CN111556510A - Base station data cooperation caching model and method based on content similarity - Google Patents

Base station data cooperation caching model and method based on content similarity Download PDF

Info

Publication number
CN111556510A
CN111556510A CN201911252355.9A CN201911252355A CN111556510A CN 111556510 A CN111556510 A CN 111556510A CN 201911252355 A CN201911252355 A CN 201911252355A CN 111556510 A CN111556510 A CN 111556510A
Authority
CN
China
Prior art keywords
base station
data
strategy
base stations
overhead
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.)
Pending
Application number
CN201911252355.9A
Other languages
Chinese (zh)
Inventor
彭进霖
张玉立
张博
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.)
National Defense Technology Innovation Institute PLA Academy of Military Science
Original Assignee
National Defense Technology Innovation Institute PLA Academy of Military Science
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 National Defense Technology Innovation Institute PLA Academy of Military Science filed Critical National Defense Technology Innovation Institute PLA Academy of Military Science
Priority to CN201911252355.9A priority Critical patent/CN111556510A/en
Publication of CN111556510A publication Critical patent/CN111556510A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • 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
    • H04W16/22—Traffic simulation tools or models
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W88/00—Devices specially adapted for wireless communication networks, e.g. terminals, base stations or access point devices
    • H04W88/08—Access point devices

Abstract

The invention discloses a base station data cooperation caching model and a method based on content similarity, which are used for intensively downloading repeated data and reducing repeated downloading of similar contents through the cooperation of neighbor base stations with similar contents when the data contents of a micro-cellular base station are acquired, thereby improving the utilization rate of frequency spectrum resources and reducing the data acquisition overhead. Firstly, constructing a generalized local game model, wherein game participants are microcellular base stations; for any femtocell base station, other microcells are divided into one-hop neighbors, two-hop neighbors and other base stations according to the communication range; each base station randomly generates a cache strategy; randomly selecting a base station, calculating the data acquisition overhead of the base station, changing a cache strategy, recalculating the overhead, and updating the strategy according to the probability; and repeating the iteration process until all the base station strategies are converged or the set iteration times are reached. The invention can more deeply mine the cooperation relationship among the base stations, so that the data caching strategy is more efficient, and the expenditure is further reduced.

Description

Base station data cooperation caching model and method based on content similarity
Technical Field
The invention belongs to the technical field of wireless communication, and provides a base station data cooperation caching model and method based on content similarity.
Background
With the development of mobile communication technology, the data traffic load of the base station also shows a tendency of explosive growth. A cache technology (m.tag, k.michinski, c.ofria, et.al., "Distributed cooperative caching in Social Wireless Networks", IEEE trans.mobile Computing, vol.12, No.6, pp.1037-1053, 2013.) can reduce the data request amount from a base station to a core network by physically storing data, and is considered to be one of effective technologies for reducing the load of the base station, which has received much attention from researchers. In the related documents (FANG Chao, YU f. richard, HUANG Tao, LIU Jiang, LIU Yunjie, "a door theoretical application for Energy-Efficient network cable in Content-centralized Networks", China Communication, vol.11, No.11, pp.135-145, November,2014.) buffered at present, most of the research focuses on the problem of buffer link between the mobile subscriber and the base station, such as improving the signal-to-noise ratio by the microcell, and the like, and the microcell selects the buffered data, thereby providing better data service for the mobile subscriber. However, the current cache research has less concern about the acquisition overhead of the data cached by the femtocell base station.
Disclosure of Invention
The invention aims to provide a cooperative caching method which can reduce the data acquisition overhead of a base station and improve the utilization rate of a frequency spectrum based on content similarity; when the data content of the micro-cellular base station is acquired, the repeated data is downloaded intensively by the cooperation of the neighbor base stations with similar content so as to reduce the data acquisition overhead.
The technical scheme of the invention is as follows:
the invention provides a base station data cooperation caching method based on content similarity, for any micro-cellular base station with content requirement in a heterogeneous network, other micro-cellular base stations are divided into neighbor base stations and non-neighbor base stations according to a communication range; when the data content of the micro-cellular base station is acquired, the neighbor base stations with similar content cooperate to download the data in a centralized way, so that repeated downloading of the similar content is reduced, the utilization rate of frequency spectrum resources is improved, and the data acquisition overhead is reduced.
Further, the method comprises the steps of:
step 1, modeling a base station cooperation cache problem into a generalized local mutual profit game model, wherein participants of the game are all microcellular base stations with data requirements in a network;
step 2, for any micro-cellular base station, dividing other base stations into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to the communication range, and defining a utility function to represent the data requirement acquisition overhead;
step 3, initializing all base stations to randomly generate a cache strategy, and calculating utility function values under the current strategy; one base station is randomly generated, another cache strategy is randomly generated, and a utility function value under the current strategy is calculated; for the selected base station, respectively acquiring the two cache strategies as the probability for updating the cache strategy at the next moment, and preferentially updating the cache strategy of the selected base station based on the probability;
and 4, circulating the step of randomly selecting the base stations to update the cache strategy in the step 3 until the selection of all the base stations is converged or the set iteration times is reached.
Further, step 1 models the base station cooperation cache problem as a game model, which is defined as:
G＝{N,A,,un} (1)
the game model G comprises four components, wherein N is a microcellular base station set participating in a game, N represents the total number of game participants, N represents the number of the game participants, A is a strategy action space of a base station, is a network topology and comprises a one-hop neighbor base station set J of the base station NnAnd two-hop neighbor base station set
Figure RE-GDA0002577652510000024
，unIs the utility function of base station n.
Further, in step 2, defining a utility function to characterize the data requirement acquisition overhead specifically includes:
step 2-1, defining the data requirement of the micro-cellular base station n as
Figure RE-GDA0002577652510000021
Wherein; k represents a data requirement number; lnIs the total data demand of the femtocell base station n; dnkRepresenting the kth data requirement of the femtocell base station n; defining the caching strategy of the data required by the base station n as
Figure RE-GDA0002577652510000022
Wherein, ankFor corresponding data demand dnkThe caching strategy of (a), which base station or macro cell base station is selected from one-hop neighbors as the caching source, isnk∈Jn∪ {0}, where 0 represents an acquisition from a macrocell base station existing in the network, JnA set of neighbor base stations representing base station n; let ankJ, then base station n data requirement dnkHas an acquisition overhead of fnk(an,a-n)：
Figure RE-GDA0002577652510000023
Wherein, a-nRepresenting the caching strategies of the rest base stations except the base station n; j > 0 indicates data acquisition from the neighbor base station j, KjIs the sum of all buffered data downloaded from the macro cell, BS, of base station jjIs the basic overhead of base station j operation, αj0Is the unit overhead for base station j to acquire data from the macro cell,j(dnk) For data request d received at base station jnkNumber of requests, BSj+Kjαj0Being the sum of the base overhead of base station j and the overall data acquisition cost,
Figure RE-GDA0002577652510000031
in order to achieve the cost of acquiring a unit of data,
Figure RE-GDA0002577652510000032
i.e. the data d of the base station j after being sharednkCost of αjnIs the transmission cost between base stations j and n; j ═ 0 represents the data acquisition overhead required to acquire the data from the macrocell; knIs the sum of all buffered data downloaded from the macro cell, BS, of base station nnIs the basic overhead of base station n operation, αn0Is the unit overhead of base station n acquiring data from the macro cell,n(dnk) For the data request d received at base station nnkNumber of requests, BSn+Knαn0Being the sum of the base overhead of base station n and the overall data acquisition cost,
Figure RE-GDA0002577652510000033
in order to achieve the cost of acquiring a unit of data,
Figure RE-GDA0002577652510000034
i.e. after sharing, the base station n data dnkThe overhead of (c);
step 2-2, calculating the sum of the data requirement acquisition overhead of the femtocell base station n:
Figure RE-GDA0002577652510000035
step 2-3, expanding the traditional local mutual interest game from the relation between the base station and the one-hop neighbor to the relation between the base station and the one-hop neighbor and the two-hop neighbor to form a generalized local mutual interest game, and establishing a utility function u of the micro-cellular base station nnComprises the following steps:
Figure RE-GDA0002577652510000036
wherein: r isnRepresenting the data requirement acquisition overhead of node n, i represents the number of one-hop neighbor base station and two-hop neighbor base station, riRepresenting the corresponding data demand acquisition overhead.
Optimization goals for the game: carrying out strategy optimization by using a utility function of an equation (4), and considering a cooperative caching mode to minimize the utility function:
Figure RE-GDA0002577652510000037
further, the policy update process of the base station caching policy optimization method described in step 3 is specifically as follows:
step 3-1, initializing, and randomly generating a cache strategy for each microcellular base station N belonging to N;
step 3-2, detection: all the femtocell base stations and one-hop neighbors interact with the caching strategy and the file requirement information, one femtocell base station n is randomly selected for operation, and other all users repeat the previous caching selection; for the selected femtocell base station n, calculating its current strategy an(t) value of utility function
Figure RE-GDA0002577652510000041
Randomly generating another policy
Figure RE-GDA0002577652510000042
Re-interacting information and calculating utility function values
Figure RE-GDA0002577652510000043
Step 3-3, strategy updating: the base station n preferentially updates the cache strategy according to the utility function values corresponding to the two different strategies, namely
Figure RE-GDA0002577652510000044
Wherein the content of the first and second substances,
Figure RE-GDA0002577652510000045
an(t) represents the strategy of base station n at the tth iteration; pr [ a ]n(t+1)＝an(t)]Indicating base station n selection strategy an(t) as the probability of the next moment strategy;
Figure RE-GDA0002577652510000046
representing base station n selection strategies
Figure RE-GDA0002577652510000047
As the probability of the next moment strategy, β is a learning factor.
Further, step 4, the loop of step 3 includes two steps of detection and policy update, and the base station optimizes the cooperative caching policy through exploration and learning until the caching policies of all the femtocell base stations converge or reach a set iteration number, which is specifically as follows:
4-1, performing information interaction on all the micro-cells;
4-2, randomly selecting one base station for operation in each iteration;
step 4-3, repeating the data requirement caching strategy before all other base stations, namely ak(t+1)＝ak(t)。
And calculating the data acquisition overhead of all the base stations after the circulation is finished.
A model adopted by a base station data cooperative caching method based on content similarity comprises the following steps:
the generalized local mutual profit game model building module: modeling a base station cooperation cache problem into a generalized local mutual profit game model, wherein participants of the game are all microcellular base stations with data requirements in a network;
a utility function definition module: for any micro-cellular base station, dividing other base stations into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to the communication range, and defining a utility function to represent the data demand acquisition overhead;
the cache strategy updating module: for all base stations, initializing and randomly generating a cache strategy, and calculating utility function values under the current strategy; one base station is randomly generated, another cache strategy is randomly generated, and a utility function value under the current strategy is calculated; for the selected base station, respectively acquiring the two cache strategies as the probability for updating the cache strategy at the next moment, and preferentially updating the cache strategy of the selected base station based on the probability;
a loop policy update module: and circularly and randomly selecting the base stations to update the cache strategy until the selection of all the base stations is converged to reach the set iteration times.
The invention has the beneficial effects that:
under the condition that the data services of the micro-cellular base stations have similarity, the situation that repeated downloading of cache data is reduced through mutual cooperation is fully considered, a generalized local mutual-profit game model with higher frequency spectrum resource utilization rate and lower data acquisition cost is provided, and the cooperation relationship among the micro-cellular base stations is better described.
The invention designs the potential energy function as the sum of the global expenses by introducing the potential energy game model, proves the existence of Nash equilibrium, and the optimal Nash equilibrium can minimize the good property of the global expenses, thereby providing theoretical support for the design of the method.
The base station data cooperation caching method based on the content similarity detects and realizes the global optimal progressive solution through information interaction, and improves the performance of the method.
Additional features and advantages of the invention will be set forth in the detailed description which follows.
Drawings
The above and other objects, features and advantages of the present invention will become more apparent by describing in more detail exemplary embodiments thereof with reference to the attached drawings, in which like reference numerals generally represent like parts throughout.
Fig. 1 is a schematic diagram of a cooperative caching system model and a caching strategy of a single base station in a heterogeneous microcellular network.
Fig. 2 is a method flow diagram.
Fig. 3 is a schematic diagram of network overhead for different numbers of base stations.
Fig. 4 is a schematic diagram of network overhead corresponding to different traffic demands of a base station.
Detailed Description
Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. While the preferred embodiments of the present invention are shown in the drawings, it should be understood that the present invention may be embodied in various forms and should not be limited to the embodiments set forth herein.
For any N micro-cellular base stations with spectrum resource requirements in the heterogeneous micro-cellular network, according to the communication distance of the base stations, other base stations are divided into one-hop neighbors, two-hop neighbors and other base stations for any base station. Fig. 1 shows a network scenario in which base stations cooperate with each other to reduce the overhead of acquiring cache data based on content similarity. For example, when the bs 1 and the bs 2 have similar buffered data, they can collectively download from the bs 1 and then adopt an internal forwarding manner, so as to reduce the frequency requirement of the femtocell for the macrocell and reduce the data acquisition overhead. As shown in the figure, the data file required by the base station 1 is [1,2,3,4], and the file required by the base station 2 is [1,3,4,5], so that the two base stations download the file with the same requirement through the base station 1 first and then share the file with the base station 2. Generally, when a base station has a plurality of data requirements, the base station can arrange the data of the base station more flexibly according to the caching condition of surrounding neighbors, and therefore the acquisition overhead is further reduced.
The invention considers a base station data cooperation cache model based on content similarity, and for any micro-cell with content requirement in a heterogeneous network, other micro-cells are divided into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to a communication range; when the data content of the micro-cellular base station is acquired, the neighbor base stations with similar content cooperate to download the data in a centralized way, so that repeated downloading of the similar content is reduced, the utilization rate of frequency spectrum resources is improved, and the data acquisition overhead is reduced. The main learning method comprises the following steps:
step 1, modeling a base station cooperation cache problem into a generalized local mutual profit game model, wherein participants of a game are all microcellular base stations with data requirements in a network, and the game model is defined as:
G＝{N,A,,un} (1)
the game model G comprises four components, wherein N is a microcellular base station set participating in a game, N represents the total number of game participants, N represents the number of the game participants, A is a base stationIs a network topology, comprises a one-hop neighbor base station set J of a base station nnAnd two-hop neighbor base station set
Figure RE-GDA0002577652510000064
，unIs the utility function of base station n.
Step 2, for any micro-cellular base station, dividing other base stations into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to the communication range, and defining a utility function to represent the data requirement acquisition overhead, specifically;
step 2-1, defining the data requirement of the micro-cellular base station n as
Figure RE-GDA0002577652510000061
Wherein; k represents a data requirement number; lnIs the total data demand of the femtocell base station n; dnkRepresenting the kth data requirement of the femtocell base station n; defining the caching strategy of the data required by the base station n as
Figure RE-GDA0002577652510000062
Wherein, ankFor corresponding data demand dnkThe caching strategy of (a), which base station or macro cell base station is selected from one-hop neighbors as the caching source, isnk∈Jn∪ {0}, where 0 represents an acquisition from a macrocell base station existing in the network, JnA set of neighbor base stations representing base station n; let ankJ, then d is required for the datankBase station n data requirement dnkHas an acquisition overhead of fnk(an,a-n)：
Figure RE-GDA0002577652510000063
Wherein, a-nRepresenting the caching strategies of the rest base stations except the base station n; j > 0 indicates data acquisition from the neighbor base station j, KjIs the sum of all buffered data downloaded from the macro cell, BS, of base station jjIs the basic overhead of base station j operation, αj0Is that base station j gets from the macro cellThe unit overhead of the data fetch is,j(dnk) For data request d received at base station jnkNumber of requests, BSj+Kjαj0Being the sum of the base overhead of base station j and the overall data acquisition cost,
Figure RE-GDA0002577652510000071
in order to achieve the cost of acquiring a unit of data,
Figure RE-GDA0002577652510000072
i.e. the data d of the base station j after being sharednkCost of αjnIs the transmission cost between base stations j and n; j ═ 0 represents the data acquisition overhead required to acquire the data from the macrocell; knIs the sum of all buffered data downloaded from the macro cell, BS, of base station nnIs the basic overhead of base station n operation, αn0Is the unit overhead of base station n acquiring data from the macro cell,n(dnk) For the data request d received at base station nnkNumber of requests, BSn+Knαn0Being the sum of the base overhead of base station n and the overall data acquisition cost,
Figure RE-GDA0002577652510000073
in order to achieve the cost of acquiring a unit of data,
Figure RE-GDA0002577652510000074
i.e. after sharing, the base station n data dnkThe overhead of (c);
step 2-2, calculating the sum of the data requirement acquisition overhead of the femtocell base station n:
Figure RE-GDA0002577652510000075
step 2-3, expanding the traditional local mutual interest game from the relation between the base station and the one-hop neighbor to the relation between the base station and the one-hop neighbor and the two-hop neighbor to form a generalized local mutual interest game, and establishing a utility function u of the micro-cellular base station nnComprises the following steps:
Figure RE-GDA0002577652510000076
wherein: r isnRepresenting the data requirement acquisition overhead of node n, i represents the number of one-hop neighbor base station and two-hop neighbor base station, riRepresenting the corresponding data demand acquisition overhead.
Optimization goals for the game: carrying out strategy optimization by using a utility function of an equation (4), and considering a cooperative caching mode to minimize the utility function:
Figure RE-GDA0002577652510000077
step 3, initializing all base stations to randomly generate a cache strategy, and calculating utility function values under the current strategy; one base station is randomly generated, another cache strategy is randomly generated, and a utility function value under the current strategy is calculated; for the selected base station, respectively acquiring the two cache strategies as the probability for updating the cache strategy at the next moment, and preferentially updating the cache strategy of the selected base station based on the probability;
the strategy updating process of the base station cache strategy optimization method specifically comprises the following steps:
step 3-1, initializing, and randomly generating a cache strategy for each microcellular base station N belonging to N;
step 3-2, detection: all the femtocell base stations and one-hop neighbors interact with the caching strategy and the file requirement information, one femtocell base station n is randomly selected for operation, and other all users repeat the previous caching selection; for the selected femtocell base station n, calculating its current strategy an(t) value of utility function
Figure RE-GDA0002577652510000081
Randomly generating another policy
Figure RE-GDA0002577652510000082
Re-interacting information and calculating utility function values
Figure RE-GDA0002577652510000083
Step 3-3, strategy updating: the base station n preferentially updates the cache strategy according to the utility function values corresponding to the two different strategies, namely
Figure RE-GDA0002577652510000084
Wherein the content of the first and second substances,
Figure RE-GDA0002577652510000085
an(t) represents the strategy of base station n at the tth iteration; pr [ a ]n(t+1)＝an(t)]Indicating base station n selection strategy an(t) as the probability of the next moment strategy;
Figure RE-GDA0002577652510000086
representing base station n selection strategies
Figure RE-GDA0002577652510000087
As the probability of the next moment strategy, β is a learning factor.
And 4, circulating the step of randomly selecting the base stations to update the cache strategy in the step 3 until the selection of all the base stations is converged or the set iteration times is reached.
4-1, performing information interaction on all the micro-cells;
4-2, randomly selecting one base station for operation in each iteration;
step 4-3, repeating the data requirement caching strategy before all other base stations, namely ai(t+1)＝ai(t)。
And calculating the data acquisition overhead of all the base stations after the circulation is finished.
Example 1
An embodiment of the invention is described in the following, wherein the system simulation adopts Matlab software, the parameter setting does not affect the generality, N10 microcellular base stations are randomly arranged in a network scene of 200m × 200m, and the acquisition overhead α of single data from the macrocell base stationn0Uniformly set to 1, base station time data sharing overhead αjn0.3, data requirement of single base station is 10, data distribution is according to uniform distribution, total sample is set to be 50, basic overhead BS of base stationnIs 5.
The invention relates to a base station cooperation caching method based on content similarity, which comprises the following specific processes:
step 1: initializing, setting the iteration number t to be 0, and randomly generating a cache strategy by each base station N belonging to N.
Step 2: cache policy selection policy update (round robin):
firstly, performing information interaction on all micro cells;
randomly selecting one base station n for operation in each iteration;
③ all other base stations repeat the previous buffering strategy selection, namely ai(t+1)＝ai(t) of (d). For the selected base station n, calculating the utility function value under the current strategy
Figure RE-GDA0002577652510000091
Randomly generating another strategy, re-exchanging information and calculating utility function value
Figure RE-GDA0002577652510000092
The caching strategy is preferentially updated according to the utility function values corresponding to two different strategies, namely
Figure RE-GDA0002577652510000093
Wherein the content of the first and second substances,
Figure RE-GDA0002577652510000094
β the learning factor is set to 0.15.
And step 3: and when the cache strategies of all the base stations realize convergence or reach a certain iteration number, the method converges.
And 4, step 4: global utility: the data acquisition overhead for all base stations in the network is calculated.
Fig. 3 shows the variation of the sum of the data download overhead of the whole network when the number of base stations increases from 7 to 11. As can be seen from the figure, as the number of base stations increases, the total data overhead of the whole network tends to increase. It is further observed that the average data acquisition overhead for a single base station decreases as the number of base stations increases. This is because as the number of base stations increases, the cooperation space and possibility between the base stations buffering data becomes larger and larger. Therefore, the gain due to cooperation is increased, so that the cache data acquisition overhead is further reduced. The patent of the invention compares SAP optimization methods formed by adopting alliances Based on base station data (reference documents: Yuli Zhang, Yuhua Xu and Qihui Wu, "Group Buying Based on social Aware D2D Networks: A Game therapeutic Approach," IEEE/CIC ICCC, Qingdao, 2017.). The comparison shows that the method provided by the invention has stronger performance.
Example 1 base station data demand ln
Consider 10 microcell base stations. As can be seen from fig. 4, in the case that the data demand of the base station increases, the proposed cooperative cache optimization method based on content similarity can also reduce the data acquisition overhead and improve the spectrum efficiency of the macro cell base station.
In summary, the base station data cooperation caching model and method based on content similarity fully consider cooperation among micro-cellular base stations, reduce downloading of repeated data, improve the utilization rate of frequency spectrum resources and optimize data acquisition overhead through a cooperation caching strategy on the basis of similar data content requirements. The method utilizes local information interaction to realize the optimization of the whole network data downloading overhead progressively. Compared with the method for forming the non-overlapping alliance with the base station in the existing document of the user, the method provided by the invention can be found out that the resource allocation can be more reasonably optimized and the cooperation efficiency can be improved by reasonably planning the cache source of single data, so that better performance can be obtained. Simulation results also indicate the effectiveness of the model and method.
Having described embodiments of the present invention, the foregoing description is intended to be exemplary, not exhaustive, and not limited to the embodiments disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments.

Claims (7)

1. A base station data cooperation caching method based on content similarity is characterized in that for any micro-cellular base station with content requirements in a heterogeneous network, other micro-cellular base stations are divided into neighbor base stations and non-neighbor base stations according to communication ranges; when the data content of the micro-cellular base station is acquired, the data is intensively downloaded through the cooperation of the neighbor base stations with similar content.
2. The cooperative buffering method for base station data based on content similarity as claimed in claim 1, wherein the method comprises the following steps:
step 1, modeling a base station cooperation cache problem into a generalized local mutual profit game model, wherein participants of the game are all microcellular base stations with data requirements in a network;
step 2, for any micro-cellular base station, dividing other base stations into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to the communication range, and defining a utility function to represent the data requirement acquisition overhead;
step 3, initializing all base stations to randomly generate a cache strategy, and calculating utility function values under the current strategy; one base station is randomly generated, another cache strategy is randomly generated, and a utility function value under the current strategy is calculated; for the selected base station, respectively acquiring the two cache strategies as the probability for updating the cache strategy at the next moment, and preferentially updating the cache strategy of the selected base station based on the probability;
and 4, circularly selecting the base stations randomly in the step 3 to update the cache strategy until the selection of all the base stations realizes convergence and the set iteration times are reached.
3. The cooperative caching method for the base station data based on the content similarity as claimed in claim 2, wherein the step 1 models the cooperative caching problem of the base station as a game model, and the game model is defined as:
G＝{N,A,,un} (1)
the game model G comprises four components, wherein N is a microcellular base station set participating in a game, N represents the total number of game participants, N represents the number of the game participants, A is a strategy action space of a base station, is a network topology and comprises a one-hop neighbor base station set J of the base station NnAnd two-hop neighbor base station set
Figure FDA0002309378760000011
unIs the utility function of base station n.
4. The cooperative caching method for base station data based on content similarity according to claim 2, wherein in step 2, defining a utility function to represent the data requirement acquisition overhead is specifically as follows:
step 2-1, defining the data requirement of the micro-cellular base station n as
Figure FDA0002309378760000012
Wherein; k represents a data requirement number; lnIs the total data demand of the femtocell base station n; dnkRepresenting the kth data requirement of the femtocell base station n; defining the caching strategy of the data required by the base station n as
Figure FDA0002309378760000021
Wherein, ankFor corresponding data demand dnkThe caching strategy of (a), which base station or macro cell base station is selected from one-hop neighbors as the caching source, isnk∈Jn∪ {0}, where 0 represents an acquisition from a macrocell base station existing in the network, JnA set of neighbor base stations representing base station n; let ankJ, then base station n data requirement dnkHas an acquisition overhead of fnk(an,a-n)：
Figure FDA0002309378760000022
Wherein, a-nRepresenting the caching strategies of the rest base stations except the base station n; j > 0 indicates data acquisition from the neighbor base station j, KjIs the sum of all buffered data downloaded from the macro cell, BS, of base station jjIs the basic overhead of base station j operation, αj0Is the unit overhead for base station j to acquire data from the macro cell,j(dnk) For data request d received at base station jnkNumber of requests, BSj+Kjαj0Being the sum of the base overhead of base station j and the overall data acquisition cost,
Figure FDA0002309378760000023
in order to achieve the cost of acquiring a unit of data,
Figure FDA0002309378760000024
i.e. the data d of the base station j after being sharednkCost of αjnIs the transmission cost between base stations j and n; j ═ 0 represents the data acquisition overhead required to acquire the data from the macrocell; knIs the sum of all buffered data downloaded from the macro cell, BS, of base station nnIs the basic overhead of base station n operation, αn0Is the unit overhead of base station n acquiring data from the macro cell,n(dnk) For the data request d received at base station nnkNumber of requests, BSn+Knαn0Being the sum of the base overhead of base station n and the overall data acquisition cost,
Figure FDA0002309378760000025
in order to achieve the cost of acquiring a unit of data,
Figure FDA0002309378760000026
i.e. after sharing, the base station n data dnkThe overhead of (c);
step 2-2, calculating the sum of the data requirement acquisition overhead of the femtocell base station n:
Figure FDA0002309378760000027
step 2-3, expanding the traditional local mutual interest game from the relation between the base station and the one-hop neighbor to the relation between the base station and the one-hop neighbor and the two-hop neighbor to form a generalized local mutual interest game, and establishing a utility function u of the micro-cellular base station nnComprises the following steps:
Figure FDA0002309378760000028
wherein: r isnRepresenting the data requirement acquisition overhead of node n, i represents the number of one-hop neighbor base station and two-hop neighbor base station, riRepresenting the corresponding data demand acquisition overhead.
5. The cooperative caching method for base station data based on content similarity according to claim 3 or 4, wherein the policy update process of the base station caching policy optimization method in step 3 is specifically as follows:
step 3-1, initializing, and randomly generating a cache strategy for each microcellular base station N belonging to N;
step 3-2, detection: all the femtocell base stations and one-hop neighbors interact with the caching strategy and the file requirement information, one femtocell base station n is randomly selected for operation, and other all users repeat the previous caching selection; for the selected femtocell base station n, calculating its current strategy an(t) value of utility function
Figure FDA0002309378760000031
Randomly generating another policy
Figure FDA0002309378760000032
Re-interacting information and calculating utility function values
Figure FDA0002309378760000033
Step 3-3, strategy updating: the base station n preferentially updates the cache strategy according to the utility function values corresponding to the two different strategies, namely
Figure FDA0002309378760000034
Wherein the content of the first and second substances,
Figure FDA0002309378760000035
an(t) represents the strategy of base station n at the tth iteration; pr [ a ]n(t+1)＝an(t)]Indicating base station n selection strategy an(t) as the probability of the next moment strategy;
Figure FDA0002309378760000036
representing base station n selection strategies
Figure FDA0002309378760000037
β is a learning factor with a value of 0.1-0.3.
6. The method according to claim 2, wherein step 4 includes two steps of detecting and policy updating in the loop step 3, and the base station optimizes the cooperative caching policy through exploration and learning until the caching policies of all the femtocell base stations converge or a set number of iterations is reached, specifically as follows:
4-1, performing information interaction on all the micro-cells;
4-2, randomly selecting one base station for operation in each iteration;
and 4-3, repeating the previous data demand caching strategy by all other base stations.
7. A model adopted by the content similarity-based base station data cooperative caching method according to one of claims 2 to 6, wherein the model comprises:
the generalized local mutual profit game model building module: modeling a base station cooperation cache problem into a generalized local mutual profit game model, wherein participants of the game are all microcellular base stations with data requirements in a network;
a utility function definition module: for any micro-cellular base station, dividing other base stations into one-hop neighbor base stations, two-hop neighbor base stations and non-neighbor base stations according to the communication range, and defining a utility function to represent the data demand acquisition overhead;
the cache strategy updating module: for all base stations, initializing and randomly generating a cache strategy, and calculating utility function values under the current strategy; one base station is randomly generated, another cache strategy is randomly generated, and a utility function value under the current strategy is calculated; for the selected base station, respectively acquiring the two cache strategies as the probability for updating the cache strategy at the next moment, and preferentially updating the cache strategy of the selected base station based on the probability;
a loop policy update module: and circularly and randomly selecting the base stations to update the cache strategy until the selection of all the base stations is converged to reach the set iteration times.
CN201911252355.9A 2019-12-09 2019-12-09 Base station data cooperation caching model and method based on content similarity Pending CN111556510A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911252355.9A CN111556510A (en) 2019-12-09 2019-12-09 Base station data cooperation caching model and method based on content similarity

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911252355.9A CN111556510A (en) 2019-12-09 2019-12-09 Base station data cooperation caching model and method based on content similarity

Publications (1)

Publication Number Publication Date
CN111556510A true CN111556510A (en) 2020-08-18

Family

ID=72007199

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911252355.9A Pending CN111556510A (en) 2019-12-09 2019-12-09 Base station data cooperation caching model and method based on content similarity

Country Status (1)

Country Link
CN (1) CN111556510A (en)

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100281208A1 (en) * 2009-04-30 2010-11-04 Qing Yang System and Method for Data Storage
CN108259628A (en) * 2018-02-28 2018-07-06 重庆邮电大学 Content caching and user-association combined optimization method in isomery cellular network
CN108834080A (en) * 2018-04-17 2018-11-16 东南大学 Distributed caching and user-association method in heterogeneous network based on multicasting technology

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100281208A1 (en) * 2009-04-30 2010-11-04 Qing Yang System and Method for Data Storage
CN108259628A (en) * 2018-02-28 2018-07-06 重庆邮电大学 Content caching and user-association combined optimization method in isomery cellular network
CN108834080A (en) * 2018-04-17 2018-11-16 东南大学 Distributed caching and user-association method in heterogeneous network based on multicasting technology

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
YULI ZHANG: "A Game-Theoretic Approach for Optimal Distributed Cooperative Hybrid Caching in D2D Networks", 《IEEE WIRELESS COMMUNICATIONS LETTERS》 *
YULI ZHANG: "A Multi-Leader One-Follower Stackelberg Game Approach for Cooperative Anti-Jamming No Pains No Gains", 《IEEE COMMUNICATIONS LETTERS》 *

Similar Documents

Publication Publication Date Title
CN108809695B (en) Distributed uplink unloading strategy facing mobile edge calculation
Zhang et al. Energy-latency tradeoff for energy-aware offloading in mobile edge computing networks
Zhang et al. Energy-efficient offloading for mobile edge computing in 5G heterogeneous networks
Wang et al. A probability preferred priori offloading mechanism in mobile edge computing
CN103702337B (en) Determining method for small-scale base station deployment position
CN106912079B (en) Combined user access selection and resource allocation method in cache heterogeneous network
Ge et al. Optimization on TEEN routing protocol in cognitive wireless sensor network
TWI753175B (en) Electronic apparatus, method and computer-readable storage medium for wireless communications
CN108712746B (en) Partially overlapped channel aggregation game model and learning method
CN112512065B (en) Method for unloading and migrating under mobile awareness in small cell network supporting MEC
Zhong et al. Energy allocation and utilization for wirelessly powered IoT networks
Zhang et al. DMRA: A decentralized resource allocation scheme for multi-SP mobile edge computing
Deng et al. Throughput maximization for multiedge multiuser edge computing systems
CN104684095A (en) Resource allocation method based on genetic operation in heterogeneous network convergence scenes
Zhang et al. A game-theoretic approach for optimal distributed cooperative hybrid caching in D2D networks
Xia et al. Near-optimal and learning-driven task offloading in a 5G multi-cell mobile edge cloud
Zhou et al. Multi-server federated edge learning for low power consumption wireless resource allocation based on user QoE
Shi et al. Services management and distributed multihop requests routing in mobile edge networks
CN112004265B (en) Social network resource allocation method based on SRM algorithm
Yu et al. Connected dominating set construction in cognitive radio networks
Lin Dynamic resource allocation strategy in mobile edge cloud computing environment
CN109104768B (en) Non-orthogonal multiple access joint bandwidth and rate allocation method based on simulated annealing algorithm
CN111556510A (en) Base station data cooperation caching model and method based on content similarity
CN115391962A (en) Communication base station and power distribution network collaborative planning method, device, equipment and medium
Li et al. Cooperative resource allocation for computation offloading in mobile-edge computing 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
WD01 Invention patent application deemed withdrawn after publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20200818