CN110691360B - Joint resource allocation method based on user clustering and Starkelberg model - Google Patents

Joint resource allocation method based on user clustering and Starkelberg model Download PDF

Info

Publication number
CN110691360B
CN110691360B CN201910775498.1A CN201910775498A CN110691360B CN 110691360 B CN110691360 B CN 110691360B CN 201910775498 A CN201910775498 A CN 201910775498A CN 110691360 B CN110691360 B CN 110691360B
Authority
CN
China
Prior art keywords
fue
mbs
interference
fbs
model
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201910775498.1A
Other languages
Chinese (zh)
Other versions
CN110691360A (en
Inventor
张若南
韦林原
蒋毅
翟道森
李彬
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Northwestern Polytechnical University
Original Assignee
Northwestern Polytechnical University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Northwestern Polytechnical University filed Critical Northwestern Polytechnical University
Priority to CN201910775498.1A priority Critical patent/CN110691360B/en
Publication of CN110691360A publication Critical patent/CN110691360A/en
Application granted granted Critical
Publication of CN110691360B publication Critical patent/CN110691360B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/04—Wireless resource allocation
    • H04W72/044—Wireless resource allocation based on the type of the allocated resource
    • H04W72/0453—Resources in frequency domain, e.g. a carrier in FDMA
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models
    • G06N5/042—Backward inferencing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/02—Resource partitioning among network components, e.g. reuse partitioning
    • H04W16/10—Dynamic resource partitioning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/14—Spectrum sharing arrangements between different networks
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/04—TPC
    • H04W52/06—TPC algorithms
    • H04W52/14—Separate analysis of uplink or downlink
    • H04W52/146—Uplink power control
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/04—TPC
    • H04W52/18—TPC being performed according to specific parameters
    • H04W52/24—TPC being performed according to specific parameters using SIR [Signal to Interference Ratio] or other wireless path parameters
    • H04W52/242—TPC being performed according to specific parameters using SIR [Signal to Interference Ratio] or other wireless path parameters taking into account path loss
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/53—Allocation or scheduling criteria for wireless resources based on regulatory allocation policies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W72/00—Local resource management
    • H04W72/50—Allocation or scheduling criteria for wireless resources
    • H04W72/54—Allocation or scheduling criteria for wireless resources based on quality criteria
    • H04W72/542—Allocation or scheduling criteria for wireless resources based on quality criteria using measured or perceived quality

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Theoretical Computer Science (AREA)
  • Computing Systems (AREA)
  • Evolutionary Computation (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention discloses a joint resource allocation method based on user clustering and a Starkelberg model, which comprises the steps of constructing a Femtocell double-layer heterogeneous network formed by mixing MBS and a plurality of FBS; establishing a path loss model according to the Femtocell double-layer heterogeneous network; then, a distance measurement method based on path loss is provided based on a channel model, a clustering algorithm is adopted to cluster the FBS which is densely deployed in the cell, users are grouped according to a clustering result, different subcarrier channels are distributed to different user groups, and the FBS which is densely deployed is converted into a FBS scene which is sparsely deployed; the method comprises the steps that a game model for controlling power between an MBS and an FUE is built according to a Stackelberg game model, the MBS is a game leader pricing interference power from the FUE, the FUE is an interference pricing given by game participants to the MBS to dynamically adjust own transmitting power, and when the game reaches a balanced position, interference benefits of the MBS and the overall link rate of the FUE reach an optimal solution. The coverage area and the communication performance of the Femtocell network are improved.

Description

Joint resource allocation method based on user clustering and Starkelberg model
Technical Field
The invention belongs to the technical field of wireless communication, and particularly relates to a joint resource allocation method based on user clustering and a Starkelberg model.
Background
With the development of mobile communication technology and the rapid increase of the demand of indoor communication capacity and rate, how to solve indoor signal coverage and improve communication capacity and rate becomes a hot point of research. Meanwhile, enhanced mobile broadband (eMBB), one of the most important application scenarios of 5G, is basically deployed in a relatively high frequency band. From the knowledge of electromagnetism, it can be known that the higher the frequency of an electromagnetic signal is, the greater the loss of the electromagnetic signal in space propagation is, and therefore, in order to meet the communication demand, the transmission power of a base station must be increased or the deployment density of the base station must be increased. The deployment of a large number of Femtocell base stations (femtocells) in the home or office to optimize indoor coverage and increase communication capacity and rate is now a widely accepted approach by the industry and academia.
In a typical deployment scenario, a Femtocell heterogeneous network mainly consists of a Macro Base Station (MBS) and a plurality of home base stations (FBSs) deployed by home users. In the cell range, basic signal coverage is provided by MBS, and signal enhancement is performed locally by FBS. The heterogeneous network is very effective for improving the communication experience of the cell home users, and the advantages are also very obvious. Compared with the traditional MBS, in the Femtocell network where the MBS and the FBS are deployed cooperatively, the MBS can properly reduce the transmission power of the MBS, thereby effectively reducing the interference to the MBS in other cells, and the FBS is mainly deployed by a user, thereby effectively reducing the deployment cost. However, the Femtocell network formed by combining the MBS and the FBS changes the structure of the original network and increases the complexity of the network. Meanwhile, due to limited spectrum resources and a complex channel environment between the MBS and the FBS, the FBS and the MBS need to adopt a reasonable channel model to perform network deployment optimization and resource allocation scheme to avoid severe interference. Therefore, it is necessary to perform channel measurement and modeling on the cell home network, evaluate and optimize the performance of the network according to the channel model, and provide a practical spectrum resource allocation and power control algorithm, thereby reducing interference between base stations in the network, improving the capacity and transmission rate of the network, and implementing higher reliability and wider access technology.
In recent years, as the demand for indoor user capacity and rate has increased dramatically, channel measurement and modeling of a Femtocell network formed by MBS and FBS do not consider the scenario of communication between indoor FBS and FBS. The FBS is clustered by adopting a K-means clustering algorithm, and the consideration of the channel propagation characteristic to the clustering effect is lacked. In addition, the network performance and the resource allocation effect under multi-spectrum resources are not considered when the Stackelberg game model is adopted for resource allocation.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a joint resource allocation method based on user clustering and a starkeberg model for overcoming the defects in the prior art, which can be used for base station deployment of Femtocell heterogeneous networks and design of radio access networks in homes and offices, and effectively reduce the system complexity and the overall communication performance.
The invention adopts the following technical scheme:
a Femtocell double-layer heterogeneous network formed by mixing MBS and a plurality of FBS is constructed based on a combined resource allocation method of user clustering and a Starkelberg model; establishing a path loss model according to the Femtocell double-layer heterogeneous network; then, a distance measurement method based on path loss is provided based on a channel model, a clustering algorithm is adopted to cluster the FBS which is densely deployed in the cell, users are grouped according to a clustering result, different subcarrier channels are distributed to different user groups, and the FBS which is densely deployed is converted into a FBS scene which is sparsely deployed; the method comprises the steps that a game model for controlling power between an MBS and an FUE is built according to a Stackelberg game model, the MBS is a game leader pricing interference power from the FUE, the FUE is an interference pricing given by game participants to the MBS to dynamically adjust own transmitting power, and when the game reaches a balanced position, interference benefits of the MBS and the overall link rate of the FUE reach an optimal solution.
Specifically, the construction of the Femtocell double-layer heterogeneous network formed by mixing MBS and a plurality of FBS specifically comprises the following steps:
s101, only considering an uplink in the Femtocell network joint resource allocation;
s102, defining the number of MBS in the network as 1 and the number of FBS as N, and setting the number of FUE as N if each FBS serves a home user equipment;
s103, defining a frequency spectrum bandwidth available for the network to be M, and the number of subcarrier channels to be C, where the bandwidth of each subcarrier channel is B ═ M/C;
s104, defining the transmitting power of FUEi as SiChannel gain to MBS uplink cross-layer interference is giUplink gain to FBSj is hi,j(ii) a When i is j, hi,jIndicating the channel gain between the FUE and its dedicated FBS; when i ≠ j, hi,jRepresents the same layer interference of FUEi to the uplink of FUEj;
s105, defining the price of the interference power from FUEi to MBS to be pi；
S106, calculating link capacity of FUE
Figure BDA0002174892960000031
And overall throughput V of all FUEsFUE；
S107, calculating the interference gain V of the MBSMBS。
Further, in step S106,
Figure BDA0002174892960000032
the calculation formula is as follows:
Figure BDA0002174892960000033
wherein σ2Representing a gaussian white noise power;
VFUEthe calculation formula is as follows:
Figure BDA0002174892960000034
further, in step S107, VMBSThe calculation formula is as follows:
Figure BDA0002174892960000035
specifically, the channel model PL is specifically:
PL＝A·log10(d3D)+B·log10(fc)+C(1-cosθ)2+0.5din+Ψ
wherein A, B, C represent parameters of each influencing factor, d3DIndicating the distance of signal transmission, fcDenotes the center frequency of the carrier, θ denotes the pitch angle of the signal transmission between MBS to FBS or FUE, dinDenotes the distance of the FBS or FUE from the wall, Ψ denotes shadow fading, and is a variable conforming to the log-normal distribution.
Specifically, grouping users by using a clustering algorithm specifically comprises:
s301, defining K user clusters,
Figure BDA0002174892960000046
the number of users in each user cluster is L,
Figure BDA0002174892960000041
s302, defining the geometric coordinate of the cluster center of the kth user cluster as (c)k,x,ck,y)
Figure BDA0002174892960000042
S303, defining the geometric coordinate of the first user FUE in the kth user cluster as (x)k,l,yk,l)；
S304, defining the distance between FUE i and FUE j as path loss distance PLDi,j；
S305, defining a cost function of user clustering as J (U, V);
Figure BDA0002174892960000043
wherein, let U and V be the optimization results of K and L;
s306, determining the optimized objective function of the clustering algorithm as
Figure BDA0002174892960000044
S307, C user groups are defined in total, the number of users in each group is N, and sigma N is N; the same sub-carrier channel is used within a group and different sub-carrier channels are used by users between groups.
Further, in step S304, the path loss distance PLDi,jThe method specifically comprises the following steps:
Figure BDA0002174892960000045
where PL represents a calculation formula of the path loss.
Further, the constraint conditions of step S306 are:
Figure BDA0002174892960000056
C2:0＜L≤Lmax
C3:Kmin≤K≤Kmax
wherein, in order to ensure load balance and QoS, the maximum value of L cannot exceed Lmax。
Specifically, a game model for performing power control between the MBS and the FUE is constructed according to the Stackelberg game model, and specifically includes:
s401, defining the interference threshold of MBS in any sub-carrier frequency band as Q, and the interference of any FUE to MBS as Ii；
Ii＝gisi；
S402, OFDMA access mode is adopted among user groups adopting different subcarriers, and a power control algorithm is set for users in the group adopting spectrum sharing.
S403, defining the interference gain of MBS to the ith user group as
Figure BDA0002174892960000051
Comprises the following steps:
Figure BDA0002174892960000052
s404, determining the optimized objective function of the MBS interference gain as
Figure BDA0002174892960000053
The constraint is that C1 is:
Figure BDA0002174892960000054
s405, defining signal noise interference ratio SNIR of ith FUEi(si,s-i) Comprises the following steps:
Figure BDA0002174892960000055
s406, defining the rate of the ith FUE as
Figure BDA0002174892960000061
Comprises the following steps:
Figure BDA0002174892960000062
wherein λ isiRepresents the rate gain of the ith FUE;
s407, determining an optimized objective function for the uplink rate of FUE as
Figure BDA0002174892960000063
S408, rewriting the optimization problem in the step S407 into maxf (S)i)＝maxλlog(1+SNIRi(si,s-i))-piIi；
S409, for f (S) in step S408i) Solving a first derivative to obtain:
Figure BDA0002174892960000064
S410、for f (S) in step S408i) And solving a second derivative to obtain:
Figure BDA0002174892960000065
the optimal solution of the optimization problem of S411 and S408 meets the KKT limitation, and the optimal transmitting power of the ith FUE is obtained by using the standard optimization technology and the KKT limitation condition as follows:
Figure BDA0002174892960000066
s412, defining an optimization objective function of the sum of all FUE overall uplink rates;
s413, solving the optimization problem in S412 to obtain the pricing p of the MBS to the ith FUE uplink interference poweriBy calculating
Figure BDA0002174892960000067
Obtain the transmission power s of each FUEiAnd obtaining a solution of the optimization problem in the step S404, and finally obtaining an optimal solution of the interference gain of the MBS and the overall link rate of the FUE to reach a game equilibrium point.
Further, the optimization objective function of step S412 is:
Figure BDA0002174892960000068
the constraint condition C1 is:
Figure BDA0002174892960000071
compared with the prior art, the invention has at least the following beneficial effects:
the invention relates to a novel resource allocation method in a Femtocell network, which constructs a Femtocell double-layer heterogeneous network formed by mixing MBS and a plurality of FBS. Channel characteristics in a deployment scene of the double-layer heterogeneous network are deeply explored, and a simple and effective path loss model is established; then, based on a channel model, a new distance measurement mode is provided, namely a path loss distance, based on the distance, a clustering algorithm is adopted to cluster the FBS densely deployed in the cell, users are grouped according to a clustering result, and different subcarrier channels are allocated to different user groups, so that the FBS densely deployed is converted into a FBS sparsely deployed scene, and the same frequency interference is effectively reduced; and then constructing a game model for controlling power between the MBS and the FUE according to the Stackelberg game model, wherein the MBS is a game leader to price interference power from the FUE, the FUE is an interference price given by a game participant to the MBS to dynamically adjust own transmitting power, and when the game reaches a balanced position, the interference gain of the MBS and the overall link rate of the FUE reach an optimal solution.
Furthermore, the network design of joint resource allocation in the Femtocell improves the signal coverage effect, reduces the MBS load, maximizes the overall FUE throughput, provides a stable, reliable and efficient communication link for the family user, and adopts a multi-carrier subchannel, so that the overall power consumption of the communication system is greatly reduced while the frequency spectrum resources are fully utilized.
Furthermore, based on a WINNER + channel model and the analysis of actually measured data, the indoor loss correction factor is added in consideration of the actual deployment environment of the base station, the channel model in WINNER + is popularized to the channel model of the Femtocell network, and the model can well fit actual measured data in the scene. Compared with a traditional channel model from a macro cellular base station to a cellular user, the channel model of the Femtocell network fully considers the influence of the incident angle of signals and the complexity of indoor environment on the accuracy of the channel model, so that the channel model is more consistent with the actual deployment scene of the base station, and a model basis is provided for network resource allocation and optimization algorithm design.
Furthermore, due to the performance limitation of the Femtocell network adopting the spectrum resource sharing mode, the link capacity of the system is improved by adopting the multi-carrier sub-channel. In the problem of sub-channel resource allocation, a clustering and grouping algorithm based on a path loss distance is provided, FBS with a close distance is isolated on a frequency spectrum and allocated to different user groups, the uplink interference of the FBS by FUE is effectively reduced by adopting a frequency division mode among the groups, and the interference is reduced by adopting a frequency spectrum sharing and transmission power game control mode for users in the groups. The spectrum resource allocation algorithm not only makes full use of limited spectrum resources, but also reduces the overall power consumption of a communication system, so that the communication is more environment-friendly.
Further, the users in the group use a spectrum resource sharing mode, and need to control the uplink interference power of the FUE. On the power control problem, a Stackelberg game model is constructed. Because the FBS dense deployment is converted into the FBS sparse deployment by using the user clustering and grouping algorithm, the same frequency interference among users in a group does not need to be considered in the game model, so that the complexity of the game model is greatly reduced, and the operation speed of the iterative algorithm is effectively improved. The power control problem is converted into two optimization problems of MBS benefit maximization and FUE overall link rate maximization, firstly, a solution of the FUE overall rate maximization problem is obtained through a standard optimization technology and a KKT condition, and then the solution is utilized to calculate the maximum benefit of MBS.
In conclusion, the method provides a JRAF scheme based on an actually measured channel model and used for Femtocell network resource allocation, and solves the challenging problem that a cell with high network throughput is covered on a heterogeneous access network by using MBS and FBS under the scene that the frequency spectrum resources are limited. In order to improve the coverage area and the communication performance of the Femtocell network, a user clustering and grouping algorithm is further designed to maximize the utilization rate of frequency spectrum resources and a power control algorithm to maximize the network throughput.
The technical solution of the present invention is further described in detail by the accompanying drawings and embodiments.
Drawings
FIG. 1 is a system model of a Femtocell network;
fig. 2 is a practical deployment of the transmitting and receiving antennas in channel measurement, in which a) is a transmitting terminal device deployed on the roof, and b) is a receiving device deployed in a cell room;
fig. 3 is a satellite diagram and a partial detail diagram of a measurement scenario, wherein a) is the satellite diagram of the measurement scenario, b) is a detail diagram of a cell room, c) is a detail diagram of a cell elevator room, d) is a detail diagram of a cell corridor, and e) is a plan view of a cell room distribution;
FIG. 4 is a graph comparing measured path loss results with existing standard model path loss values;
FIG. 5 is a diagram of a model result obtained by modifying the WINNER + model to conform to actual measurement data;
FIG. 6 is a shadow fading fit of a channel model;
FIG. 7 is a schematic diagram of FUE clustering;
FIG. 8 is a graph of interference power cost versus interference threshold for different FUEs;
FIG. 9 is a performance of MBS benefit performance in different channel models by the resource allocation method proposed by the present invention;
FIG. 10 is a diagram illustrating the performance of the overall FUE rate in different channel models according to the resource allocation method of the present invention;
FIG. 11 is a performance comparison of MBS benefit in the resource allocation method proposed by the present invention and other resource allocation modes;
FIG. 12 is a comparison of the overall FUE rate performance in the resource allocation method proposed by the present invention with that in other resource allocation schemes.
Detailed Description
The invention provides a joint resource allocation method based on user clustering and a Starkelberg model, which comprises the steps of firstly constructing an actual Femtocell network structure, namely adopting a Macro Base Station (MBS) deployed at the top of a central building and a plurality of Femtocell Base Stations (FBS) deployed in various residences in a typical cell environment. Channel measurement and modeling are carried out under the scene; then, based on a channel model obtained by actual measurement, an overall framework of the joint resource allocation method is established, and user clustering and clustering algorithms based on 'path loss distance' are adopted in spectrum resource allocation to reasonably allocate spectrum resources, so that the same-layer interference between FBS is effectively reduced; an algorithm based on a Stackelberg game model is adopted in power control, so that cross-layer interference benefits of the MBS and the overall throughput of all users in the FBS are maximized and a game balance point is reached.
The invention discloses a joint resource allocation method based on user clustering and a Starkelberg model, which is called Femtocell network Joint Resource Allocation (JRAF), and comprises cell Femtocell network fading channel measurement and modeling, a user clustering and grouping algorithm based on path loss distance and a multi-spectrum power control algorithm based on a Stackelberg game model.
S1, establishing a system model of Femtocell base station (Femtocell) network joint resource allocation;
s101, only considering an uplink in the Femtocell network joint resource allocation;
s102, defining the number of MBS to be 1 and the number of FBS to be N in the network, and assuming that each FBS serves one home user equipment (FUE) without loss of generality, setting the number of FUE to be N;
s103, defining a frequency spectrum bandwidth available for the network to be M, and the number of subcarrier channels to be C, where the bandwidth of each subcarrier channel is B ═ M/C;
s104, defining the transmitting power of FUEi as SiChannel gain of its uplink cross-layer interference to MBS is giUplink gain to FBSj is hi,j. When i is j, hi,jIndicating the channel gain between the FUE and its dedicated FBS; when i ≠ j, hi,jRepresents the same layer interference of FUEi to the uplink of FUEj;
s105, defining the price of the interference power from FUEi to MBS to be pi；
S106, calculating link capacity of FUE
Figure BDA0002174892960000101
And overall throughput V of all FUEsFUE；
Figure BDA0002174892960000102
The calculation formula is as follows:
Figure BDA0002174892960000103
wherein σ2Representing a gaussian white noise power;
VFUEthe calculation formula is as follows:
Figure BDA0002174892960000104
s107, calculating the interference gain V of the MBSMBS；
VMBSThe calculation formula is as follows:
Figure BDA0002174892960000105
it is considered that the interference that MBS can accept in Femtocell networks is limited, and it is also necessary to guarantee basic communication service per FUE. Meanwhile, how to cluster and group users and further allocate spectrum resources is one of the key problems.
S2, obtaining a channel model under the scene through a path loss modeling result measured by the Femtocell network channel
Comparing the existing path loss models, such as WINNER II and WINNER +, with the path loss difference obtained by actual testing, it is indicated that the existing models lack accuracy and need to be corrected.
The channel model is as follows:
PL＝A·log10(d3D)+B·log10(fc)+C(1-cosθ)2+0.5din+Ψ
wherein A, B, C represent parameters of each influencing factor, d3DIndicating the distance of signal transmission, fcDenotes the center frequency of the carrier, θ denotes the pitch angle of the signal transmission between MBS to FBS or FUE, dinDenotes the distance of the FBS or FUE from the wall, Ψ denotes shadow fading, and is a variable conforming to the log-normal distribution.
S3, carrying out user clustering and grouping method of spectrum resource allocation by using the path loss channel model;
s301, defining a total of K user clusters,
Figure BDA0002174892960000111
where the number of users in each user cluster is L,
Figure BDA0002174892960000112
s302, defining the geometric coordinate of the cluster center of the kth user cluster as (c)k,x,ck,y)；
Figure BDA0002174892960000113
S303, defining the geometric coordinate of the first user FUE in the kth user cluster as (x)k,l,yk,l)；
S304, defining the distance between FUE i and FUE j as "path loss" distance PLDi,j；
Figure BDA0002174892960000114
Where PL represents a calculation formula of the path loss.
S305, defining a cost function of user clustering as J (U, V);
Figure BDA0002174892960000115
and making U and V as the optimization results of K and L.
S306, determining the optimized objective function of the clustering algorithm as
Figure BDA0002174892960000121
The constraint conditions are C1-C3: wherein, in order to ensure load balance and QoS, the maximum value of L cannot exceed Lmax；
Figure BDA0002174892960000122
C2:0＜L≤Lmax
C3:Kmin≤K≤Kmax
S307, a total of C user groups (the number of user groups is equal to the number of subcarrier channels) are defined, and the number of users in each group is N (Σ N ═ N). The same subcarrier channel is used in the group, and different subcarrier channels are used by users among the groups;
s4, after the frequency spectrum resource allocation is completed, a power control algorithm based on a Stackelberg model is provided;
s401, defining the interference threshold of MBS in any sub-carrier frequency band as Q, and the interference of any FUE to MBS as Ii；
Ii＝gisi
S402, because the user groups adopting different subcarriers adopt an OFDMA access mode, no spectrum interference exists. Therefore, the power control algorithm is mainly set for users in a group that employ spectrum sharing.
S403, defining the interference gain of MBS to the ith user group as
Figure BDA0002174892960000123
Figure BDA0002174892960000124
S404, determining the optimized objective function of the MBS interference gain as
Figure BDA0002174892960000125
The constraint is C1;
Figure BDA0002174892960000126
s405, defining the signal-to-noise-and-interference ratio (SINR) of the ith FUE as SNIRi(si,s-i)；
Figure BDA0002174892960000127
After user grouping, the geometrical distance between users in the group is relatively far, and the co-channel interference between the users is small, so the SNIRi(si,s-i) Can be approximately calculated as:
Figure BDA0002174892960000131
s406, defining the rate of the ith FUE as
Figure BDA0002174892960000132
Figure BDA0002174892960000133
Wherein λ isiRepresenting the rate gain of the ith FUE.
S407, determining an optimized objective function for the uplink rate of FUE as
Figure BDA0002174892960000134
S408, thus, the optimization problem in S407 can be rewritten as maxf (S)i)＝maxλlog(1+SNIRi(si,s-i))-piIi；
S409, pair f (S)i) Obtaining a first derivative to obtain
Figure BDA0002174892960000135
S410, pair f (S)i) Obtaining a second derivative
Figure BDA0002174892960000136
S411 and S410 show that the optimization problem in S408 is convex optimization, so the optimal solution of the problem must meet the KKT limitation, and the optimal transmission power of the ith FUE is obtained by using the standard optimization technology and the KKT limitation condition
Figure BDA0002174892960000137
S412, the optimization objective function of the sum of the overall uplink rates of all FUEs is
Figure BDA0002174892960000138
With the constraint of C1;
Figure BDA0002174892960000139
s413, by solving the optimization problem in S412, the pricing p of the MBS to the ith FUE uplink interference power can be obtainediAnd can further be calculated
Figure BDA00021748929600001310
The transmit power s of each FUE can be obtainediAnd thus a solution to the optimization problem in S404 can be obtained. And finally obtaining the interference gain of the MBS and the optimal solution of the overall link rate of the FUE to reach the game equilibrium point.
A cell macro base station and a plurality of home micro base stations form a Femtocell heterogeneous network, and channel measurement activities are carried out in an actual Femtocell network deployment scene; then, based on the obtained channel model, a user clustering and grouping algorithm is provided, and the problem of reasonably and effectively performing spectrum resource allocation under the scene that spectrum resources are limited is solved; and then, a power control algorithm under a multi-carrier sub-channel is given based on the Stackelberg game theory and a user grouping result, the same-layer and cross-layer interference between base stations is reduced, and the network throughput of user equipment and the interference gain of a macro base station are maximized. The invention solves the challenging problem of providing the cell coverage heterogeneous access network with high network throughput by using the Macro Base Station (MBS) and the Femtocell Base Station (FBS), and the provided joint resource allocation method provides a new technical scheme for the large-scale commercial deployment of the Femtocell network.
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. The components of the embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, considering a typical cellular environment of a cell, an MBS is deployed on the roof of a central floor, and in the coverage area of the MBS, due to environmental factors such as building shading, several FBSs need to be deployed locally to enhance signal coverage. Without loss of generality, assuming that each FBS serves one FUE, the FBS operates in a private mode, i.e., the MUE cannot access the wireless network provided by the FBS. Assuming that there are a lot of building interference barriers in the cell, and the like, the FBS needs to be densely deployed, and if an access network is constructed using a spectrum sharing mode, severe co-channel interference is caused, and spectrum resources need to be spread. Secondly, due to the limitation of spectrum resources, part of the FBSs still work in the same frequency band, and the part of the FBSs not only have the same-layer interference with each other, but also have the cross-layer interference with the MBS, so that power control is required. While the channel gain of the communication link between either the user or the base station has a huge impact on the spectrum resources and power control.
Based on the system model and analysis, the patent provides a new joint resource allocation scheme aiming at the difficult problem of the Femtocell network. The scheme mainly comprises three parts:
(1) establishing a Femtocell network system model and performance indexes;
(2) measuring and modeling a Femtocell network deployment scene channel;
(3) a user clustering and grouping algorithm based on a path loss model;
(4) and a multi-carrier sub-channel power control algorithm based on the Stackelberg game model.
The invention has the following implementation steps:
step 1, establishing a Femtocell network system model and performance indexes.
(1a) Assuming that the Femtocell network system model and the performance index analysis only consider the uplink;
(1b) in a typical cell network, the number of MBS is 1, the number of FBSs is N, and without loss of generality, assuming that each FBS serves one home user equipment (FUE), the number of FUE is also set to N;
(1c) because the spectrum resources are limited, the frequency spectrum bandwidth available for the network is M, the number of subcarrier channels is C, and the bandwidth of each subcarrier channel is B ═ M/C;
(1d) due to the limited transmission power, the transmission power of FUE i is siIts uplink channel gain to MBS is giUplink gain to FBSj is hi,j. When i is j, hi,jRepresents the channel gain between FUE and the private FBS; when i ≠ j, hi,jRepresents the same layer interference of FUE i to the uplink of FUE j;
(1e) in the Stackelberg game model, MBS is the leader of the game and FUE is the participant of the game. MBS pricing the transmitting power of FUE, and assuming that the price of the interference power from FUE i to MBS is piThe larger the interference power of the FUE to the FBS is, the higher the interference price is;
(1f) calculating link capabilities of FUE
Figure BDA0002174892960000161
Wherein σ2Representing gaussian white noise power. Then, the overall throughput of all FUEs
Figure BDA0002174892960000162
(1g) Calculating interference benefits for MBS
Figure BDA0002174892960000163
And 2, acquiring a channel model under the Femtocell network through actual channel measurement, wherein relevant pictures of the channel measurement are shown in the attached figures 2 and 3.
As shown in fig. 2, it is a set of multi-band path loss channel measurement system developed by northwest university of industry. a) The transmitting terminal consists of a radio frequency antenna, a power amplifier and a vector signal generator, the transmitting antenna is arranged on the top layer of a cell center building with the height of 60 meters and is hung high by an iron tower to simulate a center MBS; b) the receiving end is composed of a radio frequency antenna, a frequency spectrograph, a power supply and a notebook computer, the receiving antenna is fixed on the trolley at the height of 1.2m and used for simulating the height of a hand-held mobile phone of a person, and the receiving equipment is arranged in the space in the cell and used for simulating indoor FUE or FBS. The receiving end is in an apartment opposite to the transmitting end, and the height variation range of the receiving end is 5m-50m variation.
This is a satellite map and a local detail map of the actual channel measurements, as shown in fig. 3.
a) In the satellite diagram of the actual test, it can be seen that the transmitting end is arranged on the roof of a building and the receiving end is arranged in an opposite apartment, which is a typical environment suitable for deploying a Femtocell network;
b) c) d) are respectively an indoor room, an indoor elevator car and an indoor corridor;
e) is a schematic plan structure diagram of an indoor room, and points in the diagram are points actually tested.
Fig. 4 shows a graph of path loss versus distance obtained from practical tests, wherein three commonly used path loss models, WINNER II, WINNER + and 3GPP R38.900, are selected. As can be seen from the figure, the results of the three channel models are greatly different from the actual test results. On one hand, the three standards are dominant in European and American countries, the scene based on actual measurement is a building in the European and American countries, the indoor space of the building in the European and American countries is generally wide, and the attenuation effect of the barriers such as walls on signals is weaker than that of the building in the China, which has a relatively narrow indoor space; on the other hand, the channel model is specific to a specific environment, and under the condition that the condition allows, the actual environment should be subjected to channel modeling before network planning and base station deployment are carried out. Therefore, the existing model needs to be modified, and a modification factor in the actual environment is added to make the model more conform to the actual channel environment.
FIG. 5 shows the channel model after applying correction factors to WINNER +, as given in the following equation
PL＝A·log10(d3D)+B·log10(fc)+C(1-cosθ)2+0.5din+Ψ
Wherein A, B, C represent parameters of each influencing factor, d3DIndicating the distance of signal transmission, fcDenotes the center frequency of the carrier, θ denotes the pitch angle of the signal transmission between MBS to FBS or FUE, dinDenotes the distance of the FBS or FUE from the wall, Ψ denotes shadow fading, and is a variable conforming to the log-normal distribution. FIG. 6 is a modeling of the shadow fading Ψ, and a model of Ψ can be obtained based on measured data and the Kolmogorov-Schmilnough test (K-S). It can be seen from the figure that the variable conforms to a lognormal distribution.
And 3, after a path loss channel model of the Femtocell network is established, designing a spectrum resource allocation algorithm for user clustering and grouping based on the model.
(3a) Since the coverage area of the FBS is limited, the FUE activity range served by the FBS is also limited, and thus the FUE position can be regarded as fixed when users are clustered and grouped. Assuming that there are a total of K user clusters,
Figure BDA0002174892960000171
wherein each userThe number of users in a cluster is L,
Figure BDA0002174892960000172
the total number of users is N;
(3b) each user cluster has a cluster center, and the geometric coordinate of the cluster center of the kth user cluster is (c)k,x,ck,y) The calculation formula is as follows:
Figure BDA0002174892960000173
(3c) the geometric coordinate of the first user FUE in the k-th user cluster is randomly generated as (x)k,l,yk,l)；
(3d) Since the path loss value determines the interference power, the distance between fuei and fuej is defined as the "path loss" distance PLD by considering the path loss as a key factor when users clusteri,jThe calculation formula is as follows:
Figure BDA0002174892960000181
where PL represents a calculation formula of the path loss.
(3e) In order to optimize the clustering effect of users, the clustering cost function of users is assumed to be J (U, V), and the calculation formula is as follows:
Figure BDA0002174892960000182
and making U and V as the optimization results of K and L.
(3f) Determining an optimal objective function for the clustering algorithm as
Figure BDA0002174892960000183
The constraint conditions are as follows:
Figure BDA0002174892960000184
C2:0＜L≤Lmax
C3:Kmin≤K≤Kmax
wherein, in order to ensure load balance and QoS, the maximum value of L cannot exceed Lmax。
(3g) And after the users are clustered, grouping the users. The user grouping is to randomly extract users from the user cluster, and ensure that the extracted users come from different user clusters, and the users in each user group share the frequency spectrum resources. A total of C user groups are defined (the number of user groups equals to the number of subcarrier channels), and the number of users in each group is N (Σ N ═ N). The same sub-carrier channel is used within a group and different sub-carrier channels are used by users between groups.
Fig. 7 shows a schematic diagram of user clustering. It can be seen that users at a short distance are classified into a cluster, and because the distance between the users in the cluster is too short and the path loss of the communication link between the users is too small, the users in the cluster use different sub-channels. And after clustering, extracting users from each cluster to form a user group, wherein the members in the user group are from different user clusters. Compared with the random distribution of frequency spectrum resources, the clustering method can effectively reduce the interference of FUE to the uplink of the MBS, simultaneously reduce the overall power consumption of the system and meet the requirement of green communication.
The specific user clustering algorithm is described in the following table:
Figure BDA0002174892960000191
Figure BDA0002174892960000201
step 4, after the frequency spectrum resource allocation of the user is completed, a power control algorithm based on a Stackelberg game is constructed;
(4a) in order to guarantee the communication quality, the interference acceptable by the MBS is limited. The interference threshold of MBS in any sub-carrier frequency band is assumed to beQ, interference of any FUE on MBS is IiThe calculation formula is Ii＝gisi；
(4b) Because the OFDMA access mode is adopted among the user groups adopting different subcarriers, no spectrum interference exists. Therefore, the power control algorithm is mainly set for users in a group that employ spectrum sharing.
(4c) Defining the interference gain of MBS to the ith user group as
Figure BDA0002174892960000202
Is calculated by the formula
Figure BDA0002174892960000203
(4d) An optimized objective function for determining the interference gain of MBS is
Figure BDA0002174892960000204
The constraint conditions are as follows:
Figure BDA0002174892960000211
(4e) defining the signal-to-noise-and-interference ratio (SINR) of the ith FUE as SNIRi(si,s-i) The calculation formula is as follows:
Figure BDA0002174892960000212
after user grouping, the geometrical distance between users in the group is relatively far, and the co-channel interference between the users is small, so the SNIRi(si,s-i) Can be approximately calculated as:
Figure BDA0002174892960000213
(4f) define the ith FUE rate as
Figure BDA0002174892960000214
The calculation formula is as follows:
Figure BDA0002174892960000215
wherein λ isiRepresenting the rate gain of the ith FUE.
(4g) An optimization objective function to determine the uplink rate of FUE is
Figure BDA0002174892960000216
(4h) The optimization problem in (4g) above can be rewritten as
maxf(si)＝maxλlog(1+SNIRi(si,s-i))-piIi；
(4i) For f(s)i) Obtaining a first derivative to obtain
Figure BDA0002174892960000217
(4j) For f(s)i) Obtaining a second derivative
Figure BDA0002174892960000218
(4k) From (4j), the optimization problem in (4g) is a convex optimization, so the optimal solution of the problem must satisfy the KKT constraint, and by using the standard optimization technique and the KKT constraint, the optimal transmit power of the ith FUE is obtained as follows:
Figure BDA0002174892960000221
(4l) the optimization objective function of the sum of the overall uplink rates of all FUEs, as available from (4k), is:
Figure BDA0002174892960000222
the constraint conditions are as follows:
Figure BDA0002174892960000223
(4m) by solving the optimization problem in (4l), the pricing p of the MBS to the ith FUE uplink interference power can be obtainediAnd can further be calculated
Figure BDA0002174892960000224
The transmit power s of each FUE can be obtainediAnd thus a solution to the optimization problem in (4d) can be obtained. And finally obtaining the interference gain of the MBS and the optimal solution of the overall link rate of the FUE to reach the game equilibrium point.
The specific joint resource allocation algorithm is described in the following table:
Figure BDA0002174892960000225
Figure BDA0002174892960000231
the effects of the invention can be further illustrated by simulations:
(1) simulation conditions
The simulation is realized by MATLAB software.
And evaluating the performance of the joint resource allocation algorithm in the Femtocell network according to the simulation result.
The network setup and simulation parameters are listed in table 3 below:
Figure BDA0002174892960000241
(2) simulation result
The simulation result is mainly divided into three parts. The first part is to verify the validity of the proposed power control algorithm, namely whether the MBS is in accordance with the expectation of the condition that the interference power price of the FUE is changed along with the MBS interference threshold; the second part, the influence of different channel models on the performance of the joint resource allocation algorithm is compared, wherein the influence comprises a measured fading channel model, a WINNER + fading channel model, a WINNER II fading channel model and 3GPP R38.900; and the third part is to compare the performance of the joint resource allocation algorithm provided by the patent with that of other resource allocation algorithms, wherein the performance comprises two modes that the spectrum resources are randomly allocated by sub-channels and are distributed by single channels, and the power control is realized by the algorithm of the patent.
Referring to fig. 8, in the sparse deployment scenario, the interference prices of different FUEs vary with the interference threshold Q. Since FUE1 is closest to MBS, the interference price of FUE1 is the highest. FUE, however, to acquire higher speeds, games between the interference price and the transmission power, and when the equilibrium point of the game is reached, the overall optimal solution is obtained. It can also be seen that, as Q increases, the interference price of FUE is continuously reduced, because Q is relatively large, the total amount of interference that can be accepted by MBS is relatively large, in order to satisfy "supply-demand balance", the interference price of FUE needs to be properly reduced, and when the game equilibrium point is reached again, the overall benefit is the greatest.
Referring to fig. 9 and fig. 10, the impact of different fading channel models on the performance of the resource allocation algorithm, including the interference gain of MBS and the overall rate of FUE, is shown. As can be seen from fig. 9, the MBS gain differences in different channel models are huge, which indicates that it is crucial to improve performance to adopt a specific channel model for a specific environment. If an inappropriate channel model is adopted in a Femtocell network, not only communication resources are wasted and the cost is increased, but also the user experience and the maintainability are greatly influenced. The phenomenon shown in fig. 10 is substantially identical to that in fig. 9.
Referring to fig. 11 and 12, the performance of the joint resource allocation algorithm of the present invention is shown in comparison with other resource allocation algorithms. Two comparison groups are arranged in the two graphs, and in the spectrum resource allocation strategy, a first algorithm adopts a subchannel random allocation mode, and a second algorithm adopts a spectrum sharing mode, so that the spectrum resource bandwidths of the two algorithms are ensured to be the same; in terms of power control strategies, both adopt the resource allocation strategy given in the patent. It can be seen that the joint resource allocation algorithm provided by the patent is obviously superior to other algorithms in two performance indexes, namely, the interference gain of the MBS and the overall rate of the FUE. It is noted that the performance of the random subchannel allocation mode is better than that of the spectrum sharing mode, which indicates that in the Femtocell network, the utilization rate and the use effect of the spectrum resources can be significantly improved by using a plurality of subchannels.
Meanwhile, it can be seen that, as the interference threshold Q of the MBS is larger and larger, the interference gain of the MBS and the overall rate of the FUE are both continuously increased. This is because the interference price decreases with increasing Q, but for obtaining a larger link rate, FUE will increase the transmit power appropriately, thereby counteracting the negative impact of the decrease in interference price on MBS interference gain. Finally, MBS and FUE reach a new game balance point in the Femtocell network, so that the overall benefit is dynamically changed and kept optimal.
The above-mentioned contents are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical idea of the present invention falls within the protection scope of the claims of the present invention.

Claims (10)

1. The joint resource allocation method based on the user clustering and the Starkelberg model is characterized by constructing a Femtocell double-layer heterogeneous network formed by mixing a macro base station MBS deployed at the top of a central building and a plurality of Femtocell base stations FBS deployed in various residences; establishing a path loss model according to the Femtocell double-layer heterogeneous network; then, a distance measurement method based on path loss is provided based on a channel model, a clustering algorithm is adopted to cluster the FBS which is densely deployed in the cell, users are grouped according to a clustering result, different subcarrier channels are distributed to different user groups, and the FBS which is densely deployed is converted into a FBS scene which is sparsely deployed; the method comprises the steps that a Starkelberg model for controlling power between an MBS and a household user equipment FUE is built according to the Starkelberg model, the MBS is a game leader and pricing interference power from the FUE, the FUE is an interference pricing given by game participants to the MBS and dynamically adjusting own transmitting power, when the game reaches a balanced position, interference benefits of the MBS and the overall link rate of the FUE reach an optimal solution, and joint resource allocation is realized by maximizing the utilization rate of spectrum resources and maximizing the throughput of a network.
2. The method of claim 1, wherein the building of the Femtocell dual-layer heterogeneous network formed by mixing MBS and several FBSs specifically comprises:
s101, only considering an uplink in the Femtocell network joint resource allocation;
s102, defining the number of MBS in the network as 1 and the number of FBS as N, and setting the number of FUE as N if each FBS serves a home user equipment;
s103, defining a frequency spectrum bandwidth available for the network to be M, and the number of subcarrier channels to be C, where the bandwidth of each subcarrier channel is B ═ M/C;
s104, defining the transmitting power of FUEi as SiChannel gain to MBS uplink cross-layer interference is giUplink gain to FBSj is hi,j(ii) a When i is j, hi,jIndicating the channel gain between the FUE and its dedicated FBS; when i ≠ j, hi,jRepresents the same layer interference of FUE i to the uplink of FUE j;
s105, defining the price of the interference power from FUEi to MBS to be pi；
S106, calculating link capacity of FUE
Figure FDA0003364109070000011
And overall throughput V of all FUEsFUE；
S107, calculating MInterference yield V of BSMBS。
3. The method of claim 2, wherein in step S106,
Figure FDA0003364109070000021
the calculation formula is as follows:
Figure FDA0003364109070000022
wherein σ2Representing a gaussian white noise power;
VFUEthe calculation formula is as follows:
Figure FDA0003364109070000023
4. the method of claim 2, wherein V107 is a step of assigning resources based on the user clustering and the Starkelberg modelMBSThe calculation formula is as follows:
Figure FDA0003364109070000024
5. the joint resource allocation method based on user clustering and Starkelberg model according to claim 1, wherein the channel model PL is specifically:
PL＝A·log10(d3D)+B·log10(fc)+C(1-cosθ)2+0.5din+Ψ
wherein A, B, C represent parameters of each influencing factor, d3DIndicating the distance of signal transmission, fcDenotes the center frequency of the carrier, θ denotes MBS to FBS or FUPitch angle of signal transmission between E, dinDenotes the distance of the FBS or FUE from the wall, Ψ denotes shadow fading, and is a variable conforming to the log-normal distribution.
6. The joint resource allocation method based on user clustering and the Starkelberg model according to claim 1, wherein grouping users by using a clustering algorithm specifically comprises:
s301, defining K user clusters,
Figure FDA0003364109070000036
the number of users in each user cluster is L,
Figure FDA0003364109070000035
s302, defining the geometric coordinate of the cluster center of the kth user cluster as (c)k,x,ck,y)
Figure FDA0003364109070000031
S303, defining the geometric coordinate of the first user FUE in the kth user cluster as (x)k,l,yk,l)；
S304, defining the distance between FUE i and FUE j as path loss distance PLDi,j；
S305, defining a cost function of user clustering as J (U, V);
Figure FDA0003364109070000032
wherein, let U*And V*The optimization results are K and L;
s306, determining the optimized objective function of the clustering algorithm as
Figure FDA0003364109070000033
S307, C user groups are defined in total, the number of users in each group is N, and sigma N is N; the same sub-carrier channel is used within a group and different sub-carrier channels are used by users between groups.
7. The method of claim 6, wherein in step S304, the PLD is a distance of Path Loss (PLD)i,jThe method specifically comprises the following steps:
Figure FDA0003364109070000034
wherein PL represents a calculation formula of path loss, specifically:
PL＝A·log10(d3D)+B·log10(fc)+C(1-cosθ)2+0.5din+Ψ
wherein A, B, C represent parameters of each influencing factor, d3DIndicating the distance of signal transmission, fcDenotes the center frequency of the carrier, θ denotes the pitch angle of the signal transmission between MBS to FBS or FUE, dinIndicating the distance of the FBS or FUE to the wall, and Ψ indicates a shadow fade.
8. The method of claim 6, wherein the constraint conditions in step S306 are:
Figure FDA0003364109070000045
C2:0＜L≤Lmax
C3:Kmin≤K≤Kmax
wherein, in order to ensure load balance and QoS, the maximum value of L cannot exceed Lmax。
9. The joint resource allocation method based on the user clustering and the starkeberg model according to claim 1, wherein a game model for power control between MBS and FUE is constructed according to a Stackelberg game model, and specifically:
s401, defining the interference threshold of MBS in any sub-carrier frequency band as Q, and the interference of any FUE to MBS as Ii；
Ii＝gisi；
S402, OFDMA access mode is adopted among user groups adopting different subcarriers, and a power control algorithm is set for users in the group adopting spectrum sharing;
s403, defining the interference gain of MBS to the ith user group as
Figure FDA0003364109070000041
Comprises the following steps:
Figure FDA0003364109070000042
s404, determining the optimized objective function of the MBS interference gain as
Figure FDA0003364109070000043
The constraint is that C1 is:
Figure FDA0003364109070000044
s405, defining signal noise interference ratio SNIR of ith FUEi(si,s-i) Comprises the following steps:
Figure FDA0003364109070000051
s406, defining the rate of the ith FUE as
Figure FDA0003364109070000052
Comprises the following steps:
Figure FDA0003364109070000053
wherein λ isiRepresents the rate gain of the ith FUE;
s407, determining an optimized objective function for the uplink rate of FUE as
Figure FDA0003364109070000054
S408, rewriting the optimization problem in the step S407 into maxf (S)i)＝maxλlog(1+SNIRi(si,s-i))-piIi；
S409, for f (S) in step S408i) Solving a first derivative to obtain:
Figure FDA0003364109070000055
s410, for f (S) in step S408i) And solving a second derivative to obtain:
Figure FDA0003364109070000056
s411, obtaining the optimal transmitting power of the ith FUE as follows:
Figure FDA0003364109070000057
s412, defining an optimization objective function of the sum of all FUE overall uplink rates;
s413, solving the optimization problem in S412 to obtain the pricing p of the MBS to the ith FUE uplink interference poweriBy calculating
Figure FDA0003364109070000058
Obtain the transmission power s of each FUEiTo obtainAnd step S404, solving the optimization problem, and finally obtaining the optimal solution of the interference gain of the MBS and the overall link rate of the FUE to reach the game equilibrium point.
10. The method of claim 9, wherein the optimization objective function of step S412 is:
Figure FDA0003364109070000061
the constraint condition C1 is:
Figure FDA0003364109070000062
CN201910775498.1A 2019-08-21 2019-08-21 Joint resource allocation method based on user clustering and Starkelberg model Active CN110691360B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910775498.1A CN110691360B (en) 2019-08-21 2019-08-21 Joint resource allocation method based on user clustering and Starkelberg model

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910775498.1A CN110691360B (en) 2019-08-21 2019-08-21 Joint resource allocation method based on user clustering and Starkelberg model

Publications (2)

Publication Number Publication Date
CN110691360A CN110691360A (en) 2020-01-14
CN110691360B true CN110691360B (en) 2022-03-22

Family

ID=69108574

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910775498.1A Active CN110691360B (en) 2019-08-21 2019-08-21 Joint resource allocation method based on user clustering and Starkelberg model

Country Status (1)

Country Link
CN (1) CN110691360B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111682915B (en) * 2020-05-29 2021-10-08 北京交通大学 Self-allocation method for frequency spectrum resources
CN111726192B (en) * 2020-06-12 2021-10-26 南京航空航天大学 Communication countermeasure medium frequency decision optimization method based on log linear algorithm
CN113411757B (en) * 2021-06-23 2022-05-06 西北工业大学 User grouping design method and system in communication and computing integrated cognitive cloud network

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103856996A (en) * 2014-02-12 2014-06-11 南京邮电大学 Power control-access control combined method
CN105848296A (en) * 2016-06-01 2016-08-10 南京邮电大学 Stackelberg game-based resource allocation method
CN107567034A (en) * 2017-08-07 2018-01-09 吉林大学 A kind of D2D resource share methods in isomery cellular network
CN109618351A (en) * 2019-01-09 2019-04-12 南京邮电大学 Resource allocation methods in heterogeneous network based on stackelberg game

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103856996A (en) * 2014-02-12 2014-06-11 南京邮电大学 Power control-access control combined method
CN105848296A (en) * 2016-06-01 2016-08-10 南京邮电大学 Stackelberg game-based resource allocation method
CN107567034A (en) * 2017-08-07 2018-01-09 吉林大学 A kind of D2D resource share methods in isomery cellular network
CN109618351A (en) * 2019-01-09 2019-04-12 南京邮电大学 Resource allocation methods in heterogeneous network based on stackelberg game

Also Published As

Publication number Publication date
CN110691360A (en) 2020-01-14

Similar Documents

Publication Publication Date Title
Yang et al. Cooperation for spectral and energy efficiency in ultra-dense small cell networks
Sun et al. D2D enhanced heterogeneous cellular networks with dynamic TDD
CN110691360B (en) Joint resource allocation method based on user clustering and Starkelberg model
Giupponi et al. Distributed interference control in OFDMA-based femtocells
US8369863B2 (en) Method for interference suppression
CN103856947B (en) The disturbance coordination method that a kind of combined channel selects and power controls
CN101977386A (en) Interference coordination method of femtocell grouping in cellular network
CN103338455B (en) Communication resource allocation method, system, terminal and network side equipment
CN107708157A (en) Intensive small cell network resource allocation methods based on efficiency
Yu et al. Dynamic resource allocation in TDD-based heterogeneous cloud radio access networks
Zeng et al. System performance of self-organizing network algorithm in WiMAX femtocells
CN105490794B (en) The packet-based resource allocation methods of the Femto cell OFDMA double-layer network
Lan et al. Resource allocation and performance study for LTE networks integrated with femtocells
Adejo et al. Interference modelling for soft frequency reuse in irregular heterogeneous cellular networks
CN105764068B (en) A kind of small Base Station Compatibility and coverage optimization method based on TABU search
Taranetz et al. Sensitivity of OFDMA-based macrocellular LTE networks to femtocell deployment density and isolation
Sousa et al. Impact of propagation model on capacity in small-cell networks
Ye et al. Hybrid-clustering game Algorithm for resource allocation in macro-femto hetnet
Li et al. Distributed power control for two-tier femtocell networks with QoS provisioning based on Q-learning
Chen et al. Clustering-based co-tier interference coordination in dense small cell networks
Sung et al. Is multicell interference coordination worthwhile in indoor wireless broadband systems?
WO2019062822A1 (en) Resource allocation method and server
CN106102151A (en) The interference management method controlled based on channel distribution and power in family base station system
Mahonen et al. From protocol stack to technology circle: exploring regulation, efficiency metrics, and the high-dimensional design space of wireless systems
Guillet et al. Downlink femto-macro icic with location-based long-term power setting

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant