WO2023159986A1 - 一种分层网络体系结构中的协作缓存方法 - Google Patents
一种分层网络体系结构中的协作缓存方法 Download PDFInfo
- Publication number
- WO2023159986A1 WO2023159986A1 PCT/CN2022/126959 CN2022126959W WO2023159986A1 WO 2023159986 A1 WO2023159986 A1 WO 2023159986A1 CN 2022126959 W CN2022126959 W CN 2022126959W WO 2023159986 A1 WO2023159986 A1 WO 2023159986A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- nodes
- node
- file
- cluster
- cache
- 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.)
- Ceased
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/10—Flow control between communication endpoints
- H04W28/14—Flow control between communication endpoints using intermediate storage
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/46—Cluster building
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/44—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for communication between vehicles and infrastructures, e.g. vehicle-to-cloud [V2C] or vehicle-to-home [V2H]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/40—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P]
- H04W4/46—Services specially adapted for particular environments, situations or purposes for vehicles, e.g. vehicle-to-pedestrians [V2P] for vehicle-to-vehicle communication [V2V]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
- H04W40/32—Connectivity information management, e.g. connectivity discovery or connectivity update for defining a routing cluster membership
-
- 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
Definitions
- the invention relates to the technical field of vehicle traffic, in particular to a collaborative caching method in a layered network architecture under an Internet of Vehicles environment.
- V2V vehicle-to-vehicle
- V2I vehicle-to-infrastructure
- MBS macro base station
- the patent No. CN112839082A mentions a heterogeneous edge cache allocation method, system, medium and application. While considering content cache allocation, it also takes power allocation into account, and allocates different transmission power for different content. In order to achieve the optimization of user experience. But this patent does not involve the problem of encoding cache.
- the invention with the patent number CN113543071A mentions a heterogeneous node cooperative edge caching method in the Internet of Vehicles environment, aiming at minimizing the delay, cost and high hit rate required for the vehicle to obtain the total content, and collaboratively place the content. But what this invention adopted is the fountain code.
- the present invention provides a collaborative caching method in a layered network architecture under the Internet of Vehicles environment, aiming at selecting content caching nodes and coordinatingly placing content while minimizing energy consumption. Has higher content delivery efficiency.
- the present invention adopts the following technical solutions:
- the embodiment of the present invention proposes a cooperative caching method in a layered network architecture, and the cooperative caching method includes the following steps:
- the network edge cache model is set within the coverage of a macro base station, including a typical layered network architecture of two layers of core layer and edge layer;
- the macro base station and each vehicle it contains are regarded as one of the cache nodes, and the nodes in the edge layer cooperate to perform content caching;
- the core layer including the macro base station supports large bandwidth and fast data packet transmission At the same time, cache the uncached part of the edge layer;
- the nodes in the edge layer are divided into multiple clusters according to the region, and the number of nodes in each cluster does not exceed the threshold ⁇ ; all nodes belonging to a specific cluster cooperate to perform caching by updating the cache information to their cluster heads, and the node threshold ⁇ related to the communication overhead within the cluster nodes;
- S3 model a set of popular files through zipper distribution, and find the most popular files in different vehicle nodes for different users;
- the content is divided into separate encoding packages, and the nth file is divided into l n file segments, and each file is equal in size; at the same time, the files in the popular file set are ranked from high to high according to their popularity The lower order is cached in turn, and the content placement matrix m is constructed; when the user u 1 requests the nth file, the m kn file segments are obtained from the kth vehicle node in the locally associated cluster, and the remaining l n -m kn file segments will be obtained through the coded multicast transmission of the macro base station;
- step S1 nodes in the edge layer cooperate to perform content caching, which means that only selected vehicles are allowed to cache other vehicles' caching event information.
- step S2 the process of dividing the nodes in the edge layer into multiple clusters by region includes the following steps:
- the vehicle nodes in the same area are divided into groups, and the number of nodes in each cluster does not exceed the threshold ⁇ .
- Each group corresponds to an edge cluster, and each cluster is further divided into various node types: cluster head Nodes, gateway nodes, and cluster members; all nodes belonging to a particular cluster perform caching cooperatively by updating cache information to their cluster head nodes, and there is no direct communication between cluster members.
- the selection process of the cluster head node includes the following steps:
- step S3 a group of popular file sets are modeled through zipper distribution, and the process of finding the most popular files in different vehicle nodes for different users includes the following steps:
- q kn represents the probability that a user of the k-th SBS requests the n-th file
- ⁇ k is the compression index, which is used to describe the skewness in the request pattern, and a larger ⁇ k means that content requests are more concentrated in high-ranking on the content file
- ⁇ kn is the ranking order of the nth file in the kth node
- step S4 the content placement matrix m is:
- m kn represents the number of file segments that the kth vehicle node caches the nth file in each cluster; the value of m satisfies the following constraints:
- M k represents the maximum capacity that each vehicle node can cache
- M represents the size of the total file
- step S5 the modeling of the optimized content placement matrix is:
- the constraints C1 and C2 are the restrictions on the cache capacity, s 0 is the size of each file segment, M is the size of the total file; E C is the energy consumption of the content cache, is the energy consumption of each vehicle node transmission, Indicates the probability that the user requests the file n times within the range of the kth node to be zero; is the transmission energy consumption of MBS, Indicates the energy consumption per bit of data transmitted by MBS, B indicates the total number of bits transmitted by MBS through coded multicast; N is the total number of files in the popular file set.
- the cooperative caching method in the layered network architecture proposed by the present invention for coded caching, in the placement stage, the content is not coded and stored in the edge server, but in the transmission stage, different users can be satisfied at the same time through coded multicast packets ask.
- the present invention can realize the best memory rate trade-off.
- FIG. 1 is a schematic structural diagram of a network edge cache model according to an embodiment of the present invention.
- Fig. 2 is a flowchart of a cooperative caching method in a layered network architecture according to an embodiment of the present invention.
- Fig. 2 is a flowchart of a cooperative caching method in a layered network architecture according to an embodiment of the present invention.
- the caching method includes the following steps:
- Figure 1 shows a two-layer hierarchical vehicle network architecture with caching capabilities.
- the model construction principle of this embodiment includes: designing a two-layer typical layered network architecture composed of a core layer and an edge layer, the core layer includes macro base stations and remote servers, and the edge layer is composed of vehicle users; The macro base station and each vehicle contained in it are regarded as one of the caching nodes. The nodes in the edge layer cooperate to perform content caching.
- the core layer supports large bandwidth and fast data packet transmission, while caching the uncached part of the edge layer.
- Each cluster is further divided into various node types, Cluster Head Node (CH), Gateway Node (GN) and Cluster Member (CM). All nodes belonging to a particular cluster perform caching cooperatively by updating cache information to their cluster head (CH), where, unlike traditional cooperative caching, no direct communication between members of a single cluster is possible.
- CH Cluster Head Node
- GN Gateway Node
- CM Cluster Member
- CH is mainly responsible for caching the updated cached event information of all its CMs, and provides communication links between nodes in its own cluster and nodes in adjacent clusters through GN.
- the main responsibility of GN is to transmit data packets between clusters and participate in the selection process of cluster heads.
- the rest of the nodes are regarded as cluster members, and the CM is the node that actually caches according to the popularity of the content, and all the CMs participate in the selection process of the cluster head.
- Clustering head CH selection method The traditionally designed WCA algorithm considers four factors to calculate the weight function between the node degree of a specific node and the node threshold ⁇ . For WCA, a weight function is computed using all factors together, which in turn produce a value to score each network node. The node with the lowest score among all nodes is selected as the cluster head. This patent also follows the WCA method for clustering, while taking into account the different factors shown in the following formula:
- w 1 , w 2 , w 3 , w 4 , and w 5 are weight factors corresponding to different parameters, expressed as:
- the first factor corresponds to the degree d of node k (i.e., the nodes within the transmission range of node k, also called neighbors of node k), and is calculated as:
- T k is calculated as:
- S k is the set of nodes within the range of k and t q, k, k represents the transmission time from node k to any node q, and the average transmission time of all network nodes is expressed by express.
- the third factor is the weighted sum H k of the distances between node k and neighbor hops.
- the hop count from CH to CM should be kept to a minimum.
- the value of H k is used In the formula, count m is the count of the m-th hop neighbor of node k, and ⁇ m is the weight factor.
- the average weighted sum of all system wide hops is calculated with express.
- the fourth factor is to designate node k as the cumulative power P k during CH, the average power utilization of all network nodes can be calculated as
- the fifth factor is the cache utilization Ck of node n, mainly in CH, the average cache space utilization of all network nodes can be calculated as
- Cluster member CM selection method In order to reduce the communication overhead within the cluster nodes, a node threshold ⁇ is set for each cluster, that is, the upper limit of the maximum number of nodes that can be accommodated in a cluster except CH. This limits the maximum number of nodes the cluster can manage.
- S3 models a set of popular file sets through a zipper distribution, looking for different users' preferences for files in different vehicle nodes.
- the probability of a user requesting the nth file in the kth node is calculated according to the following formula:
- q kn represents the probability that the user of the kth SBS requests the nth file
- ⁇ k is the compression index, which is used to describe the skewness in the request pattern, and the larger ⁇ k means that the content requests are more concentrated in the high-ranking on content files
- ⁇ kn is the ranking order of the nth file in the kth node.
- the number of user requests for the nth file in the coverage of the kth vehicle node follows a Poisson process. Therefore, within the multicast period t, the probability that all users within the coverage of the kth node request a file for a total of j times can be written as:
- each node in the edge layer can only cache some files.
- the content is divided into separate encoded packets, and the nth file is divided into l n file segments, and the size of each file segment is equal to s 0 , m kn represents the kth edge device in each cluster
- the content placement matrix m can be expressed as:
- nth file segments When user u 1 requests the nth file, it needs to obtain its l n file segments, m kn file segments are obtained from the kth vehicle node in the locally associated cluster, and the remaining l n -m kn file segments will be obtained through The coded multicast transmission of the macro base station is obtained.
- the specific value of the content placement matrix m can be obtained through optimization to form a content placement strategy. Due to the cache size of the vehicle node, the value of m should satisfy the following constraints:
- M k represents the maximum capacity that each vehicle node can cache
- M represents the size of the total file
- each file is partially cached in the vehicle node, resulting in energy consumption.
- cache power consumption is proportional to the number of file bits stored on each node.
- the cache energy consumption can be obtained by calculating the number of data bits cached by each node. The calculation is expressed as:
- e ca is the energy coefficient of the vehicle node cache in joules/bit
- s 0 is the size of each file segment.
- the energy consumption of vehicle node transmission is analyzed, that is, In the transfer phase, when a user initiates a request, the associated node will first transfer its cached file segments to the user. The remaining file segments are transmitted by the MBS to the relevant vehicle nodes, and then the associated nodes transmit them to the user. Therefore, the energy consumption of transferring files from the vehicle node includes two parts: one is the energy consumption of transmitting its own cached file segments to the user, and the other is the energy consumption of uncached file segments obtained from the MBS. The total energy consumption is calculated as:
- the energy consumption of transferring files on MBS is mainly proportional to the number of bits of files transferred on MBS, and the energy consumption of MBS transmission can be expressed as:
- K j represents a vehicle node set composed of j, 1 ⁇ j ⁇ K
- the kth vehicle node content placement optimization problem is modeled as:
- constraints C1 and C2 are limitations on the cache capacity.
- the content placement of the kth vehicle node By calculating the content placement of the kth vehicle node, the content placement of all vehicle nodes in each cluster is deduced in this way, and the specific value of the content placement matrix m is obtained.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Information Transfer Between Computers (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
一种分层网络体系结构中的协作缓存方法,包括:构建核心服务器、宏基站以及各车辆之间的网络边缘缓存模型;将边缘层中的节点按区域划分为多个集群,每个集群节点数不超过阈值β;将一组流行文件集通过拉链分布建模,寻找不同车辆节点中最流行的文件;通过使用编码缓存,把内容分割为分离的编码包,构建内容放置矩阵m;以最小化总能耗为优化目标,进一步优化内容放置矩阵m的具体值,制定内容放置策略。本方法旨在最小化能耗的情况下,选择内容缓存节点,并对内容进行协作放置,具有更高的内容传递效率。
Description
本发明涉及车辆交通技术领域,具体而言涉及一种车联网环境下分层网络体系结构中的协作缓存方法。
智能交通系统在当今社会发挥着至关重要的作用,大多数自动驾驶汽车用户的兴趣更加关注娱乐服务。用户在互联网上的请求出现了前所未有的增长,通常包括在不同的平台上观看和分享视频、音频和图像。为了获得更好的服务质量,无缝的内容交付显得至关重要。然而,在高度动态的环境下,很难实现长期的车对车(V2V)和车对基础设施(V2I)的连接。此外,路侧单元(RSU)和基站(BS)等基础设施组件的短传输范围将为V2I通信提供低效的解决方案。由于范围有限和高移动性,频繁的连接丢失和连接重建将带来大量的网络流量开销和内容检索延迟。
在物联网中,设备将连接到网络,并将收集到的数据共享到云服务器。在云服务器上进行数据分析和处理后,用户经常会请求有用的信息。不同的用户可以在同一时间段内请求相同的内容,从而导致相同的数据内容的重复传输。这促使运营商重新考虑他们当前的网络架构,并寻求更先进的技术,以低延迟和节能的方式使内容更接近最终用户。移动边缘计算/缓存在宏基站(MBS)覆盖范围内的设备,通过分层异构网络架构,使内容缓存降低能耗。
然而,由于边缘设备内存有限,以及车辆的高移动速度,在异构车辆网络中多层边缘缓存服务器(例如MBS和RSU)的交错覆盖范围内,不同服务器之间的跨层协作可以提供无缝连接,以方便内容下载。
专利号为CN112839082A的发明中提及一种异构边缘缓存分配方法、系统、介质及应用,在考虑内容缓存分配的同时也将功率分配问题考虑在内,为不同的内容分配不同的传输功率,以达到用户体验的最优化。但该专利不涉及编码缓存的问题。
专利号为CN113543071A的发明中提及一种车联网环境下异构节点协作边缘缓存方法,旨在最小化车辆获取总内容所需延迟、成本以及高命中率的情况下,对内容进行协作放置。但该发明采用的是喷泉码。
发明内容
本发明针对现有技术中的不足,提供一种车联网环境下分层网络体系结构中的协作缓存方法,旨在最小化能耗的情况下,选择内容缓存节点,并对内容进行协作放置,具有更高的 内容传递效率。
为实现上述目的,本发明采用以下技术方案:
本发明实施例提出了一种分层网络体系结构中的协作缓存方法,所述协作缓存方法包括以下步骤:
S1,构建核心服务器、宏基站以及各车辆之间的网络边缘缓存模型,该网络边缘缓存模型设置在一个宏基站覆盖范围内,包含核心层和边缘层的两层典型的分层网络体系结构;其中,将该宏基站和其所包含的每个车辆均视为其中一个缓存节点,边缘层中的节点协作执行内容缓存;包括宏基站在内的核心层支持大带宽和快速的数据包传输的同时,缓存边缘层未缓存部分;
S2,将边缘层中的节点按区域划分为多个集群,每个集群节点数不超过阈值β;属于一个特定集群的所有节点通过将缓存信息更新到其集群头来协同执行缓存,节点阈值β与集群节点内的通信开销相关;
S3,将一组流行文件集通过拉链分布建模,寻找不同的用户对不同车辆节点中最流行的文件;
S4,通过使用编码缓存,把内容分割为分离的编码包,将第n个文件划分成l
n个文件段,且每个文件大小相等;同时将流行文件集中的文件按照受欢迎程度由高到低的顺序依次进行缓存,构建内容放置矩阵m;当用户u
1请求第n个文件时,其中m
kn个文件段从本地关联的簇中第k个车辆节点获得,剩余的l
n-m
kn个文件段将通过宏基站的编码多播传输获得;
S5,以最小化总能耗为优化目标,进一步优化内容放置矩阵m的具体值,制定内容放置策略。
进一步地,步骤S1中,边缘层中的节点协作执行内容缓存是指,只允许选定的车辆缓存其他车辆缓存事件的信息。
进一步地,步骤S2中,将边缘层中的节点按区域划分为多个集群的过程包括以下步骤:
将整个边缘网络进行划分,同一片区域里的车辆节点分为一组,且每个集群节点数不超过阈值β,每组对应一个边缘簇,每个簇进一步分为各种节点类型:集群头节点、网关节点和集群成员;属于一个特定集群的所有节点通过将缓存信息更新到其集群头节点来协同执行缓存,集群成员之间不存在直接通信。
进一步地,集群头节点的选择过程包括以下步骤:
采用下述公式计算集群内每个节点的优先级评估值:
式中,w是不同参数对应的权重因子,w
1+w
2+w
3+w
4+w
5=1;d
k是节点k范围内的节点数,
是所有网络节点的节点度的平均逆;T
k是节点k与其范围内的节点的传输时间,
是所有网络节点的平均传输时间,H
k是节点k与其范围内所有节点距离的加权和,
是所有系统宽跳的平均加权和,p
k是节点k的传输功率,
是所有网络节点的平均功率利用率,c
k是节点k的高速缓存利用率,
是所有网络节点的平均缓存空间利用率。
进一步地,步骤S3中,将一组流行文件集通过拉链分布建模,寻找不同的用户对不同车辆节点中最流行的文件的过程包括以下步骤:
S31,根据下述公式计算一个用户请求第k个节点中的第n个文件的概率:
式中,q
kn表示第k个SBS的用户请求第n个文件的概率;α
k是压缩指数,用于描述请求模式中的偏态,α
k越大意味着内容请求更集中在高排名的内容文件上;γ
kn是第k个节点中第n个文件的排名顺序;
S32,在每个传输周期t中,第k个车辆节点将平均接收来自其覆盖区域内所有相关用户的U
k请求;在第k个节点的覆盖范围内,第n个文件的平均请求数定义为ω
kn,用ω
kn=U
kq
kn表示;
S33,在多播周期t内,在第k个节点覆盖范围内的所有用户请求文件共j次的概率为:
进一步地,步骤S4中,所述内容放置矩阵m为:
式中,m
kn表示每个簇内第k个车辆节点缓存第n个文件的文件段的数量;m的值满足以下约束条件:
式中,M
k表示每个车辆节点可以缓存的最大容量,M表示总文件的大小。
进一步地,步骤S5中,优化内容放置矩阵的建模为:
式中,限制条件C1和C2是对缓存容量的限制,s
0是每个文件段的大小,M是总文件的大小;E
C是内容缓存的能量消耗,
是各车辆节点传输的能耗,
表示用户在第k个节点范围内请求文件n次数为零的概率;
是MBS的传输能耗,
表示MBS传输的每比特数据的能量消耗,B表示MBS通过编码组播传输的总比特数;N为流行文件集的文件总数。
本发明的有益效果是:
本发明提出的分层网络体系结构中的协作缓存方法,对于编码缓存,在放置阶段,内容未编码并存储在边缘服务器中,但在传输阶段,可以通过编码的多播包同时满足不同的用户请求。本发明可以实现最佳的内存率权衡。
图1是本发明实施例的网络边缘缓存模型的结构示意图。
图2为本发明实施例的分层网络体系结构中的协作缓存方法流程图。
现在结合附图对本发明作进一步详细的说明。
需要注意的是,发明中所引用的如“上”、“下”、“左”、“右”、“前”、“后”等的用语,亦仅为便于叙述的明了,而非用以限定本发明可实施的范围,其相对关系的改变或调整,在 无实质变更技术内容下,当亦视为本发明可实施的范畴。
图2为本发明实施例的分层网络体系结构中的协作缓存方法流程图。参见图2,该缓存方法包括以下步骤:
S1,构建核心服务器、宏基站以及各车辆之间的网络边缘缓存模型,该模型设置在一个宏基站覆盖范围内。
图1为具有缓存能力的、两层分层的车辆网络架构。基于图1,本实施例的模型构建原理包括:设计由核心层和边缘层组成的两层典型的分层网络体系结构,核心层包含宏基站以及远程服务器,边缘层由车辆用户组成;将该宏基站和其所包含的每个车辆均视为其中一个缓存节点,边缘层中的节点协作执行内容缓存,核心层支持大带宽和快速的数据包传输,同时缓存边缘层未缓存部分。
将所有的网络用户用集合U={u
1,u
2,...,u
M}表示,该网络的所有节点都用K表示,其中,k∈K指定任何网络节点,为了有效地利用资源,只允许选定的车辆缓存其他车辆缓存事件的信息;将流行文件集用Ν={1,2,...,N}表示;对应的文件大小用s={s
1,s
2,...,s
N}表示。
由于边缘层节点缓存空间有限,网络车辆速度快,用户只能接收所请求的多媒体文件的一部分,通过编码传输将剩余部分缓存在核心层。
S2,划分边缘层中的所有节点为多个集群。
每个集群进一步分为各种节点类型,集群头节点(CH)、网关节点(GN)和集群成员(CM)。属于一个特定集群的所有节点通过将缓存信息更新到其集群头(CH)来协同执行缓存,其中,与传统的协作缓存不同,单个集群的成员之间不可能存在直接通信。
CH主要负责缓存其所有CM的更新后的缓存事件信息,并且通过GN提供自己集群中的节点和邻近集群中的节点的通信链接。GN的主要职责是集群间的数据包传输,并且参与集群头的选择过程。剩余节点都被视为集群成员,CM是根据内容的流行程度而实际进行缓存的节点,所有的CM都参与了集群头的选择过程。
聚类头CH选择方法:传统设计的WCA算法考虑四个因素来计算特定节点的节点度与节点阈值β之间的权重函数。对于WCA,一个权重函数是使用所有因素一起计算的,这些因素反过来产生一个值来对每个网络节点进行评分。选择所有节点中得分最低的节点作为集群头。本专利也遵循WCA方法进行聚类,同时考虑了下述公式所示的不同因素:
式中,w
1、w
2、w
3、w
4、w
5是不同参数对应的权重因子,表现为:
w
1+w
2+w
3+w
4+w
5=1。
第一个因素对应于节点k的d
k度(即节点k的传输范围内的节点,也称为节点k的邻居),并计算表示为:
第二个因素是从节点k到k范围内所有节点的平均传输时间T
k。这样,T
k被计算表示为:
第三个因素是节点k与邻居跳点距离的加权和H
k。为了减少发送和接收缓存信息的网络负载,从CH到CM的跳点计数应保持在最低限度。H
k的值是用
表示,式中count
m为节点k的第m跳个邻居的计数,γ
m为权值因子。同样,所有系统宽跳的平均加权和的计算用
表示。
集群成员CM选择方法:为了减少集群节点内的通信开销,为每个集群设置一个节点阈值β,即,一个集群内除CH外可容纳的最大节点数的上限。从而限制了集群可以管理的最大节点数量。
S3,将一组流行文件集通过拉链分布建模,寻找不同的用户对不同车辆节点中文件的首选项。
不同的用户对不同车辆节点中的文件有不同的首选项,因此一个车辆节点中的流行文件在其他节点中可能不流行。假设文件的流行度分布遵循拉链式分布,根据下述公式计算一个用户请求第k个节点中的第n个文件的概率:
式中,q
kn表示第k个SBS的用户请求第n个文件的概率,α
k是压缩指数,用于描述请求模式中的偏态,α
k越大意味着内容请求更集中在高排名的内容文件上;γ
kn是第k个节点中第n个文件的排名顺序。
在每个传输周期t中,第k个车辆节点将平均接收来自其覆盖区域内所有相关用户的U
k请求。在第k个节点的覆盖范围内,第n个文件的平均请求数定义为ω
kn,用ω
kn=U
kq
kn表示。
用户在第k个车辆节点覆盖范围中对第n个文件的请求数遵循泊松过程。因此,在多播周期t内,在第k个节点覆盖范围内的所有用户请求文件共j次的概率可以写为:
S4,通过使用编码缓存,把内容分割为分离的编码包。
由于每个集群节点缓存大小的限制和文件的异构性,边缘层中每个节点只能缓存部分文件。通过使用编码缓存,把内容分割为分离的编码包,将第n个文件划分成l
n个文件段,且每个文件段大小相等为s
0,m
kn表示每个簇内第k个边缘设备缓存第n个文件的文件段的数量,因此,内容放置矩阵m可以表示为:
用户u
1请求第n个文件,则需要得到它的l
n个文件段,m
kn个文件段从本地关联的簇中第k个车辆节点获得,剩余的l
n-m
kn个文件段将通过宏基站的编码多播传输获得。通过优化可以得到内容放置矩阵m的具体值,形成一种内容放置策略。由于车辆节点的高速缓存大小,m的值应该满足以下约束条件:
式中,M
k表示每个车辆节点可以缓存的最大容量,M表示总文件的大小。
S5,以最小化用户获取内容能量消耗为优化目标,优化网络边缘缓存模型,优化内容放置矩阵m的具体值,对车辆传输内容进行放置;其中,将流行文件集中的流行文件按照人气由高到低的顺序依次进行缓存。
首先分析内容缓存的能耗,定义为E
c,在放置阶段,每个文件都被部分缓存在车辆节点中,从而导致能源消耗。一般来说,缓存能耗与每个节点上存储的文件位数成正比。然后,只要已知内容放置矩阵m,就可以通过计算每个节点缓存的数据位数来获得缓存能耗,计算表示为:
式中,e
ca为焦耳/比特的车辆节点缓存的能量系数,s
0是每个文件段的大小。
其次分析车辆节点传输的能耗,即
在传输阶段,当用户发起一个请求时,关联的节点将首先向用户传输其高速缓存的文件段。而其余的文件段由MBS传输给相关的车辆节点,然后关联的节点将其传输给用户。因此,从车辆节点传输文件的能耗包括两部分:一是向用户传输自己缓存文件段的能耗,另一部分是从MBS获得的未缓存文件段的能耗,总能耗计算为:
由于车辆节点的缓存大小的限制,不能从集群中获得部分文件段,这些文件段将通过回程链路从MBS传输到车辆节点。因此,在MBS上传输文件的能耗主要与在MBS上传输的文件的位数成正比,MBS的传输能耗可以表示为:
假设每个车辆节点的缓存大小已知,将第k个车辆节点内容放置优化问题建模为:
式中,限制条件C1和C2是对缓存容量的限制。
通过求出第k个车辆节点的内容放置,以此类推出每个簇群中所有车辆节点的内容放置,得到内容放置矩阵m的具体值。
以上仅是本发明的优选实施方式,本发明的保护范围并不仅局限于上述实施例,凡属于本发明思路下的技术方案均属于本发明的保护范围。应当指出,对于本技术领域的普通技术人员来说,在不脱离本发明原理前提下的若干改进和润饰,应视为本发明的保护范围。
Claims (7)
- 一种分层网络体系结构中的协作缓存方法,其特征在于,所述协作缓存方法包括以下步骤:S1,构建核心服务器、宏基站以及各车辆之间的网络边缘缓存模型,该网络边缘缓存模型设置在一个宏基站覆盖范围内,包含核心层和边缘层的两层典型的分层网络体系结构;其中,将该宏基站和其所包含的每个车辆均视为其中一个缓存节点,边缘层中的节点协作执行内容缓存;包括宏基站在内的核心层支持大带宽和快速的数据包传输的同时,缓存边缘层未缓存部分;S2,将边缘层中的节点按区域划分为多个集群,每个集群节点数不超过阈值β;属于一个特定集群的所有节点通过将缓存信息更新到其集群头来协同执行缓存,节点阈值β与集群节点内的通信开销相关;S3,将一组流行文件集通过拉链分布建模,寻找不同的用户对不同车辆节点中最流行的文件;S4,通过使用编码缓存,把内容分割为分离的编码包,将第n个文件划分成l n个文件段,且每个文件大小相等;同时将流行文件集中的文件按照受欢迎程度由高到低的顺序依次进行缓存,构建内容放置矩阵m;当用户u 1请求第n个文件时,其中m kn个文件段从本地关联的簇中第k个车辆节点获得,剩余的l n-m kn个文件段将通过宏基站的编码多播传输获得;S5,以最小化总能耗为优化目标,进一步优化内容放置矩阵m的具体值,制定内容放置策略。
- 根据权利要求1所述的分层网络体系结构中的协作缓存方法,其特征在于,步骤S1中,边缘层中的节点协作执行内容缓存是指,只允许选定的车辆缓存其他车辆缓存事件的信息。
- 根据权利要求1所述的分层网络体系结构中的协作缓存方法,其特征在于,步骤S2中,将边缘层中的节点按区域划分为多个集群的过程包括以下步骤:将整个边缘网络进行划分,同一片区域里的车辆节点分为一组,且每个集群节点数不超过阈值β,每组对应一个边缘簇,每个簇进一步分为各种节点类型:集群头节点、网关节点和集群成员;属于一个特定集群的所有节点通过将缓存信息更新到其集群头节点来协同执行缓存,集群成员之间不存在直接通信。
- 根据权利要求1所述的分层网络体系结构中的协作缓存方法,其特征在于,步骤S3中,将一组流行文件集通过拉链分布建模,寻找不同的用户对不同车辆节点中最流行的文件的过程包括以下步骤:S31,根据下述公式计算一个用户请求第k个节点中的第n个文件的概率:式中,q kn表示第k个SBS的用户请求第n个文件的概率;α k是压缩指数,用于描述请求模式中的偏态,α k越大意味着内容请求更集中在高排名的内容文件上;γ kn是第k个节点中第n个文件的排名顺序;S32,在每个传输周期t中,第k个车辆节点将平均接收来自其覆盖区域内所有相关用户的U k请求;在第k个节点的覆盖范围内,第n个文件的平均请求数定义为ω kn,用ω kn=U kq kn表示;S33,在多播周期t内,在第k个节点覆盖范围内的所有用户请求文件共j次的概率为:
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210185518.1 | 2022-02-28 | ||
| CN202210185518.1A CN114666843B (zh) | 2022-02-28 | 2022-02-28 | 一种分层网络体系结构中的协作缓存方法 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023159986A1 true WO2023159986A1 (zh) | 2023-08-31 |
Family
ID=82026888
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2022/126959 Ceased WO2023159986A1 (zh) | 2022-02-28 | 2022-10-24 | 一种分层网络体系结构中的协作缓存方法 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114666843B (zh) |
| WO (1) | WO2023159986A1 (zh) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116866995A (zh) * | 2023-09-05 | 2023-10-10 | 江西师范大学 | 一种基于移动车辆的内容分层边缘缓存策略 |
| CN117460001A (zh) * | 2023-10-20 | 2024-01-26 | 山东科技大学 | 一种车联网中基于先验知识驱动的主动缓存决策方法 |
| CN118354370A (zh) * | 2024-04-24 | 2024-07-16 | 金陵科技学院 | 一种基于网络编码的车联网边缘节点内容缓存方法及系统 |
| CN119906971A (zh) * | 2025-01-15 | 2025-04-29 | 南京邮电大学 | 一种可感知车辆移动的路边单元内容编码放置系统及方法 |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114666843B (zh) * | 2022-02-28 | 2025-02-18 | 南京邮电大学 | 一种分层网络体系结构中的协作缓存方法 |
| CN116702904A (zh) * | 2023-04-28 | 2023-09-05 | 上海孚典智能科技有限公司 | 分阶段的人工智能算法推理计算方法及系统 |
| CN116761152B (zh) * | 2023-08-14 | 2023-11-03 | 合肥工业大学 | 路侧单元边缘缓存放置及内容交付方法 |
| CN118656352B (zh) * | 2024-08-19 | 2024-12-24 | 天津理工大学 | 一种基于文件分组和镜像流行度的数据删除重组方法 |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111741495A (zh) * | 2020-06-22 | 2020-10-02 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | 异构网络中高能效编码缓存内容放置方案设计方法 |
| CN112839082A (zh) * | 2020-12-31 | 2021-05-25 | 西安电子科技大学 | 一种异构边缘缓存分配方法、系统、介质及应用 |
| CN113543071A (zh) * | 2021-06-30 | 2021-10-22 | 南京邮电大学 | 一种车联网环境下异构节点协作边缘缓存方法 |
| CN114666843A (zh) * | 2022-02-28 | 2022-06-24 | 南京邮电大学 | 一种分层网络体系结构中的协作缓存方法 |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104750935A (zh) * | 2015-04-03 | 2015-07-01 | 福建省纳金网信息技术有限公司 | 一种用于拉链快速设计与服装快速展示的方法及系统 |
| CN113015105B (zh) * | 2021-03-08 | 2022-06-03 | 哈尔滨工业大学(深圳) | 空天地一体化网络的编码缓存内容放置与资源联合优化方法 |
-
2022
- 2022-02-28 CN CN202210185518.1A patent/CN114666843B/zh active Active
- 2022-10-24 WO PCT/CN2022/126959 patent/WO2023159986A1/zh not_active Ceased
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111741495A (zh) * | 2020-06-22 | 2020-10-02 | 哈尔滨工业大学(深圳)(哈尔滨工业大学深圳科技创新研究院) | 异构网络中高能效编码缓存内容放置方案设计方法 |
| CN112839082A (zh) * | 2020-12-31 | 2021-05-25 | 西安电子科技大学 | 一种异构边缘缓存分配方法、系统、介质及应用 |
| CN113543071A (zh) * | 2021-06-30 | 2021-10-22 | 南京邮电大学 | 一种车联网环境下异构节点协作边缘缓存方法 |
| CN114666843A (zh) * | 2022-02-28 | 2022-06-24 | 南京邮电大学 | 一种分层网络体系结构中的协作缓存方法 |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116866995A (zh) * | 2023-09-05 | 2023-10-10 | 江西师范大学 | 一种基于移动车辆的内容分层边缘缓存策略 |
| CN116866995B (zh) * | 2023-09-05 | 2023-12-01 | 江西师范大学 | 一种基于移动车辆的内容分层边缘缓存策略 |
| CN117460001A (zh) * | 2023-10-20 | 2024-01-26 | 山东科技大学 | 一种车联网中基于先验知识驱动的主动缓存决策方法 |
| CN118354370A (zh) * | 2024-04-24 | 2024-07-16 | 金陵科技学院 | 一种基于网络编码的车联网边缘节点内容缓存方法及系统 |
| CN118354370B (zh) * | 2024-04-24 | 2026-04-28 | 金陵科技学院 | 一种基于网络编码的车联网边缘节点内容缓存方法及系统 |
| CN119906971A (zh) * | 2025-01-15 | 2025-04-29 | 南京邮电大学 | 一种可感知车辆移动的路边单元内容编码放置系统及方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114666843A (zh) | 2022-06-24 |
| CN114666843B (zh) | 2025-02-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114666843B (zh) | 一种分层网络体系结构中的协作缓存方法 | |
| CN114338504B (zh) | 一种基于网络边缘系统的微服务部署和路由方法 | |
| CN110213627B (zh) | 一种基于多小区用户移动性的流媒体缓存分配方法 | |
| De Rango et al. | HED-FL: A hierarchical, energy efficient, and dynamic approach for edge Federated Learning | |
| WO2023168824A1 (zh) | 一种基于联邦学习的移动边缘缓存优化方法 | |
| CN111031102A (zh) | 一种多用户、多任务的移动边缘计算系统中可缓存的任务迁移方法 | |
| CN111741495B (zh) | 异构网络中高能效编码缓存内容放置方案设计方法 | |
| Li et al. | A survey of edge caching: Key issues and challenges | |
| Wu et al. | Federation-based deep reinforcement learning cooperative cache in vehicular edge networks | |
| CN113283177A (zh) | 一种基于异步联邦学习的移动感知缓存方法 | |
| CN114827198A (zh) | 一种应用于车联网的多层中心异步联邦学习方法 | |
| CN110730471A (zh) | 基于区域用户兴趣匹配的移动边缘缓存方法 | |
| CN115297170A (zh) | 一种基于异步联邦和深度强化学习的协作边缘缓存方法 | |
| CN113543071B (zh) | 一种车联网环境下异构节点协作边缘缓存方法 | |
| CN111885648A (zh) | 基于边缘缓存的能量有效的网络内容分发机制构建方法 | |
| Majidi et al. | HFDRL: An intelligent dynamic cooperate cashing method based on hierarchical federated deep reinforcement learning in edge-enabled IoT | |
| CN106686655A (zh) | 一种异构网络联合用户关联及内容缓存方法 | |
| Li et al. | Collaborative caching strategy based on optimization of latency and energy consumption in MEC | |
| CN112565377B (zh) | 车联网中一种面向用户服务体验的内容分级优化缓存方法 | |
| CN115278708A (zh) | 一种面向联邦学习的移动边缘计算资源管理方法 | |
| CN114374741A (zh) | Mec环境下基于强化学习的动态分组车联网缓存方法 | |
| CN118690873A (zh) | 面向边缘智能的联邦学习客户端资源异构方法及系统 | |
| CN112804103A (zh) | 区块链赋能物联网中联合资源分配与控制的智能计算迁移方法 | |
| CN108848395A (zh) | 基于果蝇优化算法的边缘协作缓存布置方法 | |
| CN117459112A (zh) | 基于图卷积网络的leo卫星网络中的移动边缘缓存方法及设备 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22928248 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 22928248 Country of ref document: EP Kind code of ref document: A1 |







