CN111556518B - Resource allocation method and system for improving network quality in multi-slice network - Google Patents

Resource allocation method and system for improving network quality in multi-slice network Download PDF

Info

Publication number
CN111556518B
CN111556518B CN202010533456.XA CN202010533456A CN111556518B CN 111556518 B CN111556518 B CN 111556518B CN 202010533456 A CN202010533456 A CN 202010533456A CN 111556518 B CN111556518 B CN 111556518B
Authority
CN
China
Prior art keywords
slice
resource allocation
tenant
network
subcarriers
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
CN202010533456.XA
Other languages
Chinese (zh)
Other versions
CN111556518A (en
Inventor
江璟
王玉东
辛培哲
马腾滕
袁思雨
张勇
郑伟军
邵炜平
王文华
陈鼎
方景辉
吴国庆
唐锦江
杨鸿珍
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.)
State Grid Economic And Technological Research Institute Co LtdB412 State Grid Office
State Grid Corp of China SGCC
State Grid Zhejiang Electric Power Co Ltd
Jiaxing Power Supply Co of State Grid Zhejiang Electric Power Co Ltd
Original Assignee
State Grid Economic And Technological Research Institute Co LtdB412 State Grid Office
State Grid Corp of China SGCC
State Grid Zhejiang Electric Power Co Ltd
Jiaxing Power Supply Co of State Grid Zhejiang Electric Power Co Ltd
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 State Grid Economic And Technological Research Institute Co LtdB412 State Grid Office, State Grid Corp of China SGCC, State Grid Zhejiang Electric Power Co Ltd, Jiaxing Power Supply Co of State Grid Zhejiang Electric Power Co Ltd filed Critical State Grid Economic And Technological Research Institute Co LtdB412 State Grid Office
Priority to CN202010533456.XA priority Critical patent/CN111556518B/en
Publication of CN111556518A publication Critical patent/CN111556518A/en
Application granted granted Critical
Publication of CN111556518B publication Critical patent/CN111556518B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/06—Testing, supervising or monitoring using simulated traffic
    • 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
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Quality & Reliability (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

The invention relates to a resource allocation method and a system for improving network quality in a multi-slice network, which are characterized by comprising the following contents: 1) slicing the heterogeneous network to be tested, and establishing a resource allocation model of each slice; 2) obtaining approximate convex functions of each resource allocation model; 3) under the condition that tenants corresponding to the slices equally divide total subcarriers, solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm, and determining an optimal slice resource allocation strategy and a scoring standard of each slice; 4) under the condition that tenants corresponding to the slices do not equally divide total subcarriers, a non-cooperative game method is adopted, the approximate convex objective function of the slice resource allocation model is solved according to the number of subcarriers of the slices, and the optimal slice resource allocation strategy and the scoring standard of the slices are determined.

Description

Resource allocation method and system for improving network quality in multi-slice network
Technical Field
The invention relates to a resource allocation method and a resource allocation system for improving network quality in a multi-slice network, and belongs to the technical field of network quality evaluation.
Background
The evaluation of the comprehensive index of the communication network mainly comprises two aspects of network overall utility quality and network comprehensive performance grading, the evaluation index of the network overall utility quality reflects the utility value of the current network from the perspective of the network quality, and the complex work of checking indexes one by one is avoided; the network comprehensive performance grading integrates various network indexes through normalization, presents various network performances, and can accelerate positioning of specific KPI (key performance index) influencing network quality.
At present, in the prior art, a certain amount of research has been conducted on performance evaluation of a network, but the utility function considered is limited to a single aspect such as frequency efficiency, energy efficiency and throughput, and is generally evaluated for a network with a single network characteristic. The fifth generation mobile communication mode adopts a wireless virtualization network, slices of the types including eMBB (enhanced mobile broadband), URLLC (ultra-high reliable ultra-low time delay communication), mMTC (mass machine type communication) and the like can be accommodated in a slicing mode, the network protocol, the design target and the service characteristic of each slice are not identical, and the evaluation cannot be comprehensively and clearly carried out by adopting a single-index evaluation method. The QoS (quality of service assurance) evaluation system is widely used for network quality evaluation, QoS indexes mainly comprise network throughput, time delay, packet loss rate, disconnection rate, success rate and the like, the indexes can reflect network conditions and network quality, specific KPIs are refined according to different functions and serve as quantitative embodiment of overall network performance, and the KPIs can effectively reflect current network conditions. The first-level index comprises network capacity, network quality and the like, the second-level index is refinement of the first-level index, wherein the network capacity comprises indexes such as resource utilization rate, system throughput and the number of concurrent users, and the network quality comprises indexes such as building capacity, maintaining capacity and mobility capacity. Commonly used evaluation algorithms include a standard deviation method (SD), a gray scale correlation analysis (GRA) and the like, and a comprehensive evaluation index value of the evaluated network is calculated by calculating the overall utility value and the comprehensive performance score of the network.
As described above, since the network quality evaluation depends on the network rate, the time delay, and other indicators, the network quality evaluation is an important evaluation criterion after the network is constructed. In the fifth generation mobile communication system, different slices provide services for different network application scenarios, and have different network evaluation indexes. As an infrastructure operator, the quality of the entire infrastructure network needs to be evaluated, and under the limited network frequency and power limitation, it is necessary to improve the network quality score. With the continuous development of the mobile internet and the continuous increase of the network demand, the development of the wireless network is limited by network resources such as computing capacity, bandwidth, cache and the like at present, and the effective resource allocation method can enable the wireless resources to be utilized more effectively, so that the resource utilization rate is improved, and the network quality is improved. However, there is no resource allocation method that can effectively utilize radio resources in the related art.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide a resource allocation method and system for improving network quality in a multi-slice network, which can make efficient use of radio resources.
In order to realize the purpose, the invention adopts the following technical scheme: the resource allocation method for improving the network quality in the multi-slice network comprises the following steps: 1) slicing the heterogeneous network to be tested, modeling aiming at the resource distribution problem influencing the quality of each sliced network, and establishing a resource distribution model of each slice; 2) processing the non-convex function in each resource allocation model to obtain an approximate convex function of each resource allocation model; 3) under the condition that tenants corresponding to the slices equally divide total subcarriers, solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm, and determining an optimal slice resource allocation strategy and a scoring standard of each slice; 4) under the condition that the tenants corresponding to the slices do not equally divide the total subcarriers, solving the approximate convex objective function of the resource allocation model of each slice by adopting a non-cooperative game method according to the number of the subcarriers of each slice, and determining the optimal slice resource allocation strategy and scoring standard of each slice.
Further, the specific process of the step 1) is as follows: 1.1) slicing the heterogeneous network to be tested into rate priority slices and delay priority slices; 1.2) setting resources of a tenant 1 for regulating and controlling rate priority slices, setting resources of a tenant 2 for regulating and controlling delay priority slices, and aiming at the problem of resource allocation influencing the network quality of each slice, establishing a resource allocation model of the tenant corresponding to each slice by taking the scores of network services of the two tenants as a target.
Further, the construction process of the resource allocation model in the step 1.2) is as follows:
the participants: tenant 1 and tenant 2;
strategy II: the optimal strategy of each tenant is the obtained optimal resource allocation scheme;
the corresponding policies of the tenants 1 and 2 are as follows:
Strategy1＝(p1,a1)
Strategy2＝(p2,a2)
wherein Strategy1 and Strategy2 are policies of tenant 1 and tenant 2, respectively; p is a radical of1,p2Power vectors allocated to corresponding users for tenant 1 and tenant 2, respectively:
Figure GDA0003649843940000021
Figure GDA0003649843940000022
wherein, the first and the second end of the pipe are connected with each other,
Figure GDA0003649843940000023
and
Figure GDA0003649843940000024
respectively for subcarriers n of base station j in regulation rate priority slice1And subcarrier n in time delay priority slice2A transmission power of; k1A set of users that are associated rate-first slices; k is2A set of users for associated latency-first slices;
a1and a2The number of subcarriers allocated to tenant 1 and tenant 2, respectively:
Figure GDA0003649843940000031
utility function: and the service scoring standard corresponding to each tenant, wherein,service scoring standard T of tenant 11Comprises the following steps:
T1＝k1r1+grade1
wherein k is1A scoring coefficient per unit rate in a rate-first slice; grade1Scoring the other indicators; parameter r1Comprises the following steps:
Figure GDA0003649843940000032
wherein the content of the first and second substances,
Figure GDA0003649843940000033
associating user k for base station jjThe channel of (2);
Figure GDA0003649843940000034
associating user k for base station jjThe channel of (2);
Figure GDA0003649843940000035
subcarrier n in regulatory rate priority slice for base station j1A transmission power of;
Figure GDA0003649843940000036
and
Figure GDA0003649843940000037
whether there is a user associated subcarrier n in base stations j and j', respectively1；δ2Is additive white gaussian noise power;
service scoring standard T of tenant 22Comprises the following steps:
T2＝k2r2+grade2
wherein k is2A score coefficient for each unit rate in the time delay priority slice; grade2Scoring other indicators; parameter r2Comprises the following steps:
Figure GDA0003649843940000038
wherein the content of the first and second substances,
Figure GDA0003649843940000039
associating user l with base station jjThe channel of (2);
Figure GDA00036498439400000310
associating user l for base station jjThe channel of (2);
Figure GDA00036498439400000311
for sub-carrier n in time delay priority slice for base station j2A transmission power of;
Figure GDA00036498439400000312
and
Figure GDA00036498439400000313
whether there is a user associated subcarrier n in base stations j and j', respectively2；
The resource allocation model of the tenant 1 is as follows:
maxk1r1+grade1
Figure GDA00036498439400000314
Figure GDA00036498439400000315
Figure GDA00036498439400000316
wherein σ0For user kjA minimum rate requirement threshold of;
Figure GDA0003649843940000041
is a non-negative real number; p1Total power for rate-first slices; parameter(s)
Figure GDA0003649843940000042
Comprises the following steps:
Figure GDA0003649843940000043
and the resource allocation model of the tenant 2 is as follows:
maxk2r2+grade2
Figure GDA0003649843940000044
Figure GDA0003649843940000045
Figure GDA0003649843940000046
wherein ε is satisfying user ljThe interruption probability of the minimum delay requirement;
Figure GDA0003649843940000047
for user ljA packet arrival rate;
Figure GDA0003649843940000048
the maximum time delay that can be tolerated by the user; p is2Total power for delay-first slicing; parameter(s)
Figure GDA0003649843940000049
Comprises the following steps:
Figure GDA00036498439400000410
further, the specific process of step 2) is as follows:
2.1) resource assignment to tenant 1 and tenant 2Adding corresponding constraints to a configuration model
Figure GDA00036498439400000411
And
Figure GDA00036498439400000412
Figure GDA00036498439400000413
2.2) adding corresponding penalty functions to the objective functions of the slice resource allocation models:
Figure GDA00036498439400000414
Figure GDA00036498439400000415
wherein, omega is a parameter tending to zero, and the parameter q belongs to (0, 1);
2.3) the approximate convex objective functions of the two slice resource allocation models are respectively as follows:
Figure GDA0003649843940000051
Figure GDA0003649843940000052
wherein:
Figure GDA0003649843940000053
Figure GDA0003649843940000054
in the formula, parameter
Figure GDA0003649843940000055
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n1An initial value of transmission power; parameter(s)
Figure GDA0003649843940000056
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n2To the initial value of the transmission power.
Further, the specific process of step 3) is as follows:
resource allocation calculation for rate-first slices:
a) set of given rate-first sliced users K1Number of subcarriers a1Channel gain
Figure GDA0003649843940000057
Noise power delta2The iteration number i is 0, the convergence precision eta of the given algorithm and a scoring coefficient k1Other scoring grades1Initial allocated power
Figure GDA0003649843940000058
b) According to the initial value, a matlab convex optimization packet is adopted to solve a convex optimization problem to obtain optimized power
Figure GDA0003649843940000059
c) According to the optimized power
Figure GDA0003649843940000061
Calculating the value of the objective function
Figure GDA0003649843940000062
d) If the value of the objective function
Figure GDA0003649843940000063
And (3) satisfying the constraint:
Figure GDA0003649843940000064
then outputting the corresponding optimal slice resource allocation strategy
Figure GDA0003649843940000065
And a scoring criterion; otherwise, let i be i +1,
Figure GDA0003649843940000066
entering into the step b);
resource allocation calculation for the latency-first slice:
A) set of given latency-first slice users K2Number of subcarriers a2Channel gain
Figure GDA0003649843940000067
Noise power delta2The iteration number i is 0, the convergence precision eta of the algorithm and a scoring coefficient k2Other scoring grades2Initial allocated power
Figure GDA0003649843940000068
B) According to the initial value, a matlab convex optimization packet is adopted to solve a convex optimization problem to obtain optimized power
Figure GDA0003649843940000069
C) According to the optimized power
Figure GDA00036498439400000610
Calculating the value of the objective function
Figure GDA00036498439400000611
D) If the value of the objective function
Figure GDA00036498439400000612
And (3) satisfying the constraint:
Figure GDA00036498439400000613
then outputting the corresponding optimal slice resource allocation strategy
Figure GDA00036498439400000614
And a scoring criterion; otherwise, let i be i +1,
Figure GDA00036498439400000615
Figure GDA00036498439400000616
entering the step B).
Further, the specific process of the step 4) is as follows:
4.1) user set K corresponding to given rate priority slice and time delay priority slice1And K2The convergence accuracy eta of the algorithm, and the initialization of the parameter channel gain in the heterogeneous network to be tested
Figure GDA00036498439400000617
And
Figure GDA00036498439400000618
noise power delta2When the iteration number i is equal to 0, distributing power
Figure GDA00036498439400000619
And
Figure GDA00036498439400000620
4.2) respectively traversing and calculating the resource allocation strategy of the speed priority slice and the time delay priority slice of different subcarriers
Figure GDA00036498439400000621
And
Figure GDA00036498439400000622
and corresponding scoring criteria
Figure GDA00036498439400000623
And
Figure GDA00036498439400000624
wherein the content of the first and second substances,
Figure GDA00036498439400000625
indicates that the number of subcarriers allocated to the rate-first slice is a1The optimal power vector of the time of day,
Figure GDA00036498439400000626
the number of subcarriers allocated to the delay priority slice is represented as a2An optimal power vector of time;
4.3) calculating the number a of the subcarriers of the tenant 1 according to the calculation result of the step 4.2)1 *And the number of subcarriers a of tenant 22 *：
Figure GDA0003649843940000071
Figure GDA0003649843940000072
a2 *＝a-a1 *
4.4) number of subcarriers a according to tenant 11 *And the number of subcarriers a of tenant 22 *And the calculation result of the step 4.2) is used for obtaining the optimal slice resource allocation strategy of the rate priority slice and the time delay priority slice
Figure GDA0003649843940000073
And
Figure GDA0003649843940000074
and scoring criteria
Figure GDA0003649843940000075
And
Figure GDA0003649843940000076
a resource allocation system for improving network quality in a multi-slice network, comprising: the resource allocation model building module is used for slicing the heterogeneous network to be tested, modeling aiming at the resource allocation problem influencing the quality of each slice network and building a resource allocation model of each slice; the approximate convex function determining module is used for processing the non-convex functions in the resource distribution models to obtain the approximate convex functions of the resource distribution models; the continuous convex approximation module is used for solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm under the condition that the total subcarriers are equally divided by tenants corresponding to each slice, and determining an optimal slice resource allocation strategy and a scoring standard of each slice; and the non-cooperative game module is used for solving the approximate convex objective function of the resource allocation model of each slice according to the number of the subcarriers of each slice by adopting a non-cooperative game method under the condition that the total subcarriers are not equally divided by the tenants corresponding to each slice, and determining the optimal slice resource allocation strategy and the scoring standard of each slice.
Further, the resource allocation model building module comprises: the heterogeneous network slicing unit is used for slicing the heterogeneous network to be tested into a rate priority slice and a time delay priority slice; and the model establishing unit is used for setting resources of the tenant 1 for regulating and controlling the rate priority slice, the tenant 2 for regulating and controlling the resources of the delay priority slice, and establishing a resource distribution model of the tenant corresponding to each slice by taking the scores of network services of the two tenants as a target aiming at the problem of resource distribution influencing the network quality of each slice.
A computer program comprising computer program instructions, wherein the computer program instructions, when executed by a processor, are adapted to implement the steps corresponding to the above-mentioned resource allocation method for improving network quality in a multi-slice network.
A computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, are configured to implement the steps corresponding to the above-mentioned resource allocation method for improving network quality in a multi-slice network.
Due to the adoption of the technical scheme, the invention has the following advantages:
1. the invention considers the problem that two types of tenants rent infrastructure operators in a fifth generation mobile communication system respectively carry out resource allocation under the speed priority and time delay priority scenes, and the two types of tenants compete wireless network resources, so that the invention adopts a continuous convex approximation algorithm and a non-cooperative game method to obtain the optimal slice resource allocation strategy of the two types of tenants corresponding to the heterogeneous network slices to be tested, and ensures the network quality balance of the two types of tenants, thereby meeting the practical requirements of the multi-tenant leased infrastructure network and enabling the wireless resources of the heterogeneous network to be tested to be effectively utilized.
2. The invention adopts the weighting scoring method widely used in the current mobile communication network to optimize the target by speed and time delay, and respectively models the resource allocation problem influencing the network quality of each slice of the heterogeneous network to be tested so as to solve the optimal slice resource allocation strategy of each slice of the heterogeneous network to be tested, improve the network quality of the multi-slice network, and can be widely applied to the technical field of network quality evaluation.
Drawings
Fig. 1 is a schematic diagram illustrating correspondence between tenants, users and a heterogeneous network;
FIG. 2 is a flow chart of the method of the present invention.
Detailed Description
The present invention is described in detail below with reference to the attached drawings. It is to be understood, however, that the drawings are provided solely for the purposes of promoting an understanding of the invention and that they are not to be construed as limiting the invention.
Since the resource allocation method for improving network quality in a multi-slice network according to the present invention relates to the related content of the infrastructure network, the related content will be described below so that the content of the present invention will be more apparent to those skilled in the art.
As shown in fig. 1, for correspondence between a tenant and a multi-service scoring policy, a downlink OFDMA (orthogonal frequency division multiple access) virtual wireless heterogeneous network having N Base Stations (BS) is considered, and an infrastructure provider (InP) and the tenant subscribe to a network service. The resources of the base station are virtualized and sliced according to the service requirements of users, and each tenant manages the resource allocation of one type of slice respectively. Considering the situation of two tenants, tenant 1 regulates and controls the allocation of eMBB slice (rate priority slice, slice 1) resources, and tenant 2 regulates and controls the allocation of URLLC slice (delay priority slice, slice 2) resources. Two tenants cannot coordinate with each other, but compete for wireless resources (subcarriers) in a non-cooperative game manner.
Network tenants are relative to infrastructure providers, and the infrastructure providers possess the basic software and hardware resources of the network, including base stations, transmission networks, core networks and the like. The tenant may be seen as a virtual operator, including an industry user such as a power grid, etc., and the equipment of the rental infrastructure provider develops network services. The user is an end user of the tenant, and is generally an individual user or an end user such as an internet of things terminal.
And the infrastructure provider allocates virtual bandwidth resources to the tenants according to the slice optimized scale factor. Based on the bandwidth requirement of the tenant, the tenant can adjust the resource allocation strategy strategically to maximize the score, and the invention expresses the resource allocation problem of two slices as a slice bidding game scheme.
Example one
Based on the above description, as shown in fig. 2, the resource allocation method for improving network quality in a multi-slice network provided by the present invention includes the following steps:
1) slicing the heterogeneous network to be tested according to the service requirements of users, modeling aiming at the resource allocation problem influencing the network quality of each slice, and establishing a resource allocation model of each slice, which specifically comprises the following steps:
1.1) slicing the heterogeneous network to be tested according to the service requirements of users, wherein the heterogeneous network to be tested is divided into rate priority slices and delay priority slices.
1.2) setting resources of a tenant 1 for regulating and controlling rate priority slices, setting a tenant 2 for regulating and controlling resources of delay priority slices, aiming at the problem of resource allocation influencing the network quality of each slice, and establishing a non-cooperative game model, namely a resource allocation model, of the tenant corresponding to each slice by taking the scores of network services of the two tenants as targets:
the participants: tenant 1 and tenant 2.
Strategy II: the optimal strategy of each tenant is the obtained optimal resource allocation scheme, wherein the strategy needs to meet the user requirement of the access slice and the total power constraint.
The corresponding policies of the tenants 1 and 2 are as follows:
Strategy1＝(p1,a1) (1)
Strategy2＝(p2,a2) (2)
wherein Strategy1 and Strategy2 are policies of tenant 1 and tenant 2, respectively; p is a radical of1,p2Power vectors allocated to corresponding users for tenant 1 and tenant 2, respectively:
Figure GDA0003649843940000091
Figure GDA0003649843940000092
wherein the content of the first and second substances,
Figure GDA0003649843940000093
and
Figure GDA0003649843940000094
respectively for subcarriers n of base station j in regulation rate priority slice1And subcarrier n in time delay priority slice2A transmission power of; k1A set of users that are associated rate-first slices; k2A set of users for associated latency-first slices; a is a1And a2The number of subcarriers allocated to tenant 1 and tenant 2 is a1+a2。
Figure GDA0003649843940000095
Utility function: and service scoring standard corresponding to each tenant.
According to the implementation situation of the existing communication network quality score, part of indexes of the network quality score are related to the network transmission rate, and part of indexes are unrelated to the network transmission rate.
Service scoring standard T of tenant 11Comprises the following steps:
T1＝k1r1+grade1 (6)
wherein k is1A scoring coefficient per unit rate in a rate-first slice; grade1Scoring other indexes including user offline rate, resource utilization rate, connection establishment success rate, continuous weak coverage proportion, continuous no-coverage proportion and the like; parameter r1Comprises the following steps:
Figure GDA0003649843940000101
wherein the content of the first and second substances,
Figure GDA0003649843940000102
associating user k for base station jjThe channel of (2);
Figure GDA0003649843940000103
associating user k for base station jjThe channel of (2);
Figure GDA0003649843940000104
subcarrier n in regulatory rate priority slice for base station j1A transmission power of;
Figure GDA0003649843940000105
and
Figure GDA0003649843940000106
whether there is a user associated subcarrier n in base stations j and j', respectively1If yes, the value is 1, otherwise, the value is notThen 0; delta. for the preparation of a coating2Is additive white gaussian noise power;
service scoring standard T of tenant 22Comprises the following steps:
T2＝k2r2+grade2 (8)
wherein k is2The scoring coefficient of each unit rate in the time delay priority slice; grade2Scoring other indicators; parameter r2Comprises the following steps:
Figure GDA0003649843940000107
wherein, the first and the second end of the pipe are connected with each other,
Figure GDA0003649843940000108
associating user l with base station jjThe channel of (2);
Figure GDA0003649843940000109
associating user l for base station jjThe channel of (2);
Figure GDA00036498439400001010
for sub-carrier n in time delay priority slice for base station j2A transmission power of;
Figure GDA00036498439400001011
and
Figure GDA00036498439400001012
whether there is a user associated subcarrier n in base stations j and j', respectively2If yes, the value is 1, otherwise, the value is 0.
Fourthly, the resource allocation model of the tenant 1 scores the network quality of the maximized rate priority slice, namely:
maxk1r1+grade1 (10)
Figure GDA00036498439400001013
Figure GDA00036498439400001014
Figure GDA00036498439400001015
wherein σ0For user kjA minimum rate requirement threshold of;
Figure GDA00036498439400001016
is a non-negative real number; p1Total power for rate-first slices; parameter(s)
Figure GDA00036498439400001017
Comprises the following steps:
Figure GDA00036498439400001018
equation (11) is used to guarantee the transmission rate requirement of each user; equation (12) indicates that the same subcarrier is allocated to at most a certain user in base station j in the rate-first slice; equation (13) represents the sum of all user powers for rate-first slice and the total power P that is less than or equal to rate-first slice1。
The resource allocation model of the tenant 2 is the network quality score of the maximum delay priority slice, namely:
maxk2r2+grade2 (15)
Figure GDA0003649843940000111
Figure GDA0003649843940000112
Figure GDA0003649843940000113
wherein epsilon is a number satisfying user ljThe interruption probability of the minimum delay requirement;
Figure GDA0003649843940000114
for user ljA packet arrival rate;
Figure GDA0003649843940000115
the maximum time delay which can be tolerated by the user; parameter(s)
Figure GDA0003649843940000116
Comprises the following steps:
Figure GDA0003649843940000117
formula (16) ensures lower interruption probability for user delay requirements in the delay-first slice; equation (17) indicates that the same subcarrier is allocated to at most a certain user in base station j in the delay-prioritized slice; equation (18) represents that the sum of all user powers of the delay-prioritized slices is equal to or less than the total power P of the delay-prioritized slices2。
2) Transforming and approximating the non-convex function in each resource allocation model to obtain an approximate convex function of each resource allocation model, which specifically comprises the following steps:
because the two resource allocation models are not convex optimization problems, a continuous convex approximation algorithm is required to be adopted, and the three problems that the speed expressions in the two resource allocation models are too complex, and the models have integer variables and speed expressions which are not convex are solved.
2.1) the expression is complex for speed:
adding corresponding constraints to resource allocation models corresponding to tenants 1 and 2
Figure GDA0003649843940000118
And
Figure GDA0003649843940000119
thereby removing the variation in the rate formula
Figure GDA00036498439400001110
And
Figure GDA00036498439400001111
the objective functions (10), (15) and constraints (11), (16) are simplified.
2.2) for integer variables in the resource allocation model:
adding a penalty function to an objective function of each slice resource allocation model to relax integer variables, wherein the penalty functions of the objective functions (10) and (15) are respectively as follows:
Figure GDA00036498439400001112
Figure GDA00036498439400001113
where ω is a parameter tending towards zero; the parameter q ∈ (0,1) can be set in accordance with the actual situation.
2.3) non-convex for speed expressions:
obtaining the parameter r according to Taylor expansion with Peyer's nuoenber1：
Figure GDA0003649843940000121
Wherein the parameters
Figure GDA0003649843940000122
Parameter(s)
Figure GDA0003649843940000123
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n1To the initial value of the transmission power.
Suppose that
Figure GDA0003649843940000124
The number of the carbon atoms is zero,
Figure GDA0003649843940000125
approximate convex function of resource allocation model approximated as rate-first slice
Figure GDA0003649843940000126
Figure GDA0003649843940000127
Thus, the near convex optimization problem can be solved, and variables are changed each time
Figure GDA0003649843940000128
Is optimized value of
Figure GDA0003649843940000129
Assigning to the initial value
Figure GDA00036498439400001210
And repeating the step of solving the convex optimization problem until the two optimal values are close to each other, so as to obtain an approximate optimal solution of the tenant 1.
Approximate convex function of resource allocation model of optimization homologism-obtainable time delay priority slice of tenant 2
Figure GDA00036498439400001211
Figure GDA0003649843940000131
Wherein the parameters
Figure GDA0003649843940000132
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n2To the initial value of the transmission power.
To facilitate the following algorithm description, the approximate convex objective functions defining the two slice resource allocation models are respectively:
Figure GDA0003649843940000133
Figure GDA0003649843940000134
3) under the condition that tenants 1 and 2 equally divide total subcarriers, a continuous convex approximation algorithm is adopted, an approximate convex objective function of each slice resource allocation model is solved according to the number of subcarriers of each slice, and an optimal slice resource allocation strategy and a scoring standard of each slice are determined, wherein the method specifically comprises the following steps:
3.1) resource allocation calculation for rate-first slices:
3.1.1) set K of rate-prioritized slice users1Number of subcarriers a1Channel gain
Figure GDA0003649843940000135
Noise power delta2The iteration number i is 0, the convergence precision eta of the given algorithm and a scoring coefficient k1Other scoring grades1Initial allocated power
Figure GDA0003649843940000136
3.1.2) according to the initial value, adopting matlab convex optimization packet to solve convex optimization problem to obtain optimized power
Figure GDA0003649843940000137
The matlab convex optimization packet is a method disclosed in the prior art, and the specific process is not described herein.
3.1.3) Power according to optimization
Figure GDA0003649843940000138
Calculating the value of the objective function
Figure GDA0003649843940000139
3.1.4) if objective function value
Figure GDA00036498439400001310
And (3) satisfying the constraint:
Figure GDA00036498439400001311
then the corresponding optimum power is output
Figure GDA00036498439400001312
Namely the optimal slice resource allocation strategy of the rate-first slice, and outputting a corresponding scoring standard; otherwise, let i be i +1,
Figure GDA0003649843940000141
go to step 3.1.2).
3.2) resource allocation calculation for the time delay priority slice:
3.2.1) set K of given latency-priority slice users2Number of subcarriers a2Channel gain
Figure GDA0003649843940000142
(j' denotes all base stations, including j), noise power delta2The iteration number i is 0, the convergence precision eta of the algorithm and a scoring coefficient k2Other scoring grades2Initial allocated power
Figure GDA0003649843940000143
3.2.2) according to the initial value, adopting matlab convex optimization packet to solve convex optimization problem to obtain optimized power
Figure GDA0003649843940000144
3.2.3) Power according to optimization
Figure GDA0003649843940000145
Calculating the value of an objective function
Figure GDA0003649843940000146
3.2.4) if objective function value
Figure GDA0003649843940000147
And (3) satisfying the constraint:
Figure GDA0003649843940000148
then the corresponding optimum power is output
Figure GDA0003649843940000149
The optimal slice resource allocation strategy of the instant delay priority slice is realized, and the corresponding scoring standard is output; otherwise, let i be i +1,
Figure GDA00036498439400001410
go to step 3.2.2).
4) Under the condition that the tenants 1 and 2 do not allocate subcarriers, a non-cooperative game method is adopted to solve the approximate convex objective function of each slice resource allocation model and determine the optimal slice resource allocation strategy and scoring standard of each slice, and the method specifically comprises the following steps:
4.1) user set K corresponding to given rate priority slice and time delay priority slice1And K2The convergence accuracy eta of the algorithm, and the initialization of the parameter channel gain in the heterogeneous network to be tested
Figure GDA00036498439400001411
And
Figure GDA00036498439400001412
noise power delta2When the iteration number i is equal to 0, distributing power
Figure GDA00036498439400001413
And
Figure GDA00036498439400001414
4.2) respectively traversing and calculating the resource allocation strategy of the speed priority slice and the time delay priority slice of different subcarriers
Figure GDA00036498439400001415
And
Figure GDA00036498439400001416
and corresponding scoring criteria
Figure GDA00036498439400001417
And
Figure GDA00036498439400001418
wherein, the first and the second end of the pipe are connected with each other,
Figure GDA00036498439400001419
indicates that the number of subcarriers allocated to the rate-first slice is a1The optimal power vector of the time of day,
Figure GDA00036498439400001420
the number of subcarriers allocated to the delay priority slice is represented as a2And the optimal power vector is the power vector which enables the target function to be maximum under the condition that the power vector meets all constraints.
4.3) obtaining the subcarrier number a of the tenant 1 by adopting the following game optimization formula (29) according to the calculation result of the step 4.2)1 *And the number of subcarriers a of tenant 22 *：
Figure GDA0003649843940000151
Wherein:
Figure GDA0003649843940000152
a2 *＝a-a1 *(31)
4.4) number of subcarriers a according to tenant 11 *And the number of subcarriers a of tenant 22 *And 4.2) obtaining the optimal slice resource allocation strategy of the rate priority slice and the time delay priority slice according to the calculation result of the step
Figure GDA0003649843940000153
And
Figure GDA0003649843940000154
and scoring criteria
Figure GDA0003649843940000155
And
Figure GDA0003649843940000156
in the step 1), other indexes are scored to obtain grade1And grade2Can be obtained by adopting the following method:
firstly, obtaining the objective overall utility quality and the objective comprehensive performance of the network by adopting an objective index evaluation method:
the network comprehensive performance integrates various network indexes through normalization, and the network performance in all aspects is presented. The evaluation method adopted is the standard deviation method (SD), which is an objective weight distribution method. In mathematics, the standard deviation is used to characterize the change in sample data from the corresponding mean. I.e., the larger the standard deviation, the larger the deviation from the mean. For a given decision matrix that has been normalized, if the standard deviation of index i is greater than the standard deviation of index j, it means that the former attribute is more important than the latter in evaluating the overall performance of each network. Thus, the former should be weighted more heavily than the latter and vice versa. The method comprises the following specific processes:
A) calculating the standard deviation of each index: to the normalized index value matrix DM _ Ns×n＝(rij)s×nCalculating the index standard deviation sigma by using the following formulaj：
Figure GDA0003649843940000157
Wherein DM _ Ns×n＝(rij)s×nIs a normalized index value matrix, each line of the matrix is a group of network performance indexes at a certain moment, s is the group number of the stored network performance indexes, n is the number of the network performance indexes to be evaluated, rijThe value is the normalized value of the index at the moment i and j.
B) Obtaining objective weight o _ aw of each index through index standard deviationj：
Figure GDA0003649843940000158
Through the two steps of operation, the index objective weight matrix AW can be obtained.
The network objective overall utility quality evaluation index reflects the utility value of the current network from the perspective of network quality, and the adopted evaluation mode is a gray level correlation analysis (GRA). The method firstly needs to set an ideal sequence as a reference sequence, the network data sequence to be evaluated is regarded as a comparison sequence, and then the geometric similarity between each comparison sequence and the reference sequence is respectively calculated and called as the gray level correlation degree. The greater the grey scale correlation of the comparison sequence, the closer it is to the reference sequence. Thus, a comparison sequence with a higher GRD indicates that the overall utility value of the network is higher. The specific process of the gray level correlation analysis method is as follows:
A) setting a reference sequence: determining the ideal solution vector R ═ (y) based on the principle of selecting the best value for each network metric1,y2,...,yn) The vector has been subjected to a normalization stage in the data pre-processing.
B) Calculating a gray scale correlation coefficient (GRC):
Figure GDA0003649843940000161
wherein DM _ Ns×n＝(rij)s×nIs a normalized index value matrix, each row of the matrix is a group of network performance indexes at a certain moment, s is the group number of the stored network performance indexes, n is the number of the evaluation network performance indexes, rijThe normalized value of the index at the moment i and the index j is obtained; rho epsilon [0,1 ∈ ]]Is the identification coefficient; k. t is a variable in min and max, and has no practical significance; as can be seen from the above equation, the smaller ρ is, the larger the difference between the gradation correlation coefficients is; in the present invention, ρ may take a value of 0.5.
C) Calculating a gray level correlation degree (GRD):
Figure GDA0003649843940000162
wherein, o _ awjThe objective weight of the index obtained by the standard deviation method is described above.
Through the operation, the gray level correlation degree (GRD) can be obtained, and the gray level correlation degree is directly used as the overall utility value UE of the network.
Combining the objective overall utility quality of the network and the comprehensive performance of the network to obtain index scores:
combining the objective overall utility quality of the network and the comprehensive performance of the network by adopting a linear combination method to obtain an index score EV:
EV＝α×W×DM_NT+(1-α)×UV (36)
wherein α is a weighted linear combination coefficient.
Example two
The embodiment provides a resource allocation system for improving network quality in a multi-slice network, which includes:
the resource allocation model building module is used for slicing the heterogeneous network to be tested, modeling aiming at the resource allocation problem influencing the network quality of each slice, and building a resource allocation model of each slice;
the approximate convex function determining module is used for processing the non-convex functions in the resource distribution models to obtain the approximate convex functions of the resource distribution models;
the continuous convex approximation module is used for solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm under the condition that tenants corresponding to each slice equally divide total subcarriers, and determining an optimal slice resource allocation strategy and a scoring standard of each slice;
and the non-cooperative game module is used for solving the approximate convex objective function of the resource allocation model of each slice according to the number of the subcarriers of each slice by adopting a non-cooperative game method under the condition that the total subcarriers are not equally divided by the tenants corresponding to each slice, and determining the optimal slice resource allocation strategy and the scoring standard of each slice.
In a preferred embodiment, the resource allocation model building module comprises:
the heterogeneous network slicing unit is used for slicing the heterogeneous network to be tested into a rate priority slice and a time delay priority slice;
and the model establishing unit is used for setting resources of the tenant 1 for regulating and controlling the rate priority slice, the tenant 2 for regulating and controlling the resources of the delay priority slice, and establishing a resource distribution model of the tenant corresponding to each slice by taking the scores of the network services of the two tenants as a target aiming at the resource distribution problem influencing the network quality of each slice.
EXAMPLE III
A computer program comprising computer program instructions for implementing the steps corresponding to the above-described resource allocation method for improving network quality in a multi-slice network when the computer program instructions are executed by a processor.
Example four
A computer readable storage medium having computer program instructions stored thereon, wherein the computer program instructions, when executed by a processor, are configured to implement the steps corresponding to the above-mentioned resource allocation method for improving network quality in a multi-slice network.
The above embodiments are only used for illustrating the present invention, and the structure, connection manner, manufacturing process and the like of each component can be changed, and equivalent changes and improvements made on the basis of the technical scheme of the present invention should not be excluded from the protection scope of the present invention.

Claims (6)

1. A resource allocation method for improving network quality in a multi-slice network, comprising:
1) slicing the heterogeneous network to be tested, modeling aiming at the resource allocation problem influencing the quality of each sliced network, and establishing a resource allocation model of each slice, wherein the concrete process is as follows:
1.1) slicing the heterogeneous network to be tested into rate priority slices and delay priority slices;
1.2) setting resources of a tenant 1 for regulating and controlling rate priority slices, setting resources of a tenant 2 for regulating and controlling delay priority slices, aiming at the problem of resource allocation influencing the network quality of each slice, establishing a resource allocation model of the tenant corresponding to each slice by taking the scores of network services of two tenants as a target, wherein the construction process of the resource allocation model is as follows:
the participants: tenant 1 and tenant 2;
strategy (II): the optimal strategy of each tenant is the obtained optimal resource allocation scheme;
the corresponding policies of the tenants 1 and 2 are as follows:
Strategy1＝(p1,a1)
Strategy2＝(p2,a2)
wherein Strategy1 and Strategy2 are policies of tenant 1 and tenant 2, respectively; p is a radical of formula1,p2Power vectors allocated to corresponding users for tenants 1 and 2, respectively:
Figure FDA0003649843930000011
Figure FDA0003649843930000012
wherein the content of the first and second substances,
Figure FDA0003649843930000013
and
Figure FDA0003649843930000014
respectively for subcarriers n of base station j in regulation rate priority slice1And subcarrier n in time delay priority slice2A transmission power of; k is1A set of users that are associated rate-first slices; k2A set of users for associated latency-first slices;
a1and a2The number of subcarriers allocated to tenant 1 and tenant 2, respectively:
Figure FDA0003649843930000015
utility function: service scoring standard corresponding to each tenant, wherein the service scoring standard T of tenant 11Comprises the following steps:
T1＝k1r1+grade1
wherein k is1A scoring coefficient per unit rate in the rate-first slice; grade1Scoring the other indicators; parameter r1Comprises the following steps:
Figure FDA0003649843930000016
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA0003649843930000021
associating user k for base station jjThe channel of (2);
Figure FDA0003649843930000022
associating user k for base station jjThe channel of (2);
Figure FDA0003649843930000023
for base station j' to take precedence in regulating rateSubcarrier n in slice1A transmission power of;
Figure FDA0003649843930000024
and
Figure FDA0003649843930000025
whether there is a user associated subcarrier n in base stations j and j', respectively1If yes, the value is 1, otherwise, the value is 0; delta. for the preparation of a coating2Is additive white Gaussian noise power;
service scoring standard T of tenant 22Comprises the following steps:
T2＝k2r2+grade2
wherein k is2The scoring coefficient of each unit rate in the time delay priority slice; grade2Scoring the other indicators; parameter r2Comprises the following steps:
Figure FDA0003649843930000026
wherein the content of the first and second substances,
Figure FDA0003649843930000027
associating user l with base station jjThe channel of (2);
Figure FDA0003649843930000028
associating user l for base station jjThe channel of (2);
Figure FDA0003649843930000029
for sub-carrier n in time delay priority slice for base station j2A transmission power of;
Figure FDA00036498439300000210
and
Figure FDA00036498439300000211
whether there are base stations j and j', respectivelyUser associated subcarrier n2；
The resource allocation model of the tenant 1 is as follows:
maxk1r1+grade1
Figure FDA00036498439300000212
Figure FDA00036498439300000213
Figure FDA00036498439300000214
wherein σ0For user kjA minimum rate requirement threshold of;
Figure FDA00036498439300000215
is a non-negative real number; p1Total power for rate-first slicing; parameter(s)
Figure FDA00036498439300000216
Comprises the following steps:
Figure FDA00036498439300000217
the resource allocation model of the tenant 2 is as follows:
maxk2r2+grade2
Figure FDA00036498439300000218
Figure FDA0003649843930000031
Figure FDA0003649843930000032
wherein ε is satisfying user ljThe interruption probability of the minimum delay requirement;
Figure FDA0003649843930000033
for user ljA packet arrival rate;
Figure FDA0003649843930000034
the maximum time delay that can be tolerated by the user; p2Total power for delay-first slicing; parameter(s)
Figure FDA0003649843930000035
Comprises the following steps:
Figure FDA0003649843930000036
2) processing the non-convex function in each resource allocation model to obtain an approximate convex function of each resource allocation model;
3) under the condition that tenants corresponding to the slices equally divide total subcarriers, solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm, and determining an optimal slice resource allocation strategy and a scoring standard of each slice;
4) and under the condition that the tenants corresponding to the slices do not equally divide the total subcarriers, solving an approximate convex objective function of the resource allocation model of each slice by adopting a non-cooperative game method according to the number of the subcarriers of each slice, and determining an optimal slice resource allocation strategy and a scoring standard of each slice.
2. The method for allocating resources for improving network quality in a multi-slice network as claimed in claim 1, wherein the specific process of step 2) is:
2.1) adding corresponding constraints to the resource allocation models corresponding to the tenants 1 and 2
Figure FDA0003649843930000037
And
Figure FDA0003649843930000038
Figure FDA0003649843930000039
2.2) adding corresponding penalty functions to the objective functions of the slice resource allocation models:
Figure FDA00036498439300000310
Figure FDA00036498439300000311
where ω is a parameter tending to zero, and q ∈ (0, 1);
2.3) the approximate convex objective functions of the two slice resource allocation models are respectively:
Figure FDA00036498439300000312
Figure FDA0003649843930000041
wherein:
Figure FDA0003649843930000042
Figure FDA0003649843930000043
in the formula, parameter
Figure FDA0003649843930000044
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n1An initial value of transmission power; parameter(s)
Figure FDA0003649843930000045
When solving iteration for the ith convex optimization problem, the base station j' is positioned on the subcarrier n2To the initial value of the transmission power.
3. The method for allocating resources for improving network quality in a multi-slice network as claimed in claim 2, wherein the specific process of step 3) is:
resource allocation calculation for rate-first slices:
a) set of given rate-first sliced users K1Number of subcarriers a1Channel gain
Figure FDA0003649843930000046
Noise power delta2The iteration number i is 0, the convergence precision eta of the given algorithm and a scoring coefficient k1Other scoring grades1Initial allocated power
Figure FDA0003649843930000047
b) According to the initial value, a matlab convex optimization packet is adopted to solve a convex optimization problem to obtain optimized power
Figure FDA0003649843930000048
c) According to the optimized power
Figure FDA0003649843930000051
Calculating the value of the objective function
Figure FDA0003649843930000052
d) If the value of the objective function
Figure FDA0003649843930000053
And (3) satisfying the constraint:
Figure FDA0003649843930000054
then outputting the corresponding optimal slice resource allocation strategy
Figure FDA0003649843930000055
And a scoring criterion; otherwise, let i be i +1,
Figure FDA0003649843930000056
entering into the step b);
resource allocation calculation for the latency-first slice:
A) set of given latency-first slice users K2The number of subcarriers a2Channel gain
Figure FDA0003649843930000057
Noise power delta2The iteration number i is 0, the convergence precision eta of the algorithm and a scoring coefficient k2Other scoring grades2Initial allocation of power
Figure FDA0003649843930000058
B) According to the initial value, a matlab convex optimization packet is adopted to solve a convex optimization problem to obtain optimized power
Figure FDA0003649843930000059
C) According to the optimized power
Figure FDA00036498439300000510
Calculating the value of the objective function
Figure FDA00036498439300000511
D) If the value of the objective function
Figure FDA00036498439300000512
And (3) satisfying the constraint:
Figure FDA00036498439300000513
then outputting the corresponding optimal slice resource allocation strategy
Figure FDA00036498439300000514
And a scoring criterion; otherwise, let i be i +1,
Figure FDA00036498439300000515
Figure FDA00036498439300000516
entering the step B).
4. The method for allocating resources for improving network quality in a multi-slice network as claimed in claim 3, wherein the specific process of step 4) is:
4.1) user set K corresponding to given rate priority slice and time delay priority slice1And K2The convergence accuracy eta of the algorithm, and the initialization of the parameter channel gain in the heterogeneous network to be tested
Figure FDA00036498439300000517
And
Figure FDA00036498439300000518
noise powerRate delta2The iteration number i is equal to 0, and the power is distributed
Figure FDA00036498439300000519
And
Figure FDA00036498439300000520
4.2) respectively traversing and calculating the resource allocation strategy of the speed priority slice and the time delay priority slice of different subcarriers
Figure FDA00036498439300000521
And
Figure FDA00036498439300000522
and corresponding scoring criteria
Figure FDA00036498439300000523
And
Figure FDA00036498439300000524
wherein, the first and the second end of the pipe are connected with each other,
Figure FDA00036498439300000525
indicates that the number of subcarriers allocated to the rate-first slice is a1The optimal power vector of the time of day,
Figure FDA00036498439300000526
the number of subcarriers allocated to the delay priority slice is represented as a2An optimal power vector of time;
4.3) calculating the number a of the subcarriers of the tenant 1 according to the calculation result of the step 4.2)1 *And the number of subcarriers a of tenant 22 *：
Figure FDA0003649843930000061
Figure FDA0003649843930000062
a2 *＝a-a1 *
4.4) number of subcarriers a according to tenant 11 *And the number a of subcarriers of tenant 22 *And the calculation result of the step 4.2) is used for obtaining the optimal slice resource allocation strategy of the rate priority slice and the time delay priority slice
Figure FDA0003649843930000063
And
Figure FDA0003649843930000064
and scoring criteria
Figure FDA0003649843930000065
And
Figure FDA0003649843930000066
5. a resource allocation system for improving network quality in a multi-slice network, comprising:
the resource allocation model building module is used for slicing the heterogeneous network to be tested, modeling aiming at the resource allocation problem influencing the network quality of each slice, and building a resource allocation model of each slice;
the approximate convex function determining module is used for processing the non-convex functions in the resource allocation models to obtain the approximate convex functions of the resource allocation models, and comprises the following steps:
the heterogeneous network slicing unit is used for slicing the heterogeneous network to be tested into a rate priority slice and a time delay priority slice;
the model establishing unit is used for setting resources of a tenant 1 for regulating and controlling the rate priority slice, a tenant 2 for regulating and controlling the resources of the delay priority slice, aiming at the resource distribution problem affecting the network quality of each slice, and establishing a resource distribution model of the tenant corresponding to each slice by taking the score of the network service of the two tenants as a target, wherein the construction process of the resource distribution model is as follows:
the participants: tenant 1 and tenant 2;
strategy II: the optimal strategy of each tenant is the obtained optimal resource allocation scheme;
the corresponding policies of the tenants 1 and 2 are as follows:
Strategy1＝(p1,a1)
Strategy2＝(p2,a2)
wherein Strategy1 and Strategy2 are policies of tenant 1 and tenant 2, respectively; p is a radical of1,p2Power vectors allocated to corresponding users for tenant 1 and tenant 2, respectively:
Figure FDA0003649843930000067
Figure FDA0003649843930000068
wherein the content of the first and second substances,
Figure FDA0003649843930000069
and
Figure FDA00036498439300000610
respectively for subcarriers n of base station j in regulation rate priority slice1And subcarrier n in time delay priority slice2A transmission power of; k1A set of users that are associated rate-first slices; k is2A set of users for associated latency-first slices;
a1and a2The number of subcarriers allocated to tenant 1 and tenant 2, respectively:
Figure FDA0003649843930000071
(iii) utility function: service scoring standard corresponding to each tenant, wherein, the service scoring standard T of the tenant 11Comprises the following steps:
T1＝k1r1+grade1
wherein k is1A scoring coefficient per unit rate in a rate-first slice; grade1Scoring other indicators; parameter r1Comprises the following steps:
Figure FDA0003649843930000072
wherein the content of the first and second substances,
Figure FDA0003649843930000073
associating user k for base station jjThe channel of (2);
Figure FDA0003649843930000074
associating user k for base station jjThe channel of (2);
Figure FDA0003649843930000075
subcarrier n in regulatory rate priority slice for base station j1A transmission power of;
Figure FDA0003649843930000076
and
Figure FDA0003649843930000077
whether there is a user associated subcarrier n in base stations j and j', respectively1If yes, the value is 1, otherwise, the value is 0; delta. for the preparation of a coating2Is additive white gaussian noise power;
service scoring standard T of tenant 22Comprises the following steps:
T2＝k2r2+grade2
wherein k is2For per unit rate in time delay-first slicesA scoring coefficient; grade2Scoring other indicators; parameter r2Comprises the following steps:
Figure FDA0003649843930000078
wherein the content of the first and second substances,
Figure FDA0003649843930000079
associating user l for base station jjThe channel of (2);
Figure FDA00036498439300000710
associating user l for base station jjThe channel of (2);
Figure FDA00036498439300000711
for sub-carrier n in time delay priority slice for base station j2A transmission power of;
Figure FDA00036498439300000712
and
Figure FDA00036498439300000713
whether there is a user associated subcarrier n in base stations j and j', respectively2；
The resource allocation model of the tenant 1 is as follows:
maxk1r1+grade1
Figure FDA00036498439300000714
Figure FDA0003649843930000081
Figure FDA0003649843930000082
wherein σ0For user kjA minimum rate requirement threshold of;
Figure FDA0003649843930000083
is a non-negative real number; p1Total power for rate-first slices; parameter(s)
Figure FDA0003649843930000084
Comprises the following steps:
Figure FDA0003649843930000085
and the resource allocation model of the tenant 2 is as follows:
maxk2r2+grade2
Figure FDA0003649843930000086
Figure FDA0003649843930000087
Figure FDA0003649843930000088
wherein epsilon is a number satisfying user ljThe interruption probability of the minimum delay requirement;
Figure FDA0003649843930000089
for user ljA packet arrival rate;
Figure FDA00036498439300000810
the maximum time delay which can be tolerated by the user; p2Total power of the time delay priority slice; parameter(s)
Figure FDA00036498439300000811
Comprises the following steps:
Figure FDA00036498439300000812
the continuous convex approximation module is used for solving an approximate convex objective function of each slice resource allocation model according to the number of subcarriers of each slice by adopting a continuous convex approximation algorithm under the condition that the total subcarriers are equally divided by tenants corresponding to each slice, and determining an optimal slice resource allocation strategy and a scoring standard of each slice;
and the non-cooperative game module is used for solving the approximate convex objective function of the resource allocation model of each slice according to the number of the subcarriers of each slice by adopting a non-cooperative game method under the condition that the total subcarriers are not equally divided by the tenants corresponding to each slice, and determining the optimal slice resource allocation strategy and the scoring standard of each slice.
6. A computer readable storage medium, having stored thereon computer program instructions, wherein the computer program instructions, when executed by a processor, are configured to implement the corresponding steps of the resource allocation method for improving network quality in a multi-slice network according to any one of claims 1-4.
CN202010533456.XA 2020-06-12 2020-06-12 Resource allocation method and system for improving network quality in multi-slice network Active CN111556518B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010533456.XA CN111556518B (en) 2020-06-12 2020-06-12 Resource allocation method and system for improving network quality in multi-slice network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010533456.XA CN111556518B (en) 2020-06-12 2020-06-12 Resource allocation method and system for improving network quality in multi-slice network

Publications (2)

Publication Number Publication Date
CN111556518A CN111556518A (en) 2020-08-18
CN111556518B true CN111556518B (en) 2022-07-12

Family

ID=72008709

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010533456.XA Active CN111556518B (en) 2020-06-12 2020-06-12 Resource allocation method and system for improving network quality in multi-slice network

Country Status (1)

Country Link
CN (1) CN111556518B (en)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111988818B (en) * 2020-08-19 2023-08-25 鹏城实验室 Resource allocation method, apparatus and computer readable storage medium
CN112423267B (en) * 2020-10-14 2022-04-22 南京大学 Vehicle networking heterogeneous resource dynamic slicing method based on Lyapunov random optimization
CN112333735B (en) * 2020-11-05 2023-05-12 中国联合网络通信集团有限公司 Time slot interval adjusting method and communication device
CN112333717B (en) * 2020-11-13 2022-08-30 国网安徽省电力有限公司信息通信分公司 5G access network slice resource allocation method and device considering power multi-service requirements
CN112291793B (en) * 2020-12-29 2021-04-06 北京邮电大学 Resource allocation method and device of network access equipment
CN113473498B (en) * 2021-06-15 2023-05-19 中国联合网络通信集团有限公司 Network slice resource arrangement method, slice arrangement device and arrangement system
CN114978277B (en) * 2022-04-24 2023-06-27 北京邮电大学 Cross-domain resource allocation method in non-ground network based on network slicing

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107682135A (en) * 2017-09-30 2018-02-09 重庆邮电大学 A kind of network slice adaptive virtual resource allocation method based on NOMA
CN108601087A (en) * 2018-04-27 2018-09-28 哈尔滨工业大学深圳研究生院 A kind of wireless communication resources allocation algorithm based on network slice
WO2019012735A1 (en) * 2017-07-11 2019-01-17 株式会社Ｎｔｔドコモ Ran slice resource management device and ran slice resource management method

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019012735A1 (en) * 2017-07-11 2019-01-17 株式会社Ｎｔｔドコモ Ran slice resource management device and ran slice resource management method
CN107682135A (en) * 2017-09-30 2018-02-09 重庆邮电大学 A kind of network slice adaptive virtual resource allocation method based on NOMA
CN108601087A (en) * 2018-04-27 2018-09-28 哈尔滨工业大学深圳研究生院 A kind of wireless communication resources allocation algorithm based on network slice

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Slicing Resource Allocation for eMBB and URLLC in 5G RAN;Tengteng Ma,Yong Zhang,Fanggang Wang,Dong Wang,Da Guo;《Wireless Communications and Mobile Computing》;20200131;第1-11页 *

Also Published As

Publication number Publication date
CN111556518A (en) 2020-08-18

Similar Documents

Publication Publication Date Title
CN111556518B (en) Resource allocation method and system for improving network quality in multi-slice network
CN111447619B (en) Joint task unloading and resource allocation method in mobile edge computing network
CN111182570B (en) User association and edge computing unloading method for improving utility of operator
CN111132191B (en) Method for unloading, caching and resource allocation of joint tasks of mobile edge computing server
CN109151864B (en) Migration decision and resource optimal allocation method for mobile edge computing ultra-dense network
Dai et al. Joint offloading and resource allocation in vehicular edge computing and networks
CN112231085B (en) Mobile terminal task migration method based on time perception in collaborative environment
CN109714382B (en) Multi-user multi-task migration decision method of unbalanced edge cloud MEC system
CN110233755B (en) Computing resource and frequency spectrum resource allocation method for fog computing in Internet of things
CN111182495B (en) 5G internet of vehicles partial calculation unloading method
CN111200831B (en) Cellular network computing unloading method fusing mobile edge computing
CN111836284B (en) Energy consumption optimization calculation and unloading method and system based on mobile edge calculation
CN113342409B (en) Delay sensitive task unloading decision method and system for multi-access edge computing system
CN106792995B (en) User access method for guaranteeing low-delay content transmission in 5G network
CN114615294B (en) Electric power internet of things gateway edge calculation method
CN112511652B (en) Cooperative computing task allocation method under edge computing
CN112437449B (en) Joint resource allocation method
Ikami et al. Dynamic channel allocation algorithm for spectrum sharing between different radio systems
CN111954230B (en) Computing migration and resource allocation method based on integration of MEC and dense cloud access network
Li et al. Joint access point selection and resource allocation in MEC-assisted network: A reinforcement learning based approach
Chen et al. Virtualized radio resource pre-allocation for QoS based resource efficiency in mobile networks
Poveda et al. Dynamic bandwidth allocation in wireless networks using a shahshahani gradient based extremum seeking control
Li A spectrum allocation algorithm based on proportional fairness
CN114554496A (en) 5G network slice resource allocation method based on machine learning
Bhandari et al. An efficient scheduling scheme for fronthaul load reduction in fog radio access networks

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