CN105873129B - The sensor network missing values reconstructing method of multi-node collaboration - Google Patents
The sensor network missing values reconstructing method of multi-node collaboration Download PDFInfo
- Publication number
- CN105873129B CN105873129B CN201610175992.0A CN201610175992A CN105873129B CN 105873129 B CN105873129 B CN 105873129B CN 201610175992 A CN201610175992 A CN 201610175992A CN 105873129 B CN105873129 B CN 105873129B
- Authority
- CN
- China
- Prior art keywords
- opinion
- node
- missing values
- judgement
- destination node
- 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.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
- H04W28/04—Error control
-
- 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)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Complex Calculations (AREA)
Abstract
The present invention relates to a kind of sensor network missing values reconstructing methods of multi-node collaboration, it include: according to the spatial coherence of destination node O and neighbor node A, obtain opinion of the neighbor node A about destination node O missing values: the judgement of destination node O missing values is x " (On+1), opinion is { b, u, a };Similar, calculate separately judgement and corresponding opinion of other neighbor nodes about missing values of destination node O;The judgement of multiple neighbor nodes of integration objective node O and corresponding opinion, obtain final opinion and corresponding judgement set;It is expected that eiFor judgement xiThe probability of appearance, to desired eiThe merging based on weight is carried out, missing values the x " (O of destination node O is restoredn++1).The present invention makes full use of sensor node perception data, and there are temporal correlations and spatial coherence feature, opinion of each neighbor node of objective quantification about destination node missing values, and accurately merge the opinion of multiple neighbor nodes, reduce the missing values error of reduction, user is not needed excessively to participate in, robustness is stronger, and reduction accuracy rate is higher.
Description
Technical field
The present invention relates to wireless sensor network technology field, in particular to a kind of sensor network of multi-node collaboration lacks
Mistake value reconstructing method.
Background technique
Wireless sensor network is exactly to be formed by being deployed in a large amount of cheap microsensor node in monitoring region, is passed through
The network system of the self-organizing for the multi-hop that communication is formed, the purpose is to collaboratively perceive, acquire and handle net
It is perceived the information of object in network overlay area, and is sent to observer.Node in sensor node communication range is that 1 jump is adjacent
Node is occupied, neighbor node is defaulted as 1 hop neighbor node, and measuring the most commonly seen method of neighbor node is according to the object between node
Whether reason distance is in communication range.Due to noise, collision, insecure connection, wireless sensor network often loss portion
The data that sub-sensor node is perceived.
In order to solve this problem, people often utilize node institute's perception data to have temporal correlation and spatial coherence
The characteristics of restore missing values.So-called temporal correlation, which refers to before and after the data that sensor node is perceived, has certain connection
System, using temporal correlation, can estimate (can certainly be missing from) in future according to the data of the past period
Data;So-called spatial coherence refer to data that sensor node is perceived with adjacent node often there is also centainly contacting, example
Such as two are deployed in outdoor adjacent sensors node perceived temperature, when the decline of some node perceived temperature, another node
It can sense temperature decline.However, this just draws a new problem, how temporal correlation and spatial coherence rationally to be melted
It closes, carrys out the missing data in accurate reproduction wireless sensor network.
Pan Li is strong et al. to combine temporal correlation and spatial coherence, using linear interpolation model, lacks to data
Mistake value is restored, but does not better solve the fusion problem of multiple missing values.This is mainly reflected in two aspects: first
First, author attempts directly to portray temporal correlation and spatial coherence with simple model, and it is obvious that this results in its method to exist
Defect, such as: autoregression model does not distinguish the correlation between neighbor node and present node, increases meeting the considerations of in this respect
Calculation amount is significantly greatly increased, does not solve the problems, such as this again and will lead to missing values reduction error increase;Secondly, author attempts to ask flat with weighting
Equal method merges the missing values that two kinds of distinct methods generate, the setting of this method heavy dependence weight, and weight
Setting to rely on user again expected to the subjectivity of missing values, it is easy to error amount is introduced, so that the missing values error restored increases
Add.Generally speaking, work on hand is supported without reasonable data, is easy to cause in the method for the multiple missing values of integrated survey
There are biggish errors with true value for the missing values of reduction.
Summary of the invention
Aiming at the shortcomings in the prior art, the present invention provides a kind of sensor network missing values reconstruct side of multi-node collaboration
Method restores missing values according to temporal correlation and spatial coherence, then first based on each neighbor node of destination node
The missing values that each neighbor node is restored are merged using opinion increment fusion rule, so that reduction obtains missing values;Using more
The method of a neighbor node cooperation restores missing values, reduces the error of reduction process, improves the accuracy rate of missing values reduction.
According to design scheme provided by the present invention, a kind of sensor network missing values reconstructing method of multi-node collaboration,
It comprises the following steps:
Step 1. obtains the reducing value x ' of missing values using interpolation formula according to the historical data sequence of destination node O
(On+1), and according to the reducing value x ' (O of the data sequence of neighbor node A amendment destination node On+1), obtain destination node O's
Missing values x " (On+1), the historical data sequence according to destination node O and neighbor node A calculates missing values x " (On+1) it is uncertain
Property u, conviction b and relative atom degree a, obtains opinion of the neighbor node A about destination node O missing values: destination node O missing values
Judgement be x " (On+1), opinion is { b, u, a };
Step 2. restores missing values according to spatial coherence, the current data of multiple neighbor nodes of integration objective node O,
If neighbor node is identical about the judgement of destination node O missing values, same judgement is merged according to binary opinion increment fusion rule
Opinion, obtain final opinion ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,
x2,...,xk};Otherwise, the opinion of different judgements is mapped as to using opinion with memberization the opinion of identical judgement, guarantees the opinion phase
In the case that prestige does not change, judgement set is expanded, the opinion set of single judgement is made to become the opinion collection of more judgements
It closes, is successively merged two-by-two according to polynary opinion increment fusion rule, until a remaining opinion in opinion set, is finally anticipated
See ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,x2,...,xk};
Step 3. is according to desired calculation formula ei=bi+aiU, it is expected that eiFor judgement xiThe probability of appearance, to desired eiInto
Merging of the row based on weight, merges formulaWherein, hiIt is expected eiWeight the weight item being averaging, viFor
Judgement xiMissing values size, missing values the x " (O of destination node On+1) it is reduced to value.
Above-mentioned, the reducing value of destination node O missing values in step 1According to
Neighbor node A historical data sequence x (A1),x(A2),...,x(An) reduction
And with true x (An+1) be compared, according to x ' (An+1) and x (An+1) difference, to correct x ' (On+1), obtain destination node
Missing values the x " (O of On+1)=x ' (On+1)+x′(An+1)-x(An+1);According to the n-1 and n-2 of destination node O and neighbor node A
Wheel data obtain the n-th wheel data x (On') and x (An'), and respectively with true value x (An) and x (On) difference is asked to obtain c and d, not really
Qualitative u isAnd conviction b:b=1-u is acquired according to formula;Calculate neighbor node A data sequence x (A1),x
(A2),...,x(An) and destination node O data sequence x (O1),x(O2),...,x(On) between difference, obtain new data sequence
Arrange d (t1),d(t2),...,d(tn), calculate data sequence d (t1),d(t2),...,d(tn) varianceWherein, E is the mean value of data sequence, relative atom degree a's
Calculation formulaWherein, dis=x (An+1)-x″(On+1)。
Above-mentioned, binary opinion increment fusion rule particular content in the step 2 are as follows: set ωAAnd ωBIt is neighbours respectively
The opinion that node A and B are provided, it is assumed that main body [A, B] is according to ωAAnd ωBObtain opinionIts calculation formula is as follows: situation 1
(uA≠ 0 or uB≠0):2 (u of situationA=0 and uB=0):
Wherein,Polynary opinion increment fusion rule particular content
Are as follows: more judgement X={ x1,x2,...,xkOpinion ωAAnd ωB, ωA◇BAssuming that main body [A, B] is according to ωAAnd ωBObtain opinionIts calculation formula is as follows: 1 (u of situationA≠ 0 or uB≠0):2 (u of situationA=0 and uB
=0):Wherein,
Above-mentioned, considered in increment fusion rule based on judgement blurring, by the way that judgement accuracy is arranged, by similar judgement
It merges, incrementally fusion rule merges corresponding opinion.
Beneficial effects of the present invention:
1, the present invention constructs opinion using interpolation method according to the data sequence of destination node and neighbor node, generate judgement,
Uncertain, conviction and relative atom degree;The neighbor node opinion of destination node is merged, is blurred and is anticipated by judgement
See with memberization, the parameter being averaging it is expected and weighted using judgement, the missing data of destination node is further restored, makes full use of
There is temporal correlation and spatial coherence in sensor node perception data, objectively quantify each neighbor node about
The opinion of destination node missing values, and the opinion of multiple neighbor nodes is accurately merged, the missing values error of reduction is reduced,
It does not need user excessively to participate in, robustness is stronger, and reduction accuracy rate is higher.
2, the present invention merges the missing values that multiple neighbor nodes are restored according to temporal correlation, and reduction obtains accurately
Missing values describe the missing of separate sources on the basis of using for reference sensor network time correlation and spatial coherence first
Value portrays missing values by elements such as conviction in node opinion and uncertainties in detail;Then it is blurred and anticipates using judgement
See with memberization, multiple opinions are reasonably permeated an a conclusion opinion;Finally according to conclusion opinion, conversion weighting is averaging
Relevant parameter restores missing values;Unlike the prior art, the present invention relies on multi-node collaboration, and opinion is made to merge and restore
Missing values are evidence-based.
Detailed description of the invention:
Fig. 1 is flow diagram of the invention.
Specific embodiment:
The present invention is described in further detail with technical solution with reference to the accompanying drawing, and detailed by preferred embodiment
Describe bright embodiments of the present invention in detail, but embodiments of the present invention are not limited to this.
Embodiment one, shown in Figure 1, a kind of sensor network missing values reconstructing method of multi-node collaboration, comprising such as
Lower step:
Step 1. obtains the reducing value x ' of missing values using interpolation formula according to the historical data sequence of destination node O
(On+1), and according to the reducing value x ' (O of the data sequence of neighbor node A amendment destination node On+1), obtain lacking for destination node O
Mistake value x " (On+1), the historical data sequence according to destination node O and neighbor node A calculates missing values x " (On+1) uncertainty
U, conviction b and relative atom degree a obtains opinion of the neighbor node A about destination node O missing values: destination node O missing values
Judgement is x " (On+1), opinion is { b, u, a };
Step 2. restores missing values according to spatial coherence, the current data of multiple neighbor nodes of integration objective node O,
If neighbor node is identical about the judgement of destination node O missing values, same judgement is merged according to binary opinion increment fusion rule
Opinion, obtain final opinion ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,
x2,...,xk};Otherwise, the opinion of different judgements is mapped as to using opinion with memberization the opinion of identical judgement, guarantees the opinion phase
In the case that prestige does not change, judgement set is expanded, the opinion set of single judgement is made to become the opinion collection of more judgements
It closes, is successively merged two-by-two according to polynary opinion increment fusion rule, until a remaining opinion in opinion set, is finally anticipated
See ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,x2,...,xk};
Step 3. is according to desired calculation formula ei=bi+aiU, it is expected that eiFor judgement xiThe probability of appearance, to desired eiInto
Merging of the row based on weight, merges formulaWherein, hiIt is expected eiWeight the weight item being averaging, viFor
Judgement xiMissing values size, missing values the x " (O of destination node On+1) it is reduced to value.
Embodiment two, shown in Figure 1, a kind of sensor network missing values reconstructing method of multi-node collaboration, comprising such as
Lower step:
Step 1. obtains the reducing value x ' of missing values using interpolation formula according to the historical data sequence of destination node O
(On+1), the reducing value of destination node O missing valuesIt is saved according to neighbours
Point A historical data sequence x (A1),x(A2),...,x(An) reduction
And with true x (An+1) be compared, according to x ' (An+1) and x (An+1) difference, to correct x ' (On+1), obtain target section
Missing values the x " (O of point On+1)=x ' (On+1)+x′(An+1)-x(An+1);According to the n-1 and n-2 of destination node O and neighbor node A
Wheel data obtain the n-th wheel data x (On') and x (An'), and respectively with true value x (An) and x (On) difference is asked to obtain c and d, not really
Qualitative u isAnd conviction b:b=1-u is acquired according to formula;Calculate neighbor node A data sequence x (A1),x
(A2),...,x(An) and destination node O data sequence x (O1),x(O2),...,x(On) between difference, obtain new data
Sequence d (t1),d(t2),...,d(tn), calculate data sequence d (t1),d(t2),...,d(tn) varianceWherein, E is the mean value of data sequence, relative atom degree a's
Calculation formulaWherein, dis=x (An+1)-x″(On+1), it obtains
Opinion of the neighbor node A about destination node O missing values: the judgement of destination node O missing values is x " (On+1), opinion be b, u,
a};
Step 2. restores missing values according to spatial coherence, the current data of multiple neighbor nodes of integration objective node O,
If neighbor node is identical about the judgement of destination node O missing values, same judgement is merged according to binary opinion increment fusion rule
Opinion, obtain final opinion ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,
x2,...,xk};Otherwise, the opinion of different judgements is mapped as to using opinion with memberization the opinion of identical judgement, guarantees the opinion phase
In the case that prestige does not change, judgement set is expanded, the opinion set of single judgement is made to become the opinion collection of more judgements
It closes, is successively merged two-by-two according to polynary opinion increment fusion rule, until a remaining opinion in opinion set, is finally anticipated
See ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,x2,...,xk};
Step 3. is according to desired calculation formula ei=bi+aiU, it is expected that eiFor judgement xiThe probability of appearance, to desired eiInto
Merging of the row based on weight, merges formulaWherein, hiIt is expected eiWeight the weight item being averaging, viFor
Judgement xiMissing values size, missing values the x " (O of destination node On+1) it is reduced to value.According to the history number of destination node O
The reducing value x ' (O of missing values is obtained using interpolation formula according to sequencen+1), and mesh is corrected according to the data sequence of neighbor node A
Mark the reducing value x ' (O of node On+1), obtain missing values the x " (O of destination node On+1), according to destination node O and neighbor node A
Historical data sequence calculate missing values x " (On+1) uncertain u, conviction b and relative atom degree a, obtain neighbor node A pass
In the opinion of destination node O missing values: the judgement of destination node O missing values is x " (On+1), opinion is { b, u, a }.
Binary opinion increment fusion rule particular content in step 2 are as follows: set ωAAnd ωBNeighbor node A and B respectively to
Opinion out, it is assumed that main body [A, B] is according to ωAAnd ωBObtain opinionIts calculation formula is as follows: 1 (u of situationA≠ 0 or
uB≠0):2 (u of situationA=0 and uB=0):Wherein, It is blurred and is considered based on judgement, by setting
Judgement accuracy is set, similar judgement is merged, merges corresponding opinion according to binary opinion increment fusion rule, such as:
Judgement 1 is missing from value should be for 21.123, and judgement 2, which is missing from value, to be 21.129, can be with based on judgement blurring
Setting accuracy is 2 significant digits, so that judgement A and judgement B are considered as the same judgement, and then can incrementally be melted
Normally merge corresponding opinion.It is further illustrated with example: setting existing judgement as { 1.214,1.214,1.213,1.32 },
Corresponding opinion collection is combined into { w1,w2,w3,w4, then opinion w1And w2Increment fusion can be directly carried out, at this time opinion collection
Conjunction is updated toIt, can be by 1.214 and 1.213 etc. when accuracy is set 2 significant digits by user
It is considered as the same judgement, can be further by corresponding opinion fusion, opinion set is updated at this time
Polynary opinion increment fusion rule particular content in step 2 are as follows: more judgement X={ x1,x2,...,xkOpinion
ωAAnd ωB, Assuming that main body [A, B] is according to ωAAnd ωBObtain opinion
Its calculation formula is as follows: 1 (u of situationA≠ 0 or uB≠0):2 (u of situationA=0 and uB=0):Wherein, It is specific:
It include original these judgements X={ x there are a new judgement X if existing k opinion corresponds to k different judgements1,
x2,...,xk, then some judgement xiCorresponding opinion can be mapped as from { b, u, a } 0,0 ..., b (xi),...,0},
u,{0,0,...,a(xi),...,0}}.Such as: it should be 1.1 according to the data-speculative missing values of neighbor node A, it is corresponding to anticipate
See { 0,4,0.6,0.3 };It should be 1.2 according to the data-speculative missing values of neighbor node B, corresponding opinion 0,5,0.5,
0.4};By arranging and mapping, judgement be changed to missing values may for 1.1 or 1.2, corresponding opinion be { 0.4,0 },
0.6, { 0.3,0 } }, and { { 0,0.5 }, 0.5, { 0,0.4 } }, the opinion after arrangement and mapping can continue incrementally
Fusion rule is merged.
The present invention makes full use of sensor node perception data to there are the characteristics that temporal correlation and spatial coherence, objective
Opinion of each neighbor node of quantization about destination node missing values, and accurately merge the opinion of multiple neighbor nodes,
Reduce the missing values error of reduction.Although existing method also using sensor node perception data there are temporal correlation and
Spatial coherence, but in the data sequence for how investigating neighbor node, how by temporal correlation and two kinds of spatial coherence
Characteristic data obtained carry out the key takeaway such as merging, also excessive to rely on parameter set by user, only in ideal feelings
Accurate reducing value can be just obtained under condition.Compared to the prior art, the method given by the present invention, it is excessive not need user
It participates in, robustness is stronger, and reduction accuracy rate is higher.
The invention is not limited to above-mentioned specific embodiment, those skilled in the art can also make a variety of variations accordingly,
But it is any all to cover within the scope of the claims with equivalent or similar variation of the invention.
Claims (2)
1. a kind of sensor network missing values reconstructing method of multi-node collaboration, it is characterised in that: comprise the following steps:
Step 1. obtains the reducing value x ' (O of missing values using interpolation formula according to the historical data sequence of destination node On+1),
And the reducing value x ' (O of destination node O is corrected according to the data sequence of neighbor node An+1), obtain the missing values x " of destination node O
(On+1), the historical data sequence according to destination node O and neighbor node A calculates missing values x " (On+1) uncertain u, conviction
B and relative atom degree a, obtains opinion of the neighbor node A about destination node O missing values: the judgement of destination node O missing values is
x″(On+1), opinion is { b, u, a };
Step 2. is according to spatial coherence, and the current data of multiple neighbor nodes of integration objective node O restores missing values, if adjacent
The judgement that node is occupied about destination node O missing values is identical, then the meaning with judgement is merged according to binary opinion increment fusion rule
See, obtains final opinion ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,
x2,...,xk};Otherwise, the opinion of different judgements is mapped as to using opinion with memberization the opinion of identical judgement, guarantees the opinion phase
In the case that prestige does not change, judgement set is expanded, the opinion set of single judgement is made to become the opinion collection of more judgements
It closes, is successively merged two-by-two according to polynary opinion increment fusion rule, until a remaining opinion in opinion set, is finally anticipated
See ω={ { b1,b2,...,bk},u,{a1,a2,...,akAnd corresponding judgement collection be combined into { x1,x2,...,xk};
Step 3. is according to desired calculation formula ei=bi+aiU, it is expected that eiFor judgement xiThe probability of appearance, to desired eiCarry out base
In the merging of weight, merge formulaWherein, hiIt is expected eiWeight the weight item being averaging, viFor judgement
xiMissing values size, missing values the x " (O of destination node On+1) it is reduced to value;
The reducing value of destination node O missing values in step 1According to neighbours
Node A historical data sequence x (A1),x(A2),...,x(An) reduction
And with true x (An+1) be compared, according to x ' (An+1) and x (An+1) difference, to correct x ' (On+1), obtain destination node
Missing values the x " (O of On+1)=x ' (On+1)+x′(An+1)-x(An+1);It is taken turns according to the n-1 and n-2 of destination node O and neighbor node A
Data obtain the n-th wheel data x (O 'n) and x (A 'n), and respectively with true value x (An) and x (On) difference is asked to obtain c and d, it does not know
Property u isAnd conviction b:b=1-u is acquired according to formula;Calculate neighbor node A data sequence x (A1),x
(A2),...,x(An) and destination node O data sequence x (O1),x(O2),...,x(On) between difference, obtain new data sequence
Arrange d (t1),d(t2),...,d(tn), calculate data sequence d (t1),d(t2),...,d(tn) varianceWherein, E is the mean value of data sequence, relative atom degree a's
Calculation formulaWherein, dis=x (An+1)-x″(On+1);
Binary opinion increment fusion rule particular content in step 2 are as follows: set ωAAnd ωBIt is the meaning that neighbor node A and B are provided respectively
See, it is assumed that main body [A, B] is according to ωAAnd ωBObtain opinionIts calculation formula is as follows:
1 (u of situationA≠ 0 or uB≠0):2 (u of situationA=0 and uB=0):Wherein,
Polynary opinion increment fusion rule particular content are as follows: more judgement X={ x1,x2,...,xkOpinion ωAAnd ωB,
Assuming that main body [A, B] is according to ωAAnd ωBObtain opinionIts calculation formula is as follows: 1 (u of situationA≠ 0 or uB≠0):2 (u of situationA=0 and uB=0):Wherein,
2. the sensor network missing values reconstructing method of multi-node collaboration according to claim 1, it is characterised in that: increment
Considered in fusion rule based on judgement blurring, by the way that judgement accuracy is arranged, similar judgement is merged, is incrementally melted
Normally merge corresponding opinion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610175992.0A CN105873129B (en) | 2016-03-24 | 2016-03-24 | The sensor network missing values reconstructing method of multi-node collaboration |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610175992.0A CN105873129B (en) | 2016-03-24 | 2016-03-24 | The sensor network missing values reconstructing method of multi-node collaboration |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105873129A CN105873129A (en) | 2016-08-17 |
CN105873129B true CN105873129B (en) | 2018-12-18 |
Family
ID=56625909
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610175992.0A Expired - Fee Related CN105873129B (en) | 2016-03-24 | 2016-03-24 | The sensor network missing values reconstructing method of multi-node collaboration |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105873129B (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101516099A (en) * | 2009-04-07 | 2009-08-26 | 华中科技大学 | Test method for sensor network anomaly |
CN103079288A (en) * | 2013-01-04 | 2013-05-01 | 中国电子科技集团公司第二十八研究所 | Method for collecting data of wireless sensor network on basis of high-speed train |
CN103619009A (en) * | 2013-10-25 | 2014-03-05 | 河海大学常州校区 | A method for establishing a trust model in an underwater sensor network |
CN104410992A (en) * | 2014-10-30 | 2015-03-11 | 重庆邮电大学 | Trust-based situation data fusion method of distributed sensor network |
-
2016
- 2016-03-24 CN CN201610175992.0A patent/CN105873129B/en not_active Expired - Fee Related
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101516099A (en) * | 2009-04-07 | 2009-08-26 | 华中科技大学 | Test method for sensor network anomaly |
CN103079288A (en) * | 2013-01-04 | 2013-05-01 | 中国电子科技集团公司第二十八研究所 | Method for collecting data of wireless sensor network on basis of high-speed train |
CN103619009A (en) * | 2013-10-25 | 2014-03-05 | 河海大学常州校区 | A method for establishing a trust model in an underwater sensor network |
CN104410992A (en) * | 2014-10-30 | 2015-03-11 | 重庆邮电大学 | Trust-based situation data fusion method of distributed sensor network |
Non-Patent Citations (2)
Title |
---|
A Trust-Based Framework for Fault-Tolerant Data Aggregation in Wireless Multimedia Sensor Networks;Y. Sun, H. Luo, S. K. Das;《IEEE Transactions on Dependable Secure Computing》;20120806;第9卷(第6期);全文 * |
Subjective Logic-Based Anomaly Detection Framework in Wireless Sensor Networks;Jinhui Yuan, Hongwei Zhou, Hong Chen;《Smart Sensor Networks: Theory and Practice》;20120101;第8卷(第1期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN105873129A (en) | 2016-08-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Tuanmu et al. | Temporal transferability of wildlife habitat models: implications for habitat monitoring | |
CN104076809B (en) | Data processing equipment and data processing method | |
Van Beveren et al. | How catch underreporting can bias stock assessment of and advice for northwest Atlantic mackerel and a possible resolution using censored catch | |
CN104820876A (en) | Short-term load forecasting method and system | |
CN110246037B (en) | Transaction characteristic prediction method, device, server and readable storage medium | |
MX2021001991A (en) | Identifying image aesthetics using region composition graphs. | |
JP6052278B2 (en) | Motion determination device, motion determination system, and motion determination method | |
CN105809264B (en) | Power load prediction method and device | |
Ivits et al. | Land-Productivity Dynamics: Towards integrated assessment of land degradation at global scales | |
Alippi et al. | On-line reconstruction of missing data in sensor/actuator networks by exploiting temporal and spatial redundancy | |
CN111461812A (en) | Object recommendation method and device, electronic equipment and readable storage medium | |
CN112148557A (en) | Method for predicting performance index in real time, computer equipment and storage medium | |
CN109376919A (en) | A kind of prediction technique of coal mine fully-mechanized mining working gas emission | |
Ni et al. | Bayesian selection of non-faulty sensors | |
Beverelli et al. | Domestic foundations of global value chains | |
CN104184601B (en) | The acquisition methods and device of user's online hours | |
CN105873129B (en) | The sensor network missing values reconstructing method of multi-node collaboration | |
CN108764348A (en) | Collecting method based on multiple data sources and system | |
Hori et al. | Inflation expectations of Japanese households: Micro evidence from a Consumer Confidence Survey | |
CN114219328A (en) | Method and device for determining inter-enterprise relationship | |
CN104468715B (en) | A kind of manufacturing industry Internet of things node collaboration storage method | |
CN109360111A (en) | Questionnaire data modification method, device, computer equipment and storage medium | |
CN110310476B (en) | Road congestion degree evaluation method and device, computer equipment and storage medium | |
CN102082737A (en) | Method for replacing Web services based on service priority | |
Adeloye | The relative utility of regression and artificial neural networks models for rapidly predicting the capacity of water supply reservoirs |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20181218 |
|
CF01 | Termination of patent right due to non-payment of annual fee |