CN103648123A - Diagnosis system deployment method based on multidimensional space evidence information - Google Patents
Diagnosis system deployment method based on multidimensional space evidence information Download PDFInfo
- Publication number
- CN103648123A CN103648123A CN201310686959.0A CN201310686959A CN103648123A CN 103648123 A CN103648123 A CN 103648123A CN 201310686959 A CN201310686959 A CN 201310686959A CN 103648123 A CN103648123 A CN 103648123A
- Authority
- CN
- China
- Prior art keywords
- evidence
- node
- network
- expense
- collected
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a diagnosis system deployment method based on multidimensional space evidence information. The method comprises steps that: firstly, a multidimensional evidence set E = {E1, E2, .., Ek} ,|E|=k is projected to a unified data space Eg = {e1, e2, ..et}; an object set Cmotes = phi and a node candidate set Mnow = M are initiated; when each network diagnosis evidence is collected by at least one node, and the Eg is not equal to the phi, a most-demanded evidence etop belonging to the Eg is found out according to a cost index; in the node candidate set Mnow, a node which can collect the evidence etop and has the smallest cost is taken as the selected node m, and the selected node m is added to the object set Cmotes; collected evidence of the selected node m is deleted from the Eg; steps above are repeated till the Eg is empty; and the object set Cmotes is outputted. Through the method, feasible solution for diagnosis deployment can be acquired in high efficiency.
Description
Technical field
The present invention relates to wireless self-organization network and sensor network field,, especially proposed a kind of diagnostic system based on hyperspace evidence information and disposed design and concrete implementation method thereof, in real time wireless sensor network diagnosis being provided to deployment scheme.
Background technology
Increasingly mature along with communication, transducer manufacture, embedded calculating, large-scale wireless Sensor Network technology develops and is widely applied to the every aspect of people's life, the numerous areas such as environmental monitoring, national defence, health examination, traffic control, disaster relief and rescue, city management rapidly.A typical wireless sensor network forms (resource-constrained hardware device and the working method of self-organizing) by a large amount of cheap sensor nodes, they can be deployed rapidly in different environment, and by the wireless form networking from forming, can not be subject to the restriction of existing cable network infrastructure.The perception data of sensor node is pooled to data processing centre by radio communication in multi-hop relay mode.Wireless sensor network makes the people can be in the restriction that is not subject to time, place, and a large amount of accurate and reliable environmental informations of Real-time Obtaining make " calculating " can be ubiquitous.
Here we mainly pay close attention to the basic conception that diagnosis node is selected (DNS) problem.If a wireless sense network application faces the problem under performance or in not expected state, keeper need to dispose as the case may be a corresponding diagnostic system and solve problem.An optional lightweight technology is to utilize network reprogrammed that some normal nodes of network are changed into the node with diagnostic function.In other words, keeper has selected original normal node in some networks, and has changed their script role, makes it have diagnostic function, and these nodes are with the evidence in cooperative mode collection network, for diagnostic work below afterwards.
Does is yet key issue how the most effectively to carry out above-mentioned diagnosis node to select? a typical solution is to select problem model to turn to a covering problem (coverage problem) this diagnosis node.Specifically describe as follows: first we suppose the acquisition that has some diagnostic messages to be continued in network, and only with the node of minimum number, obtain these evidences.So this just means that keeper only has and use the node of minimum number could be the most effective, bantamweight set up a diagnostic system, and obtain network evidence.
Summary of the invention
The present invention in order to achieve the above object, provides a kind of diagnostic system dispositions method based on hyperspace evidence information.The technical solution used in the present invention is:
The set of network diagnosis evidence to be collected is E={E
1, E
2..., E
k, | E|=k, k is positive integer; This evidence set is distributed on k data dimension; Each network diagnosis evidence subclass E
kcomprise one group of network diagnosis evidence set { e that belongs to Same Latitude
1, e
2..., e
j, j is positive integer; Whole network M={m
1, m
2..., m
n, | M|=n comprises n node, and n is positive integer;
Making diagnosis node set is that goal set is C
motes, its summation can be collected the evidence set E needing;
(a). first by multidimensional evidence set E={E
1, E
2..., E
k, | E|=k projects to a unified data space E
g={ e
1, e
2..., e
t;
Making the expense summation of the network diagnosis evidence that all needs are collected is 100%, sets in advance original network diagnosis evidence expense set and is
t is positive integer; Each network diagnosis evidence expense is normalized into the value between 0 to 100, and concrete value meets following formula:
Obtaining a network diagnosis evidence expense set after standardization is:
(b). initialization C
motes=φ, node candidate collection M
now=M; φ is empty set;
(c). when each network diagnosis evidence is at least collected once by a node, and E
gduring ≠ φ, according to expense index, find one by the evidence e of demand
top∈ E
g;
(d). at node candidate collection M
nowin, the selection e that can collect evidence
topthat node of expense minimum as selecting node m, selected node m to join goal set C this
motesin;
(e). from E
gin leave out and selected the collected evidence of node m;
(f). repeating step (c)~(e), until E
gtill sky;
(g). export target set C
motes.
The invention has the advantages that the character of using for evidence hyperspace, give a standardisation process, utilize afterwards the optimal solution that obtains of heuritic approach iteration based on greedy.Such design has brought following advantage: can process the diagnostic evidence of hyperspace attribute, obtain the feasible solution that diagnosis is disposed efficiently.
Accompanying drawing explanation
Fig. 1 is that hyperspace diagnosis node of the present invention is selected problem-instance schematic diagram.
Embodiment
Below in conjunction with concrete drawings and Examples, the invention will be further described.
The present invention is based on forest and the application of Urban Large scale wireless sensor network needs the demand of a lightweight, the little diagnostic system deployment scheme for feature of operation expense to produce.Forest and Urban Large scale wireless sensor network are mainly collected diversified environment sensing data, such as: temperature, humidity, illumination and dense carbon dioxide degrees of data.Such application system needs administrative staff to use the scheme of a strong deployment diagnostic system, guarantees the variety of event that function and reply with back-up system run into.
For the wireless sense network system of a long-time running, keeper always faces some diagnostic tasks, such as: find low basic reason or the better understanding to measuring results of performance.In a wireless sensor network system (such as CitySee system) of reality, keeper mainly pays close attention to the network diagnosis evidence of 22 types.These network diagnosis evidences comprise: temperature, illumination, humidity, received signal strength (Received Signal Strength Indicator, RSSI), the number of transmissions of expectation (Expected Transmission number, ETX), father node variation and flow information etc.These network diagnosis evidences can be divided into 3 classes: the evidence based on probability (probability evidence), the evidence based on reasoning (inference evidence) and physical proof (physical evidence).Illustrate: in general, physical proof comprises temperature, humidity and illumination the index of this class description natural world physical characteristic.Be different from physical proof, the evidence based on probability is only paid close attention to those indexs of describing with probabilistic model.Real grid diagnostic evidence is always distributed in hyperspace thus.Diagnosis node selects problem just need to expand under the environment of hyperspace so.
First, provide the formal definition that hyperspace diagnosis node is selected problem (MDNS).The network diagnosis evidence set E={E that exists the needs of k latitude to be collected in assumed wireless sensor network
1, E
2..., E
k, | E|=k, k is positive integer; Each network diagnosis evidence subclass E wherein
kcomprise one group of network diagnosis evidence set { e that belongs to Same Latitude
1, e
2..., e
j, j is positive integer; Whole network M={m
1, m
2..., m
n, | M|=n comprises n node, and n is positive integer.
At evidence set E
1in, evidence e
m1can be by node m
1catch, its corresponding expense is c
m1.So a hyperspace diagnosis node selects problem (MDNS) to be described to, and from a network n node, selects the combination C of a group node, this group node can effectively be collected whole diagnostic evidence set E in hyperspace.That is to say, C is one group of U that satisfies condition on k information latitude
mi ∈ C{ e
mithe node set of }=E.We need to minimize the expense min that last diagnostic node is selected set
c∑
mi ∈ Cc
mi.
In a true application, node often has different sign abilities at different information data latitudes.In figure, 1 wireless sense network is deployed in the region of 500m*500m.The perception of a node humidity of latticed region representation characterizes scope, the sign ability of node channel link quality of region representation (based on Received Signal Strength Indicator, RSSI) of net-point shape.Therefore for a node, its ability of collecting diagnostic evidence, in different latitude, has different abilities.
In hyperspace diagnosis node, select in problem (MDNS), network diagnosis evidence is distributed in E on multidimensional information space
1, E
1..., E
kso first we need one of them latitude to solve this problem.Instinctively, we can be from for most important evidence latitude current diagnostic task.Yet the target of MDNS is to collect all effective evidences, and this just means that we can project to the network diagnosis evidence of multidimensional on a new unified latitude, and does not change the solution of primal problem.
Projection process part is based on keeper's Heuristics.First, we are machine-processed for the network diagnosis evidence that need to be collected defines corresponding expense assignment.Main idea of the present invention is the normal state standardisation process based on an evidence set.We suppose that the expense summation of the network diagnosis evidence that all needs are collected is 100%.Each network evidence needs to be composed a rational value in standardisation process.Original network diagnosis evidence expense set is set in advance by keeper
t is positive integer; In technical scheme of the present invention, each network diagnosis evidence expense is normalized to the value between 0 to 100, and concrete value meets following formula:
Network diagnosis evidence expense set after definition standardization is:
Use aforesaid way, we obtain a standardized network evidence expense.
Hyperspace diagnosis node selects the concrete solution of problem as follows:
The set of network diagnosis evidence to be collected is E={E
1, E
2..., E
k, | E|=k, k is positive integer; This evidence set is distributed on k data dimension; Each network diagnosis evidence subclass E
kcomprise one group of network diagnosis evidence set { e that belongs to Same Latitude
1, e
2..., e
j, j is positive integer; Whole network M={m
1, m
2..., m
n, | M|=n comprises n node, and n is positive integer;
Making diagnosis node set is that goal set is C
motes, its summation can be collected the evidence set E needing;
(a). first by multidimensional evidence set E={E
1, E
2..., E
k, | E|=k projects to a unified data space E
g={ e
1, e
2..., e
t;
Making the expense summation of the network diagnosis evidence that all needs are collected is 100%, sets in advance original network diagnosis evidence expense set and is
t is positive integer; Each network diagnosis evidence expense is normalized into the value between 0 to 100, and concrete value meets following formula:
Obtaining a network diagnosis evidence expense set after standardization is:
(b). initialization C
motes=φ, node candidate collection M
now=M; φ is empty set;
(c). when each network diagnosis evidence is at least collected once by a node, and E
gduring ≠ φ, according to expense index, find one by the evidence e of demand
top∈ E
g;
(d). at node candidate collection M
nowin, the selection e that can collect evidence
topthat node of expense minimum as selecting node m, selected node m to join goal set C this
motesin;
(e). from E
gin leave out and selected the collected evidence of node m;
(f). repeating step (c)~(e), until E
gtill sky;
(g). export target set C
motes, this set is the node set with minimal-overhead that can collection network diagnostic evidence set E.
Claims (1)
1. the diagnostic system dispositions method based on hyperspace evidence information, is characterized in that, comprising:
The set of network diagnosis evidence to be collected is E={E
1, E
2..., E
k, | E|=k, k is positive integer; This evidence set is distributed on k data dimension; Each network diagnosis evidence subclass E
kcomprise one group of network diagnosis evidence set { e that belongs to Same Latitude
1, e
2..., e
j, j is positive integer; Whole network M={m
1, m
2..., m
n, | M|=n comprises n node, and n is positive integer;
Making diagnosis node set is that goal set is C
motes, its summation can be collected the evidence set E needing;
(a). first by multidimensional evidence set E={E
1, E
2..., E
k, | E|=k projects to a unified data space E
g={ e
1, e
2..., e
t;
Making the expense summation of the network diagnosis evidence that all needs are collected is 100%, sets in advance original network diagnosis evidence expense set and is
t is positive integer; Each network diagnosis evidence expense is normalized into the value between 0 to 100, and concrete value meets following formula:
Obtaining a network diagnosis evidence expense set after standardization is:
(b). initialization C
motes=φ, node candidate collection M
now=M; φ is empty set;
(c). when each network diagnosis evidence is at least collected once by a node, and E
gduring ≠ φ, according to expense index, find one by the evidence e of demand
top∈ E
g;
(d). at node candidate collection M
nowin, the selection e that can collect evidence
topthat node of expense minimum as selecting node m, selected node m to join goal set C this
motesin;
(e). from E
gin leave out and selected the collected evidence of node m;
(f). repeating step (c)~(e), until E
gtill sky;
(g). export target set C
motes.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310686959.0A CN103648123A (en) | 2013-12-13 | 2013-12-13 | Diagnosis system deployment method based on multidimensional space evidence information |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310686959.0A CN103648123A (en) | 2013-12-13 | 2013-12-13 | Diagnosis system deployment method based on multidimensional space evidence information |
Publications (1)
Publication Number | Publication Date |
---|---|
CN103648123A true CN103648123A (en) | 2014-03-19 |
Family
ID=50253261
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310686959.0A Pending CN103648123A (en) | 2013-12-13 | 2013-12-13 | Diagnosis system deployment method based on multidimensional space evidence information |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103648123A (en) |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101556604A (en) * | 2009-05-06 | 2009-10-14 | 北京大学 | Method for automatically generating optimization strategy orientating complex data warehouse environment |
-
2013
- 2013-12-13 CN CN201310686959.0A patent/CN103648123A/en active Pending
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101556604A (en) * | 2009-05-06 | 2009-10-14 | 北京大学 | Method for automatically generating optimization strategy orientating complex data warehouse environment |
Non-Patent Citations (1)
Title |
---|
LIAN SHUO 等: "Near-Optimal Diagnosis System Deployment in Wireless Sensor Networks", 《INTERNATIONAL JOURNAL OF DISTRIBUTED SENSOR NETWORKS》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Giorgetti et al. | A robust wireless sensor network for landslide risk analysis: System design, deployment, and field testing | |
US8923882B2 (en) | Determining a deployment of a wireless network | |
Al Qundus et al. | Wireless sensor network for AI-based flood disaster detection | |
Paredes-Parra et al. | An alternative internet-of-things solution based on LOra for PV power plants: Data monitoring and management | |
Liu et al. | CitySee: Not only a wireless sensor network | |
Dohare et al. | Wireless communication and environment monitoring in underground coal mines–review | |
Yang et al. | Integration of wireless sensor networks in environmental monitoring cyber infrastructure | |
Arsiwala et al. | Digital twin with Machine learning for predictive monitoring of CO2 equivalent from existing buildings | |
CN106687773A (en) | System and methods for sensor node localization and sensor network organization based on contextual event detection | |
KR102191763B1 (en) | Scenario based Disaster Prediction System and the Method | |
CN103310388A (en) | Method for calculating composite index of grid operation based on information source entropy | |
CN102724681A (en) | Sensor network coverage hole detection method combining with energy efficiency | |
Zabasta et al. | Automatic wireless meters reading for water distribution network in Talsi city | |
Tundjungsari et al. | Android-based application using mobile adhoc network for search and rescue operation during disaster | |
Muhendra et al. | Development of digital water meter infrastructure using wireless sensor networks | |
Zrelli et al. | Collect Tree Protocol for SHM system using wireless sensor networks | |
CN105338661A (en) | Environment monitoring method and device taking cloud computing as configuration and employing data fusion calculation design | |
Arjannikov et al. | Using markov chains to model sensor network reliability | |
CN103648123A (en) | Diagnosis system deployment method based on multidimensional space evidence information | |
Walia et al. | The characteristics of smart energy information management systems for built environments | |
Vo | Online working condition monitoring system integrated power saving and security using zigbee wireless sensor network | |
Tang et al. | An evaluation on sensor network technologies for AMI associated mudslide warning system | |
CN110986378A (en) | Boiler safety monitoring platform | |
Fu et al. | Managing your trees: Insights from a metropolitan-scale low-power wireless network | |
Shitumbapo et al. | Simulation of a smart water metering network in Tsumeb East, Namibia |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20140319 |
|
WD01 | Invention patent application deemed withdrawn after publication |