CN106211319A - A kind of non-fingerprint passive type localization method based on WI FI signal - Google Patents

Abstract

The invention discloses a kind of non-fingerprint passive type localization method based on WI FI signal, the method comprises the steps: step 1, builds WI FI transceiver network；Step 2, gathers the CSI value in WI FI signal；Step 3, filter preprocessing process；Step 4, estimates effective depth and the match parameter of target to be measured；Step 5, utilizes match parameter to obtain the pad value of target to be measured, determines region residing for target to be measured by its pad value；Step 6, positions target.The present invention, compared to fingerprint positioning method, expends time and the problem of manpower during solving its fingerprint collecting, and robustness is good, practical；The present invention, by extracting the CSI information in WI FI signal, solves the problem that existing non-fingerprint positioning method positioning precision is low.

Description

A kind of non-fingerprint passive type localization method based on WI-FI signal

Technical field

The present invention relates to the application of location, the non-fingerprint passive type localization method in a kind of faced chamber.

Background technology

Along with people become increasingly complex and finely in the activity of the interior space, demand day based on Indoor Location Information service Tend to become strong strong.City's inner position is broadly divided into active location and passive type positions two kinds at present, and its main distinction is to be positioned target Whether carrying relevant device, active positioning requirements is decided to be target Portable device, positions device location, and passive type Location is that target need not Portable device, positions the interference of characteristic information according to target.Owing to applying ring at some Under border, target is not intended to carry relevant device, and therefore, passive type scene becomes the focus of research.

In passive type positions, according to the location information obtained, it is based on non-electromagnetic signal passive to be divided into location Formula location and passive type based on electromagnetic signal location.Location based on non-electromagnetic signal is to use such as infrared ray, photographic head, biography Sensors etc. carry out the determination of target location, and these equipment have certain particularity due to the selection of its feature, and application scenarios is also subject to Arrive a lot of restriction.And in location of based on electromagnetic signal, it is most commonly used that use RSS information positions, and is divided into fingerprint side Method location and non-fingerprint method position.Wherein fingerprint method carries out fingerprint collecting due to needs before location, expends time and people Power, and robustness is bad, and therefore practicality is the strongest；Rather than fingerprint location is due to the restriction of model and asking of RSS information itself Topic, ratio of precision is poor.To this, there is researcher to use special equipment to extract stable amplitude and phase information positions, but Special equipment needs special (such as aerial array) or the fancy price of making (such as USRP (Universal Software Radio Peripheral, general software radio peripheral hardware)).Therefore part researcher is had CSI information to be extracted and determines Position, but still there is problems in that 1, location based on CSI information does not has ready-made model to use, and can only carry out fingerprint collecting and carry Take；2, separate due to multi-subcarrier based on CSI information, it is impossible to find general information expressing method.

Summary of the invention

For above-mentioned problems of the prior art, the invention provides below scheme:

A kind of non-fingerprint passive type localization method based on WI-FI signal, comprises the following steps:

Step 1, builds WI-FI transceiver network, including transmitting terminal a and receiving terminal b；

Step 2, in the WI-FI transceiver network built, gathers the CSI value in WI-FI signal, the CSI value of described collection Being divided into test CSI value and actual CSI value, wherein test CSI value is:

(2-1-1) WI-FI transceiver network does not has initial CSI value during target；

(2-1-2) in WI-FI transceiver network, put into the CSI value during target of the first known altitude and position；

(2-1-3) in WI-FI transceiver network, put into the CSI value during target of the second known altitude and position, Qi Zhong The height of two known target and the height of the first known target are different；

Actual CSI value refers to, the CSI value gathered when there is target to be measured in WI-FI transceiver network；

The all CSI values collected in step 2 are utilized 3 σ filtering methods to be filtered, after Filtering Processing by step 3 CSI value carries out pretreatment, eliminates multipath effect；

Step 4, utilizes pretreated CSI value in step 3, estimates the effective depth of target to be measured, pass through to be estimated out Target to be measured effective depth estimate match parameter；

Step 5, utilizes match parameter to obtain the pad value of target to be measured, determines district residing for target to be measured by its pad value Territory；

Step 6, according to region residing for the target to be measured obtained in step 5, positions target.

Further, the WI-FI transceiver network described in step 1 at least needs both links.

Further, the pre-treatment step described in step 3 includes:

(3-1) the multipath fading signal in described filtered signal is carried out enhancement process, processes formula as follows:

Wherein, n is the group number of a period of time interior CSI value gathered, and (1, n), each group of CSI value is by 30 subcarriers for i ∈ Composition, k ∈ (1,30), csi_mkiThe CSI value collected on i-th time point for kth subcarrier, ρ_kiFor kth subcarrier The CSI value that upper i-th time point collects accounts for the proportion of the CSI value that on this subcarrier, all time points collect, CSI_mkpIt is CSI value after enhancement process on k subcarrier；

Wherein, m takes 1,2,3,4 respectively, and it represents (2-1-1), (2-1-2), (2-1-3) and WI-FI in step 2 respectively Transceiver network exists four kinds of situations of target to be measured；

(3-2) equation below is utilized, the too strong signal in CSI value after rejecting (3-1) process:

Work as ε_abk＜ 1 and ε_abkTime maximum, take the input as system of the kth subcarrier；Reject its remaining sub-carriers；

Wherein, ab represents that the receiving-transmitting chain of signal, k ∈ (1,30), m take 2,3,4 respectively and represent the in step 2 respectively (2-1-2), there is three kinds of situations of target to be measured, CSI in (2-1-3) and WI-FI transceiver network_1kpRepresent kth after enhancement process Initial CSI value on subcarrier, CSI_mkpAll CSI values during target jamming are had on kth subcarrier after representing enhancement process, ε_abkRepresenting the ratio of each CSI value and initial CSI value when having target jamming in ab link in kth subcarrier, ab link is Refer to the information transmitting terminal direct line of sight link to receiving terminal.

Further, match parameter δ described in step 4₁And δ_t, concrete calculating process is as follows:

(4-1) utilize height and the positional information of the first known target, calculate theoretical attenuation value D on ab link_ab, formula As follows:

Wherein, D_abBe the theoretical attenuation value on ab link, c (v) and s (v) be fresnel integral, v is Fresnel-Kiel Hough diffraction parameter, its calculating formula is:

Wherein, t ∈ (-1,1), λ are signal wavelength, and h1 is the height of the first known target, d_a2Be the first known target away from Transmitting terminal distance, d_b2It is that the first known target refers to information transmitting terminal directly regarding to receiving terminal away from receiving terminal distance, ab link Away from link；

(4-2) D obtained by (4-1)_ab, utilize following formula to calculate match parameter δ₁:

Wherein, CSI_2kpFor the CSI value of the first known target after pretreatment, CSI_1kpFor through pretreated initially CSI value, D_abIt is the theoretical attenuation value of ab link, δ₁For match parameter；

(4-3) known h1 and δ₁, obtain following ratio relation according to formula (1):

Wherein, h1 is the height of the first known target, and ω is ratio parameter；

(4-4) when there is target to be measured in WI-FI network, the method for utilization index weighted moving average (EWMA) is estimated The height of target to be measured, concrete formula is as follows:

Wherein, factor alpha represents that the change of weight, α ∈ (0,1), t are that CSI value gathers moment, S_tEstimate to be measured for t The height value of target, h1, h2 are respectively height and the height of the second known target of the first known target；

(4-5) according to the ratio parameter ω obtained in the height of the target to be measured estimated in (4-4) and (4-3), utilize Equation below, it is thus achieved that the match parameter of differing heights target:

Wherein, δ_tFor estimating the match parameter of differing heights target.

Further, the concrete steps of step 5 include:

(5-1) variances sigma of initial CSI value after pretreatment is set₁As threshold value, to be measured by what step 4 obtained The height of target and match parameter, utilize formula (1) calculated D_abIf, D_abMore than σ₁Time, represent that this target to be measured is in Detection region；Otherwise, it means that this target to be measured to be in detection zone overseas；

If (5-2) target to be measured is in detection region, it is handled as follows:

Wherein, f ∈ (1, t1), t1 are the number of target to be measured, D on ab link_abfRepresent and treat for f on ab link Survey the theoretical attenuation value of target, min (D_ab) represent on the 8th link the minima of theoretical attenuation value in target to be measured.

Further, the process positioning target in step 6 includes:

If (6-1) target to be measured is in MTA region, following formula is utilized to be calculated the range information d of target to be measured_a4、d_b4, then According to the deployment scenario of transmitting terminal receiving terminal, obtaining target position information, computing formula is as follows:

Wherein, c (v) and s (v) is fresnel integral, and its calculating formula is:

Wherein, h estimates height S in being step (4-4)_t, λ is signal wavelength, d_a4For target to be measured away from transmitting terminal distance, d_b4 For target to be measured away from receiving terminal distance；

(6-2) when target is in OMTA region, utilize the emphasis algorithm of influence area to position target to be measured in OMTA district Position in territory.

Further, the influence area described in step (6-2) emphasis algorithm particularly as follows:

(6-2-1) there is L bar link in the WI-FI transceiver network described in setting, determine in L bar link and can surround region Chain travel permit number n₁, calculate the pad value D of each of the links_q, then its weights are:

Wherein, q ∈ (1, n₁), D_qIt is the pad value of q article of link, ε_qIt is that the pad value of q article of link is at n₁In bar link Shared weights；

(6-2-2) it is weighted according to the point midway of each of the links, is shown below, obtain the position of OMTA zone location Confidence ceases:

Wherein, x_obj、y_objFor the position coordinates of target to be measured, x_q、y_qIt is the q article link point midway coordinate.

Compared with prior art, the present invention has a following technique effect:

1. the present invention is compared to fingerprint positioning method, expends asking of time and manpower during solving its fingerprint collecting Topic, and robustness is good, practical；

2. the present invention is by extracting the CSI information in WI-FI signal, solves existing non-fingerprint positioning method positioning precision Low problem.

Accompanying drawing explanation

Fig. 1 is CSI non-fingerprint location system flow chart；

Fig. 2 is CSI non-fingerprint location system hardware schematic diagram；

Fig. 3 is that CSI positions equipment schematic diagram, and wherein (a) is industrial control mainboard schematic diagram, and (b) is transmission antenna schematic diagram；

Fig. 4 is CSI non-fingerprint location actual deployment figure, and wherein (a) is floor map, and (b) is realistic picture；

Fig. 5 is CSI non-fingerprint location statistical result figure；

Fig. 6 is CSI non-fingerprint experiment positioning precision figure under distinct device quantity；

Fig. 7 is that the non-fingerprint location of CSI tests loss figure.

Detailed description of the invention

With embodiment, the present invention is described in detail below in conjunction with the accompanying drawings.

The present embodiment describes a kind of non-fingerprint passive type localization method based on WI-FI signal, and it comprises the following steps:

Step 1, according to indoor situations, builds WI-FI transceiver network, and including transmitting terminal a and receiving terminal b, and this WI-FI receives Hairnet network at least includes both links；In order to ensure experiment effect, this network needs information transmitting apparatus to collide with each other, and And try not and other WI-FI equipment clash.

Step 2, in the WI-FI transceiver network built, gathers the CSI value in WI-FI signal, the CSI value of described collection Being divided into test CSI value and actual CSI value, wherein test CSI value is:

(2-1-1) WI-FI transceiver network does not has initial CSI value during target；

(2-1-2) in WI-FI transceiver network, put into the CSI value during target of the first known altitude and position；

(2-1-3) in WI-FI transceiver network, put into the CSI value during target of the second known altitude and position, Qi Zhong The height of two known target and the height of the first known target are different；

Actual CSI value refers to, when target to be measured enters the optional position of WI-FI transceiver network, pauses 10 seconds, is gathered CSI value, and this method is also shown emphatically is how to position this target to be measured；

The above-mentioned all data collected all are sent in data processor.

The all CSI values collected in step 2 are utilized 3 σ filtering methods to be filtered, after Filtering Processing by step 3 CSI value carries out pretreatment, eliminates multipath effect；

The problem probably occurring wave distortion for the data collected, this programme uses 3 σ filtering methods to make abnormal Become waveform try not to occur in subsequent step, wherein 3 σ filtering methods particularly as follows:

The data received are rayleigh distributed, but for each subcarrier, its value is theoretically certain, it is simply that Say the subcarrier data being an actually-received be theoretical value and noise and.And in the case of environment is more stable, multipath value is also Can be more stable, so, the result received be a stationary value and noise and.Its result meets Gauss distribution.Cause Here select the 3 σ filtering methods generally used in engineering to filter waveform.

Owing to, during receiving signal, indoor have a lot of reflections, it may appear that the multipath effect of transmission, this effect It is divided into multipath to strengthen and multipath fading.This programme is existing to be strengthened multipath signal, then enhancing signal is rejected, concrete grammar For:

(3-1) the multipath fading signal in described filtered signal is carried out enhancement process, processes formula as follows:

Wherein, n is the group number of a period of time interior CSI value gathered, and (1, n), each group of CSI value is by 30 subcarriers for i ∈ Composition, k ∈ (1,30), csi_mkiThe CSI value that kth subcarrier collects on i-th time point, ρ_kiFor on kth subcarrier The CSI value that i-th time point collects accounts for the proportion of the CSI value that on this subcarrier, all time points collect, CSI_mkpFor kth CSI value after enhancement process on individual subcarrier；

Wherein, m takes 1,2,3,4 respectively, and it represents (2-1-1), (2-1-2), (2-1-3) and WI-FI in step 2 respectively Transceiver network exists four kinds of situations of target to be measured；

(3-2) equation below is utilized, the too strong signal in CSI value after rejecting (3-1) process:

Work as ε_abk＜ 1 and ε_abkTime maximum, take the input as system of the kth subcarrier；Reject its remaining sub-carriers；

Wherein, ab represents that the receiving-transmitting chain of signal, k ∈ (1,30), m take 2,3,4 respectively and represent the in step 2 respectively (2-1-2), there is three kinds of situations of target to be measured, CSI in (2-1-3) and WI-FI transceiver network_1kpRepresent kth after enhancement process Initial CSI value on subcarrier, CSI_mkpAll CSI values during target jamming are had on kth subcarrier after representing enhancement process, ε_abkRepresent the ratio of each CSI value and initial CSI value, ab link when having target jamming in ab link in kth subcarrier Refer to that information transmitting terminal arrives the direct line of sight link of receiving terminal.

Step 4, utilizes pretreated CSI value in step 3, estimates the effective depth of target to be measured, pass through to be estimated out Target to be measured effective depth estimate match parameter δ₁And δ_t；

Owing to tooth shape model is model based on RSS information, and used herein be CSI information, here according to first Know position and the target effective elevation information of target, determine match parameter δ₁, by match parameter δ₁Obtain match parameter with to be measured The ratio relation of object height, estimates object height to be measured by EWMA method, finally obtains corresponding to differing heights target to be measured Match parameter, it is concrete that to calculate process as follows:

(4-1) utilize height and the positional information of the first known target, calculate theoretical attenuation value D on ab link_ab, formula As follows:

Wherein, D_abBe the theoretical attenuation value on ab link, c (v) and s (v) be fresnel integral, v is Fresnel-Kiel Hough diffraction parameter, its calculating formula is:

Wherein, t ∈ (-1,1), λ are signal wavelength, and h1 is the height of the first known target, d_a2Be the first known target away from Transmitting terminal distance, d_b2Be the first known target away from receiving terminal distance, ab link refers to that transmitting terminal is to directly regarding between receiving terminal Away from link；

(4-2) D obtained by (4-1)_ab, utilize following formula to calculate match parameter δ₁:

Wherein, CSI_2kpFor the CSI value under height target conditions single after pretreatment, CSI_1kpFor after pretreatment Initial CSI value, D_abIt is the theoretical attenuation value of ab link, δ₁For match parameter；

(4-3) known h1 and δ₁, obtain following ratio relation according to formula (1):

Wherein, h1 is the height of the first known target, and ω is ratio parameter；

(4-4) when there is target to be measured in WI-FI network, the method for utilization index weighted moving average (EWMA) is estimated The height of target to be measured, concrete formula is as follows:

Wherein, factor alpha represents that the change of weight, α ∈ (0,1), t are that CSI value gathers moment, S_tEstimate to be measured for t The height value of target, h1, h2 are respectively height and the height of the second known target of the first known target；

In order to reach to distinguish, the height of the first known target and the second known target is preferably according to experiment area here Altitudes selects, and if local average height is 1.7m, height variance is 0.2m, then select a height at about 1.7m Target and a height test the people of about 1.9m height；

(4-5) after carrying out twice known collection, can be by being derived by later height h value, for the most to be measured H value when target location calculates, and the δ made new advances by the h value derivation derived_tValue, makes theoretical attenuation and at that time actual decline Subtract coupling.

The ratio parameter ω obtained in height according to the target to be measured estimated in (4-4) and (4-3), utilizes following public Formula, it is thus achieved that the match parameter of differing heights target:

Wherein, δ_tFor estimating the match parameter of differing heights target.

There are certain error and cumulative errors in this method of estimation, it is therefore proposed that carry out after carrying out the experiment of about 10 times Error concealment, in order to reach positioning precision, concrete removing method is exactly to carry out initializing height collection again, so can directly pick Except error.

Step 5, utilizes match parameter to obtain the pad value of target to be measured, determines district residing for target to be measured by its pad value Territory；

Before location, it is impossible to the data according to gathering intuitively provide whether there is target to be measured, it is therefore desirable to before location First judge whether target to be measured, it is judged that method is:

(5-1) variances sigma of initial CSI value after pretreatment is set₁As threshold value, to be measured by what step 4 obtained The height of target and match parameter, utilize formula (1) calculated D_abIf, D_abMore than σ₁Time, represent that this target to be measured is in Detection region；Otherwise, it means that this target to be measured to be in detection zone overseas；

It is hypothetical owing to using tooth shape model: when the 55% of target to be measured in first Fresnel zone just to other Region effective, the half of target the most to be measured will in first Fresnel zone time could use above method.The most here need It is given at being to use in what situations, is handled as follows:

Wherein, f ∈ (1, t1), t1 are the number of target to be measured, D on ab link_abfRepresent and treat for f on ab link Survey the theoretical attenuation value of target, min (D_ab) represent on the 8th link the minima of theoretical attenuation value in target to be measured；

Wherein: if target to be measured location can use model, model Free Region is defined as model and can position region (Model Touchable Area, MTA), if target to be measured location can not use model, for model unusable area (Out Of Model Touchable Area, OMTA).

Step 6, according to region residing for the target to be measured obtained in step 5, positions target to be measured, wherein positions target to be measured Process include:

If (6-1) target to be measured is in MTA region, following formula is utilized to be calculated the range information d of target_a4、d_b4, further according to The deployment scenario of transmitting terminal receiving terminal, obtains target position information, and computing formula is as follows:

Wherein, c (v) and s (v) is fresnel integral, and its calculating formula is:

Wherein, h estimates height S in being step (4-4)_t, λ is signal wavelength, d_a4For target to be measured away from transmitting terminal distance, d_b4 For target to be measured away from receiving terminal distance, D_abfIt is the theoretical attenuation value of f target to be measured, D on ab link_abf1It it is ab chain The actual attenuation value of Lu Shang f target to be measured；

From above computational methods, the core concept of this method is the actual attenuation value in known link, by coupling Parameter δ_tTheoretical attenuation value D can be made_abfWith actual attenuation value D_abf1Approximately equal, match parameter δ_tSpy is there is with object height to be measured Fixed ratio relation, if can the height of known target to be measured, then can be obtained by this coupling corresponding to target to be measured ginseng Number, this method have employed known twice known target height and position, and the method for the height estimating follow-up target to be measured obtains Match parameter, thus position target to be measured.

In order to verify the feasibility of the method, applicant has also carried out confirmatory experiment, and applicant puts in WI-FI network One, target to be measured of Location-Unknown the most known, is positioned the position of this target to be measured, verifies above-mentioned algorithm by the method Accuracy.

In the present embodiment, using the WI-FI signal of 2.4GHz, therefore λ is 0.125m, can release d according to above-mentioned formula_aWith d_bAnd and product, when exist two with uplink time, concrete d can be obtained_aAnd d_b, in the deployment scenario according to sending and receiving end, Obtain the position of target to be measured；

(6-2) when target is in OMTA region, utilize the emphasis algorithm of influence area to position target to be measured in OMTA district Position in territory；

Wherein influence area emphasis algorithm particularly as follows:

(6-2-1) there is L bar link in the WI-FI transceiver network described in setting, determine in L bar link and can surround region Chain travel permit number n₁, calculate the pad value D of each of the links_q, then its weights are:

Wherein, q ∈ (1, n₁), D_qIt is the pad value of q article of link, ε_qIt is that the pad value of q article of link is at n₁In bar link Shared weights；

(6-2-2) it is weighted according to the point midway of each of the links, is shown below, obtain the position of OMTA zone location Confidence ceases:

Wherein, x_obj、y_objFor the position coordinates of target to be measured, x_q、y_qIt is the q article link point midway coordinate.

The present embodiment additionally provides experimental verification:

This sample plot point is selected in 8th floors laboratorys of Northwest University's Information Institute, and Experimental Area size is 6*12m, such as Fig. 3 institute Showing, the subscriber station point of five-pointed star in the drawings carries out positioning experiment.Testing for convenience, the transmitting and receiving device selected here is all Carrying the machine of Intel5300AGN wireless network card, operating system is Ubuntu10.04.4, can serve as signal and sends and connect Receiving, such as Fig. 3, data processor is the computer of windowsOS.When signal sends, simply simulate wlan device, do not do it It changes.And in order to ensure to cover, only choose one group of transmission here every time, remaining is respectively as the equipment of reception.This experiment is altogether Gather the situation of 88 some positions.The position carrying out distinct methods determines.

Simultaneously herein for the effect of embodiment the method, and existing method compares.

For RASS system, RTI system and Alico system, use RSS information, this information in CSI information also Can obtain, and calculating can be extracted.RASS system is that equipment is divided into delta-shaped region, is analyzed, by 3 in experiment The CSI transmitter-receiver of antenna is modeled to 3 sending and receiving devices, so can form multiple delta-shaped region, and so set The software of meter also can realize.RTI system is to divide link impact, uses changing of RSS information in experiment To result, and owing to RTI system is to obtain result according to the figure affecting link, due to the deployment on only both sides in experiment, It is assumed that the midpoint affecting link is final goal point.And Alico system is system immediate with article, here dispose and Its deployment way is the most about the same, only because plant issue, it is impossible to excessively dense deployment, it may appear that do not account in article First Fresnel zone outside region, and herein for convenience of calculation and effectively contrast, know all of device location here, do not have There is the estimation to position.Pilot system be the information of CSI is learnt after carry out Characteristic Contrast location, feature is subcarrier Cross-correlation function.And CWF system uses CSI information to solve, use the mode of poll to carry out data acquisition here, make system Under the conditions of limited deployment facility, there is link as much as possible, and determine each link acquisition 20 data, after completing more Change sending end.

As shown in Figure 4, the statistical result of distinct device quantity is as shown in Figure 5 for experimental result.

RTI system effect is worst as seen from the figure, and middle position error reaches 3.5m, and RASS system is secondly, and middle position error is The effect of 3m, Alico system is reasonable in RSS information, and middle position error reaches 2m, Pilot system in curve, essence Spend the highest, but middle position error ratio CWF system is lower.Pilot system is 1.4m, and CWF system is 1.3m.And according to above Narration, the location of CWF system is divided into MTA and OMTA region, and OMTA region is to estimate with midpoint method, such error Very big, meanwhile, some region belonging to point only has 1 group of link to cover, and accuracy rate so can be caused to decline.Therefore at CDF In curve, some effect can be deteriorated.But its result is much better than to use the location of RSS information fingerprint, and also ratio uses similar side The Alico locating effect of method to be got well, and therefore the positioning precision of this system is in higher level.

This illustrates no matter multipath is how many in indoor, and CWF system in this paper has reasonable precision, though the method Poorer than Pilot system based on finger print information, but the fingerprint collecting that need not Pilot method early stage due to it, save Substantial amounts of time.

From fig. 5, it can be seen that compare here is the result of middle position error.Through contrast it is found that distinct methods Position error is different, and error reduces along with the increase of transmission equipment quantity, and this is due to along with the increase of number of devices, Overlay area is increasing, and the locating effect of above several method is all relevant with overlay area.Under any deployment density, RTI system is worst, and this is owing to RTI system is when proposing, and is to need around region one, equipment is enclosed deployment, merely just deploys Both sides, and also 8 points are outside region, and and when experiment, the mode that RTI system is not disposed for two rows is improved, result Can be far short of what is expected.RASS system effect is the best, but compares other several method and still have gap, and mainly its dispositions method is little by three Dihedral, although triangle can also be formed here, but can not be as the equilateral triangle in literary composition and the hexagon especially formed.Remaining three Plant method effect, at some points, there is similar effect, be owing to, on some point, three kinds of methods can obtain more accurately Result.Alico system is due to multipath and the instability of RSS information, and effect is general, but owing to there being certain process, ratio uses RSS The method of information fingerprint detection to be got well.And for Pilot system, its effect is generally best, reason and a upper experiment phase With, owing to acquiring fingerprint, the impact of multipath need not be considered.But it is also due to gather fingerprint, very labor intensive.Finally, right In CWF system in this paper, under low-density is disposed, effect can be poorer than Pilot, when this disposes mainly due to low-density, CWF system has a lot of position to be in OMTA region, and error so can be caused to increase.And Pilot system is finger print information, this The region, effectively location of the system of kind positions region greatly than the CSI ellipse typically drawn, therefore its effect when low-density is disposed Better than CWF system.

Here definition: be defined as False Negative (FN) in Computer Subject:

POM=Σ target in region but the probability being judged as not in region

Result as shown in Figure 6, finds along with the reduction of coverage, and sharply increasing does not occurs in the result of CWF system, Reason is the location in its OMTA region, and in an experiment, a lot of the putting near deployment region does not meets positioning requirements, but also There is certain change, therefore cannot detect and can detect in CWF in Pilot.

Under maximum coverage rate, the loss of CWF is higher than Pilot, uses when location mainly due to CWF here Boundary be as object variations, and Pilot is by arranging in advance.The localization method of Pilot can detect more Point, but simultaneously positioning precision also can decline.

After compared for existing method, it can be deduced that system can obtain the positioning precision that comparison is high.

Claims (7)

1. a non-fingerprint passive type localization method based on WI-FI signal, it is characterised in that comprise the following steps:

Step 1, builds WI-FI transceiver network, including transmitting terminal a and receiving terminal b；

Step 2, in the WI-FI transceiver network built, gathers the CSI value in WI-FI signal, and described CSI value is divided into test CSI value and actual CSI value, wherein test CSI value is:

(2-1-1) WI-FI transceiver network does not has initial CSI value during target；

(2-1-2) in WI-FI transceiver network, put into the CSI value during target of the first known altitude and position；

(2-1-3) putting into the CSI value during target of the second known altitude and position in WI-FI transceiver network, wherein second Know that the height of the height of target and the first known target is different；

Actual CSI value refers to: the CSI value gathered when there is target to be measured in WI-FI transceiver network；

The all CSI values collected in step 2 are utilized 3 σ filtering methods to be filtered, to the CSI after Filtering Processing by step 3 Value carries out pretreatment, eliminates multipath effect；

Step 4, utilizes pretreated CSI value in step 3, estimates the effective depth of target to be measured, and that passes through to be estimated out treats The effective depth surveying target estimates match parameter；

Step 5, utilizes match parameter to obtain the pad value of target to be measured, determines region residing for target to be measured by its pad value；

Step 6, according to region residing for the target to be measured obtained in step 5, positions target.

2. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 1, it is characterised in that in step 1 Described WI-FI transceiver network at least needs both links.

3. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 1, it is characterised in that in step 3 Described pre-treatment step includes:

(3-1) the multipath fading signal in described filtered signal is carried out enhancement process, processes formula as follows:

$\left\{\begin{array}{c}{\mathrm{\ρ}}_{ki}=\frac{{\mathrm{csi}}_{mki}}{{\Σ}_{i=1}^n{\mathrm{csi}}_{mki}}\\ {\mathrm{CSI}}_{mkp}=\sqrt{{\mathrm{\ρ}}_{k1}*{\mathrm{csi}}_{mk1}^2+{\mathrm{\ρ}}_{k2}*{\mathrm{csi}}_{mk2}^2+\mathrm{...}+{\mathrm{\ρ}}_{kn}*{\mathrm{csi}}_{mkn}^2}\end{array}\right.$

Wherein, n is the group number of a period of time interior CSI value gathered, i ∈ (1, n), each group of CSI value is made up of 30 sub-carrier waves, K ∈ (1,30), csi_mkiRepresent the CSI value that kth subcarrier collects on i-th time point, ρ_kiFor on kth subcarrier The CSI value that i-th time point collects accounts for the proportion of the CSI value that on this subcarrier, all time points collect, CSI_mkpFor kth CSI value after enhancement process on individual subcarrier；

Wherein, m takes 1,2,3,4 respectively, and it represents (2-1-1), (2-1-2), (2-1-3) in step 2 and WI-FI transmitting-receiving respectively Network exists four kinds of situations of target to be measured；

(3-2) equation below is utilized, the too strong signal in CSI value after rejecting (3-1) process:

${\mathrm{\ϵ}}_{abk}=\frac{{\mathrm{CSI}}_{mkp}}{{\mathrm{CSI}}_{1kp}}$

Work as ε_abk＜ 1 and ε_abkTime maximum, take the input as system of the kth subcarrier；Reject its remaining sub-carriers；

Wherein, ab represents that the receiving-transmitting chain of signal, k ∈ (1,30), m take 2,3,4 respectively and represent (the 2-1-in step 2 respectively 2), there is three kinds of situations of target to be measured, CSI in (2-1-3) and WI-FI transceiver network_1kpAfter representing enhancement process, kth carries Initial CSI value on ripple, CSI_mkpAll CSI values during target jamming, ε is had on kth subcarrier after representing enhancement process_abkTable Showing the ratio of each CSI value and initial CSI value when having target jamming in ab link in kth subcarrier, ab link refers to letter Breath transmitting terminal is to the direct line of sight link of receiving terminal.

4. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 3, it is characterised in that in step 4 The concrete calculating process of described match parameter is as follows:

(4-1) utilize height and the positional information of the first known target, calculate theoretical attenuation value D on ab link_ab, formula is such as Under:

$D_{ab}=20\log \[\frac{1}{\sqrt{2}}\·e^{j\frac{\mathrm{\π}}{4}}\[(\frac{1}{2}-c(v\left)\right)-j(\frac{1}{2}-s(v\left)\right)\]\]---\left(1\right)$

Wherein, D_abBe the theoretical attenuation value on ab link, c (v) and s (v) be fresnel integral, v is Fresnel-Kirchhoff Diffraction parameter, its calculating formula is:

$c\left(v\right)={\∫}_0^{v}cos\left(t^2\right)dt$

$s\left(v\right)={\∫}_0^v\sin \left(t^2\right)dt$

$v=h1\sqrt{\frac{2\·(d_{a2}+d_{b2})}{\mathrm{\λ}\·d_{a2}\·d_{b2}}}$

Wherein, t ∈ (-1,1), λ are signal wavelength, and h1 is the height of the first known target, d_a2It is that the first known target is away from transmission End distance, d_b2Be the first known target away from receiving terminal distance, ab link refers to that information transmitting terminal arrives the direct sighting distance chain of receiving terminal Road；

(4-2) D obtained by (4-1)_ab, utilize following formula to calculate match parameter δ₁:

$\frac{1}{{\mathrm{\δ}}_1}mean({\mathrm{CSI}}_{1kp}-{\mathrm{CSI}}_{2kp})\≈D_{ab}$

Wherein, CSI_2kpFor the CSI value of the first known target after pretreatment, CSI_1kpFor through pretreated initial CSI Value, D_abIt is the theoretical attenuation value of ab link, δ₁For match parameter；

(4-3) known h1 and δ₁, obtain following ratio relation according to formula (1):

${\log }_{10}\sqrt{\frac{1}{h1}}\∝\frac{1}{{\mathrm{\δ}}_1}\⇒\frac{\frac{1}{{\mathrm{\δ}}_1}}{{\log }_{10}\sqrt{\frac{1}{h1}}}=\mathrm{\ω}$

Wherein, h1 is the height of the first known target, and ω is ratio parameter；

(4-4) when there is target to be measured in WI-FI network, the method for utilization index weighted moving average (EWMA) is estimated to be measured The height of target, concrete formula is as follows:

$\left\{\begin{array}{c}{S}_1=h1\\ S_2=h2\\ S_t=\mathrm{\α}\·S_{t-1}+(1-\mathrm{\α})\·S_{t-2}\end{array}\right.$

Wherein, factor alpha represents that the change of weight, α ∈ (0,1), t are that CSI value gathers moment, S_tTarget to be measured is estimated for t Height value, h1, h2 are respectively height and the height of the second known target of the first known target；

(4-5) according to the ratio parameter ω obtained in the height of the target to be measured estimated in (4-4) and (4-3), utilize as follows Formula, it is thus achieved that match parameter δ of differing heights target_t:

${\log }_{10}\sqrt{\frac{1}{h}}\∝\frac{1}{{\mathrm{\δ}}_t}\⇒\frac{\frac{1}{{\mathrm{\δ}}_t}}{{\log }_{10}\sqrt{\frac{1}{S_t}}}=\mathrm{\ω}$

Wherein, δ_tFor estimating the match parameter of differing heights target.

5. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 4, it is characterised in that step 5 Concrete steps include:

(5-1) variances sigma of initial CSI value after pretreatment is set₁As threshold value, by the target to be measured that obtains in step 4 Height and match parameter, utilize formula (1) to be calculated D_abIf, D_abMore than σ₁Time, represent that this target to be measured is in detection region； Otherwise, it means that this target to be measured to be in detection zone overseas；

If (5-2) target to be measured is in detection region, it is handled as follows:

Wherein, f ∈ (1, t1), t1 are the number of target to be measured, D on ab link_abfRepresent f mesh to be measured on ab link Target theoretical attenuation value, min (D_ab) represent on ab link the minima of theoretical attenuation value in target to be measured.

6. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 5, it is characterised in that in step 6 The process of location target includes:

If (6-1) target to be measured is in MTA region, following formula is utilized to be calculated the range information d of target to be measured_a4、d_b4, further according to sending out The deployment scenario of sending end receiving terminal, obtains target position information, and computing formula is as follows:

$\left\{\begin{array}{c}{D}_{abf}=\frac{1}{{\mathrm{\δ}}_t}mean({\mathrm{CSI}}_{1kp}-{\mathrm{CSI}}_{4kp})\\ D_{abf1}\≈D_{abf}\\ D_{abf1}=20log\[\frac{1}{\sqrt{2}}\·e^{j\frac{\mathrm{\π}}{4}}\[(\frac{1}{2}-c(v\left)\right)-j(\frac{1}{2}-s(v\left)\right)\]\]\\ v=h\sqrt{\frac{2\·(d_{a4}+d_{b4})}{\mathrm{\λ}\·d_{a4}\·d_{b4}}}\end{array}\right.$

Wherein, c (v) and s (v) is fresnel integral, and its calculating formula is:

$c\left(v\right)={\∫}_0^{v}cos\left(t^2\right)dt$

$s\left(v\right)={\∫}_0^v\sin \left(t^2\right)dt;$

Wherein, h estimates height S in being step (4-4)_t, λ is signal wavelength, d_a4For target to be measured away from transmitting terminal distance, d_b4For treating Survey target away from receiving terminal distance；

(6-2) when target is in OMTA region, utilize the emphasis algorithm of influence area to position target to be measured in OMTA region Position.

7. non-fingerprint passive type localization method based on WI-FI signal as claimed in claim 6, it is characterised in that step (6- 2) the emphasis algorithm of the influence area described in particularly as follows:

(6-2-1) there is L bar link in the WI-FI transceiver network described in setting, determine the chain that can surround region in L bar link Travel permit number n₁, calculate the pad value D of each of the links_q, then its weights are:

${\mathrm{\ϵ}}_q=\frac{D_q}{{\mathrm{\Σ}}_{q=1}^{n_1}{D}_q}$

Wherein, q ∈ (1, n₁), D_qIt is the pad value of q article of link, ε_qIt is that the pad value of q article of link is at n₁Shared by bar link Weights；

(6-2-2) it is weighted according to the point midway of each of the links, is shown below, obtain the position letter of OMTA zone location Breath:

$x_{obj}=\underset{q=1}{\overset{n_1}{\Σ}}{\mathrm{\ϵ}}_q\×x_q,y_{obj}=\underset{q=1}{\overset{n_1}{\Σ}}{\mathrm{\ϵ}}_q\×y_q$

Wherein, x_obj、y_objFor the position coordinates of target to be measured, x_q、y_qIt is the q article link point midway coordinate.