CN104066180B - A kind of underwater sensing net node locating and update method based on time-varying error bounds - Google Patents

Abstract

The invention discloses a kind of underwater sensing net node locating and update method based on time-varying error bounds, belong to underwater wireless sensor network field of locating technology, comprise the following steps：Step one, node to be positioned build positioning group；Step 2, positioning group's on-line proving error bounds；Step 3, the set-member estimation feasible set for building node to be positioned；Step 4, rasterizing scanning feasible set, the coordinate for asking for node to be positioned are estimated；Step 5, positioning node perform the collection person's adaptive location based on time-varying error bounds and update on demand in new locating periodically.The present invention is by the on-line proving to error bounds in positioning group, reduce multi-source and do not know adverse effect of the noise to multi-hop distance estimations, improve the accuracy and reliability of position coordinates estimation, updated by performing the collection person's adaptive location based on time-varying error bounds on demand, reduce energy ezpenditure, the ageing of positioning is enhanced, the positioning performance of underwater sensing net is improve.

Description

A kind of underwater sensing net node locating and update method based on time-varying error bounds

Technical field

It is extensive under particularly complicated marine environment the invention belongs to underwater wireless sensor network field of locating technology Underwater wireless sensor network node is self-positioning, specially a kind of underwater sensing net node locating based on time-varying error bounds and more New method.

Background technology

Extensive random placement task such as execution information perception, collection, treatment and transmission in complicated marine environment is water Typical case's application of lower wireless sensor network (Underwater Wireless Sensor Networks), such as naval battle field situation Perception, marine environmental monitoring, submarine target locating and tracking etc..Node self-localization is the critical support technology of underwater sensing net, more It is the premise of such application, because the positional information that perceived information is typically required for node is corresponding with can just embody its Meaning.

Node to be positioned needs to complete to make position/task by oneself by corresponding reference information, generally includes the seat of anchor node Mark, distance of node to be positioned to anchor node etc..To based on range finding location algorithm for, accurate distance estimations be undoubtedly into The basis of row high-precision coordinate estimation.But the node range finding under complicated marine environment is inevitably dry by many source noises Disturb, such as ambient noise, beacon position error, multipath effect, multi-hop distance estimations error, the prior information of these noises is often It is insufficient and unascertained, because being limited by complicated applications environment, we are difficult to get many source noises and are accurately distributed rule Rule and characteristic parameter.Most scholars directly assume that multi-source does not know noise Gaussian distributed even standardized normal distribution, Then the error processing method of classics is taken to improve the precision of positioning, it is clear that this kind of way be it is not rigorous be also irrational. Separately there are a small number of scholars to employ set-member estimation method (Set Membership Estimation) and do not know noise processing multi-source Problem.Set-member estimation method is a kind of method of estimation assumed based on unknown but bounded (Unknown But Bounded) noise, The advantage of this method is the statistical property such as distribution, average or variance for requiring no knowledge about noise, only requires system noise Bounded and noise circle (or error bounds) are known, it is possible to find one by all having with metric data, model structure and noise Assume the set of compatible state or parameter composition in boundary, you can row collection (Feasible Solution Set).But in complicated ocean In environmental applications, the actual effect of this kind of method is also not satisfactory, because error bounds can be with the dynamic change of marine environment Produce drift, it is assumed that it is known and to immobilize be unreasonable also unpractical, if the feelings not changed over time to error bounds Condition is adopted an effective measure, and feasible set can be caused empty set occur, has a strong impact on positioning performance.

Additionally, after the completion of node self-localization, the height of updating location information mechanism efficiency will also influence whether whole system Performance.Traditional location information updating method is that node is periodically positioned, it means that Sensor Network needs to carry out Position estimation and information updating in global scope, not only increase node traffic, consume big energy, and it is fixed to also reduce Position it is ageing, this is that those applications (such as location tracking system) higher to updating location information efficiency requirements to be received 's.

Therefore, it is necessary to explore one kind can effectively overcome multi-source do not know noise and time-varying error bounds influence and with compared with The underwater sensing net node self-localization new method of strong online cognitive and dynamic self-adapting ability.

The content of the invention

It is an object of the invention to provide a kind of underwater sensing net node locating and update method based on time-varying error bounds, solution Exist under certainly complicated marine environment multi-source do not know the underwater wireless sensor network node of noise and the influence of time-varying error bounds from Orientation problem；By the on-line proving to error bounds in positioning group, reduce multi-source and do not know noise to multi-hop distance estimations Adverse effect, improves the accuracy and reliability of position coordinates estimation, by performing the collection person based on time-varying error bounds on demand Adaptive location updates, and reduces energy ezpenditure, enhances the ageing of positioning, improves the positioning performance of underwater sensing net.

The present invention proposes a kind of underwater sensing net node locating and update method based on time-varying error bounds, specifically include with Lower step：

Step one, node to be positioned build positioning group；

(1) all anchor nodes broadcast one group of " positioning comprising information such as self ID, coordinate, multi-hop threshold value, locating periodicallies Bag "；One group of all node broadcasts " range finding bag ", by information exchange, obtain the ID of all neighbor nodes in itself communication context, The information such as coordinate, distance；

(2) node statistics to be positioned itself skip to the anchor node quantity in multi-hop threshold range, if reach coordinate estimating The minimum quantity of calculation, then include positioning group, and go to step two by the common intermediate node that all anchor nodes and participation information are forwarded； Minimum quantity such as anchor node sum less than coordinate estimation, then labeled as wouldn't meet location condition node, and wait next Locating periodically；

Step 2, positioning group's on-line proving error bounds；

(1) node to be positioned sends " on-line proving " command frame to all anchor nodes in positioning group, and this command frame contains anchor Node repeats the maximum allowable frequency n of range finding；

(2) any two anchor node N in group_iWith N_jRange finding n times is repeated, the original range finding sample set that sample number is n is set up

(3) by the method for nonparametric sampling with replacement from original range finding sample set D_ijExtract the self-service sample that capacity is similarly n

(4) B self-service sample is independently extracted in succession, is obtainedWherein, b=1, 2,…,B；Then each self-service sample is averaged：

(5) anchor node N is estimated_iWith N_jBetween measurement distance standard deviation

Wherein, X_iIt is anchor node N_iCoordinate, X_jIt is anchor node N_jCoordinate；

(6) all anchor node combination of two in group, (2)~(5) in repeat step two, between trying to achieve any two anchor node Measurement distance standard deviation, and extract maximum therein

(7) demarcateIt is the error bounds of positioning group's range measurement noise；

Step 3, structure node N to be positioned_aSet-member estimation feasible set S_a(x)：

Wherein, u is the number for positioning anchor node in group, R₀It is the communication radius of node, d '_iaIt is anchor node N_iWith it is to be positioned Node N_aBetween measurement distance, ε '_aIt is positioning group's error bounds of on-line proving；

Step 4, rasterizing scanning feasible set S_aX (), asks for node N to be positioned_aCoordinate estimate：

subject to X_a∈S_a(x)

Wherein,It is feasible set S_aThe center position coordinates of (x) each grid space,It is node N to be positioned_aCoordinate Optimal point estimation；

Step 5, positioning node perform the collection person's adaptive location based on time-varying error bounds in new locating periodically Update；

(1) node N_aAfter the n moment completes positioning, extract collection person's adaptive location and update decision model：

Ψ^-(n)≤|H^T(n)W_a(n)|≤Ψ⁺(n) (4)

Wherein, Ψ^-N () is a column vector for containing n moment all range measurement estimate lower bounds, Ψ⁺N () is right Answer the column vector in n moment all range measurement estimate upper bounds, W_a(n)=[X₁ X₂ … X_m]^TRepresent that the n moment positions institute in group There is the m coordinate of sensor node；H^TN () is collection person's transition matrix, its columns is equal to the total m of positioning group's interior nodes, its row Number is equal to the total number of anchor node in positioning group, often one anchor node N of row correspondence_i, and to should anchor node ID i-th element It is ﹣ 1, corresponding node N_aA-th element be 1, remaining element be 0；

(2) it is the n+1 moment in new locating periodically, to removing node coordinate W in model (4)_aN () remaining parameter outward is carried out Update, the collection person's adaptive location for obtaining the n+1 moment updates decision model：

Ψ^-(n+1)≤|H^T(n+1)W_a(n)|≤Ψ⁺(n+1) (5)

By the node coordinate W at n moment_aN () substitutes into model (5), and check whether it sets up；

If set up, node N_aWithout updating, i.e. W_a(n+1)=W_a(n)；

Otherwise, with μ_w(n+1) for step-length finds the node location coordinate W that can meet model (6)_a(n+1)：

min|W_a(n+1)-W_a(n)|² (6)

subject to:Ψ^-(n+1)≤|H^T(n+1)W_a(n+1)|≤Ψ⁺(n+1)

Calculate W_a(n+1) specific formula is：

W_a(n+1)=W_a(n)+μ_w(n+1)e^* _a(n+1)H(n+1) (7)

The time change step length μ_w(n+1) asked for by following formula：

Wherein, For all range measurements of n+1 moment are estimated The optimal point estimation of value, ε '_a(n+1) it is positioning group's error bounds at n+1 moment.

The advantage of the invention is that：

(1) present invention proposes a kind of underwater sensing net node locating and update method based on time-varying error bounds, by right The on-line proving of error bounds in positioning group, solves by error bounds drifting problem caused by the ambient influnence of complicated ocean, reduces Multi-source does not know adverse effect of the noise to multi-hop distance estimations, improves the accuracy and reliability of position coordinates estimation；

(2) present invention proposes a kind of underwater sensing net node locating and update method based on time-varying error bounds, completes On the basis of initial alignment, updated by performing the collection person's adaptive location based on time-varying error bounds on demand, reduce energy and disappear Consumption, enhance positioning it is ageing, improve the positioning performance of underwater sensing net.

Brief description of the drawings

Fig. 1 is that a kind of underwater sensing net node locating and update method based on time-varying error bounds proposed by the present invention are implemented Flow chart of steps；

Fig. 2 is the three-dimensional deployment schematic diagram of underwater wireless sensor network in the present invention；

Fig. 3 is average localization error contrast of the new method proposed by the present invention with conventional method under heterogeneous networks degree of communication Figure.

Specific embodiment

The present invention is described in further detail below in conjunction with the accompanying drawings.The present invention is a kind of water based on time-varying error bounds Lower sensing net node positioning and update method, implementation steps especially by following steps as shown in figure 1, realize：

Step one, node to be positioned build positioning group；

(1) all anchor nodes broadcast one group of " positioning comprising information such as self ID, coordinate, multi-hop threshold value, locating periodicallies Bag "；One group of all node broadcasts " range finding bag ", by information exchange, obtain the ID of all neighbor nodes in itself communication context, The information such as coordinate, distance；

(2) node statistics to be positioned itself skip to the anchor node quantity in multi-hop threshold range, if reach coordinate estimating The minimum quantity of calculation (is 4 during three-dimensional deployment, for 3) during two dimension deployment, then by all anchor nodes and participates in the common of information forwarding Intermediate node includes positioning group, and goes to step two；Minimum quantity such as anchor node sum less than coordinate estimation, then labeled as wouldn't Meet location condition node, and wait next locating periodically；

Step 2, positioning group's on-line proving error bounds；

(1) node to be positioned sends " on-line proving " command frame to all anchor nodes in positioning group, and this command frame contains anchor Node repeats the maximum allowable frequency n of range finding；N should disclosure satisfy that sampling with replacement builds the minimum requirements of self-service sample, again must Must consider to repeat brought energy ezpenditure of finding range, so considering, normal conditions n takes 5；

(2) any two anchor node N in group_iWith N_jRange finding n times is repeated, the original range finding sample set that sample number is n is set up

(3) by the method for nonparametric sampling with replacement from original range finding sample set D_ijExtract the self-service sample that capacity is similarly nIt should be noted that due to using sampling with replacement, so in original range finding sample set Certain sample is possible to repeatedly to be drawn, it is also possible to also do not draw once；

(4) B self-service sample is independently extracted in succession, is obtainedWherein, b=1, 2,…,B；Then each self-service sample is averaged：To meet the sampling of bootstrap It is required that and control node amount of calculation, the sample number B normal conditions take 200；

(5) anchor node N is estimated_iWith N_jBetween measurement distance standard deviation

Wherein, X_iIt is anchor node N_iCoordinate, X_jIt is anchor node N_jCoordinate；

(6) all anchor node combination of two in group, (2)~(5) in repeat step two, between trying to achieve any two anchor node Measurement distance standard deviation, and extract maximum therein

(7) demarcateIt is the error bounds of positioning group's range measurement noise；

Step 3, structure node N to be positioned_aSet-member estimation feasible set S_a(x)：

Wherein, u is the number for positioning anchor node in group, R₀It is the communication radius of node, d '_iaIt is anchor node N_iWith it is to be positioned Node N_aBetween measurement distance, ε '_aIt is positioning group's error bounds of on-line proving；It is probably multi-hop in view of the range finding between node Distance estimations, therefore the lower bound of error bounds takes communication radius R₀With d '_ia-ε′_aBetween less one；

Step 4, rasterizing scanning feasible set S_aX (), asks for node N to be positioned_aCoordinate estimate：

subject to X_a∈S_a(x)

Wherein,It is feasible set S_aThe center position coordinates of (x) each grid space,It is node N to be positioned_aCoordinate Optimal point estimation；The size of each grid space depends on the granularity of rasterizing scanning, generally takes scanning granularity for node is logical 1 the percent of news radius；

Step 5, positioning node perform the collection person's adaptive location based on time-varying error bounds more in new locating periodically Newly；

(1) node N_aAfter the n moment completes positioning, extract collection person's adaptive location and update decision model：

Ψ^-(n)≤|H^T(n)W_a(n)|≤Ψ⁺(n) (4)

Wherein, Ψ^-N () is a column vector for containing n moment all range measurement estimate lower bounds, Ψ⁺N () is right Answer the column vector in n moment all range measurement estimate upper bounds, W_a(n)=[X₁ X₂ … X_m]^TRepresent that the n moment positions institute in group There is the m coordinate of sensor node；H^TN () is collection person's transition matrix, its columns is equal to the total m of positioning group's interior nodes, its row Number is equal to the total number of anchor node in positioning group, often one anchor node N of row correspondence_i, and to should anchor node ID i-th element It is ﹣ 1, corresponding node N_aA-th element be 1, remaining element be 0；

(2) it is the n+1 moment in new locating periodically, to removing node coordinate W in model (4)_aN () remaining parameter outward is carried out Update, the collection person's adaptive location for obtaining the n+1 moment updates decision model：

Ψ^-(n+1)≤|H^T(n+1)W_a(n)|≤Ψ⁺(n+1) (5)

By the node coordinate W at n moment_aN () substitutes into model (5), and check whether it sets up；

If set up, node N_aWithout updating, i.e. W_a(n+1)=W_a(n)；

Otherwise, with μ_w(n+1) for step-length finds the node location coordinate W that can meet model (6)_a(n+1)：

min|W_a(n+1)-W_a(n)|² (6)

subject to:Ψ^-(n+1)≤|H^T(n+1)W_a(n+1)|≤Ψ⁺(n+1)

W_a(n+1) computing formula is：

W_a(n+1)=W_a(n)+μ_w(n+1)e^* _a(n+1)H(n+1) (7)

The time change step length μ_w(n+1) asked for by following formula：

Wherein, For all range measurements of n+1 moment are estimated The optimal point estimation of value, ε '_a(n+1) it is positioning group's error bounds at n+1 moment.

Embodiment

As shown in Fig. 2 in the three-dimensional spatial area of 3000m × 3000m × 200m 400 sensor sections of random placement Point, wherein anchor node ratio are 10%, are represented with five-pointed star, and anchor node ID is 1~40；Node to be positioned black circle table Show, its ID is 41~400.Dotted line between node represents two nodes and can carry out Direct Communication, and the communication radius of node is adjustable, The length of dotted line represents the Euclidean distance between two nodes.During positioning group's on-line proving error bounds, anchor node repeats range finding most Big permission frequency n is 5, and double sampling number of times B is 200.The granularity of rasterizing scanning takes 1 the percent of node communication radius.Measurement Noise is the non-Gaussian noise of Rayleigh distributed, its standard deviation take respectively 2% and the 5% of actual range so as to conventional method The influence that contrast error bounds time-varying is brought.

It is by adjusting communication radius that network-in-dialing degree is incremented by successively to 13 from 4 under above-mentioned network environment, and make respectively With proposed by the invention underwater sensing net node locating and update method based on time-varying error bounds and traditional do not consider to miss The difference circle multi-hop location algorithm for producing drift affected by environment carries out node self-localization, obtains Sensor Network under heterogeneous networks degree of communication Average localization error situation of change.As shown in figure 3, the solid line for being marked with hollow square is localization method proposed by the invention Average localization error in the case of standard deviation takes actual range 2%, is marked with the solid line of open triangles for conventional method is being marked Quasi- difference takes average localization error in the case of actual range 2%, and the dotted line for being marked with open diamonds is proposed by the invention to determine Average localization error of the position method in the case of standard deviation takes actual range 5%, the dotted line for being marked with empty circles is tradition side Average localization error of the method in the case of standard deviation takes actual range 5%.

From figure 3, it can be seen that no matter standard deviation is the 2% or 5% of actual range, method proposed by the present invention can be obtained More excellent positioning precision is obtained, average localization error is lower than traditional algorithm by about more than 30%, this explanation this method is to many hop distances The rejection ability of evaluated error is better than conventional method effect.What is more important, as a result of the treatment of time-varying error bounds Mechanism, under two kinds of influence of noises of different distributions rule, the positioning precision difference for obtaining is very for localization method proposed by the present invention It is small, and traditional algorithm has obvious gap under the influence of two kinds of error bounds.This explanation localization method proposed by the invention is made an uproar The influence of sound regularity of distribution change is smaller, the second step of method (positioning group's on-line proving error bounds) effect substantially, this for Multi-source is not known for the positioning under influence of noise, it appears particularly important.Additionally, we can further be seen that the present invention is carried determining Position method also remains to obtain preferable positioning precision in the case of network-in-dialing degree is less.

It should be noted that this embodiment is merely to illustrate technical scheme and unrestricted, although with reference to preferably Embodiment has been described in detail to the present invention, it will be understood by those within the art that, can be to technology of the invention Scheme is modified or equivalent, without deviating from the spirit and scope of technical solution of the present invention.

Claims (4)

1. a kind of underwater sensing net node locating and update method based on time-varying error bounds, it is characterised in that：By following step It is rapid to realize：

Step one, node to be positioned build positioning group；

(1) all anchor nodes are broadcasted one group " positioning bag " comprising information such as self ID, coordinate, multi-hop threshold value, locating periodicallies；Institute Have one group of node broadcasts " range finding bag ", by information exchange, obtain the ID of all neighbor nodes in itself communication context, coordinate, The information such as distance；

(2) node statistics to be positioned itself skip to the anchor node quantity in multi-hop threshold range, if reaching coordinate estimation Minimum quantity, then include positioning group, and go to step two by the common intermediate node that all anchor nodes and participation information are forwarded；Such as anchor Node total number then labeled as wouldn't meet location condition node, and waits next positioning less than the minimum quantity of coordinate estimation Cycle；

Step 2, positioning group's on-line proving error bounds；

(1) node to be positioned sends " on-line proving " command frame to all anchor nodes in positioning group, and this command frame contains anchor node Repeat the maximum allowable frequency n of range finding；

(2) any two anchor node N in group_iWith N_jRange finding n times is repeated, the original range finding sample set that sample number is n is set up

(3) by the method for nonparametric sampling with replacement from original range finding sample set D_ijExtract the self-service sample that capacity is similarly n

(4) B self-service sample is independently extracted in successionB=1,2 ..., B；And to each certainly Sample is helped to average：B=1,2 ..., B；

(5) anchor node N is estimated_iWith N_jBetween measurement distance standard deviation

${\widehat{\mathrm{\σ}}}_{ij}=\sqrt{\frac{1}{B-1}\underset{b=1}{\overset{B}{\Σ}}{({{\hat{D}}_{ij}}^{*b}-||X_i-X_j|{|}_2)}^2}---\left(1\right)$

Wherein, X_iIt is anchor node N_iCoordinate, X_jIt is anchor node N_jCoordinate；

(6) all anchor node combination of two in group, the survey between any two anchor node is tried to achieve in (2)~(5) in repeat step two Span extracts maximum therein from standard deviation

(7) demarcateIt is the error bounds of positioning group's range measurement noise；

Step 3, structure node N to be positioned_aSet-member estimation feasible set S_a(x)：

$S_a\left(x\right)=\underset{i=1}{\overset{u}{\∩}}\left\{min\right\{R_0,d_{ia}^{\′}-{\mathrm{\ϵ}}_a^{\′}\}\≤||X_i-X_a|{|}_2\≤d_{ia}^{\′}+{\mathrm{\ϵ}}_a^{\′}\}---\left(2\right)$

Wherein, u is the number for positioning anchor node in group, R₀It is the communication radius of node, d '_iaIt is anchor node N_iWith node to be positioned N_aBetween measurement distance, ε '_aIt is positioning group's error bounds of on-line proving；

Step 4, rasterizing scanning feasible set S_aX (), asks for node N to be positioned_aCoordinate estimate：

${\hat{X}}_a=\underset{X_a}{\arg min}\frac{\underset{i=1}{\overset{k}{\Σ}}\sqrt{{\left(\right||X_m^{*}-X_i|{|}_2)}^2-{d_{ia}^{\′}}^2}}{\underset{i=1}{\overset{k}{\Σ}}\left|\right|X_m^{*}-X_i|{|}_2}---\left(3\right)$

subject to X_a∈S_a(x)

Wherein,It is feasible set S_aThe center position coordinates of (x) each grid space,It is node N to be positioned_aCoordinate it is optimal Point estimation；

Step 5, positioning node perform the collection person's adaptive location based on time-varying error bounds in new locating periodically and update；

(1) node N_aAfter the n moment completes positioning, extract collection person's adaptive location and update decision model：

Ψ^-(n)≤|H^T(n)W_a(n)|≤Ψ⁺(n) (4)

Wherein, Ψ^-N () is a column vector for containing n moment all range measurement estimate lower bounds, Ψ⁺When () is correspondence n n Carve the column vector in all range measurement estimate upper bounds；W_a(n)=[X₁ X₂ … X_m]^TRepresent that the n moment is positioned all m in group The coordinate of sensor node；H^TN () is collection person's transition matrix, its columns is equal to the total m of positioning group's interior nodes, and its line number is equal to The total number of anchor node in group is positioned, often one anchor node N of row correspondence_i, and to should anchor node ID i-th element be ﹣ 1, Corresponding node N_aA-th element be 1, remaining element be 0；

(2) it is the n+1 moment in new locating periodically, to removing node coordinate W in model (4)_aN () remaining parameter outward is updated, The collection person's adaptive location for obtaining the n+1 moment updates decision model：

Ψ^-(n+1)≤|H^T(n+1)W_a(n)|≤Ψ⁺(n+1) (5)

By the node coordinate W at n moment_aN () substitutes into model (5), and check whether it sets up；

If set up, node N_aWithout updating, i.e. W_a(n+1)=W_a(n)；

Otherwise, with μ_w(n+1) for step-length finds the node location coordinate W that can meet model (6)_a(n+1)：

$\begin{array}{c}\min |W_a(n+1)-W_a\left(n\right){|}^2\\ subjectto:{\mathrm{\Ψ}}^{-}(n+1)\≤\left|H^T(n+1)W_a(n+1)W_a(n+1)\right|\≤{\mathrm{\Ψ}}^{+}(n+1)\end{array}---\left(6\right)$

Calculate W_a(n+1) formula is：

W_a(n+1)=W_a(n)+μ_w(n+1)e^* _a(n+1)H(n+1) (7)

The time change step length μ_w(n+1) asked for by following formula：

${\mathrm{\μ}}_w(n+1)=\left\{\begin{array}{cc}\frac{1}{H^H(n+1)H(n+1)}(1-\frac{{\mathrm{\ϵ}}_a^{\′}(n+1)}{\left|{e_a}^{*}(n+1)\right|}),& \begin{array}{cc}if& \left|{e_a}^{*}(n+1)\right|>{\mathrm{\ϵ}}_a^{\′}(n+1).\end{array}\\ 0,& otherwise.\end{array}\right.$

Wherein, It is n+1 moment all range measurement estimates Optimal point estimation, ε '_a(n+1) it is positioning group's error bounds at n+1 moment.

2. a kind of underwater sensing net node locating and update method based on time-varying error bounds according to claim 1, it is special Levy and be：

Coordinate estimation minimum quantity in the step one (2), is equal to 4, under two-dimentional deployment scenario under three-dimensional deployment scenario Equal to 3.

3. a kind of underwater sensing net node locating and update method based on time-varying error bounds according to claim 1, it is special Levy and be：

Anchor node in the step 2 (1) the maximum allowable frequency n normal conditions that repeat to find range are not more than 5；(4) the self-service sample in This number B normal conditions take 200.

4. a kind of underwater sensing net node locating and update method based on time-varying error bounds according to claim 1, it is special Levy and be：

The size of each grid space in the step 4 depends on the granularity of rasterizing scanning, and it is section generally to take scanning granularity 1 the percent of point communication radius.