CN108596778B - Community division method based on interest space - Google Patents

Community division method based on interest space Download PDF

Info

Publication number
CN108596778B
CN108596778B CN201810436814.8A CN201810436814A CN108596778B CN 108596778 B CN108596778 B CN 108596778B CN 201810436814 A CN201810436814 A CN 201810436814A CN 108596778 B CN108596778 B CN 108596778B
Authority
CN
China
Prior art keywords
node
community
nodes
interest
neighbor
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810436814.8A
Other languages
Chinese (zh)
Other versions
CN108596778A (en
Inventor
蒋凌云
冯莹
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.)
Nanjing University of Posts and Telecommunications
Original Assignee
Nanjing University of Posts and Telecommunications
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 Nanjing University of Posts and Telecommunications filed Critical Nanjing University of Posts and Telecommunications
Priority to CN201810436814.8A priority Critical patent/CN108596778B/en
Publication of CN108596778A publication Critical patent/CN108596778A/en
Application granted granted Critical
Publication of CN108596778B publication Critical patent/CN108596778B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/01—Social networking

Abstract

The invention discloses a community division method based on an interest space, which comprises the following steps: setting an interest space vector of a node, distributing community labels for each node, determining a neighbor node table of the node by sending and receiving a message, calculating the similarity of the interest space vector according to a cosine similarity theorem, continuously updating a back node label of the neighbor node table of the node according to the value, and finally dividing the nodes with the same interest into a community, namely the same community label. The method has the advantages that in the network with the human nodes, the characteristic of interest attributes of people is considered, the nodes with the same interest can be divided into the same community, the interest of people has stability within a certain time, so that the interest community also has certain stability, meanwhile, the method has better convergence, and the nodes with the same interest can be quickly converged into the same community.

Description

Community division method based on interest space
Technical Field
The invention relates to a network interest community division method, in particular to a community division method based on an interest space.
Background
The concept of the opportunistic Network is originally derived from a Delay Tolerant Network (DTN), and the opportunistic Network is a Network that realizes communication between nodes in a Network by using an encounter opportunity brought by node movement when a communication link does not exist between a source node and a destination node. Because of the mobility of the nodes, the opportunistic network is in a disconnected state most of the time, the nodes realize the forwarding of data by means of meeting opportunities brought by the movement, and because a complete path does not exist between a source node and a destination node usually, the data can be transmitted from the source node to the destination node in a multi-hop forwarding mode, so that the opportunistic network has larger delay.
With the continuous development of intelligent communication equipment, smart phones are essential in life, people can form an opportunity network with people as the center by carrying the smart phones, the behavior characteristics and social attributes of people play an important role in researching the opportunity network, human dynamics research shows that human movement behaviors are a frequent and complex phenomenon, and the movement behaviors of individual or group of human beings have the characteristics of regularity, aggregation, small universe, sociality and the like. The research on the human network relationship structure finds that nodes in the network not only have individuality, but also have commonality, so the concept of a community structure is proposed, and in a network, because the behavior tracks of the nodes have periodicity and similarity, or the nodes have the same interest and hobbies and other factors, a group of nodes with similar structures or close connection appears in the network structure, and the group of nodes is called a community. In the community, the nodes are dense, the nodes have more contact opportunities, and in the community, the nodes are sparse, and the nodes may not have the contact opportunities for a long time. The GN algorithm proposed by Girvan and Newman in 2002 is widely accepted, and research results of the GN algorithm prove that community structures exist in a plurality of real networks, the research of the community discovery algorithm in the field of computer science is gradually developed, and the existing community discovery algorithms mainly comprise the following types:
selecting a community center node by taking the ID as a standard: each node in the network is endowed with a unique identification number ID, the node broadcasts the ID number of the node to the network in a certain time period, and the node selects the node with the largest (or the smallest) ID number as a central node according to the comparison between the received ID number and the ID number of the node. The division algorithm takes the identification number ID as a standard for selecting the central node, the selection mode is single, and other characteristics of the node, such as the number of neighbor nodes of the node, the energy of the node and the like, are not considered;
selecting a community center node by taking the connectivity of the node as a standard: the connectivity of a node is one of algorithms which are effective in dividing communities, the connectivity of the node refers to the number of the nodes connected with other nodes and the number of neighbor nodes of the node, and the greater the connectivity of a certain node is, the greater the number of the neighbor nodes of the node is, the more the node is, and the node can be used as a central node of the community. The method has the disadvantages that the excessive connection of the central node can cause the energy of the node to be exhausted quickly, thereby causing the instability of the community;
community division of global positioning system: the nodes are provided with portable equipment with a GPS function, the physical distance between the nodes can be acquired through the GPS, the nodes in the physical distance range are divided into a community, the method has the greatest advantage that the topological structure of the whole network does not need to be known, and the application scene of the method is limited due to the fact that the nodes need the GPS function.
Although the above community division methods can divide the structure of the community to a certain extent, the criteria selected by these methods are the centrality, speed, GPS positioning technology, etc. of the node, these methods can divide the node into the specified community to a certain extent, but do not consider the social characteristics of the node, especially in the opportunistic network taking people as the center, the node is regarded as a person carrying the intelligent device, the movement of which is inevitably influenced by the social movement of the person, and the social movement of which is inevitably influenced by factors such as individual consciousness, demand, and hobby of the person. In real life, originally scattered people can be gathered together under a certain requirement, the relationship among the gathered people is more intimate, and the communication time is longer. According to different interests, the times of going to some places are more common, and the stay time is longer. Therefore, the social characteristics of people play an important role in the division of communities.
Therefore, a community division method specially aiming at artificial nodes needs to be proposed from the perspective of social characteristics of the nodes.
Disclosure of Invention
The purpose of the invention is as follows: the invention provides a community division method based on an interest space, which can perform node community division by considering the sociality of nodes so that the community division is relatively stable.
The technical scheme is as follows: a community division method based on an interest space is characterized in that an independent label is distributed to each user in a network, communities are divided according to the labels, and users with the same labels are regarded as the same community. Initially, each user forms a community. An interest space vector is set for each user, and the interest space vector represents the interest degree of each user in each message in the network. We represent the topology of the network by an undirected graph G (V, E), where V represents a set of nodes, E represents edges between a node and neighboring nodes, and each edge is assigned a weight factor w(i，k)The weight value represents the interest of the node i and the neighbor nodesThe community label of the node with the maximum weight value is selected from the neighbor nodes every time as the community label of the node, after a period of iteration, the community labels of the nodes are converged, and at the moment, the nodes with the same interest belong to a community.
The method comprises the following specific steps:
the method comprises the following steps: the node I sets the interest space vector I of the node I according to the interest and hobbies of the node I and the interest degree of n types of messages in the networkiAnd each node is assigned with a community label CiAnd stores it in the node's cache.
The user interest space vector represents the interest degree of each message in the network by the user node; the interest space vector of the user node i is defined as: i isi＝{mi1，mi2…mij…minWherein mijRepresents the interest level of the j-th message by the node i, j is (1, 2 … n), and satisfies 0 ≦ mijLess than or equal to 1 and
Figure BDA0001653360600000031
the node inputs m according to the interest condition of the nodeij。
Community labeling: when a user node in the network is initialized, a unique community label is given to the user node, and C is usediIs represented by CiA community tag referring to a user node i;
weight factor w(i，k): weight factor w(i，k)Representing the interest similarity degree of the user node i and the neighbor node k thereof; the interest similarity between the user node i and the neighbor node k is calculated by a cosine similarity formula, namely
Figure BDA0001653360600000032
Wherein IiIs the spatial vector of interest of node I, IkIs the space vector of interest, | I, of node kiI is the norm of the space vector of interest of node I, IkAnd | is the modulus of the interest space vector of node k.
Step two: when all the nodes enter the network, the structure of the network tends to be stable, and each node broadcasts a handshake message to the network. The nodes broadcast the handshake messages and receive the handshake messages of other nodes at the same time. The content in the handshake message includes a community label of the node and an interest space of the node, and the format of the handshake message of the node i is as follows:
table 1 handshake messages for nodes i
Community labeling of nodes Interest space of a node
Ci Ii
Step three: after a period of time of broadcasting and receiving handshake messages, the node knows its own neighbor node and generates a neighbor node table, which stores neighbor nodes of the node, community labels of the neighbor nodes and weight factors w(i，k)The neighbor node table of node i is defined as follows:
TABLE 2 neighbor node table of node i
Figure BDA0001653360600000041
Step four: the node i obtains the interest space vector of each neighbor node k, and calculates the weight factor w of the node i and each neighbor node k through a calculation formula of the cosine similarity theorem(i，k)It is written into a column of the weight factor in the neighbor node table.
Step five: the nodes are sorted according to the order of coming first and coming last, the sorting table is represented by S (a, b … n), S [0] ═ a represents the node a first entering the network, and S [ n-1] ═ n represents the node n last entering the network.
Step six: starting from the first node in the sorting table S (a, b … n), the initial time node i represents the first node in the set S, and the process advances to step seven;
step seven: and (4) checking the neighbor node table of the node i by the node i, if the community labels of the nodes in the neighbor node table are not unified, turning to the step eight, and if the community labels of the neighbor nodes are unified, namely the community labels in the neighbor node table are all the same, turning to the step nine.
Step eight: and sequentially comparing the weight factors in the neighbor node table by the node i, selecting the community label of the node with the maximum weight factor as the community label of the node i, randomly selecting one from the neighbor nodes if a plurality of neighbor nodes with the same maximum value exist, and changing the community label of the node i into the community label of the neighbor node. After obtaining the new community label, the node i needs to send a notification message to its neighboring node, where the notification message is used to notify the neighboring node to update the community label of the node i. The notification message is defined as follows:
TABLE 3 Notification message
Node i New community label Ci
Step nine: the labels in the neighbor node tables are unified, namely, the labels of the neighbor nodes of the node i belong to a community, if the labels of the node i are the labels of the neighbor nodes, the node i does not update the community labels, if the community labels of the node i are not the community labels of the neighbor nodes, the community labels of the node i are updated to be the community labels of the neighbor nodes, and a notification message is sent to the neighbor nodes of the node i.
Step ten: and if the node i is the last node in the set S, turning to the step eleven, and if not, selecting the node behind the node i from the set S as a new node i and turning to the step seven.
Step eleven: and if the community labels of the nodes in the set S are not updated, turning to the step six, and if not, turning to the step twelve.
Step twelve: and after the community division based on the interest space is finished, outputting all communities and outputting all nodes of the community.
Has the advantages that: compared with the prior art, the invention has the advantages that:
1. the interest attributes of the nodes are considered, and the nodes with the same interest can be divided into a community;
2. the method has better convergence, and the nodes with the same interest and hobbies can quickly converge to unify a community;
3. the interest of the nodes does not change randomly for a long time, and the interest characteristics have certain stability, so that communities divided according to the interest space are relatively stable for a certain time.
Drawings
Fig. 1 is a network architecture diagram of the node of the present invention.
Fig. 2 is a network structure diagram of the node of the present invention after going through steps one to five.
Fig. 3 is a first time label change diagram for a node of the present invention.
Fig. 4 is a second label change diagram for the node of the present invention.
FIG. 5 is a power diagram of the partitioning of a community of nodes of the present invention.
FIG. 6 is a flow chart of the method steps of the present invention.
Detailed Description
A community division method based on an interest space assumes that there are four types of messages in a network, which are m respectively1，m2， m3，m3(ii) a There are 18 nodes in the network, which use a, b, c respectivelyD, e, f, g, h, i, j, k, l, n, m, o, p, q, x, and the network structure is shown in FIG. 1.
The method comprises the following steps: the node sets the interest space vector I of the node according to the interest of the node and the interest degree of the 4 types of messages in the networkiAnd each node is assigned with a community label CiAnd stores it in the node's cache. As shown in the following table:
Figure BDA0001653360600000051
Figure BDA0001653360600000061
step two: each node broadcasts a handshake message to the network, and the nodes receive handshake messages of other nodes while broadcasting the handshake messages.
Step three: after a period of time for broadcasting and receiving the handshake messages, the nodes know their own neighbor nodes and generate a neighbor node table.
Step four: and the node i acquires the interest space vector of each neighbor node, calculates the weight factor and writes the weight factor into one column of the weight factors in the neighbor node table.
Step five: and sequencing the nodes according to the sequence of coming first and coming last of the nodes, wherein: s ═ e (a, b, c, d, e, f, g, h, i, j, k, l, n, m, o, p, q, x)
After the second step to the fifth step are completed, the neighbor node table is obtained as shown in the following table:
Figure BDA0001653360600000071
Figure BDA0001653360600000081
Figure BDA0001653360600000091
the updated network structure diagram is shown in fig. 2: the neighbor nodes selected by the neighbor node table adopt a random distribution form, and can also adopt a fixed number form to select the neighbor nodes.
Step six: starting from the first node a in the sorted list S.
Step seven: and the node a checks the neighbor node table, if the community labels of the nodes in the neighbor node table are not unified, the step eight is carried out, and if the community labels of the neighbor nodes are unified, namely the community labels in the neighbor node table are all the same, the step nine is carried out.
Step eight: the node a compares the weight factors in the neighbor node table in sequence, selects the community label of the node with the maximum weight factor as the community label of the node a, randomly selects one from the neighbor nodes if a plurality of neighbor nodes with the same maximum value exist, and changes the community label of the node a into the community label of the neighbor node. After obtaining the new community label, the node a needs to send a notification message to its neighboring node, where the notification message is used to notify the neighboring node to update the community label of the node a.
Step nine: the labels in the neighbor node tables are unified, namely, the labels of the neighbor nodes of the node a belong to a community, if the labels of the node a are the labels of the neighbor nodes, the node a does not update the community labels, if the community labels of the node a are not the community labels of the neighbor nodes, the community labels of the node a are updated to be the community labels of the neighbor nodes i, and a notification message is sent to the neighbor nodes of the node a.
Step ten: and checking whether the node a is the last node in the set S, selecting the node b behind the node a from the set S as a new node as the new node because the node a is not the last node in the set S, and turning to the step seven.
Through the seventh step to the tenth step, the first updating of the community labels of the nodes in the sorting table S from the first node to the last node is completed. The nodes with changed community labels are shown in the following table:
Figure BDA0001653360600000092
Figure BDA0001653360600000101
the label change of the nodes in the network is shown in fig. 3:
step eleven: and continuing to step six because the labels in the set S are updated.
And after the seventh step to the tenth step, the second updating of the community labels of the nodes in the sorting table S from the first node to the last node is finished. The community label of only one node changes, as shown in the following table:
node point Old community label of node Community label answer with new nodes
i Cm Cn
The label change of the network node is shown in fig. 4:
step twelve: and after the community division based on the interest space is finished, outputting all communities and outputting all nodes of the community.
The community partition diagram of the nodes in the network is shown in FIG. 5:
whereinBelong to CbThe nodes of the community have: a, b, c, d, h, f, g, node pair message m in the community1Of interest.
Wherein belongs to CkThe nodes of the community have: e, k, p, o, node pair message m within the community4Of interest.
Wherein belongs to CxThe nodes of the community have: j, q, x, node pair message m within the community3Of interest.
Wherein belongs to CnThe nodes of the community have: i, l, m, n, node pair message m within the community2Of interest.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.

Claims (4)

1. A community division method based on an interest space is characterized in that: dividing communities according to the interest of the nodes:
the method comprises the following steps: the node I sets the interest space vector I of the node I according to the interest and hobbies of the node I and the interest degree of n types of messages in the networkiAnd each node is assigned with a community label CiAnd stores it in the node's cache;
step two: when all nodes enter the network, the structure of the network tends to be stable, each node broadcasts a handshake message to the network, and the nodes also receive handshake messages of other nodes while broadcasting the handshake messages;
step three: after a period of time for broadcasting and receiving the handshake message, the node knows the neighbor node of the node and generates a neighbor node table;
step four: one node i obtains an interest space vector of a neighbor node k, and calculates a weight factor w of the node i and the neighbor node k through a calculation formula of a cosine similarity theorem(i，k)Writing the weight factor into a column of the weight factor in the neighbor node table;
step five: sorting the nodes according to the sequence of the nodes coming first and then coming, wherein a sorting table is represented by S (a, b … n), i ═ e (a, b.. cndot) in the step four represents the node a entering the network firstly, and S [ n-1] ═ n represents the node n entering the network finally;
step six: starting from the first node in the sorting table S (a, b … n), i.e. the initial time node i represents the first node a in the set S, the process proceeds to step seven;
step seven: the node i checks the neighbor node table, if the community labels of the nodes in the neighbor node table are not unified, the step eight is carried out, and if the community labels of the neighbor nodes are unified, the step nine is carried out;
step eight: the node i compares the weight factors in the neighbor node table in sequence, selects the node with the maximum weight factor, randomly selects one neighbor node from the neighbor nodes if a plurality of neighbor nodes with the same maximum value exist, changes the community label of the node i into the community label of the neighbor node, and needs to send a notification message to the neighbor node after obtaining a new community label;
step nine: if the labels in the neighbor node tables are uniform, the neighbor nodes of the node i belong to a community, and if the labels of the node i are already the labels of the neighbor nodes, the node i does not update the community labels; if the community label of the node i is not the community label of the neighbor node, updating the community label of the node i to be the community label of the neighbor node i, and sending a notification message to the neighbor node of the node i;
step ten: judging whether the node i is the last node in the set S, if so, turning to the eleventh step, otherwise, selecting a node behind the node i from the set S as a new node i, and turning to the seventh step;
step eleven: if the community labels of the nodes in the set S are updated, turning to the step six, and if not, turning to the step twelve;
step twelve: and after the community division based on the interest space is finished, outputting all communities and outputting all nodes of the community.
2. The community partitioning method based on the interest space of claim 1, wherein: in the fourth step, the weight factor W(i，k)The calculation method of (2) is defined as follows:
the interest similarity degree of the node i and the neighbor node k thereof is calculated by a cosine similarity formula, and the specific formula is as follows:
Figure FDA0001653360590000021
weight factor W(i，k)Representing the degree of interest similarity of node I with its neighbor node k, where IiAs a spatial vector of interest, Ii＝{mi1，mi2…mij…min}，mijRepresents the interest level of the j-th message by the node i, j is (1, 2 … n), and satisfies 0 ≦ mijLess than or equal to 1 and
Figure FDA0001653360590000022
Ikis the space vector of interest, | I, of node kiI is the norm of the space vector of interest of node I, IkAnd | is the modulus of the interest space vector of node k.
3. The community partitioning method based on the interest space of claim 1, wherein: after the handshake messages are broadcast and received for a period of time in the third step, the number of neighbor nodes in the neighbor node table of each node is the same.
4. The community partitioning method based on the interest space of claim 1, wherein: after the handshake messages are broadcast and received for a period of time in the third step, the number of neighbor nodes in the neighbor node table of each node is randomly distributed.
CN201810436814.8A 2018-05-08 2018-05-08 Community division method based on interest space Active CN108596778B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810436814.8A CN108596778B (en) 2018-05-08 2018-05-08 Community division method based on interest space

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810436814.8A CN108596778B (en) 2018-05-08 2018-05-08 Community division method based on interest space

Publications (2)

Publication Number Publication Date
CN108596778A CN108596778A (en) 2018-09-28
CN108596778B true CN108596778B (en) 2022-01-28

Family

ID=63635924

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810436814.8A Active CN108596778B (en) 2018-05-08 2018-05-08 Community division method based on interest space

Country Status (1)

Country Link
CN (1) CN108596778B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111383125B (en) * 2020-03-16 2023-09-01 中国联合网络通信集团有限公司 Community dividing method, system, terminal equipment and storage medium
CN111510385B (en) * 2020-06-18 2020-09-25 中国人民解放军国防科技大学 Method and device for improving end-to-end delay of delay tolerant network based on behavior similarity model

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101916256A (en) * 2010-07-13 2010-12-15 北京大学 Community discovery method for synthesizing actor interests and network topology
CN102685255A (en) * 2012-06-01 2012-09-19 重庆邮电大学 Distributed opportunistic network community division method
CN105095403A (en) * 2015-07-08 2015-11-25 福州大学 Parallel community discovery algorithm based on mixed neighbor message propagation
CN106301888A (en) * 2016-07-27 2017-01-04 西安电子科技大学 Based on core node and the network community division method of community's convergence strategy

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101916256A (en) * 2010-07-13 2010-12-15 北京大学 Community discovery method for synthesizing actor interests and network topology
CN102685255A (en) * 2012-06-01 2012-09-19 重庆邮电大学 Distributed opportunistic network community division method
CN105095403A (en) * 2015-07-08 2015-11-25 福州大学 Parallel community discovery algorithm based on mixed neighbor message propagation
CN106301888A (en) * 2016-07-27 2017-01-04 西安电子科技大学 Based on core node and the network community division method of community's convergence strategy

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
移动自组网中节点兴趣导向的人类运动模型;陈家旭等;《西安交通大学学报》;20131231;第47卷(第12期);第110-115页 *

Also Published As

Publication number Publication date
CN108596778A (en) 2018-09-28

Similar Documents

Publication Publication Date Title
Chen et al. Leveraging social networks for P2P content-based file sharing in disconnected MANETs
CN108112049B (en) A kind of wireless sensor network efficiency optimization cluster-dividing method based on whale group algorithm
Atsan et al. SCALAR: Scalable data lookup and replication protocol for mobile ad hoc networks
CN108596778B (en) Community division method based on interest space
Tahir et al. Logical clusters in a DHT-Paradigm for scalable routing in MANETs
Xiao et al. Efficient distributed skyline queries for mobile applications
Wang et al. An intelligent UAV based data aggregation algorithm for 5G-enabled internet of things
Gupta et al. Self-stabilizing multicast protocols for ad hoc networks
Batabyal et al. Analysing social behaviour and message dissemination in human based delay tolerant network
Rajasekar et al. Energy efficient cluster formation in wireless sensor networks based on multi objective bat algorithm
Zhi et al. An improved ant colony routing algorithm for WSNs
Zhuo et al. Self-organized emergence of navigability on small-world networks
Kara A new routing algorithm in MANETs: Position based hybrid routing
Taheri et al. Achieving Performability and Reliability of Data Storage in the Internet of Things
Nghiem et al. Bichromatic reverse nearest neighbors in mobile peer-to-peer networks
Gupta et al. Self-stabilizing protocol for shortest path tree for multi-cast routing in mobile networks
CN109120438B (en) Data cooperative transmission method and system under opportunity network
Lee et al. Cluster and location based overlay multicast in mobile ad hoc and sensor networks
CN114817653A (en) Unsupervised community discovery method based on central node graph convolutional network
Merhi et al. Decentralized clustering in VANET using adaptive resonance theory
Orlinski et al. Distributed expectation-based spatio-temporal cluster detection for pocket switched networks
Rajpoot et al. An enhanced approach for energy optimization using protocol based on multi factors
Dasgupta et al. An Energy Efficient Genetic Approach for Clustering of Wireless Sensor Network
Li et al. Comparative research on the formation of backbone media of wireless self-organizing network by DLA-GF algorithm and ant colony algorithm
Melvin et al. Hitchhiker: A wireless routing protocol in a delay tolerant network using density-based clustering

Legal Events

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