CN104486807B - A kind of method for routing of small-scale wireless sensor network - Google Patents
A kind of method for routing of small-scale wireless sensor network Download PDFInfo
- Publication number
- CN104486807B CN104486807B CN201410705802.2A CN201410705802A CN104486807B CN 104486807 B CN104486807 B CN 104486807B CN 201410705802 A CN201410705802 A CN 201410705802A CN 104486807 B CN104486807 B CN 104486807B
- Authority
- CN
- China
- Prior art keywords
- node
- sensor node
- path
- sensor
- sub
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- 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
- H04W40/08—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources based on transmission power
-
- 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
- H04L45/12—Shortest path evaluation
- H04L45/124—Shortest path evaluation using a combination of metrics
-
- 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
- H04W40/10—Communication route or path selection, e.g. power-based or shortest path routing based on wireless node resources based on available power or energy
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W40/00—Communication routing or communication path finding
- H04W40/24—Connectivity information management, e.g. connectivity discovery or connectivity update
- H04W40/248—Connectivity information update
-
- 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
A kind of method for routing of small-scale wireless sensor network, the invention belongs to wireless sensor network route field, solves traditional harmony searching algorithm and cannot be directly used to solve the problem of wireless sensor network is route.The present invention includes transmission global information step, sends packet step, transmission packet step and updates dump energy information Step.The present invention is improved to the coded system and candidate's harmony generation method of existing harmonic search algorithm, has both considered that path energy consumption is also contemplated for path length, the life cycle of the energy consumption of energy active balance whole network, significantly extension whole network.
Description
Technical field
The invention belongs to wireless sensor network route field, and in particular to a kind of road of small-scale wireless sensor network
By method.
Background technology
The technologies such as computer system, sensor technology, wireless communication technology and distributed information processing are developed rapidly
The connection mankind and the wireless sensor network technology of reality physical world has been promoted to develop rapidly.Wireless sensor network can be with
By the physical message of objective world together with existing network connection, thus it can to realize that other networks can not be realized a variety of
Function, is one of key technology in next generation network, is also one of 21 century most important emerging technology, extensive utilization
In military field and economy and sphere of life, to realize the interconnection of physical world and human society.
Wireless sensor network modern economy and life in various fields application with special dimensions such as military affairs
Using being very different；In economic and life, intelligent medical, environmental monitoring, smart home, architectural environment and speedily carrying out rescue work are rescued
The fields such as calamity, the application for having small-scale wireless sensor network.
For ease of understanding the present invention, relevant term is explained below：
Path：When packet is sent to aggregation node from source node in the way of multi-hop transmission via each sensor node,
The sequence being made up of each sensor node for sending the packet；The length in path is also referred to as dimension, is that path includes source section
Sensor node number including point and aggregation node.
Distance：Euclidean distance between sensor node.
Hop count：Current sensor node transmits packets to the minimum sensor node number that aggregation node needs to pass through.
Dump energy：The electricity surplus value of sensor node.
Global information：Neighbor node set, the distance of each sensor node and neighbor node including each sensor node,
Hop count of the existing energy and each sensor node of each sensor node to aggregation node.
As shown in figure 1, small-scale wireless sensor network 1 is by aggregation node 2, tens to hundreds of structures of sensor node 3
Into each sensor node 3 (is received and transmitting circuit and transmission by processing unit, data acquisition unit, data transmission unit
Amplifier) and power supply unit composition；Environmental information where each sensor node monitoring, and related data packets are treated as, respectively
Between sensor node and neighbor node can direct communication, then, aggregation node is transferred data in multi-hop transmission mode
2, aggregation node 2 sends the data that the monitoring of sensor node 3 processing is obtained to client 5 by internet or satellite 4 and carried out
Processing；Wherein, neighbor node is that can receive the sensing of information under aggregation node or the specific transmit power of sensor node
Device node；Aggregation node has stronger computing capability, storage capacity and communication capacity, and is supplied with continual power supply
Should.In small-scale sensor network, sensor node quantity is than the sensor section in the massive wireless sensors such as military affairs
Point quantity is much smaller.Some independent small-scale wireless sensor networks can be constituted by way of being connected between cluster one it is larger
The network structure of type.
Sensor node in small-scale wireless sensor network has that limited energy, memory capacity be limited, computing capability
The features such as limited and limited communication capacity, with the development of computer hardware technology, the capacity of various storage devices becomes to get over
Come bigger, volume becomes less and less, and price is more and more cheaper；The speed of various processors also becomes increasingly faster, volume and become
Less and less, power consumption is more and more lower, price is also more and more cheaper so that the storage capacity of sensor node, computing capability become
Obtaining increasingly stronger, chip power-consumption becomes more and more lower, will be no longer as most heavy during small-scale application of higher wireless sensor network
The obstacle wanted.However, battery technology does not obtain the sensor section in breakthrough progress, small-scale wireless sensor network always
Point is mostly powered using battery, and energy is very limited, it is impossible to or be difficult charging, and sometimes sensor node once dispose it is past
Toward being difficult to reclaim, undoubtedly, energy consumption has turned into the bottleneck of most critical during small-scale application of higher wireless sensor network, therefore, right
The efficiency research of small-scale wireless sensor network is the focus and difficulties of wireless sensor network research.
Routing algorithm is played an important role in small-scale wireless sensor network, its energy consumption to each sensor node
And the life cycle of whole network plays critical effect；Moreover, small-scale wireless sensor network have it is very strong should
With correlation, routing algorithm in different application may difference it is very big, the general routing algorithm of neither one can only be directed to specific
Scenario Design and select specific routing algorithm.Therefore, the research of routing algorithm is received more and more attention.At present, road
By the routing algorithm of algorithm mainly including traditional routing algorithm and based on heuristic, traditional routing algorithm includes
Flooding, SPIN, MTE, Directed Diffusion, LEACH etc.；Routing algorithm based on heuristic includes
EEABR, SDG, EBAB, QELAR, QoS-PSO etc..Also, application heuristic design routing algorithm be following trend and
Study hotspot.
Harmony search (Harmony Search, HS) algorithm is one kind intelligence that South Korea scholar Geem in 2001 et al. is proposed
Optimized algorithm.Musicians rely on the memory of oneself in algorithm simulation musical performance, by adjusting each musical instrument in band repeatedly
Tone, is finally reached the process of a beautiful harmony state.Harmonic search algorithm has very strong ability of searching optimum, it is easy to receive
Globally optimal solution is held back, its flow chart is as shown in Fig. 2 comprise the steps：
Step 1, the initialization of algorithm relevant parameter：
Initialize the size H (i.e. row vector number), select probability C, tune of harmony data base (Harmony Memory, HM)
Whole probability A, adjustment bandwidth b, evaluation times N；
Harmony data base
Wherein, XiFor i-th harmony, i=1,2 ..., H, xi,jFor XiJ-th of tone, j=1,2 ..., n；
Step 2, each tone to every harmony in harmony data base are initialized：
In formula, R is the random number between 0 to 1,For j-th of tone xjLower bound,For j-th of tone xjThe upper bound；
Step 3, the fitness for calculating each harmony in harmony storehouse；
Degree variables t=1 is evaluated in step 4, setting；
Step 5, generation candidate harmony X'={ x1',x'2,…,x'j,…,x'n, including following processes：
(5.1) harmony length variable j=1 is set；
(5.2) j-th of tone is selected for candidate's harmony X'：
The random number R between one 0 to 1 is produced, judges whether R ＜ C, is that the jth row then from harmony data base HM are random
Select a tone as candidate harmony X' j-th of tone, process (5.3) is carried out, otherwise from the span of j-th of tone
One tone of interior random generation carries out process (5.5) as candidate harmony X' j-th of tone；
(5.3) random number R between one 0 to 1 is produced, judges whether R ＜ A, is then to carry out process (5.4), otherwise enters
Row process (5.5)；
(5.4) random number R between one 0 to 1 is produced, judges whether R ＜ 0.5, is, make x'j=x'j+ R × b, is carried out
Process (5.5)；Otherwise x' is madej=x'j- R × b, carries out process (5.5)；
(5.5) judge whether j is equal to the dimension n of harmony, be then to carry out process (5.7), otherwise carry out process (5.6)；
(5.6) j+1 value is assigned to j, turns over journey (5.2)；
(5.7) a candidate harmony X' is generated, candidate's harmony X' fitness value f (X') is calculated, judges whether f (X') is small
The fitness of the maximum harmony of fitness in harmony data base HM, is to replace in harmony data base HM candidate's harmony and fit
The maximum harmony of response, carries out step 6；Otherwise harmony storehouse is not made any changes, carries out step 6；
Step 6, check whether stopping iteration：
T+1 is assigned to t, judges whether t≤N, is, 5 are gone to step, step 7 is otherwise carried out；
Optimal harmony in step 7, record harmony storehouse.
Traditional harmony searching algorithm is for solving continuous optimization problems, it is necessary to first determine the number of design variable, i.e.,
The dimension of each harmony is identical in the dimension of each harmony, and harmony data base in harmony storehouse；And wireless sensor network routing issue
It is a discrete optimization problems of device, the sensor node number that path is included not can determine that that is, the number of design variable can not be true
Fixed, the length in each path is not necessarily identical, therefore, and traditional harmony searching algorithm cannot be directly used to wireless sensor network route
The solution of problem.
The content of the invention
The present invention provides a kind of method for routing of small-scale wireless sensor network, and solving traditional harmony searching algorithm can not
The problem of solution wireless sensor network is route is directly used in, coded system and the life of candidate's harmony to existing harmonic search algorithm
It is improved into method, has both considered that path energy consumption is also contemplated for path length, improves efficiency and extend whole small-scale wireless biography
Sensor network life.
A kind of method for routing of small-scale wireless sensor network provided by the present invention, including transmission global information step
Suddenly packet step, transmission packet step, are sent and dump energy information Step is updated, it is characterised in that：
(1) global information step is transmitted：
First, each sensor node including aggregation node is included to network to be numbered respectively, each of which is used as
Unique mark；By aggregation node be arranged to each sensor node identical transmit power, aggregation node is in network to neighbour
Node broadcasts route messages are occupied, including the hop count of each sensor node to aggregation node；Each neighbor node receives route and disappeared
After breath, after wherein itself adds 1 to the hop count of aggregation node, then the route messages updated are broadcasted to the neighbor node of itself；This
Sample, each sensor node can know itself to the hop count of aggregation node, the neighbor node of itself, the distance with neighbor node
And neighbor node is to the hop count of aggregation node；
Then, each sensor node saves the distance, the numbering of neighbor node, neighbor node of itself and neighbor node to convergence
The hop count and itself existing energy and hop count of point are sent to aggregation node, and aggregation node is by the neighbor node of each sensor node
Sorted from small to large according to numbering, form the neighbor node set of each sensor node；Again by aggregation node in network with wide
Broadcast mode transmits global information to each sensor node；
The neighbor node set of the global information including each sensor node, each sensor node and neighbor node away from
From the hop count of, the existing energy of each sensor node and each sensor node to aggregation node；
(2) packet step is sent：
When certain sensor node monitors environmental information, first it is processed to turn into packet, is then stored according to itself
Global information, calculates the optimal path for itself arriving aggregation node, and the optimal path is added into packet；Then calculate most
Other each sensor nodes forward the packet to need energy and the dump energy consumed in shortest path, and update itself storage
Global information in other each sensor nodes on optimal path dump energy；Self rest energy is added again described
After packet, packet is sent to next sensor node according to the optimal path；
(3) packet step is transmitted：
Next sensor node receives the packet, self rest energy is added into the packet, according to data
Optimal path in bag continues to deliver a packet to next sensor node, so continues, until packet is sent to convergence
Node；
(4) dump energy information Step is updated：
Aggregation node is received after the packet, and packet is handled, the ring that wherein sensor node is monitored
Environment information is by internet or satellite transmission to user；Utilize the dump energy information updating of each sensor node in packet
The global information of storage, aggregation node is with some cycles to all residue energy of node information of each sensor node broadcasts.
Described method for routing, it is characterised in that：
In the transmission packet step, calculating the optimal path includes following sub-steps：
Sub-step 1, the size H that path storehouse PM is set, minimum select probability Cmin, MAXIMUM SELECTION probability CmaxAnd evaluate secondary
Number N：
PM=[X1, X2..., Xr..., XH]T, wherein, XrFor r paths, represent that current sensor node is saved to convergence
Point path, r=1,2 ..., H, Xr={ s ..., xr,j..., d }, in formula, s is current sensor node, and d is aggregation node,
xr,jFor j-th of sensor node in r paths；
H=3~12,0≤Cmin≤Cmax≤ 1, N=200~800；
Sub-step 2, to being initialized in the PM of path storehouse per paths, including following processes：
(2.1) by path length variable j assignment 1, sensor node s is set to j-th of biography in current initialization path
Sensor node；
(2.2) j+1 value is assigned to j, is j-th of sensor node of current initialization Path selection：
Judge to whether there is non-selected mistake in the neighbor node set of current initialization path -1 sensor node of jth
Sensor node, be then carry out process (2.3), otherwise turn over journey (2.1)；
(2.3) random number R between 0 to 1 is produced first, and -1 biography of jth in current initialization path is then subtracted with R
Whether the select probability of the minimum sensor node of the numbering of non-selected mistake, judge its difference in the neighbor node set of sensor node
No more than 0, it is then to carry out process (2.4), otherwise turns over journey (2.5)；
(2.4) the minimum sensor node of the numbering of the non-selected mistake is sensed as j-th of current initialization path
Device node, turns over journey (2.6)；
(2.5) difference is continued to subtract to the neighbor node collection of -1 sensor node of jth in current initialization path
The select probability of the small sensor node of numbering time of non-selected mistake in closing, then judge that being then will be described the result is that no be not more than 0
Otherwise the small sensor node of numbering time of non-selected mistake so continues as j-th of sensor node in current initialization path
Difference to the last is not more than 0, with last in the neighbor node set of current initialization path -1 sensor node of jth
It is related to the sensor node of its select probability as j-th of sensor node in current initialization path, turns over journey (2.6)；
(2.6) whether j-th of sensor node for judging current initialization path is aggregation node, is then generated at the beginning of one
Beginning path；Otherwise journey (2.2) is turned over；
To, per paths repetitive process (2.1)~(2.6), symbiosis completes the first of PM into H bar initial paths in the PM of path storehouse
Beginningization；
Sub-step 3, the fitness f (X for calculating path Ku Zhongge pathsr)：
In formula, E (Xr) it is path XrIn the gross energy that is consumed of sensor node transmission packet,L is path length, di,i+1It is sensor node i and sensor
The ENERGY E consumed when being worked with transmitting circuit is received in the distance between node i+1, sensor nodeelecFor 50nJ/bit,
The ENERGY E consumed when transmitting amplifier operation in sensor nodeampFor 0.1nJ/bit/m2；K is transmitted by sensor node
Data package size, unit is bit；EMinFor path XrThe dump energy of the minimum sensor node of middle dump energy, EAvgFor
Path XrThe average residual energy of middle all the sensors node；
Degree variables t=1 is evaluated in sub-step 4, setting；
Sub-step 5, setting number of passes variable r=1；
Sub-step 6, generation path candidate X', including following processes：
(6.1) path length variable j is assigned to by 1, current sensor node s is set to j-th of candidate route X'
Sensor node；
(6.2) j+1 value is assigned to j, judges the neighbor node model of -1 sensor node of jth in candidate route X'
With the presence or absence of the sensor node not selected in enclosing, it is then to carry out process (6.3), otherwise turns over journey (6.1)；
(6.3) j-th of sensor node is selected for candidate route X'：
The random number R between 0 to 1 is produced, judges whether R ＜ C, is then to carry out process (6.4), otherwise carries out process
(6.7)；The select probability
(6.4) judge to whether there is the path with j-th of sensor node in the PM of path storehouse, be to carry out process
(6.5)；Otherwise process (6.7) is carried out；
(6.5) whether there is the jth -1 in path candidate in j-th of sensor node for judging each paths in the storehouse of path
In the neighbor node set of individual sensor node, and the sensor node of non-selected mistake, it is then to carry out process (6.6), otherwise enters
Row process (6.7)；
(6.6) it is random out of j-th of sensor node composition in path of each bar with j-th of sensor node set
Selection one is in the neighbor node set of -1 sensor node of jth of path candidate, and the sensor node of non-selected mistake
As j-th of sensor node of path candidate, process (6.8) is carried out；
(6.7) random selection one is non-selected in the neighbor node set of -1 sensor node of jth from path candidate
The sensor node crossed carries out process (6.8) as j-th of sensor node of path candidate；
(6.8) judge whether j-th of sensor node in path candidate is aggregation node, be then to carry out process (6.9),
Otherwise journey (6.2) is turned over；
(6.9) a candidate route X' is generated, according to the calculating formula of sub-step 3, the fitness f of candidate route is calculated
(X'), judge whether f (X') is less than the fitness in the maximum path of fitness in the storehouse of path, replaced with candidate route X'
The maximum path of fitness in the storehouse of path, carries out sub-step 7；Otherwise, path storehouse is not made any changes, carries out sub-step 7；
Sub-step 7, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 8, otherwise perform sub-step 11；
Sub-step 8, from the r paths Xr in path storehouse in addition to source node and aggregation node optional sensor node xj
Carry out neighborhood search, including following processes：
(8.1) path X is selectedrIn any sensor node x in addition to source node and aggregation nodej, judge sensor node
xjUpper hop sensor node xj-1Neighbor node set and next-hop sensor node xj+1Neighbor node intersection of sets
Whether have not in path X in collectionrIn sensor node, be then carry out process (8.2)；Otherwise to path XrDo not make any changes；
(8.2) sensor node not in the Xr of path is randomly choosed in the common factor and replaces sensor node
xj, obtain new route X'r, according to the calculating formula of sub-step 3, calculate new route X'rFitness f (X'r), judge whether f (X'r)
＜ f (Xr), it is then with X'rReplace Xr, carry out sub-step 9；Otherwise, to path XrDo not make any changes, carry out sub-step 9；
Sub-step 9, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 10, otherwise perform sub-step
11；
Sub-step 10, the value imparting r by r+1, judge whether r≤H, are then to perform sub-step 6, otherwise perform sub-step 5；
The minimum paths of fitness are used as optimal path in sub-step 11, selection path storehouse.
In the process (2.3) of the sub-step 2 or (2.5), -1 sensor node of jth in current initialization path
The numbering of non-selected mistake is minimum in neighbor node set or select probability P of the small sensor node i of numbering time (j-1, i)：
In formula, aj-1Represent the neighbor node set of -1 sensor node of jth, N (aj-1) it is set aj-1First prime number
Amount, hiRepresent aj-1Middle sensor node i hop count, hmRepresent aj-1Middle sensor node m hop count；
By probability P, (j-1 i), can make each sensor node in initial path in selection next-hop sensor node
When, with larger probability selection close to aggregation node sensor node.
Harmonic search algorithm is applied to solve small-scale wireless sensor network routing issue by the present invention, and hair is calculated every time
When sending the optimal path of packet, optimal or sub-optimal path can be found；Also, both considered in the calculating formula of fitness
The energy consumption of the minimum node of dump energy in the average energy consumption in whole path, path, it is also considered that the length in path, so as to calculate
The optimal path of gained can effectively extend the life cycle of whole network.
The simpler refining of the present invention, without excessive parameter, eliminates the adjustment probability A in traditional harmony searching algorithm
With adjustment bandwidth b, only this special parameter of select probability C eliminates adjustment process.Harmony after the present invention is improved is searched for
In algorithm, select probability C employs adaptive strategy, and the parameter increases with the increase of iterations, so that search is calculated
Method has stronger ability of searching optimum initial stage in iteration, and has stronger local search ability in the later stage, increases and converges to entirely
The probability of office's optimal solution.Compared with existing routing algorithm, the routing algorithm in the present invention has high efficiency, can effectively put down
The life cycle of the energy consumption for the whole network that weighs, significantly extension whole network.
Brief description of the drawings
Fig. 1 is small-scale wireless sensor network system structural representation；
Fig. 2 is traditional harmony searching algorithm schematic flow sheet；
Fig. 3 is schematic flow sheet of the embodiment of the present invention；
Fig. 4 is the schematic flow sheet of calculating optimal path in the embodiment of the present invention；
Fig. 5 is calculates during optimal path, the schematic flow sheet initialized per paths；
Fig. 6 is the diagram of path initialization in the embodiment of the present invention；
Fig. 7 is the path library initialization result in the embodiment of the present invention；
Fig. 8 is the schematic flow sheet of generation path candidate in the embodiment of the present invention；
Fig. 9 is the diagram of generation path candidate in the embodiment of the present invention；
The changing rule schematic diagram that Figure 10 is select probability C in the embodiment of the present invention；
Figure 11 is the schematic flow sheet of optional sensor node progress neighborhood search in the embodiment of the present invention；
Figure 12 is the diagram of optional sensor node progress neighborhood search in the embodiment of the present invention；
Figure 13 (a) is to generate at random in the embodiment of the present invention comprising 10 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (b) is to generate at random in the embodiment of the present invention comprising 20 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (c) is to generate at random in the embodiment of the present invention comprising 30 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (d) is to generate at random in the embodiment of the present invention comprising 40 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (e) is to generate at random in the embodiment of the present invention comprising 50 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (f) is to generate at random in the embodiment of the present invention comprising 60 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (g) is to generate at random in the embodiment of the present invention comprising 70 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (h) is to generate at random in the embodiment of the present invention comprising 80 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (i) is to generate at random in the embodiment of the present invention comprising 90 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 13 (j) is to generate at random in the embodiment of the present invention comprising 100 sensor nodes (containing 1 aggregation node)
Small-scale wireless sensor network；
Figure 14 (a) is when each node primary power is identical in the embodiment of the present invention, of the invention and existing algorithm EEABR's is flat
Equal dump energy contrast and experiment；
Figure 14 (b) is when each node primary power is identical in the embodiment of the present invention, of the invention and existing algorithm EEABR's is surplus
Complementary energy standard deviation contrast and experiment；
Figure 14 (c) is when each node primary power is identical in the embodiment of the present invention, of the invention and existing algorithm EEABR is most
Small dump energy contrast and experiment；
When Figure 14 (d) is that each node primary power is identical in the embodiment of the present invention, of the invention and existing algorithm EEABR life
Life cycle contrast and experiment；
Figure 15 (a) is when each node primary power is different in the embodiment of the present invention, of the invention and existing algorithm EEABR's is flat
Equal dump energy contrast and experiment；
Figure 15 (b) is when each node primary power is different in the embodiment of the present invention, of the invention and existing algorithm EEABR's is surplus
Complementary energy standard deviation contrast and experiment；
Figure 15 (c) is when each node primary power is different in the embodiment of the present invention, of the invention and existing algorithm EEABR is most
Small dump energy contrast and experiment；
When Figure 15 (d) is that each node primary power is different in the embodiment of the present invention, of the invention and existing algorithm EEABR life
Life cycle contrast and experiment.
Embodiment
The present invention is described in more detail with reference to the accompanying drawings and examples.
As shown in figure 3, embodiments of the invention include transmission global information step, send packet step, transmission data
Bag step and renewal dump energy information Step：
(1) global information step is transmitted：
First, each sensor node including aggregation node is included to network to be numbered respectively, each of which is used as
Unique mark；By aggregation node be arranged to each sensor node identical transmit power, aggregation node is in network to neighbour
Node broadcasts route messages are occupied, including the hop count of each sensor node to aggregation node；Each neighbor node receives route and disappeared
After breath, after wherein itself adds 1 to the hop count of aggregation node, then the route messages updated are broadcasted to the neighbor node of itself；This
Sample, each sensor node can know itself to the hop count of aggregation node, the neighbor node of itself, the distance with neighbor node
And neighbor node is to the hop count of aggregation node；
Then, each sensor node saves the distance, the numbering of neighbor node, neighbor node of itself and neighbor node to convergence
The hop count and itself existing energy and hop count of point are sent to aggregation node, and aggregation node is by the neighbor node of each sensor node
Sorted from small to large according to numbering, form the neighbor node set of each sensor node；Again by aggregation node in network with wide
Broadcast mode transmits global information to each sensor node；
The neighbor node set of the global information including each sensor node, each sensor node and neighbor node away from
From the hop count of, the existing energy of each sensor node and each sensor node to aggregation node；
(2) packet step is sent：
When certain sensor node monitors environmental information, first it is processed to turn into packet, is then stored according to itself
Global information, calculates the optimal path for itself arriving aggregation node, and the optimal path is added into packet；Then calculate most
Other each sensor nodes forward the packet to need energy and the dump energy consumed in shortest path, and update itself storage
Global information in other each sensor nodes on optimal path dump energy；Self rest energy is added again described
After packet, packet is sent to next sensor node according to the optimal path；
(3) packet step is transmitted：
Next sensor node receives the packet, self rest energy is added into the packet, according to data
Optimal path in bag continues to deliver a packet to next sensor node, so continues, until packet is sent to convergence
Node；
(4) dump energy information Step is updated：
Aggregation node is received after the packet, and packet is handled, the ring that wherein sensor node is monitored
Environment information is by internet or satellite transmission to user；Utilize the dump energy information updating of each sensor node in packet
The global information of storage, aggregation node is with some cycles to all residue energy of node information of each sensor node broadcasts.
As shown in figure 4, in the transmission packet step of the present embodiment, calculating the optimal path includes following sub-steps：
Sub-step 1, the size H=5 that path storehouse PM is set, minimum select probability Cmin=0.2, MAXIMUM SELECTION probability Cmax=
0.9 and evaluate times N=500；
Sub-step 2, to being initialized in the PM of path storehouse per paths, symbiosis completes PM initialization into 5 initial paths；
Sub-step 3, the fitness f (X for calculating path Ku Zhongge pathsr)：
In formula, E (Xr) it is path XrIn the gross energy that is consumed of sensor node transmission packet,L is path length, di,i+1It is sensor node i and sensor
The ENERGY E consumed when being worked with transmitting circuit is received in the distance between node i+1, sensor nodeelecFor 50nJ/bit,
The ENERGY E consumed when transmitting amplifier operation in sensor nodeampFor 0.1nJ/bit/m2；K is transmitted by sensor node
Data package size, unit is bit；EMinFor path XrThe dump energy of the minimum sensor node of middle dump energy, EAvgFor
Path XrThe average residual energy of middle all the sensors node；
Degree variables t=1 is evaluated in sub-step 4, setting；
Sub-step 5, setting number of passes variable r=1；
Sub-step 6, generation path candidate X', judge whether its fitness f (X') is less than fitness maximum in the storehouse of path
The fitness in path, is to replace the maximum path of fitness in the storehouse of path with path candidate X', otherwise, path storehouse is not done
It is any to change, carry out sub-step 7；
Sub-step 7, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 8, otherwise perform sub-step 11；
Sub-step 8, the r paths X from path storehouserIn in addition to source node and aggregation node optional sensor node xj
Neighborhood search is carried out, new route X' is obtainedr, calculate new route X'rFitness f (X'r), judge whether f (X'r) ＜ f (Xr), it is
Then with X'rReplace Xr, carry out sub-step 9；Otherwise, to path XrDo not make any changes, carry out sub-step 9；
Sub-step 9, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 10, otherwise perform sub-step
11；
Sub-step 10, the value imparting r by r+1, judge whether r≤H, are then to perform sub-step 6, otherwise perform sub-step 5；
The minimum paths of fitness are used as optimal path in sub-step 11, selection path storehouse.
As shown in figure 5, the sub-step 2, to being initialized in the PM of path storehouse per paths, including following processes：
(2.1) by path length variable j assignment 1, sensor node s is set to j-th of biography in current initialization path
Sensor node；As shown in fig. 6, this small-scale wireless sensor network is in addition to current sensor node s and aggregation node d,
Also include 89 sensor nodes of sequence number 2~90；
(2.2) j+1 value is assigned to j, is j-th of sensor node of current initialization Path selection：
Judge to whether there is non-selected mistake in the neighbor node set of current initialization path -1 sensor node of jth
Sensor node, be then carry out process (2.3), otherwise turn over journey (2.1)；
(2.3) random number R between 0 to 1 is produced first, and -1 biography of jth in current initialization path is then subtracted with R
Whether the select probability of the minimum sensor node of the numbering of non-selected mistake, judge its difference in the neighbor node set of sensor node
No more than 0, it is then to carry out process (2.4), otherwise turns over journey (2.5)；
(2.4) the minimum sensor node of the numbering of the non-selected mistake is sensed as j-th of current initialization path
Device node, turns over journey (2.6)；
(2.5) difference is continued to subtract to the neighbor node collection of -1 sensor node of jth in current initialization path
The select probability of the small sensor node of numbering time of non-selected mistake in closing, then judge that being then will be described the result is that no be not more than 0
Otherwise the small sensor node of numbering time of non-selected mistake so continues as j-th of sensor node in current initialization path
Difference to the last is not more than 0, with last in the neighbor node set of current initialization path -1 sensor node of jth
It is related to the sensor node of its select probability as j-th of sensor node in current initialization path (as shown in fig. 6, here
Selection is 2nd sensor node of No. 27 sensor nodes as current initialization path), turn over journey (2.6)；
(2.6) whether j-th of sensor node for judging current initialization path is aggregation node, is then generated at the beginning of one
Beginning path；Otherwise journey (2.2) is turned over；
To, per paths repetitive process (2.1)~(2.6), symbiosis completes the first of PM into 5 initial paths in the PM of path storehouse
Beginningization, PM initialization result as shown in fig. 7, wherein the first paths by current sensor node s, the 27th, 23 ..., No. 86
Sensor node and aggregation node d are constituted；..., Article 5 path by current sensor node s, the 32nd, 47 ..., No. 52 sensing
Device node and aggregation node d are constituted.
As shown in figure 8, the sub-step 6, produces path candidate X', including following processes：
(6.1) path length variable j is assigned to by 1, current sensor node s is set to j-th of candidate route X'
Sensor node；Fig. 9 gives path storehouse PM as shown in Figure 7 five initialization paths；
(6.2) j+1 value is assigned to j, judges the neighbor node model of -1 sensor node of jth in candidate route X'
With the presence or absence of the sensor node not selected in enclosing, it is then to carry out process (6.3), otherwise turns over journey (6.1)；
(6.3) j-th of sensor node is selected for candidate route X'：
The random number R between 0 to 1 is produced, judges whether R ＜ C, is then to carry out process (6.4), otherwise carries out process
(6.7)；The select probability
As shown in Figure 10, select probability C is with the increase exponentially for evaluating degree variables t for select probability C changing rule
Rule increase；
(6.4) judge to whether there is the path with j-th of sensor node in the PM of path storehouse, be to carry out process
(6.5)；Otherwise process (6.7) is carried out；
(6.5) whether there is the jth -1 in path candidate in j-th of sensor node for judging each paths in the storehouse of path
In the neighbor node set of individual sensor node, and the sensor node of non-selected mistake, it is then to carry out process (6.6), otherwise enters
Row process (6.7)；
(6.6) it is random out of j-th of sensor node composition in path of each bar with j-th of sensor node set
Selection one is in the neighbor node set of -1 sensor node of jth of path candidate, and the sensor node of non-selected mistake
As j-th of sensor node of path candidate (as shown in figure 9, have selected No. 31 sensor node here as candidate road
Footpath X' the 2nd sensor node), carry out process (6.8)；
(6.7) random selection one is non-selected in the neighbor node set of -1 sensor node of jth from path candidate
The sensor node crossed is as j-th of sensor node of path candidate (as shown in figure 9, have selected No. 31 sensor here
No. 33 sensor node of neighbor node of node as candidate route X' the 3rd sensor node), carry out process (6.8)；
(6.8) judge whether j-th of sensor node in path candidate is aggregation node, be then to carry out process (6.9),
Otherwise journey (6.2) is turned over；
(6.9) generate a candidate route X', as shown in figure 9, here X'=s, 31,33,71,80,38,49,88,51,
D }, according to the calculating formula of sub-step 3, the fitness f (X') of candidate route is calculated, judges whether f (X') is less than in the storehouse of path and fits
The fitness in the maximum path of response, is to replace the maximum path of fitness in the storehouse of path with candidate route X', carry out son
Step 7；Otherwise, path storehouse is not made any changes, carries out sub-step 7.
As shown in figure 11, the sub-step 8, including following processes：
(8.1) path X is selectedrIn any sensor node x in addition to source node and aggregation nodej(as shown in figure 12, figure
12 be with the small-scale wireless sensor network of Fig. 6 identicals, for path Xr=s, 31,33,71,80,38,49,88,51,
D }, what is selected here is No. 71 sensor node), judge sensor node xjThe upper hop of (i.e. No. 71 sensor node)
Sensor node xj-1The neighbor node set of (i.e. No. 33 sensor node) and next-hop sensor node xj+1(i.e. No. 80
Sensor node) neighbor node intersection of sets collection (common factor here be { 63,64,65,71,82 }) in whether have not in path
XrIn sensor node, be then carry out process (8.2)；Otherwise to path XrDo not make any changes；
(8.2) one is randomly choosed in the common factor not in path XrIn sensor node (what is selected here is
No. 63 sensor nodes) replace sensor node xj(i.e. No. 71 sensor node), obtains new route X'r(i.e. X'r=s,
31,33,71,80,38,49,88,51, d }), according to the calculating formula of sub-step 3, new route X' is calculatedrFitness f (X'r), sentence
It is disconnected whether f (X'r) ＜ f (Xr), it is then with X'rReplace Xr, carry out sub-step 9；Otherwise, to path XrDo not make any changes, enter
Row sub-step 9；
The effect of the present invention can further be verified by following emulation experiment and relatively：
Square area shown in Figure 13 (a) includes 10 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 200m；
Square area shown in Figure 13 (b) includes 20 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 300m；
Square area shown in Figure 13 (c) includes 30 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 400m；
Square area shown in Figure 13 (d) includes 40 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 500m；
Square area shown in Figure 13 (e) includes 50 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 600m；
Square area shown in Figure 13 (f) includes 60 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 700m；
Square area shown in Figure 13 (g) includes 70 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 800m；
Square area shown in Figure 13 (h) includes 80 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 900m；
Square area shown in Figure 13 (i) includes 90 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 1000m；
Square area shown in Figure 13 (j) includes 100 sensor nodes for what is generated at random in the embodiment of the present invention
The small-scale wireless sensor network of (containing 1 aggregation node)；The square length of side is 1100m；
In Figure 13 (a)~Figure 13 (j), small square block representative sensor node, black inverted triangle block represents aggregation node；
10 small-scale wireless sensor networks of the random generation shown in Figure 13 (a)~Figure 13 (j) are chosen, by the present invention
In routing algorithm and classical high energy efficiency method for routing (the Energy-Efficient Ant-Based based on ant group algorithm
Routing, EEABR) in average residual energy, dump energy standard deviation, least residue energy and life cycle (in the present invention
Life cycle be the wheel that message is sent when there is the sensor node of first depleted of energy in small-scale sensor network
Number) it is compared in terms of this 4 evaluation indexes.Programming language is C++, and allocation of computer is：Intel I7-3610QM processing
Device, 8GB internal memories, 2GB solely show, the bit manipulation systems of windows 8.164.Experiment, which is compared, is divided into two parts, in Part I experiment,
The primary power of each sensor node of small-scale wireless sensor network is identical；In Part II experiment, small-scale wireless sensing
The primary power of each sensor node of device network is different.
Compare and map for the ease of experiment, the routing algorithm in the present invention is designated as IEEHSBR.
First, 1 is tested
In experiment 1, for preceding 3 evaluation indexes (average residual energy, dump energy standard deviation, least residue energy
Amount), each sensor node in 10 small-scale wireless sensor networks is respectively with some cycles to respective place network
Aggregation node sends packet, stops calculating when each aggregation node receives 1000 packets；For the 4th evaluation
Index (i.e. network lifecycle), just stops calculating when there is the depleted of energy of any one sensor node in network.Experiment
With algorithm relevant parameter as shown in table 1, experimental result takes the average of 10 computings.
Table 1 is tested and algorithm relevant parameter
During Figure 14 (a)~Figure 14 (d) is experimental result when each sensor node primary power is identical, each figure, black square
Shape is EEABR experimental result, the white experimental result that tiltedly line rectangle is IEEHSBR.Figure 14 (a), 14 (b), 14 (c) are respectively
It is that aggregation node in each network is received after 1000 packets, it is the average residual energy of each sensor node in network, surplus
10 operation averages of complementary energy standard deviation and least residue energy.Figure 14 (d) is first sensor of appearance in each network
When node energy exhausts, 10 operation averages of network lifecycle.Average residual energy means whole network energy
Consumption is fewer, from Figure 14 (a) as can be seen that IEEHSBR algorithms more save energy than EEABR algorithm, for 10 networks,
IEEHSBR algorithms save 12.3% than EEABR algorithm respectively, 8.4%, 21.3%, 10.9%, 12.2%, 21.8%,
28.8%th, 30.6%, 23.3%, 36.4% energy；The energy expenditure of each sensor node in whole wireless sensor network
Whether to the life cycle of whole network have critically important influence, the dump energy standard deviation of sensor node is smaller if can be averaged
Mean that the energy consumption of whole network is more uniform, Network morals also will be longer, from Figure 14 (b) as can be seen that big absolutely
In majority of network, IEEHSBR algorithms are all substantially more much smaller than the dump energy standard deviation obtained by EEABR algorithm；Network life
The minimum sensor node of sensor node dump energy has direct relation in cycle and network, and least residue energy value is got over
Greatly, Network morals are longer, from Figure 14 (c) as can be seen that in all-network, IEEHSBR algorithms are than EEABR algorithm
The least residue energy of gained is more a lot；Three above index is to weigh wireless sensor network life cycle most important three
Individual index, because performance of the IEEHSBR algorithms in these three indexs will be better than EEABR algorithms, therefore, IEEHSBR algorithms
The life cycle of gained can be more than EEABR algorithms, shown in such as Figure 14 (d), and for 10 networks, IEEHSBR algorithms are calculated than EEABR
Method extends 72.1% respectively, 127.9%, 130.7%, 160%, 145.2%, 130.7%, 147.2%, 136.2%,
163.4%th, 128.9% life cycle.
2nd, 2 are tested
When having been proven that identical in each sensor node primary power in experiment 1, IEEHSBR algorithms will be substantially better than
EEABR algorithms, this section will compare the effect of two kinds of algorithms in the case where each sensor node primary power is different.In this reality
In testing, the primary power random initializtion of each sensor node is any one in 10J, 20J and 30J in 10 networks
Value, for two kinds of algorithms, the primary power value of each sensor node is identical.Other experiment parameters and the algorithm parameter such as institute of table 1
Show.
In this experiment, similarly for preceding 3 evaluation indexes, each sensor node in 10 networks is respectively with certain
Cycle sends packet to the aggregation node of respective place network, stops when each aggregation node receives 1000 packets
Only calculate；For the 4th evaluation index, just stop calculating when there is the depleted of energy of sensor node in network.Experiment
As a result the average of 10 computings is taken.
During Figure 15 (a)~Figure 15 (d) is each asynchronous result of calculation of sensor node primary power, each figure, black square
Shape is EEABR experimental result, the white experimental result that tiltedly line rectangle is IEEHSBR.Figure 15 (a), 15 (b), 15 (c) are respectively
It is that aggregation node in each network is received after 1000 packets, it is the average residual energy of each sensor node in network, surplus
10 operation averages of complementary energy standard deviation and least residue energy.Figure 15 (d) is first sensor of appearance in each network
10 operation averages of Network morals when node energy exhausts.From Figure 15 (a)~15 (d) as can be seen that working as sensor
When node primary power is different, in terms of energy and extension network lifecycle is saved, IEEHSBR algorithms will be substantially better than
EEABR algorithms.For 10 networks, shown in such as Figure 15 (a), IEEHSBR algorithms save 6.7% than EEABR algorithm respectively,
4.3%th, 5.4%, 5.8%, 9.6%, 12.3%, 13%, 10.4%, 15.4% energy；As shown in Figure 15 (d), IEEHSBR
Algorithm extends 124.9% than EEABR algorithm respectively, 201.8%, 222.1%, 362.3%, 339.3%, 388.4%,
342.6%th, 395.1%, 394.1%, 343.5% life cycle.
In summary, for small-scale wireless sensor network, the present invention is saving whole network energy, equalising network energy
EEABR algorithms are substantially better than in terms of consumption and extension network lifecycle.
Claims (2)
1. a kind of method for routing of small-scale wireless sensor network, including transmission global information step, send packet step,
Transmit packet step and update dump energy information Step, it is characterised in that：
(1) global information step is transmitted：
First, each sensor node including aggregation node is included to network to be numbered respectively, it is unique as each of which
Mark；Aggregation node is arranged to save to neighbours in network with each sensor node identical transmit power, aggregation node
Point broadcast message, including the hop count of each sensor node to aggregation node；Each neighbor node is received after route messages,
After wherein itself adds 1 to the hop count of aggregation node, then the route messages updated are broadcasted to the neighbor node of itself；So, often
Individual sensor node can know itself to the hop count of aggregation node, the neighbor node of itself, the distance with neighbor node and
Hop count of the neighbor node to aggregation node；
Then, each sensor node is by its distance, the numbering of neighbor node, neighbor node to aggregation node with neighbor node
Hop count and itself existing energy and hop count are sent to aggregation node, aggregation node by the neighbor node of each sensor node according to
Numbering sorts from small to large, forms the neighbor node set of each sensor node；Again by aggregation node in network with broadcaster
Formula transmits global information to each sensor node；
The neighbor node set of the global information including each sensor node, the distance of each sensor node and neighbor node,
Hop count of the existing energy and each sensor node of each sensor node to aggregation node；
(2) packet step is sent：
When certain sensor node monitors environmental information, first it is processed to turn into packet, then according to the overall situation that itself is stored
Information, calculates the optimal path for itself arriving aggregation node, and the optimal path is added into packet；Then optimal road is calculated
Other each sensor nodes forward the packet to need energy and the dump energy consumed in footpath, and update the complete of itself storage
The dump energy of other each sensor nodes in office's information on optimal path；Self rest energy is added into the data again
Bao Hou, next sensor node is sent to by packet according to the optimal path；
(3) packet step is transmitted：
Next sensor node receives the packet, self rest energy is added into the packet, according in packet
Optimal path continue to deliver a packet to next sensor node, so continue, until packet is sent to aggregation node；
(4) dump energy information Step is updated：
Aggregation node is received after the packet, and packet is handled, and the environment that wherein sensor node is monitored is believed
Breath is by internet or satellite transmission to user；Utilize the dump energy information updating storage of each sensor node in packet
Global information, aggregation node is with some cycles to all residue energy of node information of each sensor node broadcasts；
In the transmission packet step, calculating the optimal path includes following sub-steps：
Sub-step 1, the size H that path storehouse PM is set, minimum select probability Cmin, MAXIMUM SELECTION probability CmaxAnd evaluate times N：
PM=[X1, X2..., Xr..., XH]T, wherein, XrFor r paths, represent current sensor node to aggregation node
Path, r=1,2 ..., H, Xr={ s ..., xr,j..., d }, in formula, s is current sensor node, and d is aggregation node, xr,jFor
J-th of sensor node in r paths；
H=3~12,0≤Cmin≤Cmax≤ 1, N=200~800；
Sub-step 2, to being initialized in the PM of path storehouse per paths, including following processes：
(2.1) by path length variable j assignment 1, sensor node s is set to j-th of sensor in current initialization path
Node；
(2.2) j+1 value is assigned to j, is j-th of sensor node of current initialization Path selection：
Judge the biography with the presence or absence of non-selected mistake in the neighbor node set of current initialization path -1 sensor node of jth
Sensor node, is then to carry out process (2.3), otherwise turns over journey (2.1)；
(2.3) random number R between 0 to 1 is produced first, and -1 sensor of jth in current initialization path is then subtracted with R
The select probability of the minimum sensor node of the numbering of non-selected mistake, judges whether its difference is little in the neighbor node set of node
In 0, it is then to carry out process (2.4), otherwise turns over journey (2.5)；
(2.4) the non-selected mistake is numbered into minimum sensor node as j-th of sensor section in current initialization path
Point, turns over journey (2.6)；
(2.5) difference is continued to subtract in current initialization path in the neighbor node set of -1 sensor node of jth
The select probability of the small sensor node of numbering time of non-selected mistake, then judge that being then will be described unselected the result is that no be not more than 0
Otherwise the small sensor node of numbering time selected so is continued until as j-th of sensor node in current initialization path
Last difference is not more than 0, with being finally referred in the neighbor node set of current initialization path -1 sensor node of jth
The sensor node of its select probability turns over journey (2.6) as j-th of sensor node in current initialization path；
(2.6) whether j-th of sensor node for judging current initialization path is aggregation node, is then to generate an initial road
Footpath；Otherwise journey (2.2) is turned over；
To, per paths repetitive process (2.1)~(2.6), symbiosis completes the initial of PM into H bar initial paths in the PM of path storehouse
Change；
Sub-step 3, the fitness f (X for calculating path Ku Zhongge pathsr)：
In formula, E (Xr) it is path XrIn the gross energy that is consumed of sensor node transmission packet,L is path length, di,i+1It is sensor node i and sensor
The ENERGY E consumed when being worked with transmitting circuit is received in the distance between node i+1, sensor nodeelecFor 50nJ/bit,
The ENERGY E consumed when transmitting amplifier operation in sensor nodeampFor 0.1nJ/bit/m2；K is transmitted by sensor node
Data package size, unit is bit；EMinFor path XrThe dump energy of the minimum sensor node of middle dump energy, EAvgFor
Path XrThe average residual energy of middle all the sensors node；
Degree variables t=1 is evaluated in sub-step 4, setting；
Sub-step 5, setting number of passes variable r=1；
Sub-step 6, generation path candidate X', including following processes：
(6.1) path length variable j is assigned to by 1, current sensor node s is set to candidate route X' j-th of sensing
Device node；
(6.2) j+1 value is assigned to j, judged in candidate route X' in the range of the neighbor node of -1 sensor node of jth
With the presence or absence of the sensor node not selected, it is then to carry out process (6.3), otherwise turns over journey (6.1)；
(6.3) j-th of sensor node is selected for candidate route X'：
The random number R between 0 to 1 is produced, judges whether R ＜ C, is then to carry out process (6.4), otherwise carries out process (6.7)；Institute
State select probability
(6.4) judge to whether there is the path with j-th of sensor node in the PM of path storehouse, be then to carry out process (6.5)；It is no
Then carry out process (6.7)；
(6.5) whether there is -1 biography of jth in path candidate in j-th of sensor node for judging each paths in the storehouse of path
In the neighbor node set of sensor node, and the sensor node of non-selected mistake, it is then to carry out process (6.6), otherwise carried out
Journey (6.7)；
(6.6) randomly choosed out of j-th of sensor node composition in path of each bar with j-th of sensor node set
One in the neighbor node set of -1 sensor node of jth of path candidate, and the sensor node conduct of non-selected mistake
J-th of sensor node of path candidate, carries out process (6.8)；
(6.7) non-selected mistake is randomly choosed in the neighbor node set of -1 sensor node of jth from path candidate
Sensor node carries out process (6.8) as j-th of sensor node of path candidate；
(6.8) judge whether j-th of sensor node in path candidate is aggregation node, be then to carry out process (6.9), otherwise
Turn over journey (6.2)；
(6.9) a candidate route X' is generated, according to the calculating formula of sub-step 3, the fitness f (X') of candidate route is calculated, sentences
Whether disconnected f (X') is less than the fitness in the maximum path of fitness in the storehouse of path, is to replace path storehouse with candidate route X'
The maximum path of middle fitness, carries out sub-step 7；Otherwise, path storehouse is not made any changes, carries out sub-step 7；
Sub-step 7, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 8, otherwise perform sub-step 11；
Sub-step 8, the r paths X from path storehouserIn in addition to source node and aggregation node optional sensor node xjCarry out
Neighborhood search, including following processes：
(8.1) path X is selectedrIn any sensor node x in addition to source node and aggregation nodej, judge sensor node xj's
Upper hop sensor node xj-1Neighbor node set and next-hop sensor node xj+1Neighbor node intersection of sets collection in
Whether have not in path XrIn sensor node, be then carry out process (8.2)；Otherwise to path XrDo not make any changes；
(8.2) one is randomly choosed in the common factor not in path XrIn sensor node replace sensor node xj, obtain
New route X'r, according to the calculating formula of sub-step 3, calculate new route X'rFitness f (X'r), judge whether f (X'r) ＜ f
(Xr), it is then with X'rReplace Xr, carry out sub-step 9；Otherwise, to path XrDo not make any changes, carry out sub-step 9；
Sub-step 9, the value imparting t by t+1, judge whether t≤N, are then to perform sub-step 10, otherwise perform sub-step 11；
Sub-step 10, the value imparting r by r+1, judge whether r≤H, are then to perform sub-step 6, otherwise perform sub-step 5；
The minimum paths of fitness are used as optimal path in sub-step 11, selection path storehouse.
2. method for routing as claimed in claim 1, it is characterised in that：
In the process (2.3) of the sub-step 2 or (2.5), the neighbours of -1 sensor node of jth in current initialization path
The numbering of non-selected mistake is minimum in node set or select probability P of the small sensor node i of numbering time (j-1, i)：
In formula, aj-1Represent the neighbor node set of -1 sensor node of jth, N (aj-1) it is set aj-1Number of elements, hi
Represent aj-1Middle sensor node i hop count, hmRepresent aj-1Middle sensor node m hop count.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410705802.2A CN104486807B (en) | 2014-11-28 | 2014-11-28 | A kind of method for routing of small-scale wireless sensor network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410705802.2A CN104486807B (en) | 2014-11-28 | 2014-11-28 | A kind of method for routing of small-scale wireless sensor network |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104486807A CN104486807A (en) | 2015-04-01 |
| CN104486807B true CN104486807B (en) | 2017-08-11 |
Family
ID=52761296
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410705802.2A Active CN104486807B (en) | 2014-11-28 | 2014-11-28 | A kind of method for routing of small-scale wireless sensor network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104486807B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10075364B2 (en) * | 2015-05-19 | 2018-09-11 | Alcatel-Lucent | Method and apparatus for self-tuned adaptive routing |
| CN105812255B (en) * | 2016-05-06 | 2021-05-11 | 网宿科技股份有限公司 | Method and device for selecting return source line |
| CN106550422B (en) * | 2016-11-08 | 2017-08-25 | 华中科技大学 | A kind of wireless sensor network clustering routing based on harmonic search algorithm |
| CN108541039B (en) * | 2018-04-24 | 2021-10-22 | 苏州市职业大学 | Low-power-consumption wireless sensor network static node routing method |
| CN108965417B (en) * | 2018-07-04 | 2021-12-21 | 梧州市兴能农业科技有限公司 | Farmland soil pollution wireless monitoring system |
| CN109151757A (en) * | 2018-09-26 | 2019-01-04 | 东莞青柳新材料有限公司 | A kind of real-time managing and control system of agricultural land soil quality |
| CN109151758A (en) * | 2018-09-27 | 2019-01-04 | 广州文搏科技有限公司 | Agricultural land soil health intelligent based on wireless sensor network monitors system |
| CN109963262B (en) * | 2019-01-28 | 2020-08-18 | 华南理工大学 | Wireless sensor scheduling optimization method in wireless sensor network |
| CN116567773B (en) * | 2023-07-10 | 2024-04-26 | 北京星科软件技术有限公司 | WSN clustering routing method and routing system based on Internet of things application |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102572717A (en) * | 2012-02-20 | 2012-07-11 | 南京中通电气有限公司 | Multipath routing reliable transmission method based on network coding |
| CN102711213A (en) * | 2012-05-14 | 2012-10-03 | 南京邮电大学 | Wireless sensor network routing method based on residual energy |
| CN103916927A (en) * | 2014-03-17 | 2014-07-09 | 华中科技大学 | Wireless sensor network routing method based on improved harmony search algorithm |
-
2014
- 2014-11-28 CN CN201410705802.2A patent/CN104486807B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102572717A (en) * | 2012-02-20 | 2012-07-11 | 南京中通电气有限公司 | Multipath routing reliable transmission method based on network coding |
| CN102711213A (en) * | 2012-05-14 | 2012-10-03 | 南京邮电大学 | Wireless sensor network routing method based on residual energy |
| CN103916927A (en) * | 2014-03-17 | 2014-07-09 | 华中科技大学 | Wireless sensor network routing method based on improved harmony search algorithm |
Non-Patent Citations (1)
| Title |
|---|
| A new meta-heuristic algorithm for continuous engineering optimization: harmony search theory and practice;K.S. Lee, Z.W. Geem;《Computer methods in applied mechanics and engineering》;20051231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104486807A (en) | 2015-04-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104486807B (en) | A kind of method for routing of small-scale wireless sensor network | |
| Kumar et al. | Improved leach protocol for wireless sensor networks | |
| CN101119303B (en) | Dynamic cluster based multi-objective programming wireless sensing network routing algorithm | |
| Safaei et al. | Effects of RPL objective functions on the primitive characteristics of mobile and static IoT infrastructures | |
| CN106550422B (en) | A kind of wireless sensor network clustering routing based on harmonic search algorithm | |
| Pattnaik et al. | Assimilation of fuzzy clustering approach and EHO‐Greedy algorithm for efficient routing in WSN | |
| Naeem et al. | Enhanced clustering based routing protocol in vehicular ad‐hoc networks | |
| Song et al. | A genetic algorithm for energy-efficient based multipath routing in wireless sensor networks | |
| CN100471166C (en) | Distribution type QoS multi-broadcast routing method based on mobile Ad Hoc network | |
| Ben Gouissem et al. | Energy efficient grid based k‐means clustering algorithm for large scale wireless sensor networks | |
| Fan et al. | Dynamic virtual network embedding of mobile cloud system based on global resources in internet of vehicles | |
| Sumathi et al. | NEWTR: a multipath routing for next hop destination in internet of things with artificial recurrent neural network (RNN) | |
| CN113411766A (en) | Intelligent Internet of things comprehensive sensing system and method | |
| Mody et al. | Cluster head selection algorithm for wireless sensor networks using Machine learning | |
| Singh et al. | Designing an energy efficient network using integration of KSOM, ANN and data fusion techniques | |
| Pang et al. | Neural Network‐Based Routing Energy‐Saving Algorithm for Wireless Sensor Networks | |
| Siyi et al. | Quality of service assessment routing protocols for performance in a smart building: A case study | |
| Kannammal et al. | A hybrid approach-based energy aware cluster head selection for IoT application | |
| Zhang et al. | An immune chaotic adaptive evolutionary algorithm for energy-efficient clustering management in LPWSN | |
| Hung | Incorporate ACO routing algorithm and mobile sink in wireless sensor networks. | |
| Li et al. | A Routing Algorithm for WiFi‐Based Wireless Sensor Network and the Application in Automatic Meter Reading | |
| Karmokar et al. | An implementation of basic ant-colony optimization based routing in wireless sensor networks | |
| Din et al. | Energy source saving approach using multi-tier network design technique | |
| Li et al. | Genetic programming for lifetime maximization in wireless sensor networks with a mobile sink | |
| Zaballos et al. | A genetic-based QoS aware routing for ubiquitous sensor networks |
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 |