CN110049526A - Based on the method for data capture and system for improving cluster algorithm in WSN - Google Patents
Based on the method for data capture and system for improving cluster algorithm in WSN Download PDFInfo
- Publication number
- CN110049526A CN110049526A CN201910294339.XA CN201910294339A CN110049526A CN 110049526 A CN110049526 A CN 110049526A CN 201910294339 A CN201910294339 A CN 201910294339A CN 110049526 A CN110049526 A CN 110049526A
- Authority
- CN
- China
- Prior art keywords
- node
- cluster
- algorithm
- network
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/46—Cluster building
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/06—Selective distribution of broadcast services, e.g. multimedia broadcast multicast service [MBMS]; Services to user groups; One-way selective calling services
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/02—Communication route or path selection, e.g. power-based or shortest path routing
- H04W40/04—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Multimedia (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present disclosure proposes the methods of data capture and system based on improvement cluster algorithm in WSN, for the wireless sensor network for being disposed with several wireless sensors compositions in a region, by threshold value in the election of cluster head formula of LEACH method multiplied by coefficient G, the G value is used to control for the election of cluster head quantity in heterogeneous networks structure, realizes from ordinary node and elects leader cluster node；After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines the group member of leader cluster node, completes clustering process；The collection of data is realized using the wireless sensor network of sub-clustering.The optimization algorithm that the disclosure improves LEACH cluster algorithm is made in a number of situations in the way of weighting, and cluster algorithm can preferably be suitable for current network.
Description
Technical field
This disclosure relates to wireless sensor network technology field, more particularly to the number based on improvement cluster algorithm in WSN
According to collection method and system.
Background technique
Wireless sensor network (Wireless Sensor Networks, WSN) be it is numerous in the form of self-organizing existing for
The network with certain coverage area that sensor node is constituted.WSN is widely used in data monitoring field, such as in army
Thing, national defence, damage control and mobile payment, smart home etc..From the point of view of network characteristic, sensor has certain WSN
The transmitting-receiving of transmission range, data can only carry out in limited range.Node in WSN tends not to whole while and aggregation node
Data are exchanged, exchange data often carry out in a manner of transmitting step by step, therefore carry out data transmission being worth probing into using sub-clustering.
Cluster algorithm is to limit wireless sensor efficiency of transmission in WSN to solve the problems, such as the data transmission efficiency in WSN
Factor mainly have energy consumption and the aspect of computing capability two.It is common to do to keep the coverage rate of network and the connection degree of network
Method is exactly the data selected in some suitable nodes assistance aggregation node collection networks using cluster algorithm, then data are unified
Hand aggregation node processing.But WSN network is often limited to the factors such as geographical location, sensor computing capability, selection less than
Most suitable leader cluster node.In response to this problem, it is contemplated that cluster algorithm is optimized, is allowed to be more suitable for WSN.
Currently, both at home and abroad research be based primarily upon on traditional cluster algorithm be added to geographical location, dump energy etc. because
Element considers to achieve the effect that local optimum.Cluster algorithm in traditional WSN has centralization, distributed and hybrid three classes.
In centralized algorithm, there are LEACH-C algorithm and LEACH-F algorithm than more typical, the former passes through in collection network
The energy information of all the sensors calculates through certain algorithm, determines cluster head；The latter combines annealing algorithm, and static state selects portion
Partial node is as cluster head.The advantage of centralized network is that base station need to only have operational capability, remaining sensor node only needs to pass
Transmission of data is without carrying out data processing.
In distributed clustering algorithm, the most classical is by Wendi Rabiner Heinzelman, Anantha
Chandrakasan, Hari Balakrishnan propose LEACH cluster algorithm, main thought be by way of rotation come
The election of cluster head is carried out, the advantage of distributed network is, mitigates the burden in terms of the operational capability of base station, is suitble in large scale
Network.
Hybrid method combines centralized and distributed feature, such as the side LEACH-KED based on LEACH algorithm
Method, main thought are to assign different weights for energy, geographical location, from three factors of base distance between sites, obtain most suitable work
For the sensor node of cluster head.
Inventor has found that the research of cluster algorithm is mainly based upon to the progress of traditional cluster algorithm in WSN under study for action
Optimization in a certain respect, to adapt to the network under different scenes, so that Network morals and efficiency of transmission maximize.
Summary of the invention
The purpose of this specification embodiment is to provide based on the method for data capture for improving cluster algorithm in WSN, in portion
Traditional cluster algorithm can be significantly better than by dividing in performance.
This specification embodiment is provided in WSN based on the method for data capture for improving cluster algorithm, passes through following technology
Scheme is realized:
Include:
For the wireless sensor network for being disposed with several wireless sensors compositions in a region, including 1 aggregation node
And N number of ordinary node；
By threshold value in the election of cluster head formula of LEACH method multiplied by coefficient G, which is used to control for heterogeneous networks structure
In election of cluster head quantity, realize from ordinary node elect leader cluster node；
After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines cluster head section
The group member of point completes clustering process；
The collection of data is realized using the wireless sensor network of sub-clustering.
This specification embodiment is provided in WSN based on the data gathering system for improving cluster algorithm, passes through following technology
Scheme is realized:
The wireless sensor network constituted including being disposed with several wireless sensors in a region, including 1 aggregation node
And N number of ordinary node；
By threshold value in the election of cluster head formula of LEACH method multiplied by coefficient G, which is used to control the wireless sensor network
System is realized from ordinary node for the election of cluster head quantity in heterogeneous networks structure and elects leader cluster node；
After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines cluster head section
The group member of point completes clustering process；
The wireless sensor network assists the data in aggregation node collection network using the leader cluster node of sub-clustering, then will
Data uniform transmission gives aggregation node processing.
Compared with prior art, the beneficial effect of the disclosure is:
The optimization algorithm that the disclosure improves LEACH cluster algorithm is made in a number of situations in the way of weighting,
Cluster algorithm can preferably be suitable for current network.
Detailed description of the invention
The Figure of description for constituting a part of this disclosure is used to provide further understanding of the disclosure, and the disclosure is shown
Meaning property embodiment and its explanation do not constitute the improper restriction to the disclosure for explaining the disclosure.
Fig. 1 is the LEACH cluster algorithm and DEEC cluster algorithm dump energy comparison diagram of embodiment of the present disclosure；
Fig. 2 is the LEACH cluster algorithm and DEEC cluster algorithm efficiency of transmission comparison diagram of embodiment of the present disclosure；
Fig. 3 is the WSN network architecture schematic diagram of embodiment of the present disclosure；
Fig. 4 is the genetic algorithm flow chart of embodiment of the present disclosure；
Fig. 5 is the LEACH algorithm, DEEC algorithm and the dump energy comparison for improving LEACH algorithm of embodiment of the present disclosure
Schematic diagram；
Fig. 6 is the LEACH algorithm, DEEC algorithm and the data transmiting contrast for improving LEACH algorithm of embodiment of the present disclosure
Schematic diagram.
Specific embodiment
It is noted that following detailed description is all illustrative, it is intended to provide further instruction to the disclosure.Unless another
It indicates, all technical and scientific terms that the disclosure uses have logical with disclosure person of an ordinary skill in the technical field
The identical meanings understood.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the disclosure.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Examples of implementation one
The examples of implementation introduce LEACH cluster algorithm first, and LEACH cluster algorithm is point in wireless sensor network
Cloth Clustering protocol, full name " low energy consumption and adaptive sub-clustering layered protocol ".LEACH algorithm operational process is considered as two
A different stage: sub-clustering stage and data transfer phase are established.In terms of energy consumption, the stage of data transmission is held
The continuous time is significantly larger than the duration for establishing the stage of sub-clustering.Sub-clustering establishment stage can substantially be expressed as following three process:
Selected leader cluster node, leader cluster node broadcast the sensor node in its coverage area, determine in the group of leader cluster node at
Member.
In formula (1): p is the probability that current period inner sensor node is chosen for leader cluster node, and r is algorithm operation
Periodicity, G are the set of the sensor node of not elected cluster head in the nearest 1/p algorithm cycle of operation.
In the period of each information transmission, selected leader cluster node is the first step that sub-clustering is established, firstly, choosing in network
Any node, LEACH algorithm can generate a random number, then, calculate threshold value T according to new probability formula1(n), it will generate
Random number and T1(n) it compares, if the random number generated is less than T1(n), then current wireless sensor node is chosen as cluster head
Node；Conversely, not being chosen as leader cluster node.It repeats the above process, until all nodes all traverse one time, it is complete in network cycle
Portion's leader cluster node is completely selected.
It is the second step that sub-clustering is established that leader cluster node, which carries out broadcast to the sensor node in its coverage area, is passed when common
After sensor node is chosen as leader cluster node, the sensor node into its coverage area spreads itself disappearing as leader cluster node
Breath, makes other sensors node find oneself.After other sensors node receives the message from leader cluster node, other sensings
The device highest leader cluster node of node selection signal intensity as itself leader cluster node, oneself in the cluster of the leader cluster node at
Member, then cluster inner sensor node reports itself signal strength between leader cluster node to leader cluster node feedback message, completes sub-clustering
Process.
In another examples of implementation, DEEC cluster algorithm is disclosed, DEEC cluster algorithm is carried out based on LEACH cluster algorithm
It improves, the considerations of residue of network organization energy has been added in the strategy of cluster head selection by it.DEEC cluster algorithm and LEACH sub-clustering are calculated
The operational process of method is substantially similar, and difference is that DEEC cluster algorithm has the optimization to energy consumption to consider.DEEC algorithm wishes remaining
Energy more goes to undertake the work of leader cluster node more than the node of nodes dump energy average value, and remaining in network
Little energy it is on the contrary.Since leader cluster node consumes bigger energy than non-leader cluster node, in this way, the high node of energy is more
Serving as leader cluster node can be so that the energy consumption in network tends to be average, to reduce the feelings of single node energy pre-mature exhaustion
Condition.
The operation logic of algorithm is similar with LEACH cluster algorithm, is all constantly cyclically to carry out clustering process.Difference
It is, in DEEC algorithm clustering process other than needing to know the period being presently in, it is also necessary to know the current period in network
Lower node dump energy and node average residual energy.Each round during selected leader cluster node, in LEACH cluster algorithm
The threshold value T of circulation2It (n) need to be multiplied by a weight w=E in DEEC algorithmi/EaWherein, EiIt is the residue of present node in circulation
Energy, EaIt is the average residual energy of current period node.So i.e. by examining in terms of LEACH algorithm fusion Energy Consumption Factors
Consider.
Wherein, p is the probability that current period inner sensor node is chosen for leader cluster node, and r is the period of algorithm operation
Number, G are the set of the sensor node of not elected cluster head in the nearest 1/p algorithm cycle of operation.
After the completion of cluster head selection, identical as LEACH algorithm, DEEC algorithm then covers leader cluster node in previous cycle
Sensor node in range is broadcasted, and transmitting itself becomes the information of leader cluster node.Finally, in previous cycle, sensor
The highest leader cluster node of node selection signal intensity as the cluster cluster head and transmit the information such as the residue energy of node, by cluster head
It reports to aggregation node, information is uniformly processed in aggregation node, obtains the nodes average residual energy after epicycle
Amount.
In another embodiment, the analysis of LEACH method and DEEC method is found out by front, DEEC method allows for residue
The factor of energy, it is excellent to be carried out to it by the calculation for changing threshold value T1 (n) in election of cluster head formula in LEACH algorithm
Change.Fig. 1 is emulated to LEACH algorithm and DEEC algorithm dump energy, and abscissa represents the periodicity of operation, and maximum value is
5000, ordinate represents the dump energy total in each periodic network interior nodes.
Make discovery from observation DEEC algorithm within preceding 1000 period dump energy decline it is faster than LEACH algorithm, to net
Energy consumption in network is greater than LEACH algorithm, but the rate of energy dissipation of DEEC algorithm slows down after 1000 periods.1500
Cycle, two networks almost run out of dump energy simultaneously, illustrate under two kinds of algorithms, the performance difference of network life is not
Greatly.
But in daily life, due to the available external supplement of the node energy in WSN, so during the network operation
It is generally not present the case where sensor node energy is depleted, that is to say, that in actual life, network worked before 1000 periods
LEACH algorithm is better than DEEC cluster algorithm on saving energy consumption.Meanwhile the disclosure has only counted the energy in data transmission procedure
Consumption, the energy consumption of itself when consideration network is calculated, since DEEC algorithm is improved based on LEACH algorithm
It obtains afterwards, DEEC algorithm need to count the dump energy data of all the sensors in each round, need to increase a number in each round
According to the process of collection, energy consumption will certainly be higher than LEACH algorithm.
Fig. 2 is emulated to LEACH algorithm and DEEC algorithm data efficiency of transmission, and wherein abscissa represents the week of operation
Issue, maximum value 5000, ordinate represent the number variation that aggregation node receives data packet.
It makes discovery from observation, DEEC cluster algorithm has clear superiority in data transmission, its volume of transmitted data one
Directly it is higher than LEACH cluster algorithm close to twice.
Examples of implementation two
The examples of implementation are disclosed based on the method for data capture for improving cluster algorithm, and the disclosure is research wireless sensor
Network cluster dividing algorithm, the characteristics of fully taking into account multiplicity of network, enable the sensor node random distribution in WSN, construct mathematics
Model, for the use of cluster algorithm.
In specific embodiment, data collection model construction, referring to shown in attached drawing 3, firstly, by model construction specific
Geographical environment in, model is arranged in a region with a certain size and shape here, is denoted as A, whole network operation
Period is denoted as R.Then, wireless sensor is placed in a certain way in region a, the quantity of wireless sensor is denoted as N+1,
In include 1 aggregation node, be denoted as I；N number of ordinary node, is denoted as i；It can be elected as the ordinary sensors note of leader cluster node
For c.
Next the primary power information of setting wireless sensor node i, is denoted as Ei.Each wireless sensor energy is set
The formula set is as follows:
Ei=E0*(1+rand*a) (4)
Wherein E0For average energy value；Rand is the random number between 0~1；A represents energy fluctuation range.It can will pass in this way
Sensor node energy is initialized in E0Fluctuation nearby, meets the requirement of sensor isomery in network model.Finally, due to converge section
The position of point can be designated in network deployment phase, this model is by the N number of common wireless sensor of remaining in addition to aggregation node
Node is randomly dispersed on whole map.Wherein, xi, yi respectively represent the cross of sensor, ordinate.One can be constructed above
A geographic area with a certain size and shape, wireless sensor quantity are N+1 (including 1 aggregation node and N number of
Ordinary node), wireless sensor position random distribution, the WSN that wireless sensor energy fluctuates in a certain range.
It is a kind of by process searches of selecting the superior and eliminating the inferior in simulation nature about genetic algorithm (Genetic Algorithm)
In biological evolution process and Darwinian evolutionism in the method for globally optimal solution, and simulation Darwin's genetic mechanisms
Natural selection computation model.For genetic algorithm, one initial population of creation is first had to, fitness function meter is then passed through
The fitness to current environment is calculated, to realize the evolutionary process that biology interior is selected the superior and eliminated the inferior in natural imitation circle.It passes through
Fitness function calculates again and again, then is screened again and again, satisfactory individual is left, gradually close to global optimum
Solution, genetic algorithm flow chart is referring to shown in attached drawing 4.
Improvement of the genetic algorithm to cluster algorithm is utilized in embodiment of the disclosure, since traditional LEACH sub-clustering is calculated
Method and DEEC cluster algorithm do not carry out adaptability configuration to heterogeneous networks, and election of cluster head formula from beginning to end will not be because of network
It is different and generate adaptive change, it is relatively inflexible.It is desirable to be directed to heterogeneous networks, by changing the threshold in election of cluster head formula
Value makes each round cluster head quantity in network and network structure generate the changes of some adaptability.
Original DEEC method is exactly to be generated by changing the threshold value in LEACH method in election of cluster head formula to the performance of network
It influences.Thus it gains enlightenment, network performance is promoted by the threshold value in the election of cluster head formula of modification LEACH method.The disclosure
Consider threshold value in the election of cluster head formula of LEACH method multiplied by the coefficient M between one 0 to 2, this M, value is for controlling for not
With the election of cluster head quantity in network structure.M=0.8 is such as enabled, i.e. cluster head in network can be reduced under the present conditions；Enable M=
1.2 are that cluster head in network increases under the present conditions.M value how is chosen in from 0 to 2 just can be to the shadow that network has generated
It rings, this just needs to be screened using optimization algorithm.So problem is converted into one the problem of seeking globally optimal solution, it is therefore an objective to
Choose the weight between one 0 to 2, so that network more adapts to different node resource and geographical location distribution, the performance of network
Reach best.
Here, each node in network can both be optimized by optimization algorithm, cluster algorithm can also be directed to
In some significant variables optimize, can both achieve the purpose that reduce energy consumption, the disclosure selects the latter, the reason is that
The quantity of wireless sensor node is very huge in WSN, if optimizing to nodes all in whole network, then genetic algorithm
The Space-time Complexity of total cluster algorithm all can be very high when optimizing, and existing design conditions do not allow.For the latter's
Optimization, by the inspiration of DEEC cluster algorithm, it is believed that the algorithm to threshold value T1 (n) in the cluster head of LEACH algorithm selection formula (1) into
The factors such as energy consumption, data transmission efficiency, cluster head selection are taken into consideration, have obtained good effect by row change.But LEACH points
The cluster head of cluster algorithm selects formula, and different networks might not be all suitble to, and therefore, carries out T2 (n) multiplied by certain
The coefficient of size is allowed to be more suitable for current network.This coefficient is sought by genetic algorithm, is converted one for problem and is sought entirely
The mathematical problem of office's optimal solution.
Examples of implementation three
The examples of implementation disclose algorithm simulating example, introduce map and sensor relative parameters setting first, need to be by mould
Type constructs in certain geographical environment, therefore is provided with the cartographic model of a 200*200.The whole network cycle of operation is denoted as R,
The maximum value R=5000 in period, data transmission will emulate in this 5000 periods.If the quantity of wireless sensor is N
=101, including 1 aggregation node I and 100 common wireless sensor node i.In view of converging wireless sensing in reality
Aggregation node in device network is often positioned in the position in the bosom of network, is in coordinate by the setting of this model aggregation node
The map center of (100,100), remaining 100 ordinary node are randomly dispersed in whole map.
Then, the primary power information of wireless sensor node is set, each wireless sensor i has different energy,
Energy setting method is as follows, according to formula (4): EA=E0* (1+rand*a), takes E0Value be 0.5；Rand be 0~1 between with
Machine number；A is 0.05, and thus initialization realizes that sensor node energy fluctuates near 0.5.Finally, by addition to aggregation node
100 wireless sensor node i are randomly dispersed on map.Sensor node is horizontal, ordinate is respectively xi=rand*200, yi
=rand*200, the random number between wherein rand takes 0~1.
Loss model parameter setting and model relevant evaluation index in experiment, need to carry out specific numerical value to node energy
It calculates, the dump energy summation E of all nodes in current periodACalculation formula is as follows:
EA=∑ (Ei-EL) (5)
The dump energy of present node i is changed to Et, and the primary power information of wireless sensor node i is Ei, ELFor current week
Phase transmits energy consumed by data, wherein ELCalculation formula it is as follows:
As d < d0When,
EL=L* (ETX+ERX)+L*Efs*d2 (6)
As d >=d0When,
EL=L* (ETX+ERX)+L*Efs*d4 (7)
Wherein distance of the d between two data transmission nodals, d0For distance threshold, calculation formula is as follows:
Parameter L, E in formula 6,7,8TX、ERX、Efs、EmpMeaning and the length L for being provided that data packet in MATLAB
=4*103；Emit unit message loss of energy ETX=50*10-9；Recruiting unit message loss of energy ERX=50*10-9；It is freely empty
Between ENERGY Efs=10*10-12；Attenuating space ENERGY Emp=0.0013*10-13；Due to EACount all nodes in current period
The sum of dump energy, for cluster algorithm, the E of all nodes in same period numberAValue is bigger, and algorithm performance is more excellent.This
The size of the open data packet number S received by aggregation node in statistics some cycles is imitated to measure the data transmission of network
Rate.The data packet that aggregation node receives in some cycles is more, and data transmission efficiency is higher.Load balancing degrees are also had chosen simultaneously
Come measure the distribution of cluster algorithm cluster head it is reasonable whether, its principal measure is difference in the cluster of different cluster heads between number of members,
If number of members gap is excessive in cluster, cluster algorithm is unreasonable.The specific formula for calculation of load balancing degrees is as follows:
Wherein, ncIt is the quantity of leader cluster node, XiIt is membership in the cluster of current cluster head node, u is all clusters in network
The average cluster head number of head node.For WSN, LBF is bigger, and network performance is more excellent.
Emulation experiment and analysis, this experiment by Matlab 2017b emulation platform complete, by emulation it is desirable that
It can be derived that the comparison diagram of network lifecycle and the comparison diagram of network data transmission rate.It is transported using the matrix in Matlab
It calculates, generates multiple sensor variables, every a line represents different sensors, the different performance data of each column representative sensor,
It include the geographical position coordinates of each sensor, the dump energy of each sensor, each sensor is in a network
It whether is leader cluster node.By Matlab by among these aggregation of data a to matrix, the classification and processing of data are facilitated.
Also with the mapping functions such as plot () in Matlab, dump energy comparison diagram and data transmiting contrast figure to network are completed
Drafting, intuitively show the performance difference of each cluster algorithm.
In an examples of implementation, discloses and realized based on the data collection strategy for improving cluster algorithm: due to the network of WSN
Form is various, threshold value T in the election of cluster head formula of LEACH algorithm1(n) not necessarily it is suitble to the sub-clustering of all scene lower networks, this
It is open to be inspired by DEEC method, consider the otherness of network itself, makes it suitable for new net by weighting a M value to T2 (n)
Network.Although DEEC cluster algorithm has certain improvement to LEACH algorithm, DEEC cluster algorithm is not yet to different network rings
Border is adapted to, therefore the disclosure is improved on DEEC cluster algorithm, to the threshold calculations formula (2) of DEEC cluster algorithm
Middle T2(n) it is weighted M optimization.
The disclosure focuses on how using genetic algorithm optimization LEACH algorithm.Firstly, the disclosure chooses aggregation node
The quantity for receiving entire packet calculates function as fitness.Set for 100 generations for genetic algebra, chromosome length is
10, every generation takes DEEC algorithm to carry out the operation in 10 periods, calculates aggregation node in a cycle by obtained data
The data packet number S inside received.S calculates the functional value of function as fitness.Fitness is selected in each generation calculates letter
The maximum individual of numerical value carries out heredity, generates follow-on population.It is finished until 100 generation genetic algebras are run, the M finally obtained
Value is the T for being most suitable for current network2(n) weighted value.The M value drawn is weighted to T2(n), it completes to cluster algorithm
Optimization, and the emulation in 5000 periods is carried out, obtain simulation result.
Since the selection that fitness calculates function is carried out for data transmission efficiency, the finger of effect of optimization is judged
Mark also should be data transmission efficiency.Meanwhile the disclosure has also inquired into the performance for improving sub-clustering from other levels, including, residual energy
The analysis of variation is measured come the case where reflecting energy consumption in algorithm implementation procedure；The load balancing degrees of algorithm come reflect algorithm execute
Cluster head distribution condition in the process.Specific procedure based on LEACH algorithm is as follows:
Examples of implementation four
In order to enable those skilled in the art can clearly understand the technical solution of the disclosure, below with reference to tool
The technical solution of the disclosure is described in detail in the embodiment and comparative example of body.
The comparison of dump energy variation of three kinds of cluster algorithms during simulation run, specific simulation result such as Fig. 5 institute
Show: abscissa indicates that cycle of operation number, ordinate indicate the sum of all residue energy of node in network in figure.Known to analysis
The energy consumption of LEACH cluster algorithm be it is most slow in three, the slope of curve is minimum and is kept approximately constant, and energy is the 2000th
Cycle is totally consumed.And DEEC cluster algorithm, energy consuming process are calculated in the ratio LEACH sub-clustering of when in 0 to 1000 period
Method is fast, and curve ratio LEACH cluster algorithm is precipitous, and slope is kept approximately constant.1000 periods opened to 2300 cyclic curve slopes
Beginning slows down, and the time point that energy is consumed completely is in 2300 cycles.Improve LEACH cluster algorithm, energy consuming process
Almost it is consistent with DEEC cluster algorithm acquired results.
The election of cluster head strategy of LEACH algorithm is carried out completely in accordance with probability, therefore node quilt during entire algorithm simulation
The probability for being selected as cluster head is hardly influenced by other factors, and energy consumption is caused to carry out at random, and algorithm, which is run, may incite somebody to action early period
It is leader cluster node that a certain node, which is repeated as many times and elects, leads to the too fast death in advance of the node energy consumption, Energy distribution in network
It is unbalanced.The node that the algorithm operation later period is chosen as cluster head may be chosen in the node that around distribution energy is depleted, and be caused
The node can not play the effect of leader cluster node completely, and energy consumption is naturally relatively low.
DEEC cluster algorithm and improvement LEACH cluster algorithm are then based on the considerations of residue energy of node, in a way
It weakens the node in network and is repeated a possibility that election is leader cluster node.So that the election of cluster head process in network is excellent as far as possible
First consider the cluster head more than dump energy, is carried out so that the data transmission of network is steady, energy consumption is more reasonable.Three kinds of sub-clusterings are calculated
The comparison of volume of transmitted data variation of method during simulation run, specific simulation result are as shown in Figure 6: abscissa indicates in figure
Cycle of operation number, ordinate indicate that aggregation node receives the total number of data packet.To LEACH algorithm, passed within 5000 periods
Defeated data volume is minimum, about 1*104A, data transmission stops in 2000 cycles；To DEEC cluster algorithm, 5000
The data volume transmitted in period is about 2.2*104It is a, it is higher than LEACH cluster algorithm, improves LEACH sub-clustering lower than the disclosure and calculate
Method, data transmission stop in 2300 cycles；To improvement LEACH cluster algorithm, the data volume transmitted within 5000 periods
About 2.4*104It is a, it is highest in three, data transmission stops in 2300 cycles, and above-mentioned three kinds of algorithm datas transmission stops
The time point that stop is totally consumed with energy in Fig. 5 is consistent
LEACH cluster algorithm does not consider factor in terms of energy, so that network cluster head election is carried out according to probability completely, it cannot
By the energy reasonable distribution of network, data transmission efficiency is low.DEEC cluster algorithm can be preferably according to the dump energy of network
Distribution, the dump energy of reasonable arrangement network improve data transmission efficiency during the network operation.The improvement that the disclosure is mentioned
LEACH cluster algorithm, not only merged DEEC cluster algorithm in terms of dump energy the considerations of, and by it is some it is potential because
Element is considered in addition, and algorithm is made to achieve the effect that more to adapt to current network.Load balancing degrees the LBF comparison such as table 1 of three kinds of algorithms
It is shown:
1 LEACH algorithm of table, DEEC algorithm and the load balancing degrees comparison for improving LEACH algorithm
The above numerical value is the average value of the load balancing degrees in 5000 periods, is that the load balancing degrees summation of each round removes again
It is drawn with always taking turns number 5000.As can be seen that the load balancing degrees highest of LEACH cluster algorithm, effect are best.DEEC sub-clustering
The load balancing degrees of algorithm can not show a candle to LEACH cluster algorithm, but be still higher than the improvement LEACH cluster algorithm that the disclosure is mentioned.Point
Its reason is analysed, LEACH cluster algorithm election of cluster head is a random process, and each sensor node is divided into the general of leader cluster node
Rate is suitable, and the quantity of member is suitable in leader cluster node cluster in each period, therefore its load balancing degrees is high.And DEEC cluster algorithm,
Preferentially select the sensor node more than dump energy as cluster head in the process of running.Cluster head selection has certain directionality,
Randomness is reduced, when causing the sensor node at edge to serve as cluster head around node may be dead, member in cluster
It reduces, membership in cluster when serving as cluster head with the sensor node of non-edge is caused to have larger difference.
Increasingly developed with WSN, the sensor node in WSN is increasing, and the isomerism of sensor is more frequent,
Certainly will there is no it is a kind of adapt to overall network cluster algorithm.In this case, it is intended to which finding one kind can be to each
The method that kind cluster algorithm optimizes, makes in a number of situations, cluster algorithm can preferably be suitable for current network.This public affairs
Open by the innovatory algorithm based on LEACH cluster algorithm --- the inspiration of DEEC cluster algorithm, formed it is a kind of completely new for
The improved though of LEACH cluster algorithm selects to use genetic algorithm as the optimization algorithm improved to LEACH cluster algorithm,
It designs and the Realization of Simulation WSN network model, has carried out rigorous setting and mathematical derivation to Model Parameter.In this model
On simulating, verifying has been carried out to the network performance of three kinds of cluster algorithms and network analysis has been carried out to simulation result.Demonstrate improvement
LEACH sub-clustering can significantly be better than traditional cluster algorithm really in partial properties, i.e. the disclosure proposes to improve LEACH
Cluster algorithm is effective.
Examples of implementation five
This specification embodiment is provided in WSN based on the data gathering system for improving cluster algorithm, passes through following technology
Scheme is realized:
The wireless sensor network constituted including being disposed with several wireless sensors in a region, including 1 aggregation node
And N number of ordinary node；
By threshold value in the election of cluster head formula of LEACH method multiplied by coefficient M, which is used to control the wireless sensor network
System is realized from ordinary node for the election of cluster head quantity in heterogeneous networks structure and elects leader cluster node；
After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines cluster head section
The group member of point completes clustering process；
The wireless sensor network assists the data in aggregation node collection network using the leader cluster node of sub-clustering, then will
Data uniform transmission gives aggregation node processing.
In specific embodiment, aggregation node sends data to host computer, remote server, control centre or control
Terminal etc. can receive the electric terminal of the data.
It is understood that in the description of this specification, reference term " embodiment ", " another embodiment ", " other
The description of embodiment " or " first embodiment~N embodiment " etc. means specific spy described in conjunction with this embodiment or example
Sign, structure, material or feature are included at least one embodiment or example of the invention.In the present specification, to above-mentioned
The schematic representation of term may not refer to the same embodiment or example.Moreover, the specific features of description, structure, material
Person's feature can be combined in any suitable manner in any one or more of the embodiments or examples.
The foregoing is merely preferred embodiment of the present disclosure, are not limited to the disclosure, for the skill of this field
For art personnel, the disclosure can have various modifications and variations.It is all within the spirit and principle of the disclosure, it is made any to repair
Change, equivalent replacement, improvement etc., should be included within the protection scope of the disclosure.
Claims (10)
- Based on the method for data capture for improving cluster algorithm in 1.WSN, characterized in that include:For the wireless sensor network for being disposed with several wireless sensors compositions in a region, by the election of cluster head of LEACH method Multiplied by coefficient G, which is used to control for the election of cluster head quantity in heterogeneous networks structure threshold value in formula, realizes from common Node elects leader cluster node；After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines leader cluster node Group member completes clustering process；The collection of data is realized using the wireless sensor network of sub-clustering.
- 2. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in that wireless sensing Data model constructed by device network are as follows:In a region with a certain size and shape, it is denoted as A, the whole network cycle of operation is denoted as R；Then, wireless sensor is placed in a certain way in region a, the quantity of wireless sensor is denoted as N+1, including 1 aggregation node；N number of ordinary node；The ordinary sensors that can be elected as leader cluster node are denoted as c；The primary power information of wireless sensor node is set；The N number of common wireless sensor node of remaining in addition to aggregation node is randomly dispersed on whole map.
- 3. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in that utilize heredity Algorithm picks G value, so that wireless sensor network adapts to different node resource and geographical location distribution.
- 4. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in that wireless sensing In device network, the dump energy summation of all nodes is all node current periods in current period dump energy and currently week Phase transmits the summation of the difference of energy consumed by data.
- 5. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in that wireless sensing In device network, current period transmit energy consumed by data between two data transmission nodals at a distance from it is related.
- 6. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in that using centainly The size for the data packet number S that aggregation node receives in period measures the data transmission efficiency of network；Choose load balancing degrees come measure cluster algorithm cluster head distribution it is reasonable whether, its principal measure be different cluster heads cluster Difference between interior number of members.
- 7. based on the method for data capture for improving cluster algorithm in WSN as claimed in claim 3, characterized in that utilize heredity The process of algorithm picks G value are as follows:It chooses aggregation node and receives the quantity of entire packet as fitness calculating function:Genetic algebra, chromosome length are subjected to parameter setting respectively, every generation takes DEEC algorithm to carry out the fortune in several periods It calculates, calculates data packet number S, S that aggregation node receives in one cycle as fitness meter by obtained data Calculate the functional value of function；Fitness is selected in each generation and calculates the maximum individual progress heredity of functional value, generates follow-on population, Zhi Daoshe Fixed to finish for genetic algebra operation, the G value finally obtained is the T for being most suitable for current network2(n) weighted value will be drawn G value weight to T2(n), the optimization to cluster algorithm is completed.
- 8. based on the method for data capture for improving cluster algorithm in WSN as described in claim 1, characterized in thatDuring selected leader cluster node, the threshold value T of each round circulation in LEACH cluster algorithm2(n) it is needed in DEEC algorithm Multiplied by a weight w=Ei/Ea, wherein EiIt is the dump energy of present node in circulation, EaIt is being averaged for current period node Dump energy, p are the probability that current period inner sensor node is chosen for leader cluster node, and r is the periodicity of algorithm operation, G For the set of the sensor node of cluster head not elected in the nearest 1/p algorithm cycle of operation.
- Based on the data gathering system for improving cluster algorithm in 9.WSN, characterized in that several wireless including being disposed in a region The wireless sensor network that sensor is constituted, including 1 aggregation node and N number of ordinary node；By threshold value in the election of cluster head formula of LEACH method multiplied by coefficient G, which is used to control pair the wireless sensor network Election of cluster head quantity in heterogeneous networks structure is realized from ordinary node and elects leader cluster node；After selected leader cluster node, leader cluster node broadcasts the sensor node in its coverage area, determines leader cluster node Group member completes clustering process；The wireless sensor network assists the data in aggregation node collection network using the leader cluster node of sub-clustering, then by data Uniform transmission gives aggregation node processing.
- 10. based on the data gathering system for improving cluster algorithm in WSN as claimed in claim 9, characterized in that the convergence Node sends data to host computer, remote server, control centre or controlling terminal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910294339.XA CN110049526A (en) | 2019-04-12 | 2019-04-12 | Based on the method for data capture and system for improving cluster algorithm in WSN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910294339.XA CN110049526A (en) | 2019-04-12 | 2019-04-12 | Based on the method for data capture and system for improving cluster algorithm in WSN |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110049526A true CN110049526A (en) | 2019-07-23 |
Family
ID=67277063
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910294339.XA Pending CN110049526A (en) | 2019-04-12 | 2019-04-12 | Based on the method for data capture and system for improving cluster algorithm in WSN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110049526A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110536373A (en) * | 2019-08-13 | 2019-12-03 | 江苏大学 | A kind of optical-wireless sensor network clustering method based on copula theory |
| CN111093201A (en) * | 2019-12-23 | 2020-05-01 | 内蒙古大学 | Wireless sensor network and clustering method thereof |
| CN111586789A (en) * | 2020-05-19 | 2020-08-25 | 北京理工大学 | Data transmission method and device, computer equipment and storage medium |
| WO2021051859A1 (en) * | 2019-09-18 | 2021-03-25 | 上海海事大学 | Adaptive genetic algorithm-based clustering and routing method for wireless sensor network |
| CN113099508A (en) * | 2021-03-16 | 2021-07-09 | 西南民族大学 | Random centralized self-organizing clustering method and system for wireless mobile node |
| CN118075814A (en) * | 2024-04-19 | 2024-05-24 | 上海创蓝云智信息科技股份有限公司 | Network message transmission method based on master node control |
| CN118075814B (en) * | 2024-04-19 | 2024-07-09 | 上海创蓝云智信息科技股份有限公司 | Network message transmission method based on master node control |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102547904A (en) * | 2012-02-28 | 2012-07-04 | 山东大学 | Leach protocol-based cluster head election improved algorithm |
| CN103024849A (en) * | 2012-09-27 | 2013-04-03 | 西安电子科技大学 | LEACH (Low-energy Adaptive Clustering Hierarchy)-based wireless sensor network clustering method |
| CN103338492A (en) * | 2013-05-20 | 2013-10-02 | 山东大学 | Heterogeneous wireless sensor network clustering method based on DEEC (distributed energy-efficient clustering) method |
| CN103607746A (en) * | 2013-12-02 | 2014-02-26 | 中国联合网络通信集团有限公司 | Method for realizing routing through wireless sensing nodes |
| 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 |
| EP2991393A1 (en) * | 2004-12-22 | 2016-03-02 | Wirepas Oy | Node device for a wireless sensor network |
| CN106102075A (en) * | 2016-08-25 | 2016-11-09 | 广东工业大学 | The cluster-dividing method divided based on hierarchical region in radio sensing network and system |
| CN109413710A (en) * | 2018-11-26 | 2019-03-01 | 珠海格力电器股份有限公司 | The cluster-dividing method and device of wireless sensor network based on genetic algorithm optimization |
-
2019
- 2019-04-12 CN CN201910294339.XA patent/CN110049526A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2991393A1 (en) * | 2004-12-22 | 2016-03-02 | Wirepas Oy | Node device for a wireless sensor network |
| CN102547904A (en) * | 2012-02-28 | 2012-07-04 | 山东大学 | Leach protocol-based cluster head election improved algorithm |
| CN103024849A (en) * | 2012-09-27 | 2013-04-03 | 西安电子科技大学 | LEACH (Low-energy Adaptive Clustering Hierarchy)-based wireless sensor network clustering method |
| CN103338492A (en) * | 2013-05-20 | 2013-10-02 | 山东大学 | Heterogeneous wireless sensor network clustering method based on DEEC (distributed energy-efficient clustering) method |
| CN103607746A (en) * | 2013-12-02 | 2014-02-26 | 中国联合网络通信集团有限公司 | Method for realizing routing through wireless sensing nodes |
| 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 |
| CN106102075A (en) * | 2016-08-25 | 2016-11-09 | 广东工业大学 | The cluster-dividing method divided based on hierarchical region in radio sensing network and system |
| CN109413710A (en) * | 2018-11-26 | 2019-03-01 | 珠海格力电器股份有限公司 | The cluster-dividing method and device of wireless sensor network based on genetic algorithm optimization |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110536373A (en) * | 2019-08-13 | 2019-12-03 | 江苏大学 | A kind of optical-wireless sensor network clustering method based on copula theory |
| CN110536373B (en) * | 2019-08-13 | 2022-10-28 | 江苏大学 | Optical wireless sensor network clustering method based on copula theory |
| WO2021051859A1 (en) * | 2019-09-18 | 2021-03-25 | 上海海事大学 | Adaptive genetic algorithm-based clustering and routing method for wireless sensor network |
| CN111093201A (en) * | 2019-12-23 | 2020-05-01 | 内蒙古大学 | Wireless sensor network and clustering method thereof |
| CN111586789A (en) * | 2020-05-19 | 2020-08-25 | 北京理工大学 | Data transmission method and device, computer equipment and storage medium |
| CN113099508A (en) * | 2021-03-16 | 2021-07-09 | 西南民族大学 | Random centralized self-organizing clustering method and system for wireless mobile node |
| CN118075814A (en) * | 2024-04-19 | 2024-05-24 | 上海创蓝云智信息科技股份有限公司 | Network message transmission method based on master node control |
| CN118075814B (en) * | 2024-04-19 | 2024-07-09 | 上海创蓝云智信息科技股份有限公司 | Network message transmission method based on master node control |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110049526A (en) | Based on the method for data capture and system for improving cluster algorithm in WSN | |
| Bagci et al. | An energy aware fuzzy unequal clustering algorithm for wireless sensor networks | |
| Bagci et al. | An energy aware fuzzy approach to unequal clustering in wireless sensor networks | |
| Liu et al. | Latency-aware path planning for disconnected sensor networks with mobile sinks | |
| CN104284386B (en) | Wireless sensor network cluster head selection method based on betweenness | |
| Arikumar et al. | EELTM: An energy efficient LifeTime maximization approach for WSN by PSO and fuzzy-based unequal clustering | |
| CN108900996A (en) | A kind of wireless sensor network data transmission method based on the double-deck fuzzy algorithmic approach | |
| Bennani et al. | Particle swarm optimization based clustering in wireless sensor networks: the effectiveness of distance altering | |
| Mohamed et al. | Coyote optimization based on a fuzzy logic algorithm for energy-efficiency in wireless sensor networks | |
| Jabeur | A firefly-inspired micro and macro clustering approach for wireless sensor networks | |
| CN110225569A (en) | A method of based on the WSNs clustering and multi-hop Routing Protocol for improving particle swarm algorithm | |
| CN109152103A (en) | A kind of pair of wireless sensor network carries out network-building method and device | |
| Hajjej et al. | A new optimal deployment model of internet of things based on wireless sensor networks | |
| Nabavi | An optimal routing protocol using multi-objective whale optimization algorithm for wireless sensor networks | |
| Wang et al. | Reinforcement learning-enabled resampling particle swarm optimization for sensor relocation in reconfigurable WSNs | |
| El Khediri et al. | Clustering algorithm in wireless sensor networks based on shortest path | |
| Wang et al. | Joint shortest chain and fair transmission design for energy-balanced PEGASIS in WSNs | |
| JP7362088B1 (en) | Distributed, entropy-efficient, energy-saving cluster-based routing method for SDWSN | |
| Revanesh et al. | Retracted article: Deep learning-based algorithm for optimum cluster head selection in sustainable wireless communication system | |
| Kaviarasan et al. | A Novel Spider Monkey Optimized Fuzzy C-Means Algorithm (SMOFCM) for Energy-Based Cluster-Head Selection in WSNs | |
| Subbaiah et al. | Mobile ad hoc network | |
| MRIU | A teaching learning based optimization algorithm for cluster head selection in wireless sensor networks | |
| Korteby et al. | Multi dimensional analysis of sensor communication processes | |
| CN108541072B (en) | Frequency spectrum decision multi-objective optimization method based on self-adaptive group search algorithm | |
| Liu et al. | Biological swarm intelligence based opportunistic resource allocation for wireless ad hoc networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190723 |
|
| RJ01 | Rejection of invention patent application after publication |