CN112399485A - CCN-based new node value and content popularity caching method in 6G - Google Patents

CCN-based new node value and content popularity caching method in 6G Download PDF

Info

Publication number
CN112399485A
CN112399485A CN202011199997.XA CN202011199997A CN112399485A CN 112399485 A CN112399485 A CN 112399485A CN 202011199997 A CN202011199997 A CN 202011199997A CN 112399485 A CN112399485 A CN 112399485A
Authority
CN
China
Prior art keywords
content
ccn
node
popularity
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.)
Withdrawn
Application number
CN202011199997.XA
Other languages
Chinese (zh)
Inventor
段玮
姜衍
季彦呈
王明星
卓碧婷
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.)
Nantong University
Original Assignee
Nantong University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Nantong University filed Critical Nantong University
Priority to CN202011199997.XA priority Critical patent/CN112399485A/en
Publication of CN112399485A publication Critical patent/CN112399485A/en
Withdrawn legal-status Critical Current

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
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W40/00—Communication routing or communication path finding
    • H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
    • H04W40/248—Connectivity information update

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

The invention discloses a method for caching new node value and content popularity based on a CCN (content centric networking) in 6G, which comprises the following steps of firstly caching content in the CCN to support various functions including content distribution and multicast; the node records corresponding state and interface information in the interest packet requesting process, and jumps back the data packet to the user according to the information; the forwarding information base FIB and the pending interest table PIT are kept inside the CCN node due to the content repository CS; the popularity of the content is estimated through the content request counting in the measuring process, so that the utilization rate of the cache space is further improved, and better cache performance is realized in the CCN; designs a variable
Figure DEST_PATH_FDA00027518445200000410
To match content popularity with a given node value, the proposed caching strategy is evaluated by the NDN-SIM simulator, followed by simulation result analysis.The method has the advantages of solving the problem of frequent replacement or redundancy of a large amount of contents in the traditional cache strategy, obviously reducing cache redundancy and content replacement frequency and improving the diversity of the contents.

Description

CCN-based new node value and content popularity caching method in 6G
Technical Field
The invention relates to a CCN-based new node value and content popularity caching method in 6G, and belongs to the technical field of communication.
Background
With the rapid development of the automobile internet in 6G (6G-IoV), the emerging wide-range network application, content-oriented and personalized information service, has become the mainstream trend of network development. Because the existing transmission control protocol/internet protocol (TCP/IP) architecture is based on endpoint addressing, frequent session connections cannot effectively guarantee reliable exchange of information, which brings about urgent problems of network resource waste, data redundancy, transmission congestion, and the like. The typical solution to solve the above problem is to build an overlay network, i.e. a Content Delivery Network (CDN) and peer-to-peer network (P2P) solution, at the application layer. Although these solutions can alleviate the problem of content distribution and sharing to some extent, it is difficult to adapt to various functions of the network layer and meet the requirement of high-speed transmission.
In addition, because the conventional Ad-hoc networks (VANETs) provide short connection distance and low delay, and the TCP/IP networking mode is difficult to meet these requirements, the content-centric networking (CCN) can be used to overcome the disadvantages, especially in 6G networks. Information-centric networks focus on content, breaking the traditional "host-to-host" communication mode. In particular, an "end-to-end" communication driven by the information provider will translate into a content retrieval driven by the recipient. The CCN, as a typical representative of a new information-centric network, uses content as a focal point and a basic unit of transmission, replaces the waist of an IP address with an hourglass structure, receives great attention, and becomes a research hotspot of a next-generation internet architecture. Unlike the Web, CDN and P2P, the main feature of CCN is in-network caching. The current research on CCN, RFC 8569, mainly includes distributing and sharing, placing policy, replacing policy and utilization rate, wherein the content placement is used to determine whether a node places content, and when the cache space of the node is saturated, the content replacing policy is used to replace old content with new content. In the CCN, when a request sent by a user finds corresponding content on a content router, the content is sent directly to the user without forwarding the request to an origin server. In this way, transmission delay and transmission pressure will be greatly reduced, improving user experience. Therefore, the formulation of the caching policy has a significant impact on the performance of the CCN, particularly with respect to the selection of cache locations and cache contents. Through reasonable selection of the cache position and the content, the user can more effectively acquire the content from the content router.
The Leave Copy Everywhere (LCE) in the prior art is the default caching policy for CCN, which requires the content router on the delivery path between the user and the server to cache each delivered content. This results in a large amount of content redundancy and less content diversity in the network. Probability (p), the probability of each content router caching content is p, and the non-caching probability is 1-p. When the content router receives the data packet, it randomly generates a number from 0 to 1. If the number is less than or equal to p, the content is cached, otherwise the content is directly forwarded to the next hop. ProbCache caches content in content routers according to a probability, where the probability is different for each content router, which is inversely proportional to the distance of the router from the user. Thus, the closer to the user, the greater the likelihood that the content will be cached. A Leave Copy Down (LCD) caches the content in the next hop content router only on a cache hit. However, the content needs to go through multiple requests to reach the edge of the network and a lot of content redundancy will result. In addition, when a cache hits, a Move Copy Down (MCD) will move the cached content from the hit node to the next hop content router (except for the server), which reduces cache redundancy. On the other hand, when the request comes from the user on different paths, the content caching position changes, and the dynamic nature generates more network overhead. Although the MCD and LCD work similarly, the MCD deletes the cache contents on the cache hit node (except for the server), thereby reducing the content redundancy, but at the same time, the dynamic nature of the cache node increases the network overhead. In addition, a cache strategy based on centrality is also provided, and the strategy utilizes the centrality of the nodes to improve the cache hit rate. Even so, caching content only at the hub node would cause other content routers to idle and content would be replaced frequently. While caching issues from content popularity have been considered, where the popularity of the content, cached content is classified as highly popular. Thus, only content with a higher popularity exists in the network, and the rest will be ignored, resulting in lower content diversity.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a CCN-based new node value and content popularity caching method in 6G, solves the problem that the traditional caching strategy is frequently replaced or a large amount of content is redundant, obviously reduces caching redundancy and content replacement frequency, and improves the diversity of the content. In this context, we also consider the impact of the neighbor nodes when selecting cache nodes, without placing content in the neighbor nodes. Simulation results show that the NVCP is superior to LCE, prob (0.5) and MPC in terms of cache hit rate, average hop count and average transmission delay.
In order to achieve the purpose, the invention adopts the technical scheme that: a new node value and content popularity caching method based on a CCN in 6G is disclosed, wherein communication of the CCN comprises two data packets: the interest data packet comprises a content name, a selector and a random number so as to forward the request of the user through the CCN node; a data packet consisting of the content name, the signature information and the data is sent to a satisfied user along the reverse path of the interest packet; the CCN uses the content name as a unique identifier for identification and transmission; comprises the following steps of;
the method comprises the following steps: content may first be cached in the CCN to support various functions, including content distribution and multicasting; the node records corresponding state and interface information in the interest packet requesting process, and jumps back the data packet to the user according to the information; the forwarding information base FIB and the pending interest table PIT are kept inside the CCN node due to the content repository CS;
step two: when the interest packet reaches the content router, whether the CS caches the content or not is firstly inquired; if yes, directly returning the data packet to the user, otherwise, inquiring the PIT;
step three: if the content has an input request, adding the corresponding arrival surface to the PIT, otherwise, performing maximum matching query in the FIB according to the information of the FIB; the interest packet is then forwarded to the next hop and a new PIT table will be built;
step four: when the data packet is sent back, please check whether the requested content item exists in the PIT, if so, forward the data packet to the user according to the arrival interface information, delete the item in the PIT, and determine whether the CS caches the content by a cache placement strategy;
step five: evaluating three node attributes of proper connectivity, middle centrality and feature vector centrality;
step six: the popularity of the content is estimated through the content request counting in the measuring process, so that the utilization rate of the cache space is further improved, and better cache performance is realized in the CCN;
step seven: in CCN/NDN, PIT records the unsatisfied requests, including the content name and the corresponding arrival interface, to ensure that the returned response packet is returned to the content requestor along the reverse path; the origin of the request is identified by the PIT; when a user requests content, calculating and normalizing betweenness centrality and feature vector centrality of nodes on a delivery path;
step eight: designs a variable
Figure RE-GDA0002904159810000031
To match content popularity with a given node value;
step nine: the proposed caching strategy is evaluated by an NDN-SIM simulator, and the simulator is an NDN simulation module based on NS-3 and realizes the basic function of the NDN; by modifying the code, the proposed caching policies can be implemented, and the results are imported into the MA TLAB to provide performance comparison between existing caching policies;
step ten: and (5) analyzing a simulation result.
Further, the content repository CS, the forwarding information base FIB and the pending interest table PIT in the first step are maintained inside the CCN node, so that each node uses the above three types of data structures for content distribution; for the three data structures, the CS is configured to cache copies of content passing through the node to satisfy subsequent content requests; the PIT is used for recording the unsatisfied interest packets, including content names and corresponding arrival interfaces, and aims to aggregate the same content requests and avoid repeatedly sending the same interest packets; for FIB, it saves the next hop interface information to the provider for interesting packet routing.
Further, in the fifth step, the node value is evaluated through three node attributes, and the condition that the shortest path information is inquired by adopting a named data link state routing protocol (NLSR) is considered; given an undirected graph G ═ (V, E) with n vertices and m edges, where V ═ V1,v2,...,vnDenotes a set of content routers, E ═ E1,e2,...,emRepresents links between content routers; wherein A ═ aij)n×nIs a contiguous matrix of G, for viDirectly with vjConnection of aij1, otherwise aij＝0。
Furthermore, the fifth step reflects three node attributes of connectivity, middle centrality and feature vector centrality in a manner of;
connectivity: different forwarding policies result in different routing paths for the requested content, and the cache nodes will play different roles in these policies; different forwarding policies result in different routing paths for the requested content, and cache nodes play different roles in these policies; therefore, we consider the number of paths of the request content through the cache node as the connectivity of the node; moreover, as the number of paths increases, the requested content becomes more important;
defining request content k by viThe number of routing paths of cs(vi) By viThe maximum number of routing paths is
Figure RE-GDA0002904159810000041
Can be prepared by mixing cs(vi) And
Figure RE-GDA0002904159810000042
the ratio of (A) to (B) is defined as Cs(vi) To obtain connectivity;
intermediate centrality: if the content router is located on the shortest path between the corresponding content routers, the content router is considered to be in an important position; this is reasonable because the content router at this location can affect the entire network by controlling or misreading the transmission of information; the ability to describe the characteristics of content router control information transfer is intermediate centrality (also known as node median); will sigmastIs set as vsAnd vtNumber of shortest paths between, σst(vi) Is from vsTo vtV passing throughiOf shortest path number, intermediate centrality viCan be expressed as:
Figure RE-GDA0002904159810000051
wherein n represents the number of content routers
Feature vector centrality: in fact, the impact of a content router is not only related to its own location, but also to its neighbors; if a content router is selected by a very popular participant, the corresponding impact will also increase; on the other hand, the pair of influences on the influence node, obviously, the influence is larger in the case of representing the influence by using the feature vector centrality; we will turn CE(vi) Defining the characteristic vector central point of the node to represent the influence of the adjacent node; it can also be said that CE(vi) Not only reflects the relative centrality of the network, but also reflects the long-term impact of the nodes. According to the existing research, the network is distributed according to the power law, and the nodes play different roles at different positions; the connectivity and the betweenness centrality consider the number of nodes in a request content routing path, and the feature vector centrality considers the adjacent influence; when the cache position is selected, the NVCP considers the three attributes at the same time; will M(vi) Defined as a composite attribute, we have:
M(vi)＝αCS(vi)+βCB(vi)+γCE(vi) (5)
where α, β, γ denote weights of connectivity, mediacy centrality, and feature vector centrality, and the sum of them is 1; it is worth noting that in our proposed scheme, the three mentioned attributes have different impacts on the choice of cache location; based on this, when different attributes are used to evaluate the importance of nodes in the same network, corresponding different results will be obtained; thus, the overall attribute M (v)i) Is determined by the relevant requirements of the 6G-CCN.
Further, in the sixth step, the popularity of the content can be estimated by measuring the content request count in the process, and the more the content request count, the greater the popularity and probability of the content being requested; suppose at viThe number of times of requesting the content k is
Figure RE-GDA0002904159810000052
And v isiThe maximum number of times of treatment is
Figure RE-GDA0002904159810000053
Finally, we can express the popularity of content k as
Figure RE-GDA0002904159810000054
Furthermore, a variable is designed in the eighth step
Figure RE-GDA0002904159810000068
To match content popularity with a given node value, as follows:
Figure RE-GDA0002904159810000061
wherein
Figure RE-GDA0002904159810000062
Is that the content k is at viThe popularity of (C) and
Figure RE-GDA0002904159810000063
and M (v)i) Is constant and less than 1; in general, there are two cases:
(1)
Figure RE-GDA0002904159810000064
therefore, caching content in the content router can achieve a higher cache hit rate;
(2)
Figure RE-GDA0002904159810000065
the node value is high, but the corresponding content popularity is low; if the content with lower popularity is cached, the cache space is wasted; in view of these two cases, in equation 3,
Figure RE-GDA0002904159810000069
is set to 1 or more.
Further, the main idea of NVCP proposed in the step eight is given in algorithms 1 and 2, and we have a fixed network topology considering that the location of the content router is not changed; thus, the network can be regarded as an undirected graph, requiring the pre-derivation of C using corresponding algorithms such as the Brande algorithm and the iterative power methodB(vi) And CE(vi) The computational complexity resulting in these two algorithms is o (ve); the algorithm 1 is a process of obtaining betweenness centrality and feature vector centrality; obviously, when the interest package arrives at the content router, the CS sends the content back to the user if it owns it, otherwise C is computed according to the network topologyB(vi) And CE(vi) (ii) a At the same time, CS(vi) The value of sum is increased by 1; algorithm 2, on the other hand, illustrates the process of selecting the appropriate cache location and cache content; from the results given in Algorithm 1, the calculation
Figure RE-GDA0002904159810000066
If it is not
Figure RE-GDA0002904159810000067
Caching the content, otherwise forwarding the data packet to the next hop; in addition, considering the fixed location of the content router, CB(vi) And CE(vi) The value of (c) need only be calculated once. In this way, the popularity of the content increases by 1 when requested.
Further, the simulation in said step nine uses a randomly generated network topology, which is composed of 50 nodes and 150 links; the network is provided with an origin server which is randomly connected to a node, and an edge node is connected to a user; generating a content request according to the Zipf-Mandelbrot distribution a being 0.7; the total number of different content to be requested in the network is 10,000; further assume that each user's interest is generated in terms of a poisson distribution at λ 100/s.
Further said step nine simulation technique compares the proposed NVCP policy with LCE, prob0.5 and MPC in terms of cache hit rate, average hop count and average transmission delay.
Further step ten simulation results analysis when the size of the cache node changes from 100 to 2,000 and the content density indicated by index a changes from 0.1 to 1, the system performance will change accordingly.
The invention has the beneficial effects that: a novel 6G-CCN cooperative caching strategy based on NVCP is researched, the problem that a traditional caching strategy is frequently replaced or a large amount of content is redundant is solved, caching redundancy and content replacement frequency are obviously reduced, and the diversity of the content is improved. In this context, we also consider the impact of the neighbor nodes when selecting cache nodes, without placing content in the neighbor nodes. Simulation results show that the NVCP is superior to LCE, prob (0.5) and MPC in terms of cache hit rate, average hop count and average transmission delay.
In NVCP, the value of a node is determined based on connectivity, intermediaries, and feature vector centrality. The importance of the content is determined by the popularity of the content, while the choice of caching location and caching content depends on the value of the node and the popularity of the content. Nodes of different values cache content of different popularity, where they are proportional.
On one hand, the NVCP utilizes the difference of the popularity of the content, ensures the uniform distribution of the cached content, reduces the redundancy of the content and increases the diversity of the content. On the other hand, the values of the nodes are evaluated from the plurality of attributes, and the difference between the content routing positions is utilized, so that the node utilization rate and the cache hit rate can be obviously improved, the content acquisition skip points and delay are reduced, and the user experience is improved.
Comparing NVCP with LCE, prob (b) and MPC in terms of cache hit rate, average hop count and average transmission delay, simulation results show that the proposed NVCP is superior to other cache policies in all respects.
Drawings
FIG. 1 is a schematic diagram of a network topology according to the present invention;
FIG. 2 is a schematic diagram of an algorithm of the present invention;
FIG. 3 is a diagram of the algorithm two of the present invention;
FIG. 4 is a table of simulation parameters according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood, however, that the description herein of specific embodiments is only intended to illustrate the invention and not to limit the scope of the invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, and the terms used herein in the specification of the present invention are for the purpose of describing particular embodiments only and are not intended to limit the present invention.
As shown in fig. 1, fig. 2 and fig. 3, a new node value and content popularity caching method based on CCN in 6G, the communication of CCN includes two data packets: the interest data packet comprises a content name, a selector and a random number so as to forward the request of the user through the CCN node; a data packet consisting of the content name, the signature information and the data is sent to a satisfied user along the reverse path of the interest packet; the CCN uses the content name as a unique identifier for identification and transmission; the method comprises the following operation steps;
the method comprises the following steps: content may first be cached in the CCN to support various functions, including content distribution and multicasting; the node records corresponding state and interface information in the interest packet requesting process, and jumps back the data packet to the user according to the information; the forwarding information base FIB and the pending interest table PIT are kept inside the CCN node due to the content repository CS;
the content repository CS, the forwarding information base FIB and the pending interest table PIT in said step one are maintained inside the CCN node, so that each node uses the above three types of data structures for content distribution; for the three data structures, the CS is configured to cache copies of content passing through the node to satisfy subsequent content requests; the PIT is used for recording the unsatisfied interest packets, including content names and corresponding arrival interfaces, and aims to aggregate the same content requests and avoid repeatedly sending the same interest packets; for FIB, it saves the next hop interface information to the provider for interesting packet routing.
Step two: when the interest packet reaches the content router, whether the CS caches the content or not is firstly inquired; if yes, directly returning the data packet to the user, otherwise, inquiring the PIT;
step three: if the content has an input request, adding the corresponding arrival surface to the PIT, otherwise, performing maximum matching query in the FIB according to the information of the FIB; the interest packet is then forwarded to the next hop and a new PIT table will be built;
step four: when the data packet is sent back, please check whether the requested content item exists in the PIT, if so, forward the data packet to the user according to the arrival interface information, delete the item in the PIT, and determine whether the CS caches the content by a cache placement strategy;
step five: evaluating three node attributes of proper connectivity, middle centrality and feature vector centrality;
since the locality of the cache has a great influence on the performance of the CCN, how to select the locality of the cache remains an open question. In this subsection, three node attributes are defined to evaluate the node's value, which are described based on graph theory. Furthermore, we also consider using named data link state routing protocol (NLSR) to query shortest path information. Given an undirected graph G ═ (V, E) with n vertices and m edges, where V ═ V1,v2,...,vnDenotes a set of content routers, E ═ E1,e2,...,emDenotes links between content routers. Wherein A ═ aij)n×nIs a contiguous matrix of G, for viDirectly with vjConnection of aij1, otherwise aij＝0。
1) Connectivity: different forwarding policies result in different routing paths for the requested content, and the caching nodes will play different roles in these policies. Different forwarding policies result in different routing paths for the requested content, and cache nodes play different roles in these policies. Therefore, we consider the number of paths of the requested content through the cache node as the connectivity of that node. Also, as the number of paths increases, the requested content becomes more important. Defining request content k by viThe number of routing paths of cs(vi) By viThe maximum number of routing paths is
Figure RE-GDA0002904159810000091
Can be prepared by mixing cs(vi) And
Figure RE-GDA0002904159810000092
the ratio of (A) to (B) is defined as Cs(vi) To obtain connectivity.
2) Intermediate centrality: a content router is considered to be in a significant position if it is located on the shortest path between the respective content routers. This is reasonable because the content router at this location can affect the entire network by controlling or misreading the transmission of informationLinking the collaterals. The ability to describe the characteristics of content router control information transfer is intermediate centrality (also called node median) [10 ]]. Will sigmastIs set as vsAnd vtNumber of shortest paths between, σst(vi) Is from vsTo vtV passing throughiOf shortest path number, intermediate centrality viCan be expressed as:
Figure RE-GDA0002904159810000093
in the formula, n represents the number of content routers.
3) Feature vector centrality: in fact, the impact of a content router is related not only to its own location, but also to its neighbors [11 ]. If a content router is selected by a very popular participant, the corresponding impact will also increase.
On the other hand, the impact pairs on the impact nodes, obviously, the impact is greater in the case of characterizing the impact using feature vector centrality. We will turn CE(vi) The feature vector central point of the node is defined and represents the influence of the adjacent node. It can also be said that CE(vi) Not only reflects the relative centrality of the network, but also reflects the long-term impact of the nodes.
According to the existing research [12 ]]The network is distributed according to a power law, and the nodes play different roles at different positions. The connectivity and the betweenness centrality consider the number of nodes in the request content routing path, and the feature vector centrality considers the influence of the neighbors. When selecting the cache location, NVCP considers the above three attributes simultaneously. Mixing M (v)i) Defined as a composite attribute, we have:
M(vi)＝αCS(vi)+βCB(vi)+γCE(vi) (8)
here, α, β, γ denote weights of connectivity, mediacy centrality, and feature vector centrality, and the sum of these is 1. It is worth noting that in our proposed solution, three are mentionedAttributes have different effects on the choice of cache location. Based on this, when different attributes are used to evaluate the importance of nodes in the same network, correspondingly different results will be obtained. Thus, the overall attribute M (v)i) Is determined by the relevant requirements of the 6G-CCN.
Step six: the popularity of the content is estimated through the content request counting in the measuring process, so that the utilization rate of the cache space is further improved, and better cache performance is realized in the CCN;
popularity is a factor in extracting content, since whether or not to cache every piece of content through a content router is another issue for CCNs. The popularity of content can be estimated by measuring the content request count in the process, which means that the more the content request count, the greater the popularity and probability that content is requested. Suppose at viThe number of times of requesting the content k is
Figure RE-GDA0002904159810000101
And v isiThe maximum number of times of treatment is
Figure RE-GDA0002904159810000102
Finally, we can express the popularity of content k as
Figure RE-GDA0002904159810000103
In particular, the present invention relates to a method for producing,
Figure RE-GDA0002904159810000104
it also takes more than a certain time frame to make sense, rather than the entire historical time. In view of the present disclosure aiming to provide a new idea to further improve the utilization of the cache space and achieve better cache performance in CCN, we will consider the time horizon in our future work.
Step seven: in CCN/NDN, PIT records the unsatisfied requests, including the content name and the corresponding arrival interface, to ensure that the returned response packet is returned to the content requestor along the reverse path; the origin of the request is identified by the PIT; when the user requests the content, the nodes on the delivery path are calculated and normalizedThe median centrality and the feature vector centrality of the vector; designs a variable
Figure RE-GDA0002904159810000105
To match content popularity with a given node value; the core idea of the NVCP is to add a table including a content name, a routing path number and a content request number to each content node based on a node value and content popularity, so as to store information of content and cache nodes. Notably, in CCN/NDN, the PIT records the unsatisfied request, including the content name and corresponding arrival interface, to ensure that the returned response packet is returned to the content requestor along the reverse path. Thus, the origin of the request is identified by the PIT. In this way, when a user requests content, the betweenness centrality and the feature vector centrality of the nodes on the delivery path will be calculated and normalized. Once the request is satisfied, the data packet is returned on the reverse delivery path. At this time, the content popularity is counted according to the content click rate. In our proposed scheme, we have designed a variable
Figure RE-GDA0002904159810000118
To match content popularity with a given node value, as follows:
Figure RE-GDA0002904159810000111
wherein
Figure RE-GDA0002904159810000112
Is that the content k is at viThe popularity of (C) and
Figure RE-GDA0002904159810000113
and M (v)i) Is constant and less than 1. In general, there are two cases: (1)
Figure RE-GDA0002904159810000114
meaning that the popularity of the content is more important than the value of the node. Thus, content routingThe cache contents in the device can obtain higher cache hit rate, and the proposed NVCP strategy is compared with LCE, Prob (0.5) and MPC in terms of cache hit rate, average hop count and average transmission delay, as detailed in fig. 4. (2)
Figure RE-GDA0002904159810000115
Indicating that the node value is high but the corresponding content popularity is low. This would result in a waste of cache space if content with lower popularity was cached. In view of these two cases, in equation 3,
Figure RE-GDA0002904159810000119
is set to 1 or more.
The main idea of the proposed NVCP is given in algorithms 1 and 2, as shown in fig. 2 and 3. In our proposed solution we have a fixed network topology, considering that the location of the content routers does not change, see fig. 1. Therefore, the network can be regarded as an undirected graph, and C is obtained in advance by using corresponding algorithms (such as Brande algorithm and power iteration method)B(vi) And CE(vi) The computational complexity that leads to these two algorithms is o (ve). Algorithm 1 is a process of obtaining betweenness centrality and feature vector centrality. Obviously, when the interest package arrives at the content router, the CS sends the content back to the user if it owns it, otherwise C is computed according to the network topologyB(vi) And CE(vi). At the same time, CS(vi) The value of sum is increased by 1. Algorithm 2, on the other hand, illustrates the process of selecting the appropriate cache location and cache contents. From the results given in Algorithm 1, the calculation
Figure RE-GDA0002904159810000116
If it is not
Figure RE-GDA0002904159810000117
The content is cached otherwise the packet is forwarded to the next hop. In addition, considering the fixed location of the content router, CB(vi) And CE(vi) Need only count the value ofAnd (5) calculating once. In this way, the popularity of the content increases by 1 when requested, which is easily achieved. Clearly, the algorithm we propose greatly improves over the existing work
Figure RE-GDA00029041598100001110
Efficiency of calculation of the values. Obviously, the computational complexity of algorithm 1 and algorithm 2 is not very high and is practical and acceptable.
Step nine: the proposed caching strategy is evaluated by an NDN-SIM simulator, and the simulator is an NDN simulation module based on NS-3 and realizes the basic function of the NDN; by modifying the code, the proposed caching policies can be implemented, and the results are imported into the MA TLAB to provide performance comparison between existing caching policies; the simulation used a randomly generated network topology as shown in fig. 1, which consisted of 50 nodes and 150 links. There is an origin server in the network that is randomly connected to a node, and edge nodes are connected to the users. The content request is generated according to a Zipf-Mandelbrot distribution (a ═ 0.7). The total number of different content to be requested in the network is 10,000. Further assume that each user's interest is generated in terms of a poisson distribution at λ 100/s. For simplicity and fairness, various properties of the nodes are considered in combination, and in the simulation results given herein, specific gravity values of α (connectivity), β (betweenness centrality), and γ (feature vector centrality) are equivalently given: 1/3. Using The Least Recently Used (LRU) [14] as a cache replacement policy, The total simulation time was 100 s. More specifically, the simulation results have been evaluated for various values of cache size and a. The main simulation parameters are shown in fig. 4, and the simulation technique compares the proposed NVCP policy with LCE, prob0.5 and MPC in terms of cache hit rate, average hop count and average transmission delay in step nine.
Wherein:
1) cache hit rate: refers to the probability that a cache node, rather than a server, will satisfy a user request. This is a typical parameter reflecting the performance of the caching strategy. The higher the cache hit rate, the greater the probability that the cache node will satisfy the user request. The number of requests satisfied by the cache nodes and the total content number requested by the user are set as N and N respectively, and the cache hit rate can be obtained from the ratio of N to N.
2) Average hop count: refers to the number of hops a user requests to reach a caching node or origin server. Which reflects the distance between the caching node and the user. The smaller the hop count, the closer the cache node is to the consumer, and the higher the overall system efficiency.
3) Average transmission delay: refers to the delay experienced by a consumer in providing a request for content to obtain data. It may reflect the speed at which a network meets consumer demand. Because the cache node is closer to the user than the origin server, smaller transmission delay and faster request response can be achieved, thereby improving quality of service (QoS).
Step ten: simulation results analysis, when the size of the cache node is changed from 100 to 2,000, and the content density indicated by the index a is changed from 0.1 to 1, the system performance will be changed accordingly.
This is resealable because the LCE requires all nodes on the delivery path to cache content without discrepancies, which results in a large amount of content redundancy and frequent replacement. It is reusable because LCEs require all nodes on the delivery path to cache content indifferently, which results in a large amount of content redundancy and frequent replacement. In addition, Prob (0.5) caches the content passing through the cache node with a fixed probability. Even if the cache space is reduced, it still results in content redundancy and low content diversity.
The MPC will not store all content on every node on the path, but only cache popular content. On the contrary, the NVCP comprehensively considers the node value and the content popularity, caches the content with higher popularity in the node with higher value, and caches the content with lower popularity in the node with lower value, thereby greatly reducing the replacement frequency, improving the content diversity, and reducing the content redundancy. Compared with LCE, Prob (0.5) and MPC schemes, the proposed NVCP cache hit rate is improved by 11% to 15%. The second and third subgraphs show that the average hop count and the average transmission delay gradually decrease as the node cache capacity increases. Furthermore, of NVCPThe performance is better than other schemes. This is because LCEs buffer content indiscriminately, Prob (0.5) performs probabilistic buffering, while MPC only buffers the most popular content without any requirement on the nodes. On the contrary, the NVCP comprehensively evaluates the node value from connectivity, betweenness centrality and feature vector centrality, and allocates different weights according to different requirements, thereby improving the response speed to the content request and reducing the network overhead. Compared with the traditional CCN caching strategy, the average hop count and the average transmission delay of the NVCP are greatly improved. The average hop count of NVCP is reduced compared to LCE, prob (0.5) and MPC
Figure RE-GDA0002904159810000132
Hop, average transmission delay is reduced
Figure RE-GDA0002904159810000131
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents or improvements made within the spirit and principle of the present invention should be included in the scope of the present invention.

Claims (10)

1. A new node value and content popularity caching method based on a CCN in 6G is disclosed, wherein communication of the CCN comprises two data packets: the interest data packet comprises a content name, a selector and a random number so as to forward the request of the user through the CCN node; a data packet consisting of the content name, the signature information and the data is sent to a satisfied user along the reverse path of the interest packet; the CCN uses the content name as a unique identifier for identification and transmission; it is characterized by comprising the following operation steps;
the method comprises the following steps: content may first be cached in the CCN to support various functions, including content distribution and multicasting; the node records corresponding state and interface information in the interest packet requesting process, and jumps back the data packet to the user according to the information; the forwarding information base FIB and the pending interest table PIT are kept inside the CCN node due to the content repository CS;
step two: when the interest packet reaches the content router, whether the CS caches the content or not is firstly inquired; if yes, directly returning the data packet to the user, otherwise, inquiring the PIT;
step three: if the content has an input request, adding the corresponding arrival surface to the PIT, otherwise, performing maximum matching query in the FIB according to the information of the FIB; the interest packet is then forwarded to the next hop and a new PIT table will be built;
step four: when the data packet is sent back, please check whether the requested content item exists in the PIT, if so, forward the data packet to the user according to the arrival interface information, delete the item in the PIT, and determine whether the CS caches the content by a cache placement strategy;
step five: evaluating three node attributes of proper connectivity, middle centrality and feature vector centrality;
step six: the popularity of the content is estimated through the content request counting in the measuring process, so that the utilization rate of the cache space is further improved, and better cache performance is realized in the CCN;
step seven: in CCN/NDN, PIT records the unsatisfied requests, including the content name and the corresponding arrival interface, to ensure that the returned response packet is returned to the content requestor along the reverse path; the origin of the request is identified by the PIT; when a user requests content, calculating and normalizing betweenness centrality and feature vector centrality of nodes on a delivery path;
step eight: designs a variable
Figure FDA0002751844520000011
To match content popularity with a given node value;
step nine: the proposed caching strategy is evaluated by an NDN-SIM simulator, and the simulator is an NDN simulation module based on NS-3 and realizes the basic function of the NDN; by modifying the code, the proposed caching policies can be implemented, and the results are imported into the MA TLAB to provide performance comparison between existing caching policies;
step ten: and (5) analyzing a simulation result.
2. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the content repository CS, the forwarding information base FIB and the pending interest table PIT in step one are maintained inside the CCN node, so that each node uses the above three types of data structures for content distribution; for the three data structures, the CS is configured to cache copies of content passing through the node to satisfy subsequent content requests; the PIT is used for recording the unsatisfied interest packets, including content names and corresponding arrival interfaces, and aims to aggregate the same content requests and avoid repeatedly sending the same interest packets; for FIB, it saves the next hop interface information to the provider for interesting packet routing.
3. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the node value is evaluated by three node attributes in the fifth step, considering that a named data link state routing protocol (NLSR) is adopted to query shortest path information; given an undirected graph G ═ (V, E) with n vertices and m edges, where V ═ V1,v2,...,vnDenotes a set of content routers, E ═ E1,e2,...,emRepresents links between content routers; wherein A ═ aij)n×nIs a contiguous matrix of G, for viDirectly with vjConnection of aij1, otherwise aij＝0。
4. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the five step nodes reflect three node attributes of connectivity, middle centrality and feature vector centrality in a manner of;
connectivity: different forwarding policies result in different routing paths for the requested content, and the cache nodes will play different roles in these policies; different forwarding policies result in different routing paths for the requested content, and cache nodes play different roles in these policies; therefore, we consider the number of paths of the request content through the cache node as the connectivity of the node; moreover, as the number of paths increases, the requested content becomes more important;
defining request content k by viThe number of routing paths of cs(vi) By viThe maximum number of routing paths is
Figure FDA0002751844520000021
Can be prepared by mixing cs(vi) And
Figure FDA0002751844520000022
the ratio of (A) to (B) is defined as Cs(vi) To obtain connectivity;
intermediate centrality: if the content router is located on the shortest path between the corresponding content routers, the content router is considered to be in an important position; this is reasonable because the content router at this location can affect the entire network by controlling or misreading the transmission of information; the ability to describe the characteristics of content router control information transfer is intermediate centrality (also known as node median); will sigmastIs set as vsAnd vtNumber of shortest paths between, σst(vi) Is from vsTo vtV passing throughiOf shortest path number, intermediate centrality viCan be expressed as:
Figure FDA0002751844520000031
wherein n represents the number of content routers
Feature vector centrality: in fact, the impact of a content router is not only related to its own location, but also to its neighbors; if a content router is selected by a very popular participant, the corresponding impact will also increase; on the other hand, the pair of influences on the influencing node, it is clear that the influence is characterized using feature vector centralityIn case, the impact is greater; we will turn CE(vi) Defining the characteristic vector central point of the node to represent the influence of the adjacent node; it can also be said that CE(vi) Not only reflects the relative centrality of the network, but also reflects the long-term impact of the nodes. According to the existing research, the network is distributed according to the power law, and the nodes play different roles at different positions; the connectivity and the betweenness centrality consider the number of nodes in a request content routing path, and the feature vector centrality considers the adjacent influence; when the cache position is selected, the NVCP considers the three attributes at the same time; mixing M (v)i) Defined as a composite attribute, we have:
M(vi)＝αCS(vi)+βCB(vi)+γCE(vi) (2)
where α, β, γ denote weights of connectivity, mediacy centrality, and feature vector centrality, and the sum of them is 1; it is worth noting that in our proposed scheme, the three mentioned attributes have different impacts on the choice of cache location; based on this, when different attributes are used to evaluate the importance of nodes in the same network, corresponding different results will be obtained; thus, the overall attribute M (v)i) Is determined by the relevant requirements of the 6G-CCN.
5. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the popularity of the content can be estimated by measuring the content request count in the process in the sixth step, and the more the content request count, the greater the popularity and probability of the content being requested; suppose at viThe number of times of requesting the content k is
Figure FDA0002751844520000041
And v isiThe maximum number of times of treatment is
Figure FDA0002751844520000042
Finally, we can express the popularity of content k as
Figure FDA0002751844520000043
6. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein a variable is designed in the eighth step
Figure FDA0002751844520000049
To match content popularity with a given node value, as follows:
Figure FDA0002751844520000044
wherein
Figure FDA0002751844520000045
Is that the content k is at viThe popularity of (C) and
Figure FDA0002751844520000046
and M (v)i) Is constant and less than 1; in general, there are two cases:
(1)
Figure FDA0002751844520000047
therefore, caching content in the content router can achieve a higher cache hit rate;
(2)
Figure FDA0002751844520000048
the node value is high, but the corresponding content popularity is low; if the content with lower popularity is cached, the cache space is wasted; in view of these two cases, in equation 3,
Figure FDA00027518445200000410
is set to 1 or more.
7. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the main idea of NVCP proposed in step eight is given in algorithms 1 and 2, and we have a fixed network topology considering that the location of the content router is not changed; thus, the network can be regarded as an undirected graph, requiring the pre-derivation of C using corresponding algorithms such as the Brande algorithm and the iterative power methodB(vi) And CE(vi) The computational complexity resulting in these two algorithms is o (ve); the algorithm 1 is a process of obtaining betweenness centrality and feature vector centrality; obviously, when the interest package arrives at the content router, the CS sends the content back to the user if it owns it, otherwise C is computed according to the network topologyB(vi) And CE(vi) (ii) a At the same time, CS(vi) The value of sum is increased by 1; algorithm 2, on the other hand, illustrates the process of selecting the appropriate cache location and cache content; from the results given in Algorithm 1, the calculation
Figure FDA00027518445200000411
If it is not
Figure FDA00027518445200000412
Caching the content, otherwise forwarding the data packet to the next hop; in addition, considering the fixed location of the content router, CB(vi) And CE(vi) The value of (c) need only be calculated once. In this way, the popularity of the content increases by 1 when requested.
8. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the simulation in said ninth step uses a randomly generated network topology, which consists of 50 nodes and 150 links; the network is provided with an origin server which is randomly connected to a node, and an edge node is connected to a user; generating a content request according to the Zipf-Mandelbrot distribution a being 0.7; the total number of different content to be requested in the network is 10,000; further assume that each user's interest is generated in terms of a poisson distribution at λ 100/s.
9. The CCN-based new node value and content popularity caching method in 6G, as recited in claim 1, wherein the simulation technique in step nine compares the proposed NVCP policy with LCE, Prob0.5 and MPC in terms of cache hit rate, average hop count and average transmission delay.
10. The CCN-based new node value and content popularity caching method in 6G according to claim 1, wherein the simulation result analysis in the step ten shows that when the size of the cache node is changed from 100 to 2,000 and the content concentration represented by index a is changed from 0.1 to 1, the system performance is changed correspondingly.
CN202011199997.XA 2020-10-30 2020-10-30 CCN-based new node value and content popularity caching method in 6G Withdrawn CN112399485A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011199997.XA CN112399485A (en) 2020-10-30 2020-10-30 CCN-based new node value and content popularity caching method in 6G

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011199997.XA CN112399485A (en) 2020-10-30 2020-10-30 CCN-based new node value and content popularity caching method in 6G

Publications (1)

Publication Number Publication Date
CN112399485A true CN112399485A (en) 2021-02-23

Family

ID=74597841

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011199997.XA Withdrawn CN112399485A (en) 2020-10-30 2020-10-30 CCN-based new node value and content popularity caching method in 6G

Country Status (1)

Country Link
CN (1) CN112399485A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113225380A (en) * 2021-04-02 2021-08-06 中国科学院计算技术研究所 Content distribution network caching method and system based on spectral clustering
CN113382053A (en) * 2021-04-23 2021-09-10 河南科技大学 Content active pushing method based on node semi-local centrality and content popularity
CN114710561A (en) * 2022-04-01 2022-07-05 中国人民解放军战略支援部队信息工程大学 In-network content caching method and device based on protocol-independent forwarding and electronic equipment
CN114745440A (en) * 2022-03-09 2022-07-12 南京邮电大学 CCN cache replacement method and device

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160021172A1 (en) * 2014-07-18 2016-01-21 Palo Alto Research Center Incorporated Reputation-based strategy for forwarding and responding to interests over a content centric network
CN110233901A (en) * 2019-06-20 2019-09-13 南通大学 A kind of content center network caching method and system

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20160021172A1 (en) * 2014-07-18 2016-01-21 Palo Alto Research Center Incorporated Reputation-based strategy for forwarding and responding to interests over a content centric network
CN110233901A (en) * 2019-06-20 2019-09-13 南通大学 A kind of content center network caching method and system

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
刘四军: "内容中心网络中高效缓存策略研究", 《中国优秀硕士学位论文全文数据库工程科技辑》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113225380A (en) * 2021-04-02 2021-08-06 中国科学院计算技术研究所 Content distribution network caching method and system based on spectral clustering
CN113382053A (en) * 2021-04-23 2021-09-10 河南科技大学 Content active pushing method based on node semi-local centrality and content popularity
CN114745440A (en) * 2022-03-09 2022-07-12 南京邮电大学 CCN cache replacement method and device
CN114745440B (en) * 2022-03-09 2023-05-26 南京邮电大学 CCN cache replacement method and device
CN114710561A (en) * 2022-04-01 2022-07-05 中国人民解放军战略支援部队信息工程大学 In-network content caching method and device based on protocol-independent forwarding and electronic equipment
CN114710561B (en) * 2022-04-01 2023-05-16 中国人民解放军战略支援部队信息工程大学 Method and device for caching content in network based on protocol independent forwarding and electronic equipment

Similar Documents

Publication Publication Date Title
CN112399485A (en) CCN-based new node value and content popularity caching method in 6G
CN104753797B (en) A kind of content center network dynamic routing method based on selectivity caching
Nour et al. A distributed cache placement scheme for large-scale information-centric networking
CN108366089B (en) CCN caching method based on content popularity and node importance
CN106911574B (en) Name data network multiple constraint routing algorithm based on population
Hou et al. Bloom-filter-based request node collaboration caching for named data networking
Wang et al. Milking the cache cow with fairness in mind
CN113783779B (en) Hierarchical random caching method in named data network
Wu et al. MBP: A max-benefit probability-based caching strategy in information-centric networking
CN105681438A (en) Centralized caching decision strategy in content-centric networking
Gui et al. A cache placement strategy based on entropy weighting method and TOPSIS in named data networking
Nguyen et al. Adaptive caching for beneficial content distribution in information-centric networking
Liu et al. A novel cache replacement scheme against cache pollution attack in content-centric networks
Liu et al. PBRS: A content popularity and betweenness based cache replacement scheme in ICN-IoT
Kamiyama et al. Dispersing content over networks in information-centric networking
Yufei et al. A centralized control caching strategy based on popularity and betweenness centrality in ccn
Jin et al. Content routing and lookup schemes using global bloom filter for content-delivery-as-a-service
Yang et al. Content-based hyperbolic routing and push mechanism in named data networking
CN107135271B (en) Energy-efficient content-centric network caching method
Li et al. Effective caching schemes for minimizing inter-ISP traffic in named data networking
Alduayji et al. PF-EdgeCache: Popularity and freshness aware edge caching scheme for NDN/IoT networks
Tseng et al. A novel cache scheme based on content popularity and user locality for future internet
Chen et al. Improving NDN forwarding engine performance by rendezvous-based caching and forwarding
Wang et al. Caching in information-centric networking
Zhou et al. Popularity and age based cache scheme for content-centric network

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20210223

WW01 Invention patent application withdrawn after publication