CN102932804A

CN102932804A - Highly reliable covering mechanism based on trust management in wireless sensor network

Info

Publication number: CN102932804A
Application number: CN2012104875855A
Authority: CN
Inventors: 李小龙; 刘洋; 冯东磊; 李宏周
Original assignee: Guilin University of Electronic Technology
Current assignee: Guilin University of Electronic Technology
Priority date: 2012-11-27
Filing date: 2012-11-27
Publication date: 2013-02-13
Anticipated expiration: 2032-11-27
Also published as: CN102932804B

Abstract

The invention discloses a highly reliable covering mechanism based on trust management in wireless sensor network. The highly reliable covering mechanism comprises a universal trust management system and a covering algorithmic dispatching program based on trust management, wherein the covering area of sensor network is virtually subdivided, the intersections of vertical and horizontal coordinates serve as mesh points, the relatively smaller value of Rc and Rs is used as the size of a virtual mesh, Rs represents a node sensing semi-diameter, the mesh points are covered in parallel by active nodes in a circle taking the mesh points as the centre of the circle and taking node communication semi-diameter Rc as semi-diameter, the quantized data of the universal trust management system is taken as node trust degree, a joint trust degree is obtained through quantification, the working condition of working nodes in each mesh is dispatched according to the node trust degree and the joint trust degree, and the covering quality and covering reliability of covering area are ensured. The highly reliable covering mechanism is not limited to a certain specific trust management system, and therefore, the mechanism has better universality.

Description

Highly reliable covering based on trust management in a kind of wireless sensor network is machine-processed

Technical field

The present invention relates to wireless communication field, be specifically related to the highly reliable covering mechanism of wireless sensor network.

Background technology

Along with the in recent years progress of science and technology, wireless senser is gradually to the development of the aspects such as miniaturization, high-effect, low-power consumption, and can realize the batch production of high performance-price ratio in radio communication field is used.Being accompanied by wireless sensor network WSN (Wireless Sensor Network) more and more uses in fields such as battlefield monitoring, environment and Traffic monitorings ^[1], in the urgent need under the condition that is subject to resource constraint, WSN fragility inevitable and attack and destruction objective reality, WSN can provide reliable service quality.But up to now, seldom there is the complete credible WSN system can the reliable network service of effective guarantee, greatly retrained the development of WSN ^[2-3]For conserve energy, researchers begin one's study and add the function of Coverage Control in traditional transducers, in order to use few node of trying one's best to finish the covering requirement.Be vulnerable to attack and cause node institute overlay area unsafe problems, researchers to propose again the concept of trust management in order to solve WSN, with the fail safe in ensuring coverage zone.At present, based on the Secure Routing Techniques of trust management ^[4,17], safe election of cluster head ^[18], safe integration technology ^[5], secure data processes ^[20], the secure localization technology ^[6,19]With the safety time simultaneous techniques ^[7]Deng some valuable achievements in research having occurred.But till now, also the safety covering based on trust management that can imitate of row is not machine-processed.Traditional covering algorithm or trust management model are two different research directions, the two are not combined research.Highly reliable soverlay technique is one of key issue of high credible WSN, and WSN is by the task of internodal mutual cooperation execution monitored area, target following, but the prerequisite that cooperates and basis are that the participation node is normal node, but not malicious node.In order to improve the reliability of network coverage quality, masked nodes lost efficacy covering quality was impacted, and researcher's research and design covers mechanism reliably, has obtained certain achievement in research.

Document [8] passes through additionally to increase the way of some monitoring nodes in network, come the behavior of monitoring node packet discard, helps failure node forwarding data bag, supports sensing data finally to arrive safely, exactly aggregation node.Document [9] has proposed a kind of reliable covering mechanism based on 2-Coverage, shields the inefficacy of individual node by increasing redundant cover, reaches fault-tolerant effect.Lost efficacy for fear of individual node and to cause network node integrated moving, document [10] to propose " Snake " node replacement scheme based on virtual grid.Document [16] Based on Probability overlay model has proposed to satisfy the heuritic approach of the default monitoring probability of target area, covers collection to obtain maximum satisfying the demands.More than these researchs reliable cover mechanism and mainly concentrate on to solve and reduce node failure for the impact of covering quality.Yet in real world applications, except node failure caused network coverage Quality Down, network attack etc. all can impact covering quality.For example, if the monitoring node in the network is captured by the enemy, become malicious node, the auxiliary intensification mechanism of monitoring node will lose effectiveness.For " Snake " node replacement scheme, if the individual nodes in the network or part of nodes are aging, the sensing data substantial deviation is in real data, these false data will cause false alarm, the interference user decision-making consumes limited Internet resources, causes serious consequence.

The proposition of trust management model is that traditional the effective of security mechanism replenishes, and has been applied to the every aspect in the Internet network security.For above-mentioned reasons, the application of trust management model has become current study hotspot.Current, general trust management model has a variety of, the RFSN that is wherein proposed by Ganeriwal-Srivastava ^[12]It is exactly a kind of typical general trust management system based on popularity.The higher energy and the storage overhead that cause in order to reduce trust management, R.A.Shaikh etc. propose a kind of packet-based lightweight trust management system ^[13], J.W.Ho has proposed the trust management system based on region unit ^[14]

Document [11] is although proposed a kind of coverage scheduling scheme based on trust management, but this scheduling scheme just simply joins trust management model in the node, only incredible node revocation, and its covering algorithm with reality is not combined during concrete scheduling.This scheduling scheme can not guarantee the covering quality of node institute overlay area, in case node sudden failure in the course of the work, its overlay area can be absorbed in partial paralysis.

Summary of the invention

The present invention studies on the basis to above document, has proposed a kind of highly reliable covering mechanism based on trust management.From document [11] based on the covering algorithm of trust management different be, emphasis on analyzing the inherent requirement of trust management system to nodes ' behavior monitoring, quantitative analysis the relation between overlay node collection size, associating degree of belief and the trust threshold, this be before the content that do not relate to of other security protocol and existing covering algorithm based on trust management.Highly reliable covering mechanism based on trust management of the present invention can be guaranteed the highly reliable covering of overlay area, can effectively reduce in running order number of nodes again.And be not limited to certain specific trust management system, therefore, have better versatility.

With the virtual subdivision in sensor network overlay area, the coordinate infall is grid point in length and breadth, will

Rc with

In less value as the size of virtual grid, Rc is the node communication radius, Rs is the node perceived radius, by take grid point as the center of circle, live-vertex in take node communication radius R c as the circle of the radius node of node trust threshold (degree of belief surpass) is to the grid point covering that walks abreast, take the quantized data of general trust management system as the node confidence initial data, through quantizing to obtain the associating degree of belief, the highly reliable mechanism that covers based on trust management of the present invention is united the operating state that degree of belief is dispatched working node in each grid according to this.

For convenience, defined the following symbol: N={n ₁, n ₂..., n _h}: expression forms the sensor node collection of sensor network, { r ₁, r ₂..., r _hIts corresponding trust value, h represents the node number.S _iExpression is positioned at grid point p _iAdjacent mesh in, participate in nerve of a covering lattice point p _iThe overlay node collection of task, k ₀Maximum node number (the k that expression overlay node collection allows ₀Be system parameters, maximum is 7), k represents the node number of current overlay node collection. Node confidence threshold value when expression overlay node number is k, and

The node that degree of belief is lower than this threshold value is judged as malicious node,

Slightly write as Tmin.Tmax represents the associating degree of belief threshold value of grid point, is system parameters, if be lower than quality and fail safe that this threshold value then can not the ensuring coverage zone after the grid point degree of belief quantizes.T _iExpression grid point p _iCorresponding overlay node collection S _iThe associating degree of belief that obtains, wherein

T _i(k) expression grid point p _iThe associating degree of belief of the node number that corresponding overlay node integrates during as k.Among Fig. 1, Na is expressed as malicious node (hostile node), and the degree of belief of node is lower than Tmin.Nb is expressed as dormancy node (sleeping node), although the degree of belief of node is higher than the node confidence threshold value, finally is judged as redundant node through dispatching algorithm.Nc is expressed as working node (working node), and the degree of belief of node is higher than Tmin, and through the active node of dispatching algorithm.Rc: node communication radius.Rs: node perceived radius.

Provide in addition the term definition of content of the present invention:

Define 1. grid point p _iCorresponding overlay node collection S _iIn have working node over half normal operation to obtain perception data probability be called the associating degree of belief T of this grid _i

If definition 2. grid point p _iCorresponding overlay node collection S _iThe associating degree of belief T that obtains _iMore than or equal to the value of Tmax, then claim p _iBy highly reliable covering.

If each grid point is then claimed this zone by highly reliable covering by highly reliable covering in definition 3. target areas.

Definition 4. is with grid point p _iCentered by, be in the adjacent mesh or frame on sensor node be called p _iInterdependent node, p _iThe associated section point set we use Ω _iRepresent.According to the virtual grid size that arranges, p _iInterdependent node all can cover p _iThe basic process of node associating nerve of a covering lattice point as shown in Figure 1, each grid point selects several working nodes as the overlay node collection in adjacent mesh, and this grid point is united execution covering task.If current nerve of a covering lattice point, greater than trust threshold

The associating degree of belief that obtains of overlay node collection be higher than predetermined value Tmax, illustrate that this grid point has reached the requirement of highly reliable covering.For grid point p _i, corresponding overlay node collection S _i={ n ₁, n ₂, n ₃, n ₄Because degree of belief is lower than

Then be judged to be broken into malicious node, therefore be excluded and uniting covering p _iTask outside, n ₅Although be normal working node, also can cover p _i, but or owing to can not help to increase p _iAssociating degree of belief T _i, or because the node number k of overlay node collection has arrived maximum k ₀(k ₀Be system parameters), also be rejected to outside the overlay node collection.If all grid points are all by highly reliable covering, then covering task is finished.If have the associating degree of belief of grid point to be lower than threshold value Tmax, then need to re-start scheduling.

In order to make grid point can have better fail safe, requiring overlay node to concentrate has working node normal operation over half to obtain the probability of perception data greater than grid point associating degree of belief threshold value Tmax, can guarantee that like this through after (merging such as the numerical value) relevant treatment such as data fusion, data have high credible.For conserve energy, participate in to cover this grid point the working node number, be that the overlay node collection is unsuitable excessive, should be limited in k ₀Individual with interior (k ₀Be system parameters).

Too much for fear of the working node number, if k ₀The probability that the individual node that surpasses or equal trust threshold Tmin satisfies the working node normal operation that surpasses half reaches the standard of associating degree of belief Tmax, just working node can be limited in k ₀In individual.The problems referred to above can be formulated as minimum Tmin:

\{\begin{matrix} x^{k} + C_{k}^{1} x^{k - 1} {(1 - x)}^{1} + L C_{k}^{\frac{k - 1}{2}} x^{k \frac{k - 1}{2}} {(1 - x)}^{\frac{k - 1}{2}} = T_{\max} \\ 0 < x < 1 \\ x &GreaterEqual; T_{\min} \end{matrix} - - - (1)

K is odd number, and x is the node confidence value, works as Tmax=0.95, k ₀=5, the threshold value Tmin that obtains can not be less than 0.8.

When x 〉=Tmin＞0.5, k is during greater than 1 odd number, Be increasing function.Therefore can obtain k ₀The individual node that surpasses or equal trust threshold Tmin satisfies the probability of the working node normal operation that surpasses half more than or equal to Tmax.In like manner can obtain Tmax=0.9, k ₀=5, Tmin=0.75.Work as Tmax=0.95, k ₀=7, Tmin=0.77.Tmax＝0.9，k ₀=7，Tmin＝0.72。

X coordinate among Fig. 2 is the number k of working node, and the Y coordinate is the value of the Tmin that chooses, and the Z coordinate is that the working node number is that k namely is satisfied with the poor of probability corresponding when being k+1 above half the probability of working node normal operation with the node number.If be in Tmin (0＜Tmin＜0.8), in this scope of k≤7, when the trust value of all working node is Tmin, when the working node number is k, the probability the when probability that satisfies the working node normal operation that surpasses half is k+1 greater than the node number, order That is:

x^{k} + C_{k}^{1} c^{k - 1} {(1 - x)}^{1} + L + C_{k}^{\frac{k - 1}{2}} x^{k - \frac{k - 1}{2}} {(1 - x)}^{\frac{k - 1}{2}}

(2)

> x^{k + 1} + C_{k + 1}^{1} x^{k} {(1 - x)}^{1} + L + C_{k + 1}^{k_{1}} x^{k + 1 - k_{1}} {(1 - x)}^{k_{1}}

Be no more than k in the node number ₀Situation under, the working node that participates in task should be odd number, if the associating trust value that the work at present set of node obtains less than Tmax, each node number that increases increases progressively with 2.

Grid point p _iThe overlay node collection

Corresponding trust value be r ' ₁, r ' ₂..., r ' _k, r ' wherein ₁〉=r ' ₂〉=r ' _kWherein k represents the size of overlay node collection, W _KoRepresent just to have in this k the different nodes set of all combinations of o node perceived error in data,

Represent j combination, wherein

For example,

The node that represents the perception data mistake has n ' ₁, n ' ₂..., n ' _oGrid point p then _iAssociating degree of belief Mathematical Modeling be:

\{\begin{matrix} T_{i} = Σ_{o = 1}^{\frac{k - 1}{2}} {Σ_{j = 1}^{C_{k}^{o}} [Π_{n_{l}^{'} &Element; W_{k, o}^{j}} (1 - r_{l}^{'}) Π_{n_{l}^{'} &NotElement; w_{k, o}^{j}} r_{l}^{'}]} + Σ_{l = 1}^{k} r_{l}^{'} \\ &ForAll; r_{l}^{'} &GreaterEqual; T_{\min}^{k} \end{matrix};

Exist several active nodes in the covering algorithm scheduling scheme requirement grid based on trust management and monitoring the target area, and allow all the other nodes enter resting state, reach the effect that not only keeps covering quality but also can prolong network lifetime.At this time is divided with timeslice, be divided into again node scheduling stage and working stage in each timeslice, dispatch by said method in the scheduling phase interior nodes, judge whether to participate in the covering task.If participate in, then remain on active state at working stage always, otherwise enter into resting state until next timeslice is arrived.

Convenient for afterwards statement, node is divided into four classes: a class is to belong to and only belong to S _iIn set of node, use Expression; One class is to belong to but not only belong to S _iIn set of node, use Expression; One class is not belong to S _iBut belong to S _y(other arbitrary grid point P of expression node place grid _yThe overlay node collection) set of node, use

Expression; One class is not belong to S _iBut do not belong to any S yet _ySet of node, use `S _iExpression.Order

{\hat{Ω}}_{i} = {\hat{S}}_{i} U S_{i}^{%} = {{\hat{n}}_{1}, {\hat{n}}_{2}, . . ., {\hat{n}}_{l}},

The trust value of their correspondences is And

{\hat{r}}_{1} &GreaterEqual; {\hat{r}}_{2} &GreaterEqual; L &GreaterEqual; {\hat{r}}_{l} .

Order

{\overset{&OverBar;}{S}}_{i} = Ω_{i} - {\hat{Ω}}_{i} = {{\overset{&OverBar;}{n}}_{1}, {\overset{&OverBar;}{n}}_{2}, . . ., {\overset{&OverBar;}{n}}_{f}},

Wherein the trust value of their correspondences is

And

{\overset{&OverBar;}{r}}_{1} &GreaterEqual; {\overset{&OverBar;}{r}}_{2} &GreaterEqual; L &GreaterEqual; {\overset{&OverBar;}{r}}_{f} .

Make Ω _i=n ' ₁, n ' ₂..., n ' _L+f, corresponding trust value be r ' ₁, r ' ₂..., r ' _L+f, r ' wherein ₁〉=r ' ₂〉=r ' _L+fObviously, if some node n _jParticipate in the grid point set W of covering _Nj={ p _i, then expression If W _Nj={ p _i, * }, * number expression quantity is greater than 1 grid point set, then expression

If obviously Then be illustrated in working stage node n _jDo not participate in the task of any nerve of a covering lattice point, it will adjust to resting state.

Performing step based on the covering algorithm scheduling scheme of trust management is as follows:

Initialization. in the node scheduling stage, according to system parameters k ₀, T _Max, calculate

L,

Value.With Ω _i,

With

In node be arranged in order according to the size order from big to small of trust value.

Step 1. makes k=1, if Ω _iIn, r ' ₁＜T _Max, then execution in step 2.Otherwise, judge whether Then allow if set up

Carry out separately and cover p _iTask, namely

In other node and

In node all no longer participate in nerve of a covering lattice point p _iTask.If be false, then order

In node and

In node all no longer participate in nerve of a covering lattice point p _iTask.

Step 2. makes k=3, with Ω _i,

With

Middle trust value less than

Node in step 2, temporarily reject.Judge Ω _iIn whether have k or k above node, if do not have, then execution in step 3.If have, then the decision node collection n ' ₁, n ' ₂, n ' ₃Whether the associating trust value that obtains more than or equal to T _MaxIf, be false, then execution in step 3.If set up, judge n ' ₁, n ' ₂, n ' ₃In which node belong to

Then at first attempt the k set of node

Judge that whether the associating trust value that obtains is more than or equal to T _MaxIf set up, then

If be false, then will attempt successively and will belong in the current k set of node

The node of trust value minimum, with do not belong to the k set of node,

In the node replacement of trust value maximum, until satisfy the associating trust value that obtains more than or equal to T _MaxTill, if current k set of node equal n ' ₁, n ' ₂, n ' ₃, then finish.For example, if

Then attempt successively

{{\hat{n}}_{1}, {\hat{n}}_{2}, {\overset{&OverBar;}{n}}_{1}} = {n_{1}^{'},

n_{3}^{'}, n_{2}^{'}};

If only

Then attempt successively

{{\hat{n}}_{1}, {\overset{&OverBar;}{n}}_{2}, {\overset{&OverBar;}{n}}_{1}} = {n_{1}^{'}, n_{3}^{'}, n_{2}^{'}} .

Step 3. makes k=5＜k ₀, carry out according to the mode of step 2.

...

Step

Work as k=k ₀, with Ω _i,

With

Middle trust value is less than T _MinNode temporarily reject in this step.Judge Ω _iIn whether have k ₀Individual or k ₀Individual above node if do not have, then shows p _iCan not realize highly reliable covering.If have, by to the definition of trust model as can be known, set of node n ' ₁, n ' ₂..., n ' _K0The associating trust value that obtains is necessarily more than or equal to T _MaxWhich node belongs in then judging in this set of node

Then at first attempt

In all node with Those nodes of trust value maximum jointly form k ₀Set of node judges that whether the associating trust value that obtains is more than or equal to T _MaxIf, set up, then obtain S _iIf, be false, then will attempt successively current k ₀Belong in the set of node

The node of trust value minimum, and do not belong to k ₀Set of node,

In the node replacement of trust value maximum, until satisfy the associating trust value that obtains more than or equal to T _MaxTill, if current k ₀Set of node equal n ' ₁, n ' ₂..., n ' _K0, then finish.

Step

Readjust Ω _iIn each node participate in the grid point set that covers, the grid point set according to each node regenerates

And `S _i

Description of drawings

Fig. 1 is the schematic diagram of node associating nerve of a covering lattice point.

Fig. 2 is when k is odd number, T _i(k+1)-T _i(k) with the graph of a relation of k, Tmin

Embodiment

The performing step of the covering algorithm scheduling scheme based on trust management of the present invention is as follows:

L,

Value.With Ω _i,

With

Step 1. makes k=1, if Ω _iIn, r ' ₁＜T _Max, then execution in step 2.Otherwise, judge whether

Then allow if set up

Carry out separately and cover p _iTask, namely

In other node and

In node and

Step 2. makes k=3, with Ω _i,

With

Middle trust value less than

Then at first attempt the k set of node

The node of trust value minimum, with do not belong to the k set of node, In the node replacement of trust value maximum, until satisfy the associating trust value that obtains more than or equal to T _MaxTill, if current k set of node equal n ' ₁, n ' ₂, n ' ₃, then finish.For example, if

Then attempt successively

{{\hat{n}}_{1}, {\hat{n}}_{2}, {\overset{&OverBar;}{n}}_{1}} = {n_{1}^{'},

n_{3}^{'}, n_{2}^{'}};

If only

Then attempt successively

{{\hat{n}}_{1}, {\overset{&OverBar;}{n}}_{2}, {\overset{&OverBar;}{n}}_{1}} = {n_{1}^{'}, n_{3}^{'}, n_{2}^{'}} .

Step 3. makes k=5＜k ₀, carry out according to the mode of step 2.

...

Step

Work as k=k ₀, with Ω _i,

With

Middle trust value is less than T _MinNode temporarily reject in this step.Judge Ω _iIn whether have k ₀Individual or k ₀Individual above node if do not have, then shows p _iCan not realize highly reliable covering.If have, set of node n ' ₁, n ' ₂..., n ' _K0The associating trust value that obtains is necessarily more than or equal to T _MaxJudge then which node belongs in this set of node Then at first attempt In all node with

Those nodes of trust value maximum jointly form k ₀Set of node judges that whether the associating trust value that obtains is more than or equal to T _MaxIf, set up, then obtain S _iIf, be false, then will attempt successively current k ₀Belong in the set of node

The node of trust value minimum, and do not belong to k ₀Set of node,

Step

And `S _i

List of references

[1] Tang Qiuling, willow is blue or green, Qin Tuanfa, Zhang Shuyi. the energy-conservation modulation scheme of PPM in the wireless sensor network. and Chinese science E collects, and 2007,37 (12): 1583-1596.

[2] woods is rushed, Peng Xuehai. trustable network research. and Chinese journal of computers, 2005,28 (5): 751-758

[3]S.Ganeriwal,B.Srivastava.Trustworthy?sensor?networks:Issues,Challenges&Solutions.NESLTechnical?Report,2004.http://www.cc.ucla.edu/～saurabh/publications/tech-report-integrity.pdf

[4]S.Tanachaiwiwat,P.Dave,R.Bhindwale,A.Helmy.Secure?locations:routing?on?trust?and?isolating?compromised?sensors?inlocation-aware?sensor?networks.In:Proc.of?the?SenSys?2003.New?York：ACM?Press,2003.324-325.

[5]M.D.Krasniewski,P.Varadharajan,B.Rabeler，S.Bagchi,Y.C.Hu.TIBFIT:Trust?index?based?fault?tolerance?for?ability?datafaults?in?sensor.In:Werner?B,ed.Proc.of?the?Int'l?Conf.on?Dependable?Systems?and?Networks(DSN).Piscataway：IEEEComputerSociety，2005.672-681.

[6]Y.Zeng,J.Cao,J.Hong,L.Xie.Secure?localization?and?location?verification?inwireless?sensor?networks.MASS2009,864-869.

[7]Y.Yang,Y.Sun.Securing?Time-Synchronization?Protocols?in?Sensor?Networks：Attack?Detection?and?Self-Healing.GlobeCom2008,1-6.

[8]J.Wang,S.Medidi.Topology?control?for?reliable?sensor-to-sink?data?transport?in?sensor?networks.In：Proceedings?of?ICC,2008.3215-3219.

[9] Xu Qiang, Wang Yun. the solution of reliable covering problem among the fault-tolerant energy-saving wireless WSN. Journal of Software, 2006,17 (11): 184-191.

[10]A.Howard,M.Mataric,G.Sukhatme.An?incremental?self-deployment?algorithm?for?mobile?sensor?networks.AutonomousRobots,2002,13(2):113-126.

[11]Z.Yin,H.Zhao,K.Lin,P.Sun,D.Luo,X.Zhang,X.Wang.A?coverage-preserving?node?scheduling?scheme?based?on?trustselection?model?in?wireless?sensor?networks.Proceeding?of?ISPCA2006,2007,8:696-698.

[12]Ganeriwal?S,Srivastava?M.Reputation-based?framework?for?high?integrity?sensor?networks.In：Proc.of?the?2nd?ACMWorkshop?on?Security?of?Ad?Hoc?and?Sensor?Network(SASN?2004).New?York：ACM?press,2004.66-77.

[13]R.A.Shaikh,H.Jameel,B.J.Auriol,etc.Group-Based?Trust?Management?Scheme?for?Clustered?Wireless?Sensor?Networks.IEEE?Transactions?on?Parallel?and?Distributed?Systems.2009,20(11):1698–1712.

[14]J.W.Ho.Zone-based?tust?management?in?sensor?networks.PerCom?2009,1-2.

[15]Tian?D,Georganas?ND.A?node?scheduling?scheme?for?energy?conservation?in?large?wireless?sensor?networks.WirelessCommunications?and?Mobile?Computing,2003,3(2):271-290.

[16]He?J.,Ji?S.,Pan?Y.,et?al.Reliable?and?Energy?Efficient?Target?Coverage?for?Wireless?Sensor?Networks.2011,16(5)：464-474.

[17] Wang good citizen, Guo Yuanbo, Zhan Yongzhao. inbreak-tolerated wireless sensor network blurs Trust Valuation Model. communication journal, 2010,31 (12): 37-44.

[18] Feng Renjian, Cheng Jian, Xu Xiaofeng, Wan Jiangwen. based on the credible election of cluster head algorithm of the wireless sensor network of Mamdani fuzzy reasoning. high-tech communication, 2010,20 (12): 1253-1259.

[19]X.Wang,L.Ding,S.Wang.Trust?Evaluation?Sensing?for?Wireless?Sensor?Networks.IEEE?transaction?on?instrumentation?andmeasurement,2011,60(6):2088-2095.

[20]H.Feng,G.Li,W.Lu,L.Huang.Trust?based?secure?in-network?data?processing?schema?in?wireless?sensor?networkks.Journal?ofnetworks,2011,6(2):295-302.

Claims

In the wireless sensor network based on the highly reliable covering mechanism of trust management, comprise general trust management system, it is characterized in that: also comprise the covering algorithm scheduling scheme based on trust management, with the virtual subdivision in sensor network overlay area, the coordinate infall is grid point in length and breadth, will
With In less value as the size of virtual grid, Rs is the node perceived radius, by take grid point as the center of circle, live-vertex in take node communication radius R c as the circle of radius is to the grid point covering that walks abreast, take the quantized data of general trust management system as node confidence, through quantizing to obtain the associating degree of belief, described operating state of dispatching working node in each grid according to node confidence and associating degree of belief based on the covering algorithm scheduling scheme of trust management.
2. highly reliable covering based on trust management is machine-processed in a kind of wireless sensor network as claimed in claim 1, it is characterized in that: described

Live-vertex be the node that degree of belief surpasses the node confidence threshold value, this node confidence threshold value is obtained by following mathematical expression:

$\{\begin{matrix} x^{k} + C_{k}^{1} x^{k - 1} {(1 - x)}^{1} + L C_{k}^{\frac{k - 1}{2}} x^{k \frac{k - 1}{2}} {(1 - x)}^{\frac{k - 1}{2}} = T_{\max} \\ 0 < x < 1 \\ x &GreaterEqual; T_{\min} \end{matrix};$

In the formula:

K is odd number;

Node trust threshold when being k0 for the overlay node number simply is expressed as Tmin;

Tmax represents the associating degree of belief threshold value of grid point, and this is worth by default.
3. highly reliable covering based on trust management is machine-processed in a kind of wireless sensor network as claimed in claim 1, and it is characterized in that: described associating degree of belief is obtained by following mathematical expression:

$\{\begin{matrix} T_{i} = Σ_{o = 1}^{\frac{k - 1}{2}} {Σ_{j = 1}^{C_{k}^{o}} [Π_{n_{l}^{'} &Element; W_{k, o}^{j}} (1 - r_{l}^{'}) Π_{n_{l}^{'} &NotElement; w_{k, o}^{j}} r_{l}^{'}]} + Σ_{l = 1}^{k} r_{l}^{'} \\ &ForAll; r_{l}^{'} &GreaterEqual; T_{\min}^{k} \end{matrix};$

In the formula:

T _iExpression grid point p _iCorresponding overlay node collection S _iThe associating degree of belief that obtains;

p _iThe expression grid point;

S _iExpression p _iCorresponding overlay node collection;

Corresponding trust value be r ' ₁, r ' ₂..., r ' _k, r ' wherein ₁〉=r ' ₂〉=r ' _k

K represents node number that current overlay node integrates and as odd number;

W _KoRepresent just to have in this k the different nodes set of all combinations of o node perceived error in data;

N represents to form the sensor node collection of sensor network;

Represent j combination, wherein
4. highly reliable covering based on trust management is machine-processed in a kind of wireless sensor network as claimed in claim 1, and it is characterized in that: described covering algorithm scheduling scheme based on trust management is realized by following steps:

Initialization. in the node scheduling stage, according to system parameters k ₀, T _Max, calculate
L,
Value, with Ω _i,
With
In node be arranged in order according to the size order from big to small of trust value;

Step 1. makes k=1, if Ω _iIn, r ' ₁＜T _Max, then execution in step 2, otherwise, judge whether Then allow if set up
Carry out separately and cover p _iTask, namely

In other node and
In node all no longer participate in nerve of a covering lattice point p _iTask, if be false, then the order

In node and
In node all no longer participate in nerve of a covering lattice point p _iTask;

Step 2. makes k=3, with Ω _i, With Middle trust value less than
Node in step 2, temporarily reject, judge Ω _iIn whether have k or k above node, if do not have, then execution in step 3.If have, then the decision node collection n ' ₁, n ' ₂, n ' ₃Whether the associating trust value that obtains more than or equal to T _MaxIf, be false, then execution in step 3.If set up, judge n ' ₁, n ' ₂, n ' ₃In which node belong to
Then at first attempt the k set of node
Judge that whether the associating trust value that obtains is more than or equal to T _MaxIf set up, then
If be false, then will attempt successively and will belong in the current k set of node The node of trust value minimum, with do not belong to the k set of node,
In the node replacement of trust value maximum, until satisfy the associating trust value that obtains more than or equal to T _MaxTill, if current k set of node equal n ' ₁, n ' ₂, n ' ₃, then finish;

Step 3. makes k=5＜k ₀, carry out according to the mode of step 2;

......

Step
Work as k=k ₀, with Ω _i,
With
Middle trust value is less than T _MinNode temporarily reject in this step; Judge Ω _iIn whether have k ₀Individual or k ₀Individual above node if do not have, then shows p _iCan not realize highly reliable covering; If have, by to the definition of trust model as can be known, set of node n ' ₁, n ' ₂..., n ' _K0The associating trust value that obtains is necessarily more than or equal to T _MaxWhich node belongs in then judging in this set of node
Then at first attempt
In all node with
Those nodes of trust value maximum jointly form k ₀Set of node judges that whether the associating trust value that obtains is more than or equal to T _MaxIf, set up, then obtain S _iIf, be false, then will attempt successively current k ₀Belong in the set of node
The node of trust value minimum, and do not belong to k ₀Set of node, In the node replacement of trust value maximum, until satisfy the associating trust value that obtains more than or equal to T _MaxTill, if current k ₀Set of node equal n ' ₁, n ' ₂..., n ' _K0, then finish;

Step
Readjust Ω _iIn each node participate in the grid point set that covers, the grid point set according to each node regenerates

And `S _i

Wherein:

N represents to form the sensor node collection of sensor network, N={n ₁, n ₂..., n _b; { r ₁, r ₂..., r _hTrust value corresponding to expression N, h represents the node number; K represents node number that current overlay node integrates and as odd number; k ₀The maximum node number that expression overlay node collection allows, k ₀By default;
Node trust threshold when being k for the overlay node number, and
Tmax represents the associating degree of belief threshold value of grid point, and this is worth by default;

p _iExpression grid point center; S _iExpression p _iCorresponding overlay node collection; Ω _iExpression p _iThe associated section point set;
Expression belongs to and only belongs to S _iIn set of node;
Expression belongs to but not only belongs to S _iIn set of node;
Expression does not belong to S _iBut belong to S _iSet of node, j ∈ { N-{i}}; `S _iExpression does not belong to S _iBut do not belong to any S yet _jSet of node;

${\hat{Ω}}_{i} = {\hat{S}}_{i} U S_{i}^{%} = {{\hat{n}}_{1}, {\hat{n}}_{2}, . . ., {\hat{n}}_{l}},$ Corresponding trust value is
And ${\hat{r}}_{1} &GreaterEqual; {\hat{r}}_{2} &GreaterEqual; L &GreaterEqual; {\hat{r}}_{l};$ ${\overset{&OverBar;}{S}}_{i} = Ω_{i} - {\hat{Ω}}_{i} = {{\overset{&OverBar;}{n}}_{1}, {\overset{&OverBar;}{n}}_{2}, . . ., {\overset{&OverBar;}{n}}_{f}},$ Corresponding trust value is
And ${\overset{&OverBar;}{r}}_{1} &GreaterEqual; {\overset{&OverBar;}{r}}_{2} &GreaterEqual; L &GreaterEqual; {\overset{&OverBar;}{r}}_{f};$ Ω _i=n ' ₁, n ' ₂..., n ' _L+f,, corresponding trust value be r ' ₁, r ' ₂..., r ' _L+f,, r ' wherein ₁〉=r ' ₂〉=r ' _L+f