CN102395192A - Method and device for locating wireless sensor terminal - Google Patents

Abstract

The embodiment of the invention discloses a method and a device for locating a wireless sensor terminal. The method comprises the following steps: on the basis of a wireless signal energy attenuation principle, mapping received signal strength indication (RSSI) obtained by all nodes onto probability weighted distance D in an off-line training manner by virtue of a Bayesian model; and more accurately locating the nodes according to a maximum likelihood estimation method and a Kalman filtering method from static and dynamic aspects. In the method, RSSI modeling is carried out statistically, which overcomes the defects such as influence on wireless terminal location caused by antenna directivity, environment and other factors for a small and medium sized mobile terminal signal as well as higher cost caused by the addition of a special sensor in the prior art; and in addition, relative to a classic single linear model or a segmented linear model, the model is updated by an on-line learning method so as to more accurately describe the situation of larger variance in a sample acquired by a small-sized wireless terminal, thus improving the practicability and applicability.

Description

A kind of localization method of wireless sensing terminal and device

Technical field

The present invention relates to the wireless sensing communication technical field, more particularly, relate to a kind of localization method and device of wireless sensing terminal.

Background technology

Advantages such as radio sensing network obtains and processing mode as brand-new information in a kind of Internet of Things industry, and simple and expansion is strong with its framework obtain promotion and application in industrial or agricultural, medical treatment and municipal monitoring field.The monitoring target information of overlay area, networking is gathered and monitored to said radio sensing network in real time through dispose a large amount of wireless sensor nodes in the monitored area, is sent to communication base station through wireless channel and carries out analyzing and processing.

The wireless terminal location technology is as one of key technology in the wireless sensor network technology; Be responsible for providing the positional information of wireless terminal node; Make communication base station source, locator data position timely and effectively; And network topological information can be provided, thereby improve router efficiency, realize distributing rationally of network configuration.Existing wireless terminal location technology is the intensity indication RSSI of reception signal commonly used relatively, and it uses wireless transmission theory or empirical model to convert distance and then location into according to receiving signal strength signal intensity.And since wireless signal for the adaptive capacity of complex environment a little less than, especially when the portable terminal miniaturization after, antenna will further strengthen and worsen final positioning result towards signal receiving strength is influenced.Use additional firmware such as angular surveying AOA or time difference TDOA to assist operation for improving above-mentioned condition, the complexity at terminal has not only increased the computational complexity at terminal, and has increased hardware cost.

Summary of the invention

In view of this, the present invention provides a kind of localization method and device of wireless sensing terminal, realizing reducing environmental factor and antenna towards the influence to the miniaturization terminal positioning, and reaches and reduces hardware cost and calculate succinct effect.

A kind of localization method of wireless sensing terminal comprises:

Reception signal strength signal intensity through the Bayesian model algorithm for estimating obtains whole nodes indicates RSSI through after the off-line training modeling, maps to the probability weight distance B, and said Bayesian model algorithm for estimating is based on the wireless energy signal attenuation principle;

Utilize maximum likelihood estimate to combine said probability weight distance B to confirm the static node position, and utilize kalman filter method that the coordinate of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes.

Above-mentioned execution mode is that the influence towards for wireless signal of antenna is carried out modeling from the statistics angle, obtains the target location through using Bayesian probability statistical model estimating target distance and maximum likelihood estimate.Utilize static node to combine Kalman filter model to confirm positions of mobile nodes; From static and dynamic two aspects all types of nodes are accurately located; The method of signal strength signal intensity being carried out the certainty modeling with respect to classical way has more robustness; Thereby realized minimizing Miniature Terminal aerial angle and environment etc., and the purpose that reduces the hardware cost of special sensor for the positioning result influence.

Preferably, said method comprises that also the Bayesian inference model is returned in the static node position that obtains to carry out iteration and upgrade, and the positions of mobile nodes that obtains is returned in said Kalman's model carry out on-line study.

Through the renewable said Bayesian inference model of on-line study model, make static node obtain belonging to the signal energy mode under the environment; And, through the iteration of Kalman filter model, can accomplish the filter tracking of the target trajectory of movement node better.

Preferably; After the said reception signal strength signal intensity indication RSSI that whole nodes is obtained through the Bayesian model algorithm for estimating passes through the off-line training modeling, map to the probability weight distance B and be specially: set up mixed Gauss model GMM and tentatively find range the back acquisition based on this reception signal strength signal intensity lower probability Weighted distance D.

Above-mentioned mixed Gauss model be with the probability of acceptance with and nodal pitch be modeled as probability distribution; And said mixed Gauss model carries out multi-modal modeling to mapping relations; Consider because signal incident angle difference is brought difference; Can accurately locate the target mapping relations, and be not limited to the above-mentioned form of enumerating.

Preferably, utilize maximum likelihood estimate to combine said probability weight distance B to confirm that the static node position is specially: said probability weight distance B to be added the minimize error cost function obtain the maximal possibility estimation positioning result under its probability right.

Maximum likelihood location algorithm in this execution mode is non-based on the higher algorithm of positional accuracy in the topological structure in the present radio sensing network, and the minimum point of difference carries out the succeeding target location between range finding distance and the estimated distance through seeking.

Preferably, said method also comprises initialization node power information and each node is carried out the linear time Filtering Processing.

In order to reduce wireless sensing terminal in actual use, receive signal strength signal intensity RSSI and receive environment and target and move and wait influence, need carry out again that data analysis is preceding carries out Filtering Processing to signal, and be not limited to the above-mentioned form of enumerating.

A kind of positioner of wireless sensing terminal comprises:

The Bayesian model range finder module; Be used for indicating through after the off-line training modeling through the reception signal strength signal intensity that the Bayesian model algorithm for estimating obtains whole nodes; Map to the probability weight distance B, said Bayesian model algorithm for estimating is based on the wireless energy signal attenuation principle;

The node locating module is used to utilize maximum likelihood estimate and combines said probability weight distance B to confirm the static node position; And utilize kalman filter method that the position coordinates of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes.

Preferably, said device also comprises: the initialization process module is used for initialization node power information and each node is carried out the linear time Filtering Processing.

Preferably, said device also comprises: the online updating study module is used for that the Bayesian inference model is returned in the static node position that obtains and carries out iteration and upgrade, and the positions of mobile nodes that obtains returned in said Kalman's model carry out on-line study.

Said apparatus and method are corresponding, with static node locating module and mobile node module combine to have reached reduce Miniature Terminal aerial angle and environment etc. to the influence of wireless sensor network positioning, and reduce the technique effect of the hardware cost of special sensor.

Can find out from above-mentioned technical scheme; The embodiment of the invention utilizes Bayesian model that the reception signal strength signal intensity indication RSSI off-line training that whole nodes obtain is mapped to the probability weight distance B; And respectively from static and dynamic two aspects; In conjunction with maximum likelihood estimate and kalman filter method location node position more exactly; With the intensity indication modeling to received signal of statistics angle, overcome middle-size and small-sizeization of prior art portable terminal signal and received the influence for wireless terminal location such as antenna direction and environment, and increased special sensor and cause the higher shortcoming of cost; And the present invention selects for use method the linear model or the piecewise linear model that more new model is more single than classics of on-line study can describe the bigger situation of miniaturization wireless terminal collection sample variance more accurately, and practicality and applicability all increase.

Description of drawings

In order to be illustrated more clearly in the embodiment of the invention or technical scheme of the prior art; To do to introduce simply to the accompanying drawing of required use in embodiment or the description of the Prior Art below; Obviously, the accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills; Under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.

Fig. 1 is the localization method flow chart of the disclosed a kind of wireless sensing terminal of the embodiment of the invention;

Fig. 2 is the localization method flow chart of the disclosed a kind of wireless sensing terminal of further embodiment of this invention;

Fig. 3 a is the positioning device structure sketch map of the disclosed a kind of wireless sensing terminal of the embodiment of the invention;

Fig. 3 b is the location algorithm flow chart of the disclosed a kind of wireless sensing terminal of the embodiment of the invention.

Embodiment

For quote and know for the purpose of, the technical term that hereinafter uses, write a Chinese character in simplified form or abridge and sum up as follows:

RSSI:Received Signal Strength Indicator receives the intensity indication of signal.

To combine the accompanying drawing in the embodiment of the invention below, the technical scheme in the embodiment of the invention is carried out clear, intactly description, obviously, described embodiment only is the present invention's part embodiment, rather than whole embodiment.Based on the embodiment among the present invention, those of ordinary skills are not making the every other embodiment that is obtained under the creative work prerequisite, all belong to the scope of the present invention's protection.

The embodiment of the invention discloses a kind of localization method and device of wireless sensing terminal, realizing reducing environmental factor and antenna, and reach and reduce hardware cost and calculate succinct effect towards influence to the miniaturization terminal positioning.

Before carrying out the embodiment description, need to prove:

Two kinds of nodes are arranged in the embodiment of the invention, and static node and mobile node carry out said static node location positioning through the method for the embodiment of the invention; And in the actual location scene; Because block, the influence of disturbing factor such as multipath effect, cause terminal node received power and relation between the distance to distribute and deviation occurs, the localization method of static node will be not suitable for the definite of such mobile node; Then at this moment; Need be to need participating in the node of running fix in the static node of having had good positioning, i.e. static object node, and position coordinates and the speed of utilizing kalman filter method to be directed against said static object node are set up model; Set up follow-up mechanism, accurately locate.

Detail with regard to execution mode at present:

Fig. 1 shows a kind of localization method of wireless sensing terminal, comprising:

Step 11: the reception signal strength signal intensity through the Bayesian model algorithm for estimating obtains whole nodes indicates RSSI through after the off-line training modeling, maps to the probability weight distance B;

It is pointed out that the wireless energy signal attenuation principle of said Bayesian model based on classics, in the present embodiment, that is, the wireless transmission model of employing receives signal strength signal intensity indication RSSI based on this:

RSSI＝-(10nlog₁₀d+A)

Theoretical according to radio signal propagation, (RSSI, 10nlog₁₀D) should be similar to linear distribution of obedience in theory, the modeling target is path loss coefficient n,, power bias A.

Preferably, the data based a certain m of estimating (x) that the establishing target mixed Gauss model will be gathered under off-line state comes the given f of family of functions is found the solution minimum functional.That is:

arg?min?Loss(f)＝∫_x∈Dm(y-f(x))dx；

$D=\left\{\mathrm{dis}\right|\∃y=\mathrm{RSSIvs}.\mathrm{dis}\};$

Linearisation wireless terminal signal receiving strength receive antenna towards influence bigger; Receive signal strength signal intensity RSSI and can't carry out modeling with simple mapping model apart from d; Model class interpolation bigger after its modeling directly influences last positioning result, therefore in the present embodiment, sets up the F (RSSI to the range finding link; D) mixed Gauss model uses the deduction method weighting in the Bayesian inference model to confirm distance thereafter.

According to above modeling method, each receives signal strength signal intensity RSSI can obtain p apart from value according to training sample, its coefficient ω_pRepresentative under current training sample current demand signal received power RSSI, euclidean distance between node pair a_pThe posterior probability that occurs.Use the mixed model modeling to solve preferably and receive signal strength signal intensity RSSI and the multi-modal problem of distance B mapping.The result that obtains of maximal possibility estimation through probability weight can be more accurate at last.

And the present invention is different with the method for estimation of classical qualitative parameter really, and the estimated parameter of Bayesian Estimation hypothesis is a stochastic variable, through setting up this gauss hybrid models characterising parameter mapping actual distribution better.

Step 12: utilize maximum likelihood estimate to combine said probability weight distance B to confirm the static node position;

The maximum likelihood location algorithm is that wireless sensor network positioning is non-based on the algorithm preferably of locating effect in the topological structure.Through seeking the minimum point of difference between a feasible range finding distance and the estimated distance, with the position that this point is a unknown node, concrete account form is following:

If the coordinate of known reference node is respectively (x₁, y₁), (x₂, y₂), (x₃, y₃) .... (x_n, y_n), establish node coordinate undetermined for (x, y), node undetermined is respectively d to the measuring distance of known reference node₁, d₂, d₃... d_n, equation group is then arranged:

$\left\{\begin{array}{c}{(x_1-x)}^2+{(y_1-y)}^2={d_1}^2\\ {(x_2-x)}^2+{(y_2-y)}^2={d_2}^2\\ \·\\ \·\\ \·\\ {(x_n-x)}^2+{(y_n-y)}^2={d_n}^2\end{array}\right.$

In the location algorithm based on range finding, because measuring distance is not an entirely accurate, after the adding error term, above-mentioned equation can be abbreviated as

AX+E＝B

When the location estimation coordinate that can obtain mobile node was following form, the minimum range error of node to be measured was square minimum down.

$\hat{X}={\left(A^{T}A\right)}^{-1}{A}^{T}b$

After received power is less than certain saturation value, the disproportionate basically relation of power and distance, then the power test variance is bigger, and is very big to the positioning result influence.Said probability weight distance B is added the minimize error cost function obtain the maximal possibility estimation positioning result under its probability right.

The residual error location sum of squares function that in this step, uses is:

F (x)=W_p(B-AX)², wherein be W_pMixed Gaussian probability weight matrix.

Step 13: utilize kalman filter method that the coordinate of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes.

In the present embodiment, use the method for Kalman filtering that the position coordinates and the speed of static object node are set up four-dimensional (x, y, v_x, v_y) Filtering Model; The tracking results that obtains not only can target of prediction next constantly possible position; And this positioning result does not receive the influence of actual environment to signal strength signal intensity; Can improve in the static immobilization algorithm because of the influence of environmental factor, obtain more smooth track near the actual motion curve for positioning result.

To sum up can know: to after the wireless terminal miniaturization; Using reception signal strength signal intensity RSSI to carry out distance map uses single model variance bigger; Modeling result is inaccurate and cause the defective of the model aspect of deviations; Combine Kalman filter model to confirm positions of mobile nodes through using Bayesian probability statistical model estimating target distance and maximum likelihood estimate to obtain the static node position, reaching, all types of nodes are accurately located from static and dynamic two aspects; The method of signal strength signal intensity being carried out the certainty modeling with respect to classical way has more robustness; Thereby realized when the location of actual wireless sensing terminal, multi-modal to influences such as Miniature Terminal aerial angle and environment, and the purpose that reduces the hardware cost of special sensor.

Fig. 2 shows the localization method of another wireless sensing terminal, comprising:

Step 21: initialization node power information is also carried out the linear time Filtering Processing to each node.

In this step, in order to reduce wireless sensing terminal in actual use, receive signal strength signal intensity RSSI and receive environment and target and move and wait influence, need carry out again that data analysis is preceding carries out Filtering Processing to signal, and be not limited to the above-mentioned form of enumerating.

Step 22: the reception signal strength signal intensity through the Bayesian model algorithm for estimating obtains whole nodes indicates RSSI through after the off-line training modeling, maps to the probability weight distance B;

Step 23: utilize maximum likelihood estimate to combine said probability weight distance B to confirm the static node position;

Step 24: utilize kalman filter method that the coordinate of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes;

Step 25: the Bayesian inference model is returned in the static node position that obtains carry out iteration and upgrade, and the positions of mobile nodes that obtains returned in said Kalman's model carry out on-line study.

Final positioning result returned carry out on-line study in said Bayesian inference model and the said Kalman filter model and upgrade; Be to upgrade the Bayesian inference model, different static node iterative learnings is obtained based on the signal energy mode under its specific environment through online greatest hope online-EM.Through the iteration of Kalman filter model, thus the filter tracking of completion movement objective orbit.

Fig. 3 a shows a kind of positioner of wireless sensing terminal, comprises;

Bayesian model range finder module 31: be used for indicating RSSI through after the off-line training modeling through the reception signal strength signal intensity that the Bayesian model algorithm for estimating obtains whole nodes; Map to the probability weight distance B, said Bayesian model algorithm for estimating is based on the wireless energy signal attenuation principle;

Node locating module 32 is used to utilize maximum likelihood estimate and combines said probability weight distance B to confirm the static node position; And utilize kalman filter method that the position coordinates of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes;

Initialization process module 33 is used for initialization node power information and each node is carried out the linear time Filtering Processing;

Online updatingstudy module 34 is used for that the Bayesian inference model is returned in the static node position that obtains and carries out iteration and upgrade, and the positions of mobile nodes that obtains is returned in said Kalman's model carry out on-line study.

Said apparatus and method are corresponding, with static node locating module and mobile node module combine to have reached reduce Miniature Terminal aerial angle and environment etc. to the influence of wireless sensor network positioning, and reduce the technique effect of the hardware cost of special sensor.

Fig. 3 b is and said method and the corresponding logical flow chart of device that as shown in the figure, detailed description can be repeated no more referring to Fig. 2 and Fig. 3 diagram and respective description here.

In sum:

The present invention is based on the wireless energy signal attenuation principle; Utilize Bayesian model that the reception signal strength signal intensity indication RSSI off-line training that whole nodes obtain is mapped to the probability weight distance B; And respectively from static and dynamic two aspects; In conjunction with maximum likelihood estimate and kalman filter method location node position more exactly; With the intensity indication modeling to received signal of statistics angle, overcome middle-size and small-sizeization of prior art portable terminal signal and received the influence of wireless terminal location such as antenna direction and environment, and increased special sensor and cause the higher shortcoming of cost;

And; The present invention selects for use method the linear model or the piecewise linear model that more new model is more single than classics of on-line study can describe the bigger situation of miniaturization wireless terminal collection sample variance more accurately; Practicality and applicability all increase, and can support the location of the wireless sensing terminal in multiple fields such as location, wireless down-hole, wireless medical monitoring and the security protection of wireless prison.

Each embodiment adopts the mode of going forward one by one to describe in this specification, and what each embodiment stressed all is and the difference of other embodiment that identical similar part is mutually referring to getting final product between each embodiment.For the disclosed device of embodiment, because it is corresponding with the embodiment disclosed method, so description is fairly simple, relevant part is partly explained referring to method and is got final product.

It will be understood by those skilled in the art that and to use in many different processes and the technology any one to come expression information, message and signal.For example, the message of mentioning in the above-mentioned explanation, information can be expressed as voltage, electric current, electromagnetic wave, magnetic field or magnetic particle, light field or above combination in any.

The professional can also further recognize; The unit and the algorithm steps of each example of describing in conjunction with embodiment disclosed herein; Can realize with electronic hardware, computer software or the combination of the two; For the interchangeability of hardware and software clearly is described, the composition and the step of each example described prevailingly according to function in above-mentioned explanation.These functions still are that software mode is carried out with hardware actually, depend on the application-specific and the design constraint of technical scheme.The professional and technical personnel can use distinct methods to realize described function to each certain applications, but this realization should not thought and exceeds scope of the present invention.

The method of describing in conjunction with embodiment disclosed herein or the step of algorithm can be directly with the software modules of hardware, processor execution, and perhaps the combination of the two is implemented.Software module can place the storage medium of any other form known in random asccess memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or the technical field.

To the above-mentioned explanation of the disclosed embodiments, make this area professional and technical personnel can realize or use the present invention.Multiple modification to these embodiment will be conspicuous concerning those skilled in the art, and defined General Principle can realize under the situation that does not break away from the spirit or scope of the present invention in other embodiments among this paper.Therefore, the present invention will can not be restricted to these embodiment shown in this paper, but will meet and principle disclosed herein and features of novelty the wideest corresponding to scope.

Claims (8)

1. the localization method of a wireless sensing terminal is characterized in that, comprising:

Reception signal strength signal intensity through the Bayesian model algorithm for estimating obtains whole nodes indicates RSSI after the off-line training modeling, maps to the probability weight distance B, and said Bayesian model algorithm for estimating is based on the wireless energy signal attenuation principle;

Utilize maximum likelihood estimate to combine said probability weight distance B to confirm the static node position, and utilize kalman filter method that the coordinate of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes.

2. method according to claim 1; It is characterized in that; Said reception signal strength signal intensity indication of whole nodes being obtained through the Bayesian model algorithm for estimating maps to the probability weight distance B and is specially after the off-line training modeling: set up the mixed Gauss model GMM back of tentatively finding range and obtain to receive signal strength signal intensity lower probability Weighted distance D based on this.

3. method according to claim 1 is characterized in that, the Bayesian inference model is returned in the static node position that obtains carry out iteration and upgrade and to carry out iteration and upgrade, and the positions of mobile nodes that obtains returned in said Kalman's model carry out on-line study.

4. method according to claim 1; It is characterized in that, utilize maximum likelihood estimate to combine said probability weight distance B to confirm that the static node position is specially: said probability weight distance B to be added the minimize error cost function obtain the maximal possibility estimation positioning result under its probability right.

5. method according to claim 1 is characterized in that, also comprises initialization node power information and each node is carried out the linear time Filtering Processing.

6. the positioner of a wireless sensing terminal is characterized in that, comprising:

Bayesian model range finder module: be used for indicating RSSI through after the off-line training modeling through the reception signal strength signal intensity that the Bayesian model algorithm for estimating obtains whole nodes; Map to the probability weight distance B, said Bayesian model algorithm for estimating is based on the wireless energy signal attenuation principle;

The node locating module is used to utilize maximum likelihood estimate and combines said probability weight distance B to confirm the static node position; And utilize kalman filter method that the position coordinates of the static node position that obtains and speed are set up four-dimensional Filtering Model and confirm positions of mobile nodes.

7. device according to claim 6 also comprises: the initialization process module is used for initialization node power information and each node signal intensity is carried out the linear time Filtering Processing.

8. device according to claim 6; Also comprise: the online updating study module; Be used for that the Bayesian inference model is returned in the static node position that obtains and carry out iteration and upgrade, and the positions of mobile nodes that obtains returned in said Kalman's model carry out on-line study.

Images