CN105898679A - Method for collecting heterogeneous sensing data based on representative nodes and multilayer fusion - Google Patents

Method for collecting heterogeneous sensing data based on representative nodes and multilayer fusion Download PDF

Info

Publication number
CN105898679A
CN105898679A CN201610196665.3A CN201610196665A CN105898679A CN 105898679 A CN105898679 A CN 105898679A CN 201610196665 A CN201610196665 A CN 201610196665A CN 105898679 A CN105898679 A CN 105898679A
Authority
CN
China
Prior art keywords
node
data
base station
jumping
value
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.)
Granted
Application number
CN201610196665.3A
Other languages
Chinese (zh)
Other versions
CN105898679B (en
Inventor
刘语欣
刘安丰
淡州阳
廖志军
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.)
Central South University
Original Assignee
Central South 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 Central South University filed Critical Central South University
Priority to CN201610196665.3A priority Critical patent/CN105898679B/en
Priority to PCT/CN2016/090088 priority patent/WO2017166539A1/en
Publication of CN105898679A publication Critical patent/CN105898679A/en
Application granted granted Critical
Publication of CN105898679B publication Critical patent/CN105898679B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/30—Services specially adapted for particular environments, situations or purposes
    • H04W4/38—Services specially adapted for particular environments, situations or purposes for collecting sensor information
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W52/00—Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02—Power saving arrangements
    • H04W52/0209—Power saving arrangements in terminal devices
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Landscapes

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

Abstract

The invention discloses a method for collecting heterogeneous sensing data based on representative nodes and multilayer fusion. The method is composed of two parts including, 1 that for each kind of sensing data, multiple data coverage sets are formed in a whole wireless sensor network, wherein the difference value between data in the same data coverage set is less than a prescribed threshold value, so that the sensing value of each data coverage set can be represented by a representative node and the data transmitted in the network is decreased; and 2 that in a process of routing the data of the representative node to a base station, the representative nodes of the same kind of data may converge on the same path as much as possible according to a routing algorithm provided by the invention so that the same kind of data is subjected to data fusion again and the data required to be transmitted to the base station is decreased again. The method may obviously prolong the service life of the network.

Description

A kind of based on representing the heterogeneous sensor method of data capture that node merges with multilayer more
Technical field
The invention belongs to wireless network isomeric data assembling sphere, particularly to a kind of sensing number that can improve network life According to collection method.
Background technology
Wireless sensor network be by substantial amounts of each other by multi-hop wireless link and the sensor node communicated with The wireless network that the mode of self-organizing and multi-hop is constituted, can apply to industry monitoring widely, agricultural, civilian, environmental monitoring, Battlefield, ocean, it is considered to be the Key technology of the important internet of things in future in the various particular surroundings such as fire and application One of.Present in wireless sensor network research, a vital problem is the most i.e. can be by the data of network aware It is sent to base and can improve again the life-span of network.The life-span of network is the matter of utmost importance of wireless sensor network research.Thus, carry The research gone out is the most, the most only introduces relevant research related to the present invention.Its background technology is discussed below:
Reduce the energy ezpenditure of node, thus just can improve network life.Thus the research saving node energy becomes An important research contents in wireless sensor network.The important method saving energy is the joint reducing and participating in perception Point quantity, only selects a part of node to participate in the monitoring of event or target from node. so can reach following 2 mesh Mark: (1) only selects a part of node to carry out perception from event or target area, thus decreases the nodes participating in perception Amount, so these nodes being not involved in perception just can enter sleep state, thus save energy. and (2) reduce accordingly to thing Part or the perception of the little node of target apperception information content, and select can to the node that event or target monitoring information quality are high On the premise of not reducing event and target monitoring quality, reduce the data volume mailing to base station, thus improve network life.
In the method having been proposed that, independent for the data of perception is mail to base station by each node, even if thus selecting Less node is monitored event and is still suffered from bigger energy ezpenditure in the case of environment. thus document [Hung C C, Peng W C, Lee W C.Energy-aware set-covering approaches for approximate data collection in wireless sensor networks.IEEE Transactions on Knowledge and Data Engineering, 2012；24 (11): 1993-2007.] proposing a kind of method representing node, this method is logical Cross in data cover set, choose a next perception value replacing all nodes of whole set for approximation of value being referred to as representative joint Method. thus, and it is not required to mail to each node perceived value base station, thus represents nodal method and can further reduce and send out Deliver to the data volume of base station, thus energy can be saved, but this requires that application system can tolerate certain data error scope.
The another kind of important method saving energy is data fusion. the node of wireless sensor network disposition is due to perception Information between there is correlation, such as: when certain event occur after, have multiple node perceived to event simultaneously, thus do not have completely It is necessary the node data of each perception events information is sent to base station, but by these node perceived before mailing to base station Data carry out data fusion, only the data that can accurately express event information are extracted, namely send out after data fusion again Toward base station. such that the data volume that node needs to send can be reduced, the most do not reduce the information acquired in system.
But, existing research there is the problem that (1) represents in nodal method, and only research can only process single The network of perception value, and the network of multiple perception value can not be processed.(2) currently proposed representative node selecting method is the most multiple Miscellaneous, need the energy consumed the most；(3) each represents node is independent base station of mailing to, thus energy ezpenditure compares Greatly.Fail to use Data fusion technique.Current technology is caused to there is also bigger improvements.Therefore, it is necessary to design Plant heterogeneous sensor method of data capture to improve network life.
Summary of the invention
The technical problem to be solved is to provide a kind of based on representing the heterogeneous sensor that node merges with multilayer more Method of data capture, should can reduce, based on representing the node heterogeneous sensor method of data capture with multilayer fusion, the data transmitted more Amount, improves network life.
The technical solution of invention is as follows:
A kind of based on representing the heterogeneous sensor method of data capture that node merges with multilayer, at wireless sensor network more In, each node can sense the different types of perception value such as multiclass perception value, such as temperature, humidity, air pressure, multiclass perception number It is stated to be isomeric data；By following two methods to improve network life, will the information of perception data at satisfied application need On the premise of asking, network life is made to maximize as far as possible:
Method 1: for each class perception data, forms multiple data cover set in whole wireless sensor network, Data cover set is designated as DCS, and the difference of this type of perception data value of each data cover set interior nodes sets less than application Fixed threshold values, is provided with a head node as representing node in each data cover set；
Method 2: for each class perception data, initiated the route to base station to transmit such perception number by representing node According to；The route of same class perception data, during routeing to base station, collects on a routed path as much as possible, passes through Base station it is transferred to after data fusion.
In method 1, the forming process of set head node is as follows:
Each node randomly chooses a number between 0～1, if the number for node A selection is less than threshold value P (A), then This node acts as set head node；Wherein being calculated as follows of P (A):
In formula, mod is remainder function；P is the probability that node is elected as set head node, according to the demand applied It is previously set；γ is to circulate the wheel number carried out at present；G is nearestWheel is not elected as the node set of set head node； EavgAnd EARepresent node A and the average energy of its neighbor node and the present energy of node A respectively.
In method 1, the forming process of the set member's node in data cover set is as follows:
(1), when initializing, it is ∞ that each node arranges the jumping figure of oneself arrival each class DCS set head node, the most infinite Greatly；
(2) corresponding any one set head node A, first broadcasts, and broadcast is CMA, represent and create set member The message of node A；Broadcast CM of node AAComprise 6 contents:
A the ID of () set head node, is designated as CMA.h；
B the ID of () node A oneself, is set to CMA.ID；
C () node arrives the jumping figure of set head node A
D () limits broadcast value Δ in certain jumping figure, be designated as CMA.Δ；
The type of (e) perception valueψ is the kind of perception value,The kind of table perception value Sum；
F the perception value of () set head node A, is designated as CMA.υ；
The broadcast of the kth class perception value that set head node A broadcasts oneself is: CMA.h=A, CMA.ID=A, Represent that the maximum hop count (its value is relevant to application, is typically set to 8) of node is initiated in the distance broadcast of broadcast, CMA. ψ=k, CMA. υ=xA；xAIt is the concrete numerical value of perception value；Any node receiving broadcast, such as node B, if receiving To broadcast packet be CMA, proceed as follows successively:
(I) CM is first readA. ψ, compares the CM oneself preservedA. ψ class data arrive CMA.h[the ID of head node is gathered] Jumping figure HAWhether it is less than
If it is, abandon this broadcast, do not do any operation, continue waiting for new message；
Set head node CM oneself is arrived if it is not, then updateA.h jumping figureAnd arrival is set The set head node of kth class data is CMA.h, the down hop arriving set head node route is CMA.ID, next step behaviour is continued Make；
(II) by the CM of oneselfA. ψ class data perception value and set head node CMA.h perception value CMA. the value of υ is carried out Compare, ifThen B sends addition message JM participating in set head node A to set head node A, and continues the (III) operation；The process adding message (JM) is fairly simple. for prior art；Receive the node of JM message with Minimal routing Method forwards JM to set head node, thus final JM arrives set head node, in set head node record data cover set Each member node；
D (.) is difference functions,RepresentDifference, εkRepresent the difference threshold of kth class data perception value Value, for set in advance oneself know value；
(III) if CMA. Δ-1=0, then all operations terminates；The node B otherwise receiving message updates broadcast packet After content, outwards broadcast carries out next circulation again；Broadcast packet content after renewal is: CMA.h=A, CMA.ID=B, CMA.Δ =CMA. Δ-1, CMA. ψ=k, CMA. υ=xA。
Fail to become set head node in the first round, also fail to enough become the node of data cover set (set member) A ' automatically becomes set head node.
In method 2, use data fusion method for routing, after being merged by homogeneous data, then by representing node to remittance Poly-node sends data；
Described data fusion method for routing, comprises the following steps:
Step S1: the jumping figure that oneself is arrived base station by base station is 0, each node arranges every class observed value and arrives the jumping of base station Number is ∞, thus totalIndividual ∞, then, base station is outwards broadcasted, and oneself kth class perception value is arrived by the node receiving broadcast Reach the jumping figure of base station to compare with the jumping figure in broadcast packet, if the kth class perception that the jumping figure in broadcast packet adds 1 to be preserved less than oneself Value arrives the jumping figure of base station, then add, with the jumping figure of broadcast packet, the jumping figure that 1 replacement preserves, and then, oneself is arrived base station by this node Jumping figure carry out same broadcast so that each node obtains arriving the minimum hop count of kth class perception value of base station；
Step S2: when kth class set syncephalon node has data to need to be transferred to base station, uses shortest route algorithm, successively The node that selected distance base station nearly is jumped route as via node, transfers data to base station；
Step S3: oneself kth class is arrived the jumping figure setting of base station by the node carrying out kth class data route in step s 2 It is 0；Then, jumping figure becomes the node of 0 and arrives the jumping figure of base station to external diffusion oneself, and its method is similar with step S1, impacted In the range of node change oneself arrive base station jumping figure so that other homogeneous data route can attracted to the road first created By upper, thus many data route can collect in after carrying out data fusion in a route, then mails to base station, required by reducing Data volume to be sent, to reduce energy ezpenditure, improves network life.
In wireless sensor network, the information of multiple node perceived may be the most close, if the information of node perceived The scope of difference is in the threshold values specified, it is possible to represents with the perception information of a node and exists with the difference with its perception information The perception value of other node in threshold range.The data volume that thus can reduce the required transmission of network improves network life.
The present invention further proposes according to above principle and represents the isomeric data receipts that node merges with multilayer more Diversity method.In the present invention, as it is shown in figure 1, not only perception value in certain threshold range node composition one set, this collection The perception value closing all nodes represents by a perception value representing node, it is only necessary to the value of this node mails to base station the most permissible ?.
Compared with prior art, it is an advantage of the current invention that:
The present invention based on representing the heterogeneous sensor method of data capture that node and multilayer merge more, be distributed representative Node selecting method, carries out representing node election from any one node just can complete as long as random, so that algorithm energy Consume little, the most effectively；The situation of multiple perception value can be processed；Data mail to base station after again merging, thus enter one Step decreases data volume, improves network life.
To studying than ever, the improvement of the present invention is as follows: (a) conventional research is only capable of the feelings processing single perception value Condition.In actual network, sensing node is often equipped with multiple sensors part, and such as one sensing node tends to simultaneously The perception value such as sense temperature, humidity, air pressure.Thus the perception value of this isomery can not represent by a typical value, and it is intended to Represent by multiple different types of typical values, thus, in the present invention, the different perception value to isomery use different typical values Represent；B () relaxes the deficiency needing suitably to select to represent node to reduce transmitted data on network in research in the past.With If toward method in the typical value that selects different, then the number representing node also differs, and namely needs to send base station to Packet number different.Thus often in order to select suitable typical value and the complexity of increasing method.The present invention relaxes This constraint, it is only necessary to it is typical value that random appointment does not becomes the value of the node in set.Because the present invention is also to generation Table node sends the data of base station to and has carried out secondary data fusion, thus on the one hand greatly reduces the data being sent to base station Amount, and decrease the complexity selecting typical value.
Another important improvement of the present invention is when the value representing node mails to base station, uses the method that route is converged, The route making same type perception value can merge together, thus carries out data fusion, so can greatly reduce and be sent to The data volume of base station, improves network life.As it is shown in figure 1, route origin 1, route origin 2, the data of route origin 3 are last all Merge in a route, thus these similar data carry out data fusion again to reduce the data volume of data transmission.And with Past method be each route be all individually to be routed to base station, thus its volume of transmitted data is very big, thus the side of the present invention Method can be greatly improved network life.Through two kinds of improvement as above, the inventive method can simplify conventional algorithm, expands The scope of application of algorithm, reduces the data volume being sent to base station, improves network life.
Inventive method is divided into two parts.1: for each class perception data, in whole wireless sensor network Forming multiple data cover set, in same data cover set, the difference between data is less than the threshold values specified.So, each Individual data cover set just can represent node with one and represent the perception value of whole set.This is the minimizing net of first aspect The data volume that network is transmitted；2: represent the data of node during routeing to base station, the representative node meeting of same class data Converge on a paths, so that same class data carry out number of times again according to the routing algorithm of present invention proposition is as much as possible According to fusion, thus again reduce and need to be transferred to the data volume of base station.Thus the inventive method can significantly improve the network longevity Life.
Accompanying drawing explanation
Fig. 1 is based on representing the heterogeneous sensor data route schematic diagram that node merges with multilayer more；
Fig. 2 is set forming process schematic diagram；
Fig. 3 is the forming process schematic diagram of route；
Fig. 4 is the comparison diagram of the data volume undertaken at distinct methods lower node；
Fig. 5 is the contrast of the Energy Expenditure Levels at distinct methods lower node
Fig. 6 is that the ceiling capacity of network consumes comparison diagram；
Fig. 7 is that the ceiling capacity under different pieces of information coefficient correlation consumes comparison diagram；
Fig. 8 is the comparison diagram of distinct methods lower network residual energy dose rate.
Detailed description of the invention
For the ease of understanding the present invention, below in conjunction with Figure of description and preferred embodiment, invention herein is done more complete Face, describe meticulously, but protection scope of the present invention is not limited to specific embodiment.
Unless otherwise defined, all technical terms used hereinafter are generally understood that implication phase with those skilled in the art With.Technical term used herein is intended merely to describe the purpose of specific embodiment, is not intended to limit the present invention's Protection domain.
Embodiment:
Such as Fig. 1-3, a kind of based on representing the heterogeneous sensor method of data capture that node merges with multilayer, at wireless sensing more In device network, each node can sense the different types of perception value such as multiclass perception value, such as temperature, humidity, air pressure, multiclass Perception data is referred to as isomeric data；By following two methods to improve network life, will perception data information meet On the premise of application demand, network life is made to maximize as far as possible:
Method 1: for each class perception data, forms multiple data cover set in whole wireless sensor network, Data cover set is designated as DCS, and the difference of this type of perception data value of each data cover set interior nodes sets less than application Fixed threshold values, is provided with a head node as representing node in each data cover set；
Method 2: for each class perception data, initiated the route to base station to transmit such perception number by representing node According to；The route of same class perception data, during routeing to base station, collects on a routed path as much as possible, passes through Base station it is transferred to after data fusion.
In method 1, the forming process of set head node is as follows:
Each node randomly chooses a number between 0～1, if the number for node A selection is less than threshold value P (A), then This node acts as set head node；Wherein being calculated as follows of P (A):
In formula, mod is remainder function；P is the probability that node is elected as set head node, according to the demand applied It is previously set；γ is to circulate the wheel number carried out at present；G is nearestWheel is not elected as the node set of set head node； EavgAnd EARepresent node A and the average energy of its neighbor node and the present energy of node A respectively.
In method 1, the forming process of the set member's node in data cover set is as follows:
(1), when initializing, it is ∞ that each node arranges the jumping figure of oneself arrival each class DCS set head node, the most infinite Greatly；
(2) corresponding any one set head node A, first broadcasts, and broadcast is CMA, represent and create set member The message of node A；Broadcast CM of node AAComprise 6 contents:
A the ID of () set head node, is designated as CMA.h；
B the ID of () node A oneself, is set to CMA.ID；
C () node arrives the jumping figure of set head node A
D () limits broadcast value Δ in certain jumping figure, be designated as CMA.Δ；
The type of (e) perception valueψ is the kind of perception value,The kind of table perception value Sum；
F the perception value of () set head node A, is designated as CMA.υ；
The broadcast of the kth class perception value that set head node A broadcasts oneself is: CMA.h=A, CMA.ID=A, Represent that the maximum hop count (its value is relevant to application, is typically set to 8) of node is initiated in the distance broadcast of broadcast, CMA. ψ=k, CMA. υ=xA；xAIt is the concrete numerical value of perception value；Any node receiving broadcast, such as node B, if receiving To broadcast packet be CMA, proceed as follows successively:
(I) CM is first readA. ψ, compares the CM oneself preservedA. ψ class data arrive CMA.h[the ID of head node is gathered] Jumping figure HAWhether it is less than
If it is, abandon this broadcast, do not do any operation, continue waiting for new message；
Set head node CM oneself is arrived if it is not, then updateA.h jumping figureAnd arrival is set The set head node of kth class data is CMA.h, the down hop arriving set head node route is CMA.ID, next step behaviour is continued Make；
(II) by the CM of oneselfA. ψ class data perception value and set head node CMA.h perception value CMA. the value of υ is carried out Compare, ifThen B sends addition message JM participating in set head node A to set head node A, and continues the (III) operation；The process adding message (JM) is fairly simple. for prior art；Receive the node of JM message with Minimal routing Method forwards JM to set head node, thus final JM arrives set head node, in set head node record data cover set Each member node；
D (.) is difference functions,RepresentDifference, εkRepresent the difference threshold of kth class data perception value Value, for given value set in advance；
(III) if CMA. Δ-1=0, then all operations terminates；The node B otherwise receiving message updates broadcast packet After content, outwards broadcast carries out next circulation again；Broadcast packet content after renewal is: CMA.h=A, CMA.ID=B, CMA.Δ =CMA. Δ-1, CMA. ψ=k, CMA. υ=xA。
Fail to become set head node in the first round, also fail to enough become the node of data cover set (set member) A ' automatically becomes set head node.
In method 2, use data fusion method for routing, after being merged by homogeneous data, then by representing node to remittance Poly-node sends data；
Described data fusion method for routing, comprises the following steps:
Step S1: the jumping figure that oneself is arrived base station by base station is 0, each node arranges every class observed value and arrives the jumping of base station Number is ∞, thus totalIndividual ∞, then, base station is outwards broadcasted, and oneself kth class perception value is arrived by the node receiving broadcast Reach the jumping figure of base station to compare with the jumping figure in broadcast packet, if the kth class perception that the jumping figure in broadcast packet adds 1 to be preserved less than oneself Value arrives the jumping figure of base station, then add, with the jumping figure of broadcast packet, the jumping figure that 1 replacement preserves, and then, oneself is arrived base station by this node Jumping figure carry out same broadcast so that each node obtains arriving the minimum hop count of kth class perception value of base station；
Step S2: when kth class set syncephalon node has data to need to be transferred to base station, uses shortest route algorithm, successively The node that selected distance base station nearly is jumped route as via node, transfers data to base station；
Step S3: oneself kth class is arrived the jumping figure setting of base station by the node carrying out kth class data route in step s 2 It is 0；Then, jumping figure becomes the node of 0 and arrives the jumping figure of base station to external diffusion oneself, and its method is similar with step S1, impacted In the range of node change oneself arrive base station jumping figure so that other homogeneous data route can attracted to the road first created By upper, thus many data route can collect in after carrying out data fusion in a route, then mails to base station, required by reducing Data volume to be sent, to reduce energy ezpenditure, improves network life.
To each class perception data, the data that the data cover set based algorithm proposed according to the present invention is formed one by one are covered Lid set, each set is represented the value of this data cover set by the perception value referred to as covering set head node.Cover collection After hop algorithm runs, each class data have multiple data cover set, and the node arbitrarily having this type of perception value must belong to Some set in this type of data cover set.Each class perception value is run same algorithm, each class in such network Perception value is all comprised by multiple data cover set.Then, each data cover set by the head node gathered initiate to The route of base station, in the present invention, the route of initiation has the characteristic that similar route can pool together, thus, at multiple collection Syncephalon node collects on Yi Tiaolu (routed path) in routing procedure, it becomes possible to carries out data fusion, thus subtracts further It is sent to the data volume of base station required for Shao, thus improves network life.
The forming process of following set head node is all referring to the set cephalomere point-shaped of kth class data cover set (DCS) One-tenth process. it is similar owing to the forming process of other class data cover set assembles process with kth class data cover, thus Have only to perform altogether following processIt is secondary, it becomes possible to the perception value of classes all in network is formed data cover set.
In the selection algorithm of set head node, each node randomly chooses a number between 0～1, for node A such as The number that fruit selects is less than threshold value P (A), then this node acts as set head node.Wherein being calculated as follows of P (A):
P ( A ) = p 1 - p ( γ mod 1 p ) E c E a v g , A ∈ G 0 , A ∉ G ,
In formula, mod is remainder function, and p is the probability that node is elected as set head node, and γ is that circulation is carried out at present Wheel number；G is nearestWheel is not elected as the node set of set head node.From formula (8), it was elected to the joint of bunch head Point can not become a bunch head in ensuing 1/p wheel circulation.EavgAnd EcRepresent the flat of node A and its neighbor node respectively All energy and present energy of node A.
(2) forming process of set member's node
When starting, it is ∞ that each node arranges the jumping figure of oneself arrival each class DCS set head node. through above Operational set head node produce after, each node becoming set head node carries out subsequent operation:
The forming process of set member's node is discussed below as a example by set head node A. first set head node A is carried out extensively Broadcast, if broadcast is CMA(representing the message creating set member node A). broadcast CM of node AAMainly comprise 6 Main contents: the ID of (a) set head node, are designated as CMA.h. the ID of (b) node A oneself, is set to CMA.ID. (c) node arrives collection The jumping figure of syncephalon node AD () limits broadcast value Δ in certain jumping figure, be designated as CMA. Δ. the type of (e) dataF the perception value of () set head node A, is designated as CMA. υ. so set head node A broadcasts oneself the The broadcast of k class perception value is: CMA.h=A, CMA.ID=A,CMA. ψ=k, CMA. υ=xA. Ren Hejie Receive the node of broadcast, such as node B, if the broadcast packet received is CMA. proceed as follows successively:
(I) CM is first readA. ψ, compares the CM oneself preservedA. ψ class data arrive CMA.h jumping figure HAWhether it is less thanSet head node CM oneself is arrived if it is not, then updateA.h jumping figureAnd arrival is set The set head node of kth class data is CMA.h, the down hop arriving set head node route is CMA.ID, (II) behaviour is continued Make. otherwise, abandon this broadcast, do not do any operation.
(II) by the CM of oneselfA. ψ class data perception value and set head node CMA.h perception value CMA. the value of υ is carried out Compare, ifThen B sends the addition message (JM) participating in set head node A to set head node A, and continues The operation of (III). the process adding message (JM) is fairly simple. receive the node of JM message with Minimal routing method to set Head node forwards JM, thus final JM arrives set head node, set head node record each member of data cover set joint Point.
(III) if CMA. Δ-1=0, then all operations terminates. the most outwards broadcasts after otherwise updating the content of broadcast packet. New broadcast packet content is: CMA.h=A, CMA.ID=B, CMA. Δ=CMA. Δ-1, CMA. ψ=k, CMA. υ=xA.
The forming process of data acquisition system member is as shown in Figure 2. set in node A competition and become set head node, then node A to Outer broadcast, after the node in broadcasting area receives the broadcast message of node A, if it find that oneself be belonging to the perception value with A Distance in error range, then to set head node A application add. at this moment node P, B, G, J, D add set head node A, Become the member (member) of node A place DCS, as shown in Fig. 2 (1). same, node P, B, G, J, D, C the most outwards broadcast, and make Node N, E, K, H, S, I, O add, such as Fig. 2 (2) if shown in. settingThen in the range of 2 jumpings, node can add set The DCS. at head node A place is same, if F also competes as gathering head node, then node F also does same operation, thus Form the DCSs. shown in Fig. 2 (3)
Routing procedure is similar to jumping figure diffusion protocol as it is shown on figure 3, Routing Protocol begins with, it is therefore an objective to make node all Know that oneself arrives the jumping figure of base station. but differ with common jumping figure diffusion protocol, each node hasClass perception value, with Class perception value could merge, thus, the purpose of weight data fusion route is intended to make similar perception value converge to same route So that data fusion. thus each node is preservedClass perception value arrives the minimum hop count of base station. when just starting, thus, Oneself is arrived the jumping figure of base station by base station be that 0. each node arrange every class observed value to arrive the jumping figure of base station is ∞, thus totalIndividual ∞. then, outwards broadcasting, oneself kth class perception value is arrived jumping figure and the broadcast of base station by the node receiving broadcast Jumping figure in bag compares, if the jumping figure in broadcast packet add 1 less than oneself preserve kth class perception value arrive base station jumping figure, Then add, with the jumping figure of broadcast packet, the jumping figure that 1 replacement preserves. then, oneself is arrived the jumping figure of base station and carries out same wide by this node Broadcast. so, each node obtains arriving the minimum hop count of the kth class perception value of base station. and we give tuple evidence by Fig. 3 Merge route sets up process. set the kind of data in networkFig. 3 (1) gives after carrying out jumping figure broadcast, Mei Gejie The minimum hop count of the every class observed value arriving base station of some record, the most same node all classes perception value arrives the value of base station Identical. then, if set head node has data to be sent to base station, then node uses this type of other minimum according to the classification of data Jumping figure routing mechanism is routed to base station. the at this moment route track as shown in Fig. 3 (1). and unlike conventional routing mechanism, at this In the heavy data fusion method for routing of invention, the jumping figure that oneself kth class arrives base station is set to 0. such as by the node in this route Black route track shown in Fig. 3 (2). then, jumping figure becomes the node of 0 node change in external diffusion, impacted scope The jumping figure arriving base station of oneself, as shown in Fig. 3 (3). and the jumping figure arriving base station due to the upper node of this route is 0, thus its Its homogeneous data route in the route that will attracted to create at first, thus many data route can collect in a route Mail to base station. as shown in Fig. 3 (4), the numerical value in circle such as 4/4 represents the jumping figure arriving base station, and each circle has two-digit, point Do not represent that two class data are to the jumping figure arriving base station；
Including the route of two class data in Fig. 3, solid line look is a class, and the route of dotted line is a class.Same class data Route to pool together as far as possible, in order to merges.Inhomogeneous data then need not converge.
By Fig. 3, routing procedure is described.Fig. 3 (1) gives after the jumping figure carrying out step S1 is broadcasted, and each node is remembered The minimum hop count of the every class observed value arriving base station of record, Fig. 3 is provided with two class perception datas.The all differences of the most same node The jumping figure value that perception data arrives base station is identical. then, if there being set head node to have data to be sent to base station, then and node foundation The classification of data uses this type of other minimum hop routing mechanism to be routed to base station. the at this moment route track as shown in Fig. 3 (1). Unlike conventional routing mechanism, in the method 2 of the present invention, oneself kth class is arrived base station by the node in this route Jumping figure is set to the 0. solid line route track as shown in Fig. 3 (2). and then, jumping figure becomes the node of 0 to external diffusion, impacted model Enclose the jumping figure arriving base station of interior node change oneself, as shown in Fig. 3 (3). owing to the upper node of this route arrives base station Jumping figure is 0, thus other homogeneous data route in the route that will attracted to first create, thus many data route can be collected A route mails to base station. as shown in Fig. 3 (4).
The method that the method for the present invention compares therewith is: (1) the 1st kind is directly to send out to base station after each node produces data Sending, the most referred to as immediate data collection method (direction data collection, DDC) [sees document: Liu A, Ren J, Li X, et al.Design principles and improvement of cost function based energy aware routing algorithms for wireless sensor networks.Computer Networks, 2012,56 (7): 1951-1967]. the 2nd kind is document [Hung C C, Peng W C, Lee W C.Energy- aware set-covering approaches for approximate data collection in wireless Sensor networks.IEEE Transactions on Knowledge and Data Engineering, 2012；24 (11) method: 1993-2007.], each data cover collection share the perception value representing its set of node on behalf, and directly Connecing to base station repeats, this method is called to represent node data collection method (R-node data collection, RNDC). this Inventive method is MRRF.
From the experimental result of Fig. 4 and 5 can be seen that MRRF method that the present invention proposes can equalising network energy ezpenditure, Nearly base station area, its energy ezpenditure is much smaller than DDC Yu RNDC method, and when remote base station area, its energy ezpenditure is but higher than DDC With RNDC method. the reason causing this phenomenon is: MRRF method due to use represent node with heavily merge method for routing thus The energy ezpenditure making nearly base station area node is significantly less than other 2 methods. and in remote base station area, MRRF method uses relatively Big DCS, thus node spends in and creates on DCS with the energy ezpenditure safeguarded much larger than other two kinds of methods, thus at remote base The energy ezpenditure of region MRRF method of standing is the highest. but, this does not reduce the performance of MRRF method, takes full advantage of on the contrary Network energy consumes, and improves the capacity usage ratio of network.
From fig. 6, it can be seen that the inventive method is much smaller than it relative to other two kinds of methods, the ceiling capacity consumption of network Its two kinds of method.And the ceiling capacity consumption of network just determines network life, say, that the method for the present invention can be significantly The raising network life of degree.It is to represent node data collection method respectively, 1.6 times of immediate data collection method lower network life-span To 3.2 times.
Fig. 7 gives the ceiling capacity consumption contrast under different pieces of information coefficient correlation, as can be seen from the figure: represent joint Point data collection method and immediate data collection method lower network ceiling capacity consumption are 2.12～3.73 times of the inventive method, 4.24～7.13 times. illustrate that the inventive method can significantly reduce the energy ezpenditure of node.
Fig. 8 gives the contrast of distinct methods lower network residual energy dose rate, it can be seen that the inventive method is off line Network residual energy dose rate is only about 10%, and other two kinds of methods are up to 70%, illustrate that the inventive method can effectively utilize energy Amount.

Claims (5)

1. based on representing the heterogeneous sensor method of data capture that node merges with multilayer, in wireless sensor network more, Each node can sense multiclass perception value, it is characterised in that by following two methods to improve network life:
Method 1: for each class perception data, forms multiple data cover set, data in whole wireless sensor network Covering set and be designated as DCS, the difference of this type of perception data value of each data cover set interior nodes is less than application settings Threshold values, is provided with a head node as representing node in each data cover set；
Method 2: for each class perception data, initiated the route to base station to transmit such perception data by representing node；With The route of one class perception data, during routeing to base station, collects on a routed path, through data as much as possible Base station it is transferred to after fusion.
The most according to claim 1 based on representing the heterogeneous sensor method of data capture that node merges with multilayer, it is special more Levying and be, in method 1, the forming process of set head node is as follows:
Each node randomly chooses a number between 0～1, if the number for node A selection is less than threshold value P (A), then this joint Point acts as set head node；Wherein being calculated as follows of P (A):
In formula, mod is remainder function；P is the probability that node is elected as set head node, prior according to the demand of application Set；γ is to circulate the wheel number carried out at present；G is nearestWheel is not elected as the node set of set head node；EavgAnd EA Represent node A and the average energy of its neighbor node and the present energy of node A respectively.
The most according to claim 1 based on representing the heterogeneous sensor method of data capture that node merges with multilayer, it is special more Levying and be, in method 1, the forming process of the set member's node in data cover set is as follows:
(1), when initializing, it is ∞ that each node arranges the jumping figure of oneself arrival each class DCS set head node, i.e. infinitely great；
(2) corresponding any one set head node A, first broadcasts, and broadcast is CMA, represents and creates set member node A Message；Broadcast CMA of node A comprises 6 contents:
A the ID of () set head node, is designated as CMA.h；
B the ID of () node A oneself, is set to CMA.ID；
C () node arrives the jumping figure of set head node A
D () limits broadcast value Δ in certain jumping figure, be designated as CMA.Δ；
The type of (e) perception valueψ is the kind of perception value,The kind sum of table perception value；
F the perception value of () set head node A, is designated as CMA.υ；
The broadcast of the kth class perception value that set head node A broadcasts oneself is: CMA.h=A, CMA.ID=A, Represent that the maximum hop count of node, CM are initiated in the distance broadcast of broadcastA.ψ=k, CMA. υ=xA；xAIt is the concrete numerical value of perception value； Any node receiving broadcast, such as node B, if the broadcast packet received is CMA, proceed as follows successively:
(I) CM is first readA.ψ, compare the CM oneself preservedA.ψClass data arrive CMA.h jumping figure HAWhether it is less than
If it is, abandon this broadcast, do not do any operation, continue waiting for new message；
Set head node CM oneself is arrived if it is not, then updateA.h jumping figureAnd arrival kth class is set The set head node of data is CMA.h, the down hop arriving set head node route is CMA.ID, next step operation is continued；
(II) by the CM of oneselfA. ψ class data perception value and set head node CMA.h perception value CMA. the value of υ compares, IfThen B sends addition message JM participating in set head node A to set head node A, and continues (III) Operation；The node receiving JM message forwards JM with Minimal routing method to set head node, thus final JM arrives collection syncephalon Node, each member node in set head node record data cover set；
D (.) is difference functions,RepresentDifference, εkRepresent the difference threshold of kth class data perception value, in advance The given value first set；
(III) if CMA. Δ-1=0, then all operations terminates；The node B otherwise receiving message updates the content of broadcast packet After more outwards broadcast carry out next circulation；Broadcast packet content after renewal is: CMA.h=A, CMA.ID=B, CMA. Δ= CMA. Δ-1, CMA. ψ=k, CMA. υ=xA。
The most according to claim 3 based on representing the heterogeneous sensor method of data capture that node merges with multilayer, it is special more Levy and be, fail to become set head node in the first round, also fail to enough become the node of data cover set (set member) A ' automatically becomes set head node.
5. according to described in any one of claim 1-4 based on representing the heterogeneous sensor Data Collection side that node and multilayer merge more Method, it is characterised in that in method 2, uses data fusion method for routing, after being merged by homogeneous data, then is saved by representative Point sends data to aggregation node；
Described data fusion method for routing, comprises the following steps:
Step S1: the jumping figure that oneself is arrived base station by base station is 0, each node arranges the jumping figure of every class observed value arrival base station and is ∞, thus totalIndividual ∞, then, base station is outwards broadcasted, and oneself kth class perception value is arrived base by the node receiving broadcast The jumping figure stood compares with the jumping figure in broadcast packet, if the kth class perception value that the jumping figure in broadcast packet adds 1 to be preserved less than oneself arrives Reaching the jumping figure of base station, then add, with the jumping figure of broadcast packet, the jumping figure that 1 replacement preserves, then, oneself is arrived the jumping of base station by this node Number carries out same broadcast so that each node obtains arriving the minimum hop count of the kth class perception value of base station；
Step S2: when kth class set syncephalon node has data to need to be transferred to base station, uses shortest route algorithm, chooses successively The node that distance base station nearly is jumped route as via node, transfers data to base station；
Step S3: the jumping figure that oneself kth class arrives base station is set to 0 by the node carrying out kth class data route in step s 2； Then, jumping figure becomes the node of 0 and arrives the jumping figure of base station to external diffusion oneself, arriving of the node change in impacted scope oneself Reach the jumping figure of base station so that other homogeneous data route can attracted in the route that first creates, thus the route meeting of many data Collect in after carrying out data fusion in a route, then mail to base station.
CN201610196665.3A 2016-03-31 2016-03-31 It is a kind of based on the heterogeneous sensor method of data capture for representing node more and being merged with multilayer Expired - Fee Related CN105898679B (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201610196665.3A CN105898679B (en) 2016-03-31 2016-03-31 It is a kind of based on the heterogeneous sensor method of data capture for representing node more and being merged with multilayer
PCT/CN2016/090088 WO2017166539A1 (en) 2016-03-31 2016-07-15 Heterogeneous sensing data collection method based on multiple representative nodes and multilevel fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610196665.3A CN105898679B (en) 2016-03-31 2016-03-31 It is a kind of based on the heterogeneous sensor method of data capture for representing node more and being merged with multilayer

Publications (2)

Publication Number Publication Date
CN105898679A true CN105898679A (en) 2016-08-24
CN105898679B CN105898679B (en) 2019-09-20

Family

ID=57011727

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610196665.3A Expired - Fee Related CN105898679B (en) 2016-03-31 2016-03-31 It is a kind of based on the heterogeneous sensor method of data capture for representing node more and being merged with multilayer

Country Status (2)

Country Link
CN (1) CN105898679B (en)
WO (1) WO2017166539A1 (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107733875A (en) * 2017-09-21 2018-02-23 深圳市盛路物联通讯技术有限公司 A kind of data transmission method and system
CN110557185A (en) * 2018-05-31 2019-12-10 成都鼎桥通信技术有限公司 method and device for determining operation mode of relay equipment
CN115714770A (en) * 2022-09-23 2023-02-24 中南大学 Cloud-edge cooperative data service network balanced computing method

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111800284B (en) * 2019-04-08 2023-06-20 阿里巴巴集团控股有限公司 Edge cloud node set selection method and device and electronic equipment
CN111417084B (en) * 2020-03-26 2022-07-12 仲恺农业工程学院 Method, system, medium and device for allocating agricultural condition acquisition tasks
CN114554566B (en) * 2022-01-27 2024-07-09 哈尔滨海能达科技有限公司 Wireless link routing method, wireless link routing device, electronic equipment and storage medium
CN115695289B (en) * 2023-01-03 2023-03-17 顺霆科技(无锡)有限公司 Heterogeneous multi-hop expansion network

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101415011A (en) * 2008-10-31 2009-04-22 北京工业大学 Safety effective data polymerization method for wireless sensor network
US20090154407A1 (en) * 2007-12-17 2009-06-18 Electronics And Telecommunications Research Institute Of Daejeon Method of performing routing and allocating resources in wireless sensor network
CN101877827A (en) * 2010-05-18 2010-11-03 浙江工业大学 Building monitoring oriented wireless sensor network data acquisition method
CN101917750A (en) * 2010-07-30 2010-12-15 南京工业大学 Cross-district virtual cluster multi-hop routing method for minimum effort delivery
CN102802175A (en) * 2012-02-24 2012-11-28 东北电力大学 Energy-consumption-based routing optimization algorithm for wireless sensor network
CN104411000A (en) * 2014-12-15 2015-03-11 南昌航空大学 Method for selecting cluster head of hierarchical routing protocol in wireless sensor network

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN201426183Y (en) * 2009-05-22 2010-03-17 广东工业大学 Cluster-based wireless sensor network routing system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090154407A1 (en) * 2007-12-17 2009-06-18 Electronics And Telecommunications Research Institute Of Daejeon Method of performing routing and allocating resources in wireless sensor network
CN101415011A (en) * 2008-10-31 2009-04-22 北京工业大学 Safety effective data polymerization method for wireless sensor network
CN101877827A (en) * 2010-05-18 2010-11-03 浙江工业大学 Building monitoring oriented wireless sensor network data acquisition method
CN101917750A (en) * 2010-07-30 2010-12-15 南京工业大学 Cross-district virtual cluster multi-hop routing method for minimum effort delivery
CN102802175A (en) * 2012-02-24 2012-11-28 东北电力大学 Energy-consumption-based routing optimization algorithm for wireless sensor network
CN104411000A (en) * 2014-12-15 2015-03-11 南昌航空大学 Method for selecting cluster head of hierarchical routing protocol in wireless sensor network

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
万健，吴健荣，徐向华: "无线传感器网络定向扩散路由协议研究", 《计算机工程与科学》 *
王康: "基于分簇路由的无线传感器网络数据融合算法研究", 《中国优秀硕士学位论文全文数据库-信息科技辑》 *

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107733875A (en) * 2017-09-21 2018-02-23 深圳市盛路物联通讯技术有限公司 A kind of data transmission method and system
CN110557185A (en) * 2018-05-31 2019-12-10 成都鼎桥通信技术有限公司 method and device for determining operation mode of relay equipment
CN110557185B (en) * 2018-05-31 2022-08-19 成都鼎桥通信技术有限公司 Method and device for determining operation mode of relay equipment
CN115714770A (en) * 2022-09-23 2023-02-24 中南大学 Cloud-edge cooperative data service network balanced computing method
CN115714770B (en) * 2022-09-23 2024-04-12 中南大学 Cloud-edge collaborative data service network equalization calculation method

Also Published As

Publication number Publication date
CN105898679B (en) 2019-09-20
WO2017166539A1 (en) 2017-10-05

Similar Documents

Publication Publication Date Title
CN105898679A (en) Method for collecting heterogeneous sensing data based on representative nodes and multilayer fusion
CN1794732B (en) Implementing method of IPv6 micro sensing routing appliance protocol stack system structure
Arjunan et al. F5N‐based unequal clustering protocol (F5NUCP) for wireless sensor networks
CN102149160B (en) Energy perception routing algorithm used for wireless sensing network
CN101808382B (en) Data communication method of wireless sensor network
CN100358311C (en) Method of implementing isomeric sensor network system supporting wake mechanism
CN110620611B (en) Cooperative spectrum sensing method based on GEO and LEO double-layer satellite network
CN101808289B (en) Method for acquiring data of wireless sensor network based on mobile sink node
CN103906164B (en) Maritime search and rescue wireless sensor network topology and topology control method
CN109673034B (en) Wireless sensor network clustering routing method based on longicorn stigma search
CN106817738B (en) Routing method based on alliance game
CN101083616A (en) Ant algorithm based wireless self-organized network energy-saving routing method on demand
CN103220747A (en) Route designing method of cognitive radio mesh network
Li et al. Energy-efficient multipath routing in wireless sensor network considering wireless interference
Tan et al. QSDN-WISE: A new QoS-based routing protocol for software-defined wireless sensor networks
CN104717238B (en) Distributed Services combined method based on ant group algorithm in mobile ad hoc network
CN106572513A (en) Wireless sensor routing algorithm based on fuzzy multi-attribute decision
CN103024857A (en) Clustering control method applied to wireless sensor networks
Liu et al. An improved energy efficient unequal clustering algorithm of wireless sensor network
CN111065142A (en) Relay selection method for energy harvesting wireless networks
CN105636144A (en) Method for prolonging life period of zigbee network
CN108566658A (en) A kind of cluster algorithm of Energy Consumption in Wireless Sensor Networks equilibrium
Naushad et al. Energy efficient clustering with reliable and load-balanced multipath routing for WSNs
CN102983948B (en) A kind of radio sensing network self-adaption cluster transmission method and device thereof
CN103929778A (en) Data grading transmission method

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20190920

CF01 Termination of patent right due to non-payment of annual fee