence signal formula received by the AP _mk Formula of (1), UE pilot pollution formula, channel estimationAnd the variance formula of the channel estimation, and can acquire the channel estimation information and the pilot pollution information by calculation.

Optionally, as shown in fig. 5, in the downlink data transmission stage, after precoding and power control, for the transmission signal from the mth AP to the kth UE in the service area, the AP service vector is multiplied due to the selective service of the AP (i.e. the AP provides service for all or part of the UEs in the area served by the AP).

For APs serving dense area UEs (APs in the first set of APs), the transmission signal of the mth AP is expressed as The acquisition can be calculated by the following "dense area AP transmit signal formula": />

Wherein,is the total signal power of the AP at the transmitting end; η (eta) _mk Is the power control coefficient between the mth AP and the kth UE; p is p _mk Representing the transmission power of the mth AP to the kth UE; />Whether an mth AP representing a service dense area has a service relationship with a kth UE of the dense area: when the value is 1, the two are in service relationship, and when the value is 0, the two are not in service relationship. U (U) ^dense Is the set of all UEs in the dense area. />Transmitting signals to kth UE in the dense area for the mth AP;

Is a precoding matrix between the mth AP and the kth UE according to the uplink channel estimation value +.>By utilizing the diversity of TDD channels, in the case of adopting a maximum ratio transmission (Maximum Ratio Transmission, MRT) precoding mode in downlink, the method comprises the steps of +.>The acquisition can be calculated by the following formula:

for the APs serving the sparse region UE, the UEs served by different AP ends may be different, and the service vectors of the corresponding APs may be different, so that the signal sent by the mth AP in the AP set (second AP set) serving the sparse regionThe acquisition can be calculated by the following "sparse area AP transmit signal formula":

wherein,is the total signal power of the AP at the transmitting end; η (eta) _mk Is the power control coefficient between the mth AP and the kth UE; p is p _mk Representing the transmission power of the mth AP to the kth UE; />Transmitting a signal for the mth AP to the kth UE,>the mth AP representing the service sparse region has a service relationship with the kth UE of the sparse region: when the value is 1, the service relationship exists between the two, and when the value is 0, the service relationship does not exist between the two. U (U) ^sparse Is the set of all UEs in the sparse region, M ^sparse Is the set of APs serving sparse area UEs.

Is a precoding matrix between the mth AP and the kth UE according to the uplink channel estimation value +. >By utilizing the diversity of TDD channels, under the condition that MRT precoding mode is adopted in the downlink, the method comprises the steps of ++>The acquisition can be calculated by the following formula:

for a downlink signal receiving end in a dense area, a signal received by a kth UEThe signal receiving formula of the UE in the dense area can be obtained by the following formula "And (3) calculating and obtaining: />

Furthermore, it can pass throughRepresenting the signal received by the kth UE +.>The corresponding desired signal can be transmitted byRepresenting the signal received by the kth UE +.>The uncertainty of the corresponding precoding gain can be determined by +.>Representing the signal received by the kth UE +.>Corresponding multi-UE interference, ">And->The acquisition can be calculated by the following formula:

and the unit bandwidth downlink reachable rate of the kth UE in the dense area can be obtained by the following 'dense area UE downlink reachable rate formula':

for a downlink signal receiving end of a sparse area, a signal received by a kth UEThe acquisition can be calculated by the following "sparse area UE received signal formula":

furthermore, it can pass throughRepresenting the signal received by the kth UE +.>The corresponding desired signal can be transmitted byRepresenting the signal received by the kth UE +.>The uncertainty of the corresponding precoding gain can be determined by +.>Representing the signal received by the kth UE +. >Corresponding multi-UE interference, ">And->The acquisition can be calculated by the following formula:

furthermore, the unit bandwidth downlink reachable rate of the kth UE in the sparse region can be obtained through the following 'sparse region UE downlink reachable rate formula':

therefore, for the downlink data transmission stage, the unit bandwidth downlink reachable rate of any one UE in the dense area can be obtained by calculating through the dense area AP sending signal formula, the dense area UE receiving signal formula and the dense area UE downlink reachable rate formula; through the sparse area AP transmission signal formula, the sparse area UE receiving signal formula and the sparse area UE downlink reachable rate formula, the unit bandwidth downlink reachable rate of any one UE in the sparse area can be obtained through calculation.

It can be understood that, in the case of uneven UE distribution in the service area, compared with the CF mimo system in the related art, the method for determining the pilot frequency in the non-cellular massive mimo system provided by the embodiment of the present invention can achieve a higher downlink reachable rate of the UE.

Specifically, in the CF mimo system in the related art, during the uplink pilot sequence transmission stage, all APs will receive pilot sequences sent by all UEs, when the APs perform channel estimation on UEs that are subjected to pilot multiplexing, the accuracy of the channel estimation result will be affected, and if the UEs that are subjected to pilot multiplexing are still served by the same AP, during the downlink data reception stage, it can be known from the above "dense area UE downlink reachable rate formula" and "sparse area UE downlink reachable rate formula": the target signal on the molecule in the formula can cause inaccurate target signal receiving of a receiving end (UE) after the signal subjected to MRT precoding treatment passes through a real channel due to inaccurate channel estimation; the interference on the denominator also causes that the signal after MRT precoding cannot counteract the influence of the real channel due to inaccurate channel estimation, so that the interference is increased. I.e. the actual target signal fraction decreases and the other UE interference fraction increases, the achievable rate of the UE decreases.

The pilot frequency determining method in the honeycomb-free large-scale multiple-input multiple-output system provided by the embodiment of the invention can realize the isolation of part of UE in the pilot frequency multiplexing process through AP selection, and can be known according to the 'dense area UE downlink reachable rate formula' and the 'sparse area UE downlink reachable rate formula': the interference part corresponding to the denominator in the formula can be reduced, and the target signal corresponding to the numerator in the formula is close to the real signal of the transmitting end, so that the higher downlink reachable rate of the UE can be realized.

Optionally, fig. 6 is a third flow chart of a method for determining a pilot in a non-cellular massive multiple-input multiple-output system according to the present invention, as shown in fig. 6, the method for determining a pilot in a non-cellular massive multiple-input multiple-output system may include steps 601 to 606:

step 601, random pilot frequency allocation;

specifically, in the case of pilot allocation for the first time, the CPU of the CF mimo system may allocate pilots for UEs in the service area based on a random pilot allocation manner.

Step 602, dividing a user equipment set in a sparse area and a user equipment set in a dense area;

specifically, the CPU of the CF mimo system may determine a dense area and a sparse area based on the historical UE distribution data in the service area, and further may divide all UEs of the dense area into a set of user equipments of the dense area and all user equipments of the sparse area into a set of user equipments of the sparse area based on the geographic location information of the UEs in the current service area in a period of time in which pilots are allocated to the UEs.

Step 603, determining an AP set of a dense service area and determining an AP set of a sparse service area;

specifically, based on the distance threshold and the distances between all APs of the service area and the center position of the dense area, an AP set (first AP set) of the service dense area may be determined; and further, it is possible to determine that the AP other than the first AP set among all APs is an AP set (second AP set) serving the sparse region.

Step 604, determining one or more pilot allocation configurations corresponding to the dense area;

specifically, the number of the plurality of orthogonal pilot sequences (τ _p ) Greater than or equal to the number of UEs in the dense area (|u) ^dense I), orthogonal pilot sequences can be allocated to UEs in the dense area, and UEs in the dense area can be configured to be served by the same AP set in a fully connected manner, and one pilot allocation configuration corresponding to the dense area can be obtained. In this case, interference of UEs in a dense area upon data reception is mainly due to the influence of UEs in a sparse area, while there is no interference at all between UEs in a dense area.

Specifically, the number of the plurality of orthogonal pilot sequences (τ _p ) Less than the number of UEs in dense areas (|u) ^dense I), pilot allocation may be performed using LHPA algorithms, in particular, a large-scale fading coefficient between each AP in the first set of APs and the user equipment in the dense area may be determined; and further, based on the large-scale fading coefficient between each AP in the first AP set and the user equipment in the dense area and the distance between each AP in the first AP set and the user equipment in the dense area, one or more pilot frequency allocation configurations corresponding to the dense area can be obtained.

Step 605, determining one or more pilot allocation configurations corresponding to the sparse region;

specifically, based on the large-scale fading coefficients between each AP of the second set of APs and the user equipment in the sparse region, a service relationship between each AP of the second set of APs and the user equipment in the sparse region may be determined, which may be represented by a service matrix;

further, a joint interference matrix (third joint matrix) may be determined based on a service relationship between each AP of the second set of APs and the user equipment in the sparse region, and channel estimation between each AP of the second set of APs and the user equipment in the sparse region;

and is further based on a target interference threshold (lambda) _threshold ) The method comprises the steps of obtaining a third combined matrix, namely obtaining a combined filtering interference matrix (fourth combined matrix), wherein the row and column number of the fourth combined matrix is the same as that of the third combined matrix, the value of a second element is 1 when the first element is larger than or equal to a target interference threshold value, the value of the second element is 0 when the first element is smaller than the target interference threshold value, the first element is any element in the third combined matrix, and the second element is an element with the same row and column number as that of the first element in the fourth combined matrix;

Further, based on the fourth joint matrix, a user association graph (target graph) can be determined, and based on the third joint matrix, the information amount corresponding to each side among all vertexes of the target graph is determined, wherein each vertex of the target graph has a unique corresponding relation with each user equipment in the sparse area, and the number of vertexes of the target graph is the same as the number of user equipment in the sparse area;

further, coloring each vertex of the target graph based on the number of the plurality of orthogonal pilot sequences, one or more coloring configurations may be determined (the number of the one or more coloring configurations may be represented as Nc);

further, one or more pilot allocation configurations corresponding to the sparse region may be determined based on the one or more coloring configurations.

Alternatively, in the case where Nc > 0, the threshold value [ |u ] may be based on the number of color uses ^sparse |/τ _p ]Screening the plurality of first coloring configurations to obtain one or more second coloring configurations according to the color use times t corresponding to each first coloring configuration, so that the color use times corresponding to each second coloring configuration are smaller than or equal to a color use times threshold, wherein the set of all UE in the sparse area can be U ^sparse All the UE numbers of the sparse region can be determined by U ^sparse Mode length |U ^sparse I represents that the number of orthogonal pilot sequences is τ _p 。

Alternatively, in the case nc=0The update target interference threshold (λ) may be adjusted by the following formula _threshold ) Furthermore, based on the updated target interference threshold value, the joint filtering interference matrix (fourth joint matrix) can be obtained again, the target graph is constructed, and coloring operation is performed on each vertex of the target graph, so that coloring configuration can be determined:

wherein θ' represents a third combining matrix, and the set of all UEs in the sparse region may be U ^sparse Tt represents how many times lambda is adjusted _threshold All the UE numbers of the sparse region can be determined by U ^sparse Mode length |U ^sparse I represents, each time lambda is adjusted _threshold The value of tt is then increased by 1.

Step 606, a target pilot allocation configuration is determined.

Specifically, the CPU of the CF mimo system may determine a target pilot allocation configuration among one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area, so as to maximize a system downlink total rate corresponding to the target pilot allocation configuration.

Optionally, fig. 7 is one of the experimental simulation diagrams provided by the present invention, fig. 8 is a second experimental simulation diagram provided by the present invention, and fig. 9 is a third experimental simulation diagram provided by the present invention, as shown in fig. 7-9, by MATLAB simulation software, it is verified that the UE velocity is severely affected due to the uneven distribution of the UE. The experimental link simulates two scenes of square distribution and uneven distribution, and simulates the downlink rate of the system under the condition that the traditional multiple pilot frequency distribution scheme only changes the position distribution of UE, wherein FIG. 7 shows the total rate of downlink users in the system, FIG. 8 shows the minimum downlink user rate, FIG. 9 shows the mean square error of the downlink user rate, and the mean square error can show the equality among different user reachable downlink rates. Simulation results show that: the performance of these several pilot allocation algorithms, whether it be the overall rate or the equalization of the user's communication quality, is consistent.

As shown in fig. 7 to 9, the graph indicated by the BalanceRPA mark represents a simulation curve corresponding to a random pilot allocation (Random Pilot Assignment, RPA) algorithm in the related art in the case of uniform UE distribution, the graph indicated by the BalanceLBGPA mark represents a simulation curve corresponding to a greedy pilot allocation (Location Based Greedy Pilot Assignment, LBGPA) algorithm in the related art based on geographic location in the case of uniform UE distribution, the graph indicated by the BalanceMI mark represents a simulation curve corresponding to a maximum increment (Maximal Increment, MI) algorithm in the related art in the case of uniform UE distribution, the graph indicated by the ubamaancerpa mark represents a simulation curve corresponding to an RPA algorithm in the case of non-uniform UE distribution, the graph indicated by the ubamacelbgpa mark in the graph represents a simulation curve corresponding to an LBGPA algorithm in the case of non-uniform UE distribution, and the graph indicated by the ubalaancemi mark in the graph in the case of non-uniform UE distribution.

The horizontal axis of the graph in fig. 7 represents the total rate of the downlink user in units of (Mbits/s/Hz), and the vertical axis of the graph in fig. 7 represents the cumulative distribution rate. The horizontal axis of the graph in fig. 8 represents the downlink minimum user rate in units of (Mbits/s/Hz), and the vertical axis of the graph in fig. 8 represents the cumulative distribution rate. The horizontal axis of the graph in fig. 9 represents the downlink user rate mean square error, and the vertical axis of the graph in fig. 9 represents the cumulative distribution rate.

Optionally, fig. 10 is a fourth experimental simulation diagram provided by the present application, fig. 11 is a fifth experimental simulation diagram provided by the present application, and fig. 12 is a sixth experimental simulation diagram provided by the present application, as shown in fig. 10-12, and by MATLAB simulation software, the improvement of the rate of the system in the UE uneven distribution scenario is verified. Fig. 10 shows the total rate of the downstream users in the system, fig. 11 shows the downstream minimum user rate, and fig. 12 shows the mean square error of the downstream user rates, which can represent the equality between the achievable downstream rates of different users. Simulation results show that: the embodiment of the application has the performance advantage under the scene of uneven user distribution, and the total rate of the system, the lower limit of the minimum rate and the user service balance are excellent.

As shown in fig. 10 to fig. 12, the curve indicated by the LBGPA mark in the drawing represents a simulation curve corresponding to the LBGPA algorithm under the condition of uneven UE distribution, the curve indicated by the MI mark in the drawing represents a simulation curve corresponding to the MI algorithm under the condition of uneven UE distribution, the curve indicated by the HSRA mark in the drawing represents a simulation curve corresponding to the HSRA algorithm provided by the embodiment of the present application under the condition of uneven UE distribution.

The horizontal axis of the graph in fig. 10 represents the total rate of the downlink user in units of (Mbits/s/Hz), and the vertical axis of the graph in fig. 10 represents the cumulative distribution rate. The horizontal axis of the graph in fig. 11 represents the downlink minimum user rate in units of (Mbits/s/Hz), and the vertical axis of the graph in fig. 11 represents the cumulative distribution rate. The horizontal axis of the graph in fig. 12 represents the downlink user rate mean square error, and the vertical axis of the graph in fig. 12 represents the cumulative distribution rate.

The pilot frequency determining device in the non-cellular large-scale multiple-input multiple-output system provided by the invention is described below, and the pilot frequency determining device in the non-cellular large-scale multiple-input multiple-output system described below and the pilot frequency determining method in the non-cellular large-scale multiple-input multiple-output system described above can be correspondingly referred to each other.

Fig. 13 is a schematic structural diagram of a pilot determining device in a honeycomb-free large-scale mimo system according to the present invention, and as shown in fig. 13, the device includes: a first determination module 1301 and a second determination module 1302, wherein:

a first determining module 1301, configured to determine, based on a first AP set serving a dense area, one or more pilot allocation configurations corresponding to the dense area, and determine, based on a second AP set serving a sparse area, one or more pilot allocation configurations corresponding to the sparse area;

a second determining module 1302, configured to determine a target pilot allocation configuration from one or more pilot allocation configurations corresponding to the dense area and one or more pilot allocation configurations corresponding to the sparse area, so as to maximize a system downlink total rate corresponding to the target pilot allocation configuration; wherein the dense region and the sparse region are determined based on historical user equipment distribution data in a service region, the service region comprising the dense region and the sparse region, the first set of APs having no intersection with the second set of APs.

The device for determining the pilot frequency in the honeycomb-free large-scale multi-input multi-output system provided by the invention can determine one or more pilot frequency distribution configurations corresponding to the dense region based on the first AP set by dividing the service region into the dense region and the sparse region, determine one or more pilot frequency distribution configurations corresponding to the sparse region based on the second AP set, and further determine the target pilot frequency distribution configuration so as to maximize the total downlink rate of the system corresponding to the target pilot frequency distribution configuration, so that the one or more pilot frequency distribution configurations corresponding to the dense region and the one or more pilot frequency distribution configurations corresponding to the sparse region can be screened, the screened pilot frequency distribution configurations can relieve the pilot frequency pollution of the dense region, and the UE (user equipment) in different geographic positions in the service region can be realized, and good communication quality can be obtained.

Optionally, the first determining module is specifically configured to:

Determining one or more pilot allocation configurations corresponding to the sparse region based on the one or more target coloring configurations; wherein the color usage number threshold is determined based on the number of user devices in the sparse region and the number of the plurality of orthogonal pilot sequences.

Optionally, the apparatus further includes a third determining module, before the first set of APs based on the dense area of service determines one or more pilot allocation configurations corresponding to the dense area, and the second set of APs based on the sparse area of service determines one or more pilot allocation configurations corresponding to the sparse area, the third determining module is configured to:

Fig. 14 is a schematic structural diagram of an electronic device according to the present invention, and as shown in fig. 14, the electronic device may include: processor 1410, communication interface (Communications Interface) 1420, memory 1430 and communication bus 1440, wherein processor 1410, communication interface 1420 and memory 1430 communicate with each other via communication bus 1440. Processor 1410 can invoke logic instructions in memory 1430 to perform a method for pilot determination in a non-cellular large-scale multiple-input multiple-output system, the method comprising:

determining a target pilot frequency allocation configuration in one or more pilot frequency allocation configurations corresponding to the dense area and one or more pilot frequency allocation configurations corresponding to the sparse area so as to maximize the system downlink total rate corresponding to the target pilot frequency allocation configuration; wherein the dense region and the sparse region are determined based on historical user equipment distribution data in a service region, the service region comprising the dense region and the sparse region, the first set of APs having no intersection with the second set of APs.

In addition, the logic instructions in the memory 1430 described above may be implemented in the form of software functional units and may be stored in a computer readable storage medium when sold or used as a stand alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In another aspect, the present invention also provides a computer program product, the computer program product comprising a computer program, the computer program being storable on a non-transitory computer readable storage medium, the computer program, when executed by a processor, being capable of performing the method for determining pilots in a honeycomb-free large-scale multiple-input multiple-output system provided by the above methods, the method comprising:

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which when executed by a processor is implemented to perform the method for pilot determination in a non-cellular massive multiple-input multiple-output system provided by the methods above, the method comprising:

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for pilot determination in a honeycomb-free large-scale multiple-input multiple-output system, comprising:

2. The method for pilot determination in a honeycomb-less massive multiple-input multiple-output system according to claim 1, wherein the determining one or more pilot allocation configurations corresponding to the dense area based on the first AP set serving the dense area comprises:

3. The method for pilot determination in a honeycomb-free massive multiple-input multiple-output system according to claim 2, wherein the obtaining one or more pilot allocation configurations corresponding to the dense area based on a large-scale fading coefficient between each AP in the first set of APs and the user equipment in the dense area and a distance between each AP in the first set of APs and the user equipment in the dense area comprises:

4. A method for pilot determination in a non-cellular massive multiple-input multiple-output system according to any of claims 1-3, wherein the determining, based on the second set of APs serving the sparse region, one or more pilot allocation configurations corresponding to the sparse region comprises:

5. The method for pilot determination in a honeycomb-less massive multiple-input multiple-output system of claim 4, wherein the first coloring operation comprises:

6. The method for pilot determination in a honeycomb-less massive multiple-input multiple-output system of claim 5, wherein the nth coloring operation comprises:

7. The method for pilot determination in a honeycomb-less massive multiple-input multiple-output system according to claim 4, wherein the determining one or more pilot allocation configurations corresponding to the sparse region based on the one or more coloring configurations comprises:

8. A method of pilot determination in a honeycomb-free large-scale multiple-input multiple-output system according to any one of claims 1-3 or 5-7, characterized in that before the determining of one or more pilot allocation configurations corresponding to dense areas based on the first set of APs serving dense areas and the determining of one or more pilot allocation configurations corresponding to sparse areas based on the second set of APs serving sparse areas, the method further comprises:

9. A pilot determination apparatus in a honeycomb-free large-scale multiple-input multiple-output system, comprising:

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the method of pilot determination in a non-cellular massive multiple-input multiple-output system as claimed in any one of claims 1 to 8 when the program is executed by the processor.

11. A non-transitory computer readable storage medium having stored thereon a computer program, which when executed by a processor implements a method of pilot determination in a non-cellular massive multiple-input multiple-output system according to any of claims 1 to 8.