WO2019062822A1

WO2019062822A1 - Resource allocation method and server

Info

Publication number: WO2019062822A1
Application number: PCT/CN2018/108061
Authority: WO
Inventors: 彭焦阳; 叶兵; 高莎莎; 孙学真
Original assignee: 中兴通讯股份有限公司
Priority date: 2017-09-27
Filing date: 2018-09-27
Publication date: 2019-04-04
Also published as: CN109561435B; CN109561435A

Abstract

Disclosed is a resource allocation method. The method comprises: acquiring a directed graph of a small cell base station (SBS) network; grouping a plurality of SBSs in the SBS network according to a set grouping strategy and the directed graph to obtain a plurality of SBS groups; and allocating a sub-channel to each SBS group. Also disclosed is a resource allocation server.

Description

Resource allocation method and server

The present application claims the priority of the Chinese Patent Application, filed on Sep. 27, 2017, the priority of

Technical field

The present disclosure relates to the field of wireless communication technologies, for example, to a resource allocation method and a server.

Background technique

Ultra Dense Heterogeneous Network (UDN) is one of the hotspots in recent years. It effectively solves the problem of signal coverage under dense user distribution, but it also causes more network interference. Since UDNs share the same spectrum resources as macro cells, they may cause severe co-channel interference when they use the same channel. At the same time, since the small cell base station (SBS) of the ultra-dense network is usually installed by the user according to his own needs, it is difficult for the communication network operator to obtain an accurate base station location, so it is difficult to carry out effective network planning to solve the complicated Interference problem.

Co-channel interference has a great relationship with the access mode of SBS. At present, there are three ways of SBS accessing the communication core network: closed access mode, open access mode and hybrid access mode. The closed access mode allows only authorized small cell users (SUEs) to access, and unauthorized users cannot access them. In the open access mode, if the SBS has sufficient resources, the operator allows all users to access the SBS arbitrarily. However, in this access mode, the unlicensed user equipment can use all spectrum resources. Greatly reduce the performance of authorized SUEs within the SBS. In the hybrid access mode, the SBS allows unauthorized users to access, but only uses resources that are not used by authorized users, which not only ensures the communication quality of authorized users, but also improves spectrum utilization, thereby improving system capacity.

In the hyper-dense network hybrid access mode, there is a macro base station (MBS). In the coverage area of the MBS, multiple small cells (Small cells) are in the running state at the same time. In order to make the MBS have no dead angle coverage for its coverage area, the configuration mode of the cellular network is adopted, that is, each MBS has three transmitting antennas, the angle between the transmitting antennas is 120 degrees, and the coverage area of each antenna is a sector of 120 degrees. . The MBS has a large transmission power and a large coverage area. The MBS service object is a large number of macro cell users (MPUs) existing in the macro cell. The transmit power of the SBS is extremely small compared to the MBS, the SBS uses an omnidirectional antenna, and the SBS serves the 2-8 authorized SUEs in their respective coverage areas. In the case where the number of subchannels is constant, if more subchannels are allocated to the SBS, many SBSs may share the same subchannel, resulting in strong interference within the network. If a small number of subchannels are allocated to each SBS, the spectrum utilization is reduced, and although the interference is avoided, the throughput of the system is lost.

In the related art, the resource allocation strategy in the hybrid access mode is generally combined with the radio technology allocation method, but the allocation method combined with the radio technology has high computational complexity, and the allocation strategy with low computational complexity causes many SBSs to share the same subchannel. As a result, interference within the network is strong.

Summary of the invention

The present disclosure provides a resource allocation method and a server, which are used to solve the problem that the computational complexity of the related technology resource allocation strategy is high, and the allocation strategy with low computational complexity is likely to cause strong co-channel interference between different SBSs.

In an embodiment, the present disclosure provides a resource allocation method, including:

Obtaining a directed graph of a small cell base station SBS network;

According to the directed graph, grouping a plurality of SBSs in the SBS network according to a set grouping policy to obtain a plurality of SBS groups;

A subchannel is assigned to each of the SBS groups.

In an embodiment, grouping a plurality of SBSs in the SBS network according to a set grouping policy includes:

Determining an initial grouping matrix according to the directed graph;

Interchanging elements in the initial grouping matrix according to a simulated annealing algorithm;

When the exit condition of the simulated annealing algorithm is satisfied, the switched grouping matrix is output.

In an embodiment, a subchannel is allocated for each of the SBS groups, including:

Determining the number of subchannels required for each of the SBS groups;

Calculating the throughput of the subchannel to be allocated in each of the SBS groups;

If the throughput of the subchannel allocated to the first SBS group is greater than the throughput when the subchannel is allocated to the second SBS group and the number of subchannels that the first SBS group has been allocated is smaller than the first SBS The number of subchannels required by the group is then assigned to the first SBS group.

In an embodiment, the second SBS group may be a group other than the first SBS group.

In an embodiment, determining the number of subchannels required for each of the SBS groups includes:

Measuring user rate requirements for each of the SBS teams;

The number of subchannels required for each of the SBS groups is determined based on the user rate requirements.

In an embodiment, calculating the throughput of the subchannels to be allocated in each of the SBS groups includes:

For each SBS group, calculating a signal to interference plus noise ratio of each SBS in the SBS group for the subchannel to be allocated;

Using the minimum value of all signal to interference plus noise ratios in the SBS group as the signal to interference plus noise ratio of the SBS group;

Calculating the throughput of the subchannel to be allocated in the SBS group according to the signal to interference plus noise ratio of the SBS group.

In an embodiment, the throughput of the SBS team is calculated as:

Where B is the channel bandwidth, j is the SBS in the SBS group, k is the subchannel assigned to j, n is the user connected to j, and F is the total number of SBSs.

For the signal-to-interference plus noise ratio of the SBS group, λ _k,l is a pointer variable, the value of the pointer variable is 0 or 1, and when λ _k,l =1, indicating that the subchannel k is allocated to the The first SBS group, when λ _{k, l} =0, indicates that subchannel k is not allocated to the first SBS group.

In an embodiment, after allocating the subchannels for each of the SBS groups, the method further comprises: allocating the subchannels to which the SBS group is assigned to a plurality of users of the SBS service in the SBS group.

In an embodiment, the subchannels to which each of the SBS groups are assigned are allocated to a plurality of users of the SBS service in the SBS group, including:

Checking whether there are free subchannels in the SBS group;

When there are a plurality of the idle subchannels, the subchannels having the largest throughput among the plurality of idle subchannels and the plurality of users are allocated to users corresponding to the subchannels having the highest throughput.

Subchannels of the SBS in the SBS group are divided into an authorized subchannel and an unlicensed subchannel according to a preset ratio;

Determining whether the user is an authorized user;

If the user is an authorized user, check whether the authorized subchannel has a first idle subchannel;

And if a plurality of the first idle subchannels exist in the authorized subchannel, a subchannel that has the largest throughput among the plurality of first idle subchannels and the authorized user is allocated to the authorized user;

If the preset user is an unauthorized user, check whether the unlicensed subchannel has a second idle subchannel;

And assigning, in the non-authorized subchannel, a plurality of the second idle subchannels, the subchannels that have the largest throughput among the plurality of second idle subchannels and the unlicensed users to the non Authorized user.

In an embodiment, the present disclosure further provides a resource allocation server, comprising at least a memory, a processor, wherein the memory stores a computer program, and the steps implemented by the processor when executing the computer program on the memory include: :

Obtaining a directed graph of a small cell base station SBS network;

A subchannel is assigned to each of the SBS groups.

In an embodiment, when the processor groups the multiple SBSs in the SBS network, the step of implementing includes: determining an initial grouping matrix according to the directed graph; The elements in the initial divided group matrix are swapped; when the exit condition of the simulated annealing algorithm is satisfied, the switched grouping matrix is output.

In an embodiment, when the processor allocates a subchannel for each of the SBS groups, the steps implemented include: determining a number of subchannels required by each of the SBS groups; and calculating subchannels to be allocated in each The throughput of the SBS group; if the throughput when the subchannel is allocated to the first SBS group is greater than the throughput when the subchannel is allocated to the second SBS group and the first SBS group has been allocated The number of channels is less than the number of subchannels required by the first SBS group, and the subchannels are allocated to the first SBS group.

In an embodiment, when the processor determines the number of subchannels required by each of the SBS groups, the steps of implementing include: measuring a user rate requirement of each of the SBS groups; determining, according to the user rate requirements, The number of subchannels required for each of the SBS groups.

In an embodiment, when the processor calculates the throughput of the subchannel to be allocated in each of the SBS groups, the step of implementing includes: calculating, for each SBS group, the subchannel to be allocated in the SBS group The signal-to-interference plus noise ratio of each SBS; the minimum value of all signal-to-interference plus noise ratios in the SBS group is used as the signal-to-interference plus noise ratio of the SBS group; according to the signal and interference of the SBS group The noise ratio is calculated, and the throughput of the subchannel to be allocated in the SBS group is calculated.

In an embodiment, the processor calculates a throughput formula for the SBS group to use:

For the signal-to-interference plus noise ratio of the SBS group, λ _k,l is a pointer variable, the value of the pointer variable is 0 or 1, and when λ _k,l =1, indicating that the subchannel k is allocated to the The first SBS group, when λ _{k, l} =0, indicates that subchannel k is not allocated to the second SBS group.

In an embodiment, after the processor allocates a subchannel for each of the SBS groups, the step further includes: allocating a subchannel to which each SBS group is assigned to an SBS in the SBS group. Serving multiple users.

In an embodiment, when the processor allocates a subchannel to which each SBS group is assigned to multiple users of the SBS service in the SBS group, the implemented steps include: viewing the SBS group Whether there is an idle subchannel; in the case where there are a plurality of the free subchannels, a child that maximizes the user rate in a case where the plurality of idle subchannels meet the minimum rate requirement of each user among the plurality of users The channel is assigned to a user corresponding to the subchannel.

In an embodiment, when the processor allocates a subchannel to which each SBS group is assigned to multiple users of the SBS service in the SBS group, the achievable steps include: The subchannels of the SBS in the SBS group are divided into an authorized subchannel and an unlicensed subchannel according to a preset ratio; whether the user is an authorized user; and if the user is an authorized user, whether the authorized subchannel is viewed There is a first idle subchannel; if there are a plurality of the first idle subchannels in the authorized subchannel, the minimum of each authorized user is satisfied between the plurality of the first idle subchannels and the authorized user The subchannel that maximizes the user rate in the case of rate demand is allocated to the authorized user; if the preset user is an unauthorized user, whether the unlicensed subchannel exists to have a second idle subchannel; If there are a plurality of the second idle subchannels in the unlicensed subchannel, satisfying the minimum rate requirement of each unauthorized user among the plurality of the second idle subchannels and the unauthorized user Maximizing the user rate of the subchannel allocated to the non-authorized user.

DRAWINGS

1 is a flowchart of a resource allocation method in an embodiment of the present disclosure;

2 is a graph showing an average throughput trend of a macro cell user in a case where an SBS arrangement density is increased under different algorithms according to an embodiment of the present disclosure;

3 is an average signal to interference plus noise ratio of a SUE in different SBS deployment densities according to an embodiment of the present disclosure;

4 is a user fairness under different algorithms in an embodiment of the present disclosure;

FIG. 5 is a satisfaction degree in a case where a user uses a different algorithm in an embodiment of the present disclosure; FIG.

FIG. 6 is a structural block diagram of a resource allocation server according to an embodiment of the present disclosure.

Detailed ways

In order to solve the problem that the resource allocation strategy in the hybrid access mode has high computational complexity and the computational complexity is low, the co-channel interference between different SBSs is strong, and the present disclosure provides a resource allocation method. The present invention will be described below in conjunction with the drawings and embodiments. The embodiments described herein are merely illustrative of the disclosure and are not limiting of the disclosure.

An embodiment of the present disclosure provides a resource allocation method. FIG. 1 is a flowchart of a resource allocation method according to an embodiment of the present disclosure.

Step 110: Obtain a directed graph of the SBS network.

Step 120: According to the directed graph, grouping a plurality of SBSs in the SBS network according to the set grouping policy to obtain a plurality of SBS groups.

Step 130, assigning a subchannel to each SBS group.

In this embodiment, the small cell Small cell is a small cell in which an SBS is deployed, that is, one Small cell corresponds to one SBS, and the SBS is grouped, that is, the Small cell is grouped. Each small cell in which the SBS is deployed has a small cell gateway, and is configured to receive a user data measurement report fed back by the SUE, where the content of the user data measurement report includes location information of the SUE, power information sent by the SBS in the small cell to the SUE, and the like. . In step 110, the directed graph G of the SBS network is first obtained according to the graph theory, G=[T, E, W], where T={t ₁ , t ₂ , . . . , t _F } is all in the network. The set of vertices, each element in the collection represents an SBS. E is a set of edges between vertices, also called a potential interference matrix. The value of the element e _i,j in the latent interference matrix is 0 or 1. When e _i,j =1, it means that there exists between SBS _i and SBS _j . The interference and the interference value exceed the preset interference threshold I _th . When e _i,j =0, it means that there is no interference or interference value between SBS _i and SBS _j does not exceed the preset interference threshold I _th . Among them, SBS _i and SBS _j are respectively different SBSs in the SBS network. W is the weight of the directed graph edge, also called the interference matrix. The element w _ij =w _ji =max(p _ij /p _ii ,p _ji /p _jj ) in the interference matrix, where p _ij is sent to SBS for SBS _j the average power of the signal _i of the user, the average power of the signal p _ii addressed to own user as SBS _i, p _ji transmitted to the SBS _j user signal is the SBS _i mean power, p _jj addressed to own user SBS _j The average power of the signal. Thus, the elements in W are the greater of the ratio of the average power of the signal transmitted by the user to the plurality of neighboring SBSs to the signal power sent to it by the SBS. In an embodiment, if the user receives the average power of a signal sent to it by a neighboring SBS, the elements in W are the average power of the signal sent to it by the neighboring SBS and the signal sent to it by the self-authorized SBS. The ratio of power.

For the purpose of minimizing intra-group interference, according to the element values of the latent interference matrix in the directed graph, the F SBSs in the network are divided into L groups, denoted as χ={1, 2,..., L}, and the initial packet matrix V, meet when e _i between _I and SBS SBS _{_j, j} = 0, the _I and may be SBS SBS SBS _J assigned to the same team, and can only be assigned to any one SBS SBS into a group . The elements in the initial grouping matrix are exchanged according to the simulated annealing algorithm to obtain a better grouping matrix to meet the minimum interference requirement between the SBSs in the group, which is the grouping result of the SBS.

In an embodiment, the steps of translating the elements in the initial grouping matrix according to the simulated annealing algorithm to obtain a better grouping matrix are as follows:

In step 110, the initial temperature Tem and the number of iterations IN are set, and Tem and IN can take appropriate large initial values according to actual problems, and the settings of Tem and IN can prevent infinite iteration.

Step 120: Acquire an initial grouping matrix V and set a maximum number Z of each group, wherein the grouping matrix is a matrix of F rows and F columns, and the maximum value Z ranges from an integer in [1, F-1], V = (v _il ) _F*L , where v _il represents an element in the grouping matrix, F represents a matrix of F rows and F columns, and L represents the number of packets. When v _il = 1, the SBS SBS assigned to Group 1 _I, v _il = 0 when the time, regardless of the SBS SBS _I Group 1 and Group SBS SBS C _l in the number of no more than one Z.

Step 130, constructing the interference function as

Where v _il and v _jl respectively represent different elements in the grouping matrix.

Step 140: Transfer one SBS in any one SBS group to another group under the preset constraint condition, that is, change one element in V. Among them, the default constraints are:

Where st represents the meaning of the constraint and C _l represents the number of SBSs within the SBS team.

This formula indicates that the same base station cannot be in a different SBS group.

v _il ∈{0,1}.

Step 150, calculating a variation value of the disturbance function Δt'=C(S')-C(S), and if Δt'<0, then using S' as a new solution of the interference function, if Δt' ≥ 0, then the probability

S' is taken as a new solution to the interference function, where C(S) is the value of the interference function calculated from the initial grouping matrix, and C(S') is the value of the interference function calculated from the grouping matrix after the switching.

Step 160: Determine whether the preset exit condition is reached. If the exit condition is reached, the packet matrix that is the new solution of the interference function is output, otherwise step 170 is performed; wherein the exit condition is one of the following conditions: the number of iterations has exceeded IN times The change value of the disturbance function lasts Δt' ≥ 0; the temperature Tem reaches the threshold.

In step 170, the temperature Tem is decreased, and step 140 is performed.

In an embodiment, after the SBS packet is completed, a subchannel is allocated for each SBS group, and the throughput of the group is maximized under the premise of satisfying the user rate requirement in the group, so the orthogonal subchannel is allocated between the groups, and the allocation thereof is performed. The steps for the subchannel are as follows:

Step 210, passing the formula

Measuring the user rate requirement of SBS Team 1, where

For the minimum rate requirement of the user of SBS group 1, |C _l | is the number of SBSs in SBS group 1, and D _j is the set of users on SBS _j .

Step 220, determining the number of subchannels Y _l required by the SBS group 1 according to the user rate requirement, and

Where K is the number of all subchannels that can be assigned to the SBS team.

Step 230: Calculate a Signal to Interference plus Noise Ratio (SINR) of each SBS of the subchannel k in the SBS group 1, wherein the SINR of the subchannel k on the SBS _j in the SBS group 1 is

among them,

with

Representing the transmit power of MBS and SBS _j on subchannel k, respectively.

with

Representing the channel gains of MBS and SBS _j to user n on subchannel k, respectively, σ ² is the noise power. In one embodiment, all of the above power information can be obtained from the user data report.

Step 240: The minimum SINR value of the SINR of each SBS in the SBS group 1 calculated by the subchannel k calculated in step 230 is taken as the SINR of the SBS group 1.

Step 250, according to the SINR of the SBS group 1, under the premise that the minimum rate requirement of the user is met, the throughput in the group is maximized, that is,

When the throughput when subchannel k is allocated to SBS group 1 is greater than the throughput when subchannel k is allocated to other SBS groups and the number of subchannels that SBS group 1 has allocated is smaller than the number of subchannels required by SBS group 1, The subchannel k is assigned to the first group, otherwise the subchannel k is not assigned to the first group; wherein λ _k,l is a pointer variable, and the value of the pointer variable can only be 0 or 1, when λ _k,l =1, Indicates that subchannel k is assigned to SBS group 1, and when λ _k,l =0, it means that subchannel k is not allocated to SBS group 1.

In step 260, it is determined whether all subchannels are allocated. When all subchannels are allocated, the subchannel allocation matrix of each group is output. When there are still subchannels unallocated, step 250 is repeated until all subchannels are allocated.

In an embodiment, the throughput within the group in step 250 meets the following constraints in the calculation:

In this embodiment, the SBS is grouped according to the interference between the SBSs, and the sub-channels are allocated to the SBS group according to the actual needs of the SBS group, which not only reduces the computational complexity, but also minimizes the SBS in the case of ensuring the user rate in the group. The interference between the two technologies solves the problem that the computational complexity of the resource allocation strategy in the hybrid access mode is high, and the allocation strategy with low computational complexity is likely to cause strong co-channel interference between different SBSs.

After completing the subchannel allocation, the SBSs in each group use the subchannels assigned to the group, avoiding interference between different groups. The method provided by this embodiment further includes allocating subchannels allocated to the group to a plurality of users of the SBS in the group to ensure that the user can perform data communication through the subchannels allocated to the user. In this embodiment, the method for allocating a subchannel to a user may be: first, whether there is an idle subchannel in the SBS group; and in the case where there is an idle subchannel, the channel condition between the idle subchannel and the user is the best. The subchannels are assigned to users of the SBS within the group. In an embodiment, the channel condition is SINR, and the smaller the SINR value between the subchannel and the user, the better the channel condition between the subchannel and the user.

In this embodiment, the subchannels allocated to the SBS group can be divided into authorized groups according to a preset ratio, in order to ensure the communication quality of the authorized users and the communication requirements of the unlicensed users. The subchannel and the unlicensed subchannel, wherein the preset ratio can be adjusted according to the real-time network environment, and the default may be 1:3. After completing the subchannel proportional allocation, first perform subchannel allocation to the authorized user, as follows:

Step 310: Check whether the first idle subchannel exists in the authorized subchannel of the SBS corresponding to the authorized user, and if there is the first idle subchannel, directly allocate the subchannel with the best channel condition among the authorized subchannel and the authorized user to the authorized User, and step 350 is performed, if there is no first idle subchannel, step 320 is performed.

Step 320: Check whether there is an idle subchannel in the MBS. If there is an idle subchannel, select the subchannel with the best user channel condition in the idle subchannel of the MBS, assign it to the authorized user, and perform step 350 if there is no free subchannel. Then, step 330 is performed.

Step 330: Check whether there is a second idle subchannel in the unlicensed subchannel of the SBS corresponding to the authorized user, and if there is a second idle subchannel, directly allocate the subchannel with the best channel condition among the unlicensed subchannel and the authorized user. To the authorized user, and step 350 is performed, if there is no second idle subchannel, step 340 is performed.

In step 340, the authorized user is placed in the waiting queue, and when there are available subchannels, the subchannel is allocated to the user in the forefront of the waiting queue.

Step 350: Check the communication rate of the authorized user who has been assigned the subchannel, determine whether the communication rate meets the minimum rate requirement, and if the communication rate meets the minimum rate requirement, do not allocate a new subchannel to the authorized user, if the communication If the rate does not meet the minimum rate requirement, step 310 is re-executed to allocate the subchannel to the authorized user. In this embodiment, when sub-channel allocation is performed for an unauthorized user, since the unauthorized user cannot use the SBS sub-grant sub-channel, the execution starts directly from step 320. Although there are fewer subchannels under the SBS that an unlicensed user can use, in fact, an unauthorized user can use an unlicensed subchannel under any SBS, and when an authorized user leaves the SBS authorized for it, it becomes an unauthorized user. . In an embodiment, when the authorized sub-channel and the unlicensed user check whether the sub-channel is idle, if the idle sub-channel is allocated to the user with poor communication quality, it is still regarded as no idle sub-channel.

The effects of the present embodiment will be described in detail below with reference to FIGS. 2 to 5.

2 is a graph showing an average throughput trend of a macro cell user in a case where different SBS layout densities are increased, and FIG. 3 is an average signal to interference plus noise ratio of a SUE in a different SBS (ie, Small cell) deployment density. Figure 4 shows user fairness under different algorithms, and Figure 5 shows satisfaction when users use different algorithms. The algorithms used in the above figures are respectively: the resource allocation method provided by the embodiment of the present disclosure in the closed mode (corresponding to the polyline with "diamond" in FIGS. 2 to 5), and the implementation of the present disclosure in the mixed mode. The resource allocation method provided by the example (corresponding to the polyline with "asterisk" in Fig. 2 to Fig. 5), Cluster-based Resource Allocation (CRA) based on the Hopfield field network (Hopfield) (corresponding to the polyline with "square" in Figures 2 to 5), the intra-group orthogonal grouping algorithm (corresponding to the polyline with "triangle" in Figures 2 to 5) and the cluster-based heuristic based on Hopfield network Heuristic Cluster-based Femto-Femto Interference minimized Subchannels Allocation Algorithm (HCFM) (corresponding to a broken line with "circle" in FIGS. 2 to 5). The above algorithms were all simulated in the simulation environment as shown in Table 1.

Table 1

参数parameter	取值Value
MBS/SBS的最大发送功率Maximum transmit power of MBS/SBS	46dBm，20dBm46dBm, 20dBm
用户数User number	10MUE/扇区，1-4SUEs/SBS10MUE/sector, 1-4SUEs/SBS
MBS/SBS的天线增益MBS/SBS antenna gain	14dBi/5dBi14dBi/5dBi
室内/室外的阴影衰落差Indoor/outdoor shadow fading	8dB，4dB8dB, 4dB

系统带宽System bandwidth	10MHz10MHz
资源块RBs数目Number of resource block RBs	5050
外墙/内墙的穿透损耗Penetration loss of external wall/interior wall	20dB，5dB20dB, 5dB

FIG. 2 is a graph showing the average throughput of macro cell users in the case where the SBS arrangement density is increased by different algorithms. As can be seen from FIG. 2, when the SBS layout density is increased, the average throughput of the macro cell user is gradually reduced as the SBS density increases. The reason for this result is that as the SBS arrangement density increases, the number of SBSs around the MUE increases, causing the MUE to suffer from cross-layer interference caused by the SBS layer being larger than when the number of SBSs is small, so the MUE The quality of communication is also reduced. In the hybrid access mode, the method provided by the embodiment of the present disclosure uses a user's throughput to be at a higher level than other methods. Since the MUE acts as an unauthorized user, the network can be switched to proximity when subjected to severe interference. The communication quality is better on the SBS, which effectively guarantees the communication quality of the MUE.

Figure 3 shows the average signal to interference and noise ratio of SUEs at different SBS deployment densities. Compared with other algorithms, the resource allocation method provided by the embodiments of the present disclosure can effectively suppress SBS peer interference in a closed mode, and obtain a higher signal to interference and noise ratio, because the above method effectively prevents resource waste, thereby Improve the quality of user communication. In the hybrid access mode, the resource allocation method SBS provided by the embodiment of the present disclosure improves the user communication quality to a greater extent, and the cross-layer interference is more suppressed in the hybrid access mode, so the SUE letter The dry noise ratio is at a higher level.

Figure 4 shows user fairness under different algorithms. It can be seen from FIG. 4 that the user fairness of the resource allocation method provided by the embodiment of the present disclosure is the highest in the hybrid access mode, and the user fairness of the resource allocation method provided by the embodiment of the present disclosure is low in the closed mode. The user fairness of the resource allocation method provided by the embodiment of the present disclosure is adopted in the hybrid mode access mode. Since the resource allocation method provided by the embodiment of the present disclosure applies the simulated annealing algorithm to the grouping, the grouping problem is solved more efficiently, and the grouping result is obtained through multiple iterations, so that the ultra-dense network is more effectively suppressed. The same layer of interference also makes the degree of interference between users less different, making the user fairness at a higher level. In the hybrid access mode, the resource allocation method is used to suppress cross-layer interference to a greater extent, so that the communication quality of the MUE is greatly improved, and thus the fairness is improved to a greater extent. Compared with the above resource allocation method, other comparison algorithms do not consider user fairness too much, and the number of SBSs in each group differs greatly in other grouping schemes, resulting in a large difference in the number of users in different groups, ultimately resulting in The fairness of users in different groups is quite different, and the fairness of users is reduced. In an embodiment, the larger the ordinate value in FIG. 4, the greater the fairness.

Figure 5 shows the user's satisfaction with different algorithms. Compared with the CRA algorithm, the intra-group orthogonal algorithm and the HCFM algorithm, the user satisfaction of the resource allocation method proposed by the embodiment of the present disclosure is higher. Due to the effective suppression of the same layer interference and cross-layer interference existing in the ultra-dense network, the spectrum resources are utilized more efficiently, and the communication rate obtained by the user is higher, thereby obtaining higher satisfaction. Because the number of SBSs in each group of the comparison algorithm is quite different, the number of subchannels obtained by different groups is quite different, which results in different numbers of channels that can be multiplexed in the group, and some users cannot allocate enough channels. This leads to poor communication quality and low user satisfaction. In the above resource allocation method, the number of SBSs in the closed access mode is relatively consistent, and the channels that can be allocated are not much different, so that the user communication quality is good and the satisfaction is high. The above resource allocation method further adopts a hybrid access mode based on the closed access mode, and better utilizes the spectrum resources, and the user communication quality is further improved, thereby further satisfying the user satisfaction. In an embodiment, the larger the ordinate value in FIG. 5, the higher the user satisfaction.

In an embodiment, an embodiment of the present disclosure provides a resource allocation server, and FIG. 6 is a structural block diagram of a resource allocation server according to an embodiment of the present disclosure. As shown in FIG. 6, the server includes at least a memory 610 and a processor. 620, a computer program is stored on the memory 610, and the processor 620 implements the steps 110 to 130 in the execution of the computer program on the memory, which is the same as the steps of the embodiment of the present disclosure, and details are not described herein again. In actual use, the resource allocation server provided by the embodiment of the present disclosure may be disposed on the MBS as an actual device that allocates a subchannel for the SBS.

In this embodiment, the Small cell is a small cell in which an SBS is deployed, that is, one Small cell corresponds to one SBS, and the SBS is grouped to group the Small cells. Each small cell in which the SBS is deployed has a small cell gateway, and is configured to receive a user data measurement report fed back by the SUE, where the content of the user data measurement report includes location information of the SUE, power information sent by the SBS in the small cell to the SUE, and the like. . When the processor runs the computer program, it first obtains the directed graph G of the SBS network according to the graph theory, G=[T, E, W], where T={t ₁ , t ₂ ,..., t _F } is A collection of all the vertices in the network. Each element in the collection represents an SBS. E is a set of edges between vertices, also called a potential interference matrix. The value of the element e _i,j in the latent interference matrix is 0 or 1. When e _i,j =1, it means that there exists between SBS _i and SBS _j . The interference and the interference value exceed the preset interference threshold I _th . When e _i,j =0, it means that there is no interference or interference value between SBS _i and SBS _j does not exceed the preset interference threshold I _th . Among them, SBS _i and SBS _j are respectively different SBSs in the SBS network. W is the weight of the directed graph edge, also called the interference matrix. The element w _ij =w _ji =max(p _ij /p _ii ,p _ji /p _jj ) in the interference matrix, where p _ij is sent to SBS for SBS _j the average power of the signal within the user _i, an average power, p _ji signal p _ii addressed to own user to SBS _i transmitted as SBS _i to the average power of the signal within the user's SBS _{_j,} p _jj SBS _j addressed to own user The average power of the signal. Thus, the elements in W are the greater of the ratio of the average power of the signal transmitted by the user to the plurality of neighboring SBSs to the signal power sent to it by the SBS. In an embodiment, if the user receives the average power of a signal sent to it by a neighboring SBS, the elements in W are the average power of the signal sent to it by the neighboring SBS and the signal sent to it by the self-authorized SBS. The ratio of power.

For the purpose of minimizing intra-group interference, the processor divides the F SBSs in the network into L groups according to the element values of the potential interference matrix in the directed graph when running the computer program, which is expressed as χ={1,2,... , L}, an initial packet is determined matrix V, meet when e _i between _I and SBS SBS _{_j, j} = 0, the _I and may be SBS SBS SBS _J assigned to the same team, and only any one SBS Assigned to an SBS team. The elements in the initial grouping matrix are exchanged according to the simulated annealing algorithm to obtain a better grouping matrix to meet the minimum interference requirement between the SBSs in the group, which is the grouping result of the SBS.

In an embodiment, the step of the processor operating the simulated annealing algorithm in the computer program to swap the elements in the initial grouping matrix to obtain a better grouping matrix is the same as steps 110 to 170 in the disclosed embodiment. I will not repeat them here.

In an embodiment, after the SBS packet is completed, a subchannel is allocated for each SBS group, and the throughput of the group is maximized on the premise of satisfying the user rate requirement in the group, so the orthogonal subchannels are allocated between the groups. In an embodiment, when the processor allocates a sub-channel to each of the SBS groups, the steps are the same as the steps 210 to 260 in the embodiment of the present disclosure, and details are not described herein again.

In this embodiment, the SBS is grouped according to the interference between the SBSs, and the sub-channels are allocated to the SBS group according to the actual needs of the SBS group, which not only reduces the computational complexity, but also minimizes the SBS in the case of ensuring the user rate in the group. The inter-channel interference solves the problem that the computational complexity of the resource allocation strategy in the hybrid access mode is high in the related art, and the co-channel interference in the allocation strategy SBS with low computational complexity is strong.

After completing the subchannel allocation, the SBS in each group can only use the subchannels allocated to the group, avoiding interference between different groups. In an embodiment, the processor of the server in the embodiment further implements the following steps: when the computer program on the memory is executed, the subchannel allocated to the SBS group is allocated to multiple users of the SBS in the group to ensure that the user can Data communication is performed by subchannels assigned to itself. When assigning the subchannel to which each SBS group is assigned to multiple users of the SBS in the group, the processor implements the following steps: first, whether there is an idle subchannel in the SBS group; in the case where there is an idle subchannel The subchannels with the best channel conditions among the idle subchannels and the users are allocated to the users of the SBS in the group. In an embodiment, the channel condition is SINR, and the smaller the SINR value between the subchannel and the user, the better the channel condition between the subchannel and the user.

In this embodiment, the subchannels allocated to the SBS group can be divided into authorized groups according to a preset ratio, in order to ensure the communication quality of the authorized users and the communication requirements of the unlicensed users. The subchannel and the unlicensed subchannel, wherein the preset ratio can be adjusted according to the real-time network environment, and the default may be 1:3. After the sub-channel ratio allocation is completed, the processor performs sub-channel allocation to the authorized user, and the implementation steps are the same as steps 310 to 350 in the embodiment of the present disclosure, and details are not described herein again. In an embodiment, when sub-channel allocation is performed for an unauthorized user, since the unauthorized sub-channel of the SBS cannot be used by the unauthorized user, execution is directly started from step 320. Although there are fewer subchannels under the SBS that an unlicensed user can use, in fact, an unauthorized user can use an unlicensed subchannel under any SBS, and when an authorized user leaves the SBS authorized for it, it becomes an unauthorized user. . In an embodiment, when the authorized sub-channel and the unlicensed user check whether the sub-channel is idle, if the idle sub-channel is allocated to the user with poor communication quality, it is still regarded as no idle sub-channel.

Claims

A resource allocation method, including:

Obtaining a directed graph of a small cell base station SBS network;

According to the directed graph, grouping a plurality of SBSs in the SBS network according to a set grouping policy to obtain a plurality of SBS groups;

A subchannel is assigned to each of the SBS groups.
The resource allocation method according to claim 1, wherein the grouping the plurality of SBSs in the SBS network according to the set grouping policy comprises:

Determining an initial grouping matrix according to the directed graph;

Interchanging elements in the initial grouping matrix according to a simulated annealing algorithm;

When the exit condition of the simulated annealing algorithm is satisfied, the switched grouping matrix is output.
The resource allocation method according to claim 1, wherein assigning a subchannel to each of said SBS groups comprises:

Determining the number of subchannels required for each of the SBS groups;

Calculating the throughput of the subchannel to be allocated in each of the SBS groups;

If the throughput of the subchannel allocated to the first SBS group is greater than the throughput when the subchannel is allocated to the second SBS group and the number of subchannels that the first SBS group has been allocated is smaller than the first SBS The number of subchannels required by the group is then assigned to the first SBS group.
The resource allocation method according to claim 3, wherein the number of subchannels required for each of said SBS groups is determined, including:

Measuring user rate requirements for each of the SBS teams;

The number of subchannels required for each of the SBS groups is determined based on the user rate requirements.
The resource allocation method according to claim 3, wherein calculating the throughput of the subchannel to be allocated in each of the SBS groups comprises:

For each SBS group, calculating a signal to interference plus noise ratio of each SBS in the SBS group for the subchannel to be allocated;

Using the minimum value of all signal to interference plus noise ratios in the SBS group as the signal to interference plus noise ratio of the SBS group;

Calculating the throughput of the subchannel to be allocated in the SBS group according to the signal to interference plus noise ratio of the SBS group.
The resource allocation method according to claim 5, wherein the calculation formula of the throughput of the SBS group is:

Where B is the channel bandwidth, j is the SBS in the SBS group, k is the subchannel assigned to j, n is the user connected to j, and F is the total number of SBSs.
For the signal-to-interference plus noise ratio of the SBS group, λ k,l is a pointer variable, the value of the pointer variable is 0 or 1, and when λ k,l =1, indicating that the subchannel k is allocated to the The first SBS group, when λ k, l =0, indicates that subchannel k is not allocated to the first SBS group.
The resource allocation method according to claim 1, wherein after allocating the subchannels for each of the SBS groups, the method further comprises:

A subchannel to which each of the SBS groups is assigned is allocated to a plurality of users of the SBS service within the SBS group.
The resource allocation method according to claim 7, wherein the subchannel to which each of the SBS groups is allocated is allocated to a plurality of users of the SBS service in the SBS group, including:

Checking whether there are free subchannels in the SBS group;

When there are a plurality of the idle subchannels, the subchannels having the largest throughput among the plurality of idle subchannels and the plurality of users are allocated to users corresponding to the subchannels having the highest throughput.
The resource allocation method according to claim 7 or 8, wherein the subchannel to which each of the SBS groups is allocated is allocated to a plurality of users of the SBS service in the SBS group, including:

Subchannels of the SBS in the SBS group are divided into an authorized subchannel and an unlicensed subchannel according to a preset ratio;

Determining whether the user is an authorized user;

If the user is an authorized user, check whether the authorized subchannel has a first idle subchannel;

And if a plurality of the first idle subchannels exist in the authorized subchannel, a subchannel that has the largest throughput among the plurality of first idle subchannels and the authorized user is allocated to the authorized user;

If the preset user is an unauthorized user, check whether the unlicensed subchannel has a second idle subchannel;

And assigning, in the non-authorized subchannel, a plurality of the second idle subchannels, the subchannels that have the largest throughput among the plurality of second idle subchannels and the unlicensed users to the non Authorized user.
A resource allocation server includes at least a memory, a processor, and a computer program stored thereon, and the steps implemented by the processor when executing the computer program on the memory include:

Obtaining a directed graph of a small cell base station SBS network;

According to the directed graph, grouping a plurality of SBSs in the SBS network according to a set grouping policy to obtain a plurality of SBS groups;

A subchannel is assigned to each of the SBS groups.
The resource allocation server according to claim 10, wherein when the processor groups a plurality of SBSs in the SBS network, the implementing steps include:

Determining an initial grouping matrix according to the directed graph;

The elements in the group matrix of the initial points are exchanged according to a simulated annealing algorithm;

When the exit condition of the simulated annealing algorithm is satisfied, the switched grouping matrix is output.
The resource allocation server according to claim 10, wherein when the processor allocates a subchannel for each of the SBS groups, the implementing steps include:

Determining the number of subchannels required for each of the SBS groups;

Calculating the throughput of the subchannel to be allocated in each of the SBS groups;

If the throughput of the subchannel allocated to the first SBS group is greater than the throughput when the subchannel is allocated to the second SBS group and the number of subchannels that the first SBS group has been allocated is smaller than the first SBS The number of subchannels required by the group is then assigned to the first SBS group.
The resource allocation server according to claim 12, wherein when said processor determines the number of subchannels required for each of said SBS groups, the steps of implementing include:

Measuring user rate requirements for each of the SBS teams;

The number of subchannels required for each of the SBS groups is determined based on the user rate requirements.
The resource allocation server according to claim 12, wherein said processor calculates a throughput of the subchannel to be allocated in each of said SBS groups, and the steps of implementing:

For each SBS group, calculating a signal to interference plus noise ratio of each SBS in the SBS group for the subchannel to be allocated;

Using the minimum value of all signal to interference plus noise ratios in the SBS group as the signal to interference plus noise ratio of the SBS group;

Calculating the throughput of the subchannel to be allocated in the SBS group according to the signal to interference plus noise ratio of the SBS group.
The resource allocation server of claim 14 wherein said processor calculates a throughput formula for said SBS team to use:

Where B is the channel bandwidth, j is the SBS in the SBS group, k is the subchannel assigned to j, n is the user connected to j, and F is the total number of SBSs.
For the signal-to-interference plus noise ratio of the SBS group, λ k,l is a pointer variable, the value of the pointer variable is 0 or 1, and when λ k,l =1, indicating that the subchannel k is allocated to the The first SBS group, when λ k, l =0, indicates that subchannel k is not allocated to the first SBS group.
The resource allocation server according to claim 10, wherein after the processor allocates a subchannel for each of the SBS groups, the implementing step further comprises:

A subchannel to which each of the SBS groups is assigned is allocated to a plurality of users of the SBS service within the SBS group.
The resource allocation server according to claim 16, wherein when said processor allocates a subchannel to which said each SBS group is assigned to a plurality of users of an SBS service within said SBS group, the steps of implementing:

Checking whether there are free subchannels in the SBS group;

And in a case where there are a plurality of the idle subchannels, a subchannel that maximizes a user rate in a case where the plurality of users are satisfied with a minimum rate requirement for each user among the plurality of idle subchannels is allocated to The user corresponding to the subchannel.
The resource allocation server according to claim 16 or 17, wherein when the processor allocates a subchannel to which each of the SBS groups is allocated to a plurality of users of the SBS service in the SBS group, The steps to achieve include:

Subchannels of the SBS in the SBS group are divided into an authorized subchannel and an unlicensed subchannel according to a preset ratio;

Determining whether the user is an authorized user;

If the user is an authorized user, check whether the authorized subchannel has a first idle subchannel;

And in a case that the authorized subchannel has a plurality of the first idle subchannels, maximizing a user in a case where a plurality of the first idle subchannels and the authorized user meet each authorized user minimum rate requirement a subchannel of the rate is allocated to the authorized user;

If the preset user is an unauthorized user, check whether the unlicensed subchannel has a second idle subchannel;

If a plurality of the second idle subchannels exist in the unlicensed subchannel, the minimum rate requirement of each unauthorized user is met between the plurality of the second idle subchannels and the unlicensed user A subchannel that maximizes the user rate is assigned to the unauthorized user.