CN105933867A - Passive real-time indoor two-anchor-point positioning method based on channel state information - Google Patents

Abstract

The invention discloses a passive real-time indoor two-anchor-point positioning method based on channel state information. The method comprises the following steps of: (1), continuously detecting CSI information of a subcarrier, judging whether a user moves or not according to the CSI information, if so, turning to the step (2), and otherwise, continuously detecting the CSI information; (2), obtaining a change rate of the signal propagation path length due to movement of the user through the CSI information, and judging a change direction of the signal propagation path length due to movement of the user; and (3), according to locations of two anchor points and a previously predicted location, obtaining a location, where the user is currently, by using the change rate of the signal propagation path length and the change direction of the signal propagation path length. On the premise of passive positioning, real-time online positioning can be realized; furthermore, real-time positioning can be realized only in need of two anchor points; and thus, the basic positioning condition is reduced to the most extent.

Description

The passive type two real-time indoor orientation method of anchor point based on channel condition information

Technical field

The invention belongs to wireless communication field, more particularly, to a kind of passive type two anchor based on channel condition information The real-time indoor orientation method of point.

Background technology

Since with location technology that global positioning system (Global positioning system, be called for short GPS) is representative Since appearance, its service efficiently, easily and fast with accurately makes the life of people there occurs huge change, has driven a collection of The fast development of application and service.Indoor positioning technologies in the continuity of indoor environment, has good application as location technology Prospect.Occur along with social networks can basis based on location-based service (Location Based Service is called for short LBS) The daily rule of life of people makes the life style of personalization.And indoor positioning technologies can improve location-based further Service, it is possible to help people to complete various numerous and diverse, time-consuming task.But traditional location technology is due to the limitation of know-why Property, the most preferable at the locating effect of indoor environment.Indoor environment is covered due to body of wall, and gps signal passes meeting after building Substantially dying down, so indoor environment is not suitable for using GPS technology to position, the indoor of the most commonly used is WiFi are fixed Method for position.

The indoor positioning being currently based on WiFi can be largely classified into Fingerprint Model and propagation model.

Fingerprint Model method is a kind of model based on study, and application mode matching technique, according to the measured value of position With having been observed by the measured value of position as a comparison, then determine position according to match condition, this model is in location Effect aspect is satisfactory, CSI information, however it is necessary that substantial amounts of work to go to a given environment and sets up location model, and And after environment changes, the model set up will be the most applicable before, need amendment.In order to solve above-mentioned to set up asking of fingerprint base Topic, has the method that Many researchers proposes to utilize mass-rent to update fingerprint base, it is allowed to positioning result is evaluated and revises by user, During the maintenance making user participate in fingerprint base while enjoying positioning result updates, this method can reduce professional The work of in-site measurement, however it is necessary that dependence user, if user does not provide information, then cannot set up and update fingerprint base.

The feature that propagation model method is dependent on analyzing in signal communication process calculates propagation distance, and conventional feature comprises and connects Receive signal strength signal intensity instruction (Received Signal Strength Indication is called for short RSSI), angle of arrival (Angel Of Arrival, is called for short AOA) and arrival time (Time of Arrival is called for short TOA).Three location, limit based on RSSI The attenuation model utilizing RSSI realizes location, but the numerical value of RSSI is not constant, does not moves even if sending and receive both sides, It is likely to that serious concussion occurs, and needs human body to carry an equipment in location.AOA is according to the incidence side receiving signal Always determine position, however it is necessary that a special antenna array is to measure angle value.TOA measures the propagation time of signal, logical Spending the propagation time and extrapolate the equipment distance to AP, but TOA needs point-device clock to synchronize, small difference also can be led Causing the biggest error, equally, the method is also required to human body and carries an equipment.RSSI and TOA method in addition to drawbacks described above, Position fixing process is required at least 3 anchor points.

Summary of the invention

For disadvantages described above or the Improvement requirement of prior art, the invention provides a kind of quilt based on channel condition information The real-time indoor orientation method of dynamic formula two anchor point, it is intended that solve excessively to rely on use present in existing indoor orientation method Family participate in in-convenience in use, need do substantial amounts of in-site measurement in advance and necessarily require user to carry the technology of smart mobile phone Problem, the present invention, by utilizing physical layer channel conditions information, both guarantees to realize passive type location, real-time online location, Have only to two anchor points just can realize positioning in real time simultaneously, thus decrease the basic condition of location to greatest extent.

For achieving the above object, according to one aspect of the present invention, it is provided that a kind of based on channel condition information passive The real-time indoor orientation method of formula two anchor point, comprises the following steps:

(1) persistently detect the CSI information of subcarrier, and judge whether this user there occurs movement according to this CSI information, as Fruit there occurs movement, then proceed to step (2), otherwise continue detection CSI information；

(2) the signal propagation path length caused owing to user moves by this CSI acquisition of information rate of change and Owing to user moves the change direction of the signal propagation path length caused；

(3) initial position before moving according to the position between two anchor points and user, and use the letter that step (2) obtains The rate of change of number diffusion path length and the change direction of signal propagation path length obtain the position that user is currently located.

Preferably, in step (1), persistently detect the CSI information of subcarrier, and judge that this user is according to this CSI information No there occurs that this process mobile specifically includes following sub-step:

(1-1) use PCA method, the CSI information on all subcarriers is split, the subcarrier in same time period Stream composition matrix H, calculates H^TThe characteristic value of × H and characteristic vector q thereof_i, wherein i is the numbering of matrix exgenvalue；

(1-2) at the H obtained^TThe characteristic value of × H and characteristic vector q thereof_iIn, select to determine that Second Largest Eigenvalue is corresponding Characteristic vector q₂With corresponding principal component h₂=H × q₂；

(1-3) the characteristic vector q determined according to step (1-2)₂Calculate its average δ_q2:

${\mathrm{\δ}}_{q_2}=\frac{1}{s-1}\underset{l=2}{\overset{s}{\Σ}}|q_2\left(l\right)-q_2(l-1)|$

Here s represents the number of all subcarriers.|q₂(l)-q₂(l-1) | it is the coefficient correlation of adjacent sub-carrier CSI information Difference.

(1-4) principal component h is calculated₂VarianceAnd according to the variance obtainedThe average obtained with step (1-3)Judge whether user is moved.

Preferably, in step (1-4), whenDuring more than a threshold value, it is judged that user starts mobile, wherein said threshold value It it is measurement gained in quiet environment2 times of average.

Preferably, step (2) specifically includes following sub-step:

(2-1) the static CFR of acquisition subcarrier:

$H_s(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}{\mathrm{\Σ}}_{k\∈P_s}{a}_k(f,t)e^{-j2{\mathrm{\πf\τ}}_k\left(t\right)}$

Wherein H_s(f, t) centered by frequency be the channel frequency response that the subcarrier of f obtains in the measurement of time t, f represents son The centre frequency of carrier wave, P_sRepresent the propagation path set that the signal propagation path length of subcarrier does not changes, a_k(f,t) Represent initial phase and the decay of amplitude, the e of kth bar propagation path^-j2πΔftRepresent the phase drift that frequency shift (FS) causes,Represent the phase drift on kth bar propagation path, τ_kT () represents the propagation delay on kth bar propagation path；

(2-2) the dynamic CFR of acquisition subcarrier:

$H_d(f,t)={\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t)e^{-j2{\mathrm{\πd}}_k\left(t\right)/\mathrm{\λ}}$

Wherein P_dRepresent the propagation path set that the signal propagation path length of subcarrier changes, d_kT () represents process The kth bars diffusion path length of human body reflection.

(2-3) dynamic CFR that the static CFR obtained according to step (1) and step (2) obtain obtains total CFR:

$H(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}\left(H_s\right(f,t)+{\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t\left)e^{-j\frac{2{\mathrm{\πd}}_k\left(t\right)}{\mathrm{\λ}}}\right);$

(2-4) rate of change of signal propagation path is determined according to the frequency response of subcarrier；

(2-5) determine that user moves the signal propagation path caused according to the rate of change of the diffusion path length determined long The change direction of degree.

Preferably, step (2-4) specifically,

First, to H (f, t) delivery square:

$\begin{array}{c}|H(f,t){|}^2=\underset{k\∈P_d}{\Σ}2|H_s\left(f\right)a_k(f,t)|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|a_k(f,t)a_l(f,t)\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}|a_k(f,t){|}^2+H_s{\left(f\right)}^2\end{array}$

Wherein k, l represent different propagation paths respectively, and λ represents wavelength, θ_skRepresent propagation path s's and propagation path k Initial phase sum, θ_klRepresent the initial phase sum of propagation path k and propagation path l；

Secondly, the power of CFR is done wavelet transformation, to obtain a series of frequency values: choose in a series of frequency values The frequency values of amplitude maximum_fmax, and calculate v_k=f_max× λ, wherein v_kRepresent the propagation path of the signal reflected by user's body Rate of change.

Preferably, step (2-5) specifically,

First initial phase and the decay of amplitude of kth bar propagation path are obtainedWherein G_rAnd G_tIt is to connect Receiving and launch the signal power gain of antenna, λ is the wavelength launching signal, d_kIt is the propagation distance on kth bar propagation path, Γ_i For reflection factor；

Secondly, this decay is substituted in the power expression of step (2-4):

$\begin{array}{c}|H(f,t){|}^2=\underset{k\∈P_d}{\Σ}2|H_s\left(f\right)\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}\frac{G_t{G}_r{\mathrm{\Γ}}_l{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_l\right)}^2}\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}{|}^2+H_s{\left(f\right)}^2\end{array}$

Finally, determine that the amplitude of CSI power signal is to reduce or increase, when the amplitude of CSI power signal reduces, then The change direction of the propagation path of signal is augment direction, when the amplitude of CSI power signal increases, then and the propagation path of signal Change direction for reduce direction.

Preferably, step (3) includes following sub-step:

(3-1) according to the rate of change of signal propagation path and the change direction of signal propagation path length determine through with The diffusion path length of the signal of family reflection:

(3-2) length of the signal propagation path determined according to position and the step (3-1) of two anchor points determines the position of user Put.

Preferably, step (3-2) specifically includes:

(3-2-1) using supporting to a wifi transmitter and wifi receiver use of putting together as the first anchor point, One the 2nd wifi receiver as the second anchor point, using the position coordinates at user place as a point on elliptic wheel profile, Using the first anchor point and the second anchor point as two focuses of ellipse, build an ellipse, in this ellipse, user to two focus Length d that distance sum is signal propagation path_k(t)；

(3-2-2) according to said method and use the anchor point 1 of wifi transmitter and a wifi receiver position to make For the center of circle, constructing a radius is d_kT the circle of ()/2, one of two intersection points between this circle and above-mentioned ellipse are user institute Position；

(3-2-3) according to determining the change direction of signal propagation path length and combining above-mentioned two intersection points determined, really Determine the position that user eventually arrives at.

Preferably, step (3-2-3) is if specifically, user arrives an intersection point above, then its change direction is to reduce, If user arrives a following intersection point, then its change direction should be to increase.

It is another aspect of this invention to provide that it is the most indoor to provide a kind of passive type based on channel condition information two anchor point Alignment system, including:

According to this CSI information, first module, for persistently detecting the CSI information of subcarrier, and judges whether this user sends out Give birth to movement, if it occur that movement, then proceeded to the second module, otherwise continue detection CSI information；

Second module, the change of the signal propagation path length for being caused owing to user moves by this CSI acquisition of information Change speed and owing to user moves the change direction of the signal propagation path length caused；

3rd module, the initial position before moving according to the position between two anchor points and user, and use second It is current that the rate of change of the signal propagation path length that module obtains and the change direction of signal propagation path length obtain user The position at place.

In general, by the contemplated above technical scheme of the present invention compared with prior art, it is possible to show under acquirement Benefit effect:

(1) present invention realizes user location by the length measuring the signal propagation path reflected by human body, it is possible to solve Existing indoor orientation method certainly excessively relies on user and participates in, need the technical problem of user's Portable device.

(2) present invention realizes user location, energy by the length calculating the signal propagation path reflected by human body in real time Enough solve the technical problem needing to do substantial amounts of measurement work the most at the scene in existing indoor orientation method.

(3) present invention is by measuring signal propagation path length change direction in real time, it is only necessary to two anchor points just can be real Location during reality, it is possible to solving the existing indoor positioning that realizes at least needs the technical problem of three anchor points.

(4) present invention fully excavates the channel frequency response information of physical layer, decreases indoor positioning to the full extent Basic condition, it is possible to solve the technical problem of the too much restrictive condition of existing indoor positioning.

Accompanying drawing explanation

Fig. 1 is the flow chart of present invention passive type based on the channel condition information two real-time indoor orientation method of anchor point.

Fig. 2 be the present invention judge the flow chart of steps that people moves.

Fig. 3 is that the correct of the present invention judges the probability schematic diagram that people moves.

Fig. 4 be the present invention judge the signal schematic representation that people moves.

Wherein: a is original CSI information schematic diagram, b is adjacent sub-carrier and the ratio schematic diagram of CSI information variance.

Fig. 5 is the flow chart of steps of the rate of change of the calculating signal propagation path length of the present invention

Fig. 6 is the signal schematic representation of the rate of change of the calculating signal propagation path length of the present invention, wherein: a is original CSI information schematic diagram, b is the time-frequency figure of CSI information.

Fig. 7 is the flow chart of steps of the calculating signal propagation path length change direction of the present invention.

Fig. 8 is the signal schematic representation of the calculating signal propagation path length change direction of the present invention, wherein: a is through PCA Filtered signal graph, b is the cumulative figure of CSI absolute value.

Fig. 9 be the present invention by two anchor points location schematic diagrames.

Detailed description of the invention

In order to make the purpose of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, right The present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, and It is not used in the restriction present invention.If additionally, technical characteristic involved in each embodiment of invention described below The conflict of not constituting each other just can be mutually combined.

It is emphasized that heretofore described anchor point, refer to the point of known self-position.

Shown in Figure 1, present invention passive type based on the channel condition information two real-time indoor orientation method of anchor point includes The following step:

(1) channel condition information (Channel State Information is called for short CSI) of subcarrier is persistently detected, and Judge whether this user there occurs movement according to this CSI information, if it occur that movement, then proceed to step (2), otherwise continue inspection Survey CSI information；

(2) the signal propagation path length caused owing to user moves by this CSI acquisition of information rate of change and Owing to user moves the change direction of the signal propagation path length caused；

(3) (other hands can be passed through in this position to the initial position before moving according to the position between two anchor points and user Section obtains, such as: place an inductor on doorway, if there being people to enter, then using this position as previous prediction position), and The rate of change of the signal propagation path length that use step (2) obtains and the change direction of signal propagation path length obtain to be used The position that family is currently located.

Below in conjunction with concrete formula principle etc., the step in the inventive method is described in detail.

As shown in Figures 2 to 4, in the step (1) in the inventive method, persistently detect the CSI information of subcarrier, and according to this CSI information judges whether this user there occurs that this process mobile specifically includes following sub-step:

(1-1) principal component analysis (Primary Component Analysis is called for short PCA) method is used, all sons CSI information on carrier wave was split with 1 second for time interval, the subcarrier stream composition matrix H in same time period, calculated H^T The characteristic value of × H and characteristic vector q thereof_i(wherein i is the numbering of matrix exgenvalue)；

The principle of this step take into account the correlation of CSI information adjacent sub-carrier, i.e. CSI when nobody moves, respectively CSI information on individual subcarrier is incoherent.But, when occurring that user moves, identical owing to comprising on each subcarrier Human motion information, therefore the CSI information on each subcarrier is relevant, according to these phenomenons be assured that people moves time Between interval.In the present invention, use Intel 5300WiFi NIC to obtain the CSI information on subcarrier, altogether can extract The CSI information of 30 subcarriers.

(1-2) at the H obtained^TThe characteristic value of × H and characteristic vector q thereof_iIn, select to determine that Second Largest Eigenvalue is corresponding Characteristic vector q₂With corresponding principal component h₂=H × q₂；

Because the interference of noise, cause eigenvalue of maximum characteristic of correspondence vector q₁Can not be used, therefore in this step Use Second Largest Eigenvalue characteristic of correspondence vector q₂With corresponding principal component h₂=H × q₂.Nobody moves when, special Levy vector q₂Random fluctuation, this is because the CSI information on subcarrier is independent, but when occurring that people moves when, respectively CSI information on individual subcarrier becomes relevant, q₂Change the most smooth.

(1-3) the characteristic vector q determined according to step (1-2)₂Calculate its average δ_q2:

${\mathrm{\δ}}_{q_2}=\frac{1}{s-1}\underset{l=2}{\overset{s}{\Σ}}|q_2\left(l\right)-q_2(l-1)|$

Here s represents the number of all subcarriers.|q₂(l)-q₂(l-1) | it is the coefficient correlation of adjacent sub-carrier CSI information Difference；

(1-4) principal component h is calculated₂VarianceAnd according to the variance obtainedThe average obtained with step (1-3)Judge whether user is moved；CSI information is temporary stabilization when nobody moves, but, when occurring that people moves Time, CSI information can start fluctuation, when nobody moves, principal component h₂There is a least fluctuation, but when the person moves, h₂Have more Big fluctuation.

WhenMore than threshold value, (this threshold value is measurement gained in quiet environment2 times of average) time, permissible Judge that user starts mobile.Fig. 4 (a) is CSI hum pattern, and Fig. 4 (b) isValue, whenDuring more than threshold value, it is determined that use Family is being moved.

The accuracy judged by the method is as it is shown on figure 3, accuracy the most very can be reached.

If Fig. 5 is to shown in 8, the step (2) of the inventive method includes following sub-step:

(2-1) static channel bin response (Channel frequency response, the abbreviation of subcarrier are obtained CFR):

$H_s(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}{\mathrm{\Σ}}_{k\∈P_s}{a}_k(f,t)e^{-j2{\mathrm{\πf\τ}}_k\left(t\right)}$

Here H_s(f, t) centered by frequency be the channel frequency response that the subcarrier of f obtains in the measurement of time t, wherein, f Represent the centre frequency of this subcarrier, P_sRepresent the propagation path collection that the signal propagation path length of subcarrier does not changes Close, a_k(f, t) is a plural number, represents initial phase and the decay of amplitude, the e of kth bar propagation path^-j2πΔftRepresent that frequency is inclined Move the phase drift caused,Represent the phase drift on kth bar propagation path, τ_kT () represents kth bar propagation path On propagation delay；

(2-2) the dynamic channel frequency response of acquisition subcarrier:

$H_d(f,t)={\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t)e^{-j2{\mathrm{\πd}}_k\left(t\right)/\mathrm{\λ}}$

Wherein, P_dRepresent the propagation path set that the signal propagation path length of subcarrier changes, d_kT () represents warp Cross the kth bars diffusion path length of human body reflection.

(2-3) frequency response of subcarrier being divided into the CFR of static state and dynamic CFR sum, dynamic CFR is expressed as H_d(f, T), static CFR can be expressed as H_s(f t), represents the CFR sum that signal propagation path does not change, the most total CFR Can be expressed as:

$H(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}\left(H_s\right(f,t)+{\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t\left)e^{-j\frac{2{\mathrm{\πd}}_k\left(t\right)}{\mathrm{\λ}}}\right)$

When user moves, H_s(f t) is a constant, H_d(f, phase and amplitude t) all can change.

(2-4) rate of change of signal propagation path is determined according to the frequency response of subcarrier；

When people moves, H_s(f t) is a constant, H_d(f, phase and amplitude t) all can change.If user with Certain speed moves, and the kth bar diffusion path length through human body reflection at short notice can be with speed v_kChange, d_k T () represents the kth bar propagation path diffusion path length at time t, thenWherein d_k(0) represent that user moves Front kth bar diffusion path length.

This step specifically, first, to H (f, t) delivery square, the power of CFR can be expressed as:

$\begin{array}{c}|H(f,t){|}^2=\underset{k\∈P_d}{\Σ}2|H_s\left(f\right)a_k(f,t)|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|a_k(f,t)a_l(f,t)\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}|a_k(f,t){|}^2+H_s{\left(f\right)}^2\end{array}$

Here k, l represent different propagation paths respectively, and λ represents wavelength, θ_skRepresent propagation path s's and propagation path k Initial phase sum, θ_klRepresent the initial phase sum of propagation path k and propagation path l.From formula above it can be seen that total CFR power is a constant and a series of sinusoidal signal sum.The frequency of sinusoidal signal is diffusion path length rate of change Function, therefore can extrapolate the rate of change of diffusion path length by the frequency of these sinusoidal signals.

Secondly, the power of CFR is done wavelet transformation, to obtain a series of frequency values:Owing to health reflects Signal amplitude maximum, therefore in a series of frequency values, choose the frequency values f of amplitude maximum_max, and calculate v_k=f_max× λ, Here v_kRepresent the rate of change of the propagation path of the signal reflected by user's body.

Fig. 6 (a) shows the untreated time-domain diagram of signal, after LPF and PCA process, enters the signal after processing Row wavelet transformation, it can be seen that the frequency content of sinusoidal signal, such as Fig. 6 (b).Rate of change due to signal propagation path length For v=f_max× λ, here f_maxRepresenting the frequency of amplitude maximum in CSI power signal, λ represents the wavelength of subcarrier, for 2.4GHZ, λ=0.125m.Due to f_maxIt is concentrated mainly between 5HZ～20HZ, the change of signal propagation path length can be obtained Changing speed is 0.625m/s～2.5m/s.Here people is at the uniform velocity to move, when people from router and notebook farther out time, signal passes Broadcasting path and change speed relatively greatly, when people is close to router and notebook, signal propagation path length changes speed relatively Little.

(2-5) determine that user moves the signal propagation path caused according to the rate of change of the diffusion path length determined long The change direction (i.e. signal propagation path length increases or reduces) of degree；

Due toWherein, G_rAnd G_tBeing the signal power gain receiving and launching antenna, λ is to launch signal Wavelength, d_kBeing the propagation distance on kth bar propagation path, reflection factor is Γ_i。

In substitution step (2-4) in the power expression of CFR, can obtain:

$\begin{array}{c}|H(f,t){|}^2=\underset{k\∈P_d}{\Σ}2|H_s\left(f\right)\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}\frac{G_t{G}_r{\mathrm{\Γ}}_l{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_l\right)}^2}\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}{|}^2+H_s{\left(f\right)}^2\end{array}$

Can be seen that the amplitude of CSI power signal is square being inversely proportional to, when CSI power signal of the propagation distance with signal Amplitude reduce time, then the change direction of the propagation path of signal is augment direction, when CSI power signal amplitude increase time, Then the change direction of the propagation path of signal is for reducing direction.

(f t) is expressed as h after Butterworth LPF and PCA process to H₂, according to h from Fig. 8 (a)₂Amplitude not Can well determine the time point that signal propagation path length change direction and direction change, system is asked Wherein size is set to 25, waveform such as Fig. 8 (b), and this change fits like a glove with the direction of travel of people, when people is away from router, Signal propagation path length presents increase tendency, and when people is near router, signal propagation path length presents reduction trend.

As it is shown in figure 9, the step of the inventive method (3) includes following sub-step:

(3-1) according to the rate of change of signal propagation path and the change direction of signal propagation path length determine through with The diffusion path length of the signal of family reflection；

Assuming that the signal that WiFi transmitter is launched arrives WiFi receiver through human body reflection, this propagation path is kth bar Propagation path, the distance that this signal is propagated is d_k(0).If in the range of time t, people moves a segment distance, causes through unusually The diffusion path length of the signal of body reflection increases (reducing similar), by d_k(0) d is become_k(t), thenIts Middle v_kT () is the rate of change of signal propagation path length.Coordinate before moving due to known people i.e. previous prediction coordinate, previously Prediction coordinate is d to the distance summation of WiFi receiver and WiFi transmitter_k(0), if v can be measured by CSI_k(t), then V can be passed through_kT () calculates d_k(t)。

In order to further determine that d_kT (), needs to judge diffusion path length change direction, be to increase or reduce.Work as letter When number diffusion path length change direction is for increasing,When signal propagation path length change direction is for reducing Time,

(3-2) length of the signal propagation path determined according to position and the step (3-1) of two anchor points determines the position of user Put.

Specifically, first supporting to a wifi transmitter and a wifi receiver putting together is used as first Anchor point, the 2nd wifi receiver is as the second anchor point, using the position coordinates at user place as on elliptic wheel profile Individual point, using the first anchor point and the second anchor point as two oval focuses, build an ellipse, in this ellipse, user to two Length d that distance sum is signal propagation path of individual focus_k(t)；

Secondly, according to said method and use wifi transmitter and anchor point 1 conduct of a wifi receiver position The center of circle, constructing a radius is d_kT the circle of ()/2, one of two intersection points between this circle and above-mentioned ellipse are user place Position；

Again, according to determining the change direction of signal propagation path length and combining above-mentioned two intersection points determined, determine The position that user eventually arrives at；Specifically, if user arrives an intersection point (intersection point 1 in Fig. 9), then its change side above To being to reduce, if user arrives a following intersection point (intersection point 2 in Fig. 9), then its change direction should be to increase.

As it is shown in figure 9, anchor point 1 is WiFi receiver and the WiFi transmitter composition being in same position, anchor point 2 is one WiFi receiver forms.Assume previous prediction coordinate and two position of intersecting point as shown in the figure.If after the Δ t time, people's is true Real position is intersection point 1, then the diffusion path length of the signal reflected through human body, and its variation tendency is constantly to reduce；If passing through After the Δ t time, the actual position of people is intersection point 2, then variation tendency is for first to reduce, and increases afterwards.Propagation to human body reflected signal The change direction of path is monitored in real time, uses two anchor points just to can confirm that the actual position of user, averagely positions Error is less than 1 meter.

As it will be easily appreciated by one skilled in the art that and the foregoing is only presently preferred embodiments of the present invention, not in order to Limit the present invention, all any amendment, equivalent and improvement etc. made within the spirit and principles in the present invention, all should comprise Within protection scope of the present invention.

Claims (10)

1. the passive type two real-time indoor orientation method of anchor point based on channel condition information, it is characterised in that include following Step:

(1) persistently detect the CSI information of subcarrier, and judge whether this user there occurs movement according to this CSI information, if sent out Give birth to movement, then proceeded to step (2), otherwise continue detection CSI information；

(2) the signal propagation path length caused owing to user moves by this CSI acquisition of information rate of change and due to User moves the change direction of the signal propagation path length caused；

(3) initial position before moving according to the position between two anchor points and user, and use the signal that step (2) obtains to pass The change direction of the rate of change and signal propagation path length of broadcasting path obtains the position that user is currently located.

The passive type two real-time indoor orientation method of anchor point the most according to claim 1, it is characterised in that in step (1), hold The CSI information of continuous detection subcarrier, and judge whether this user there occurs that this process mobile specifically includes according to this CSI information Following sub-step:

(1-1) use PCA method, the CSI information on all subcarriers is split, the subcarrier stream group in same time period Become matrix H, calculate H^TThe characteristic value of × H and characteristic vector q thereof_i, wherein i is the numbering of matrix exgenvalue；

(1-2) at the H obtained^TThe characteristic value of × H and characteristic vector q thereof_iIn, select to determine Second Largest Eigenvalue characteristic of correspondence Vector q₂With corresponding principal component h₂=H × q₂；

(1-3) the characteristic vector q determined according to step (1-2)₂Calculate its average δ_q2:

${\mathrm{\δ}}_{q_2}=\frac{1}{s-1}\underset{l=2}{\overset{s}{\Σ}}|q_2\left(l\right)-q_2(l-1)|$

Here s represents the number of all subcarriers.| q2 (l)-q2 (l-1) | is the coefficient correlation of adjacent sub-carrier CSI information Difference.

(1-4) principal component h is calculated₂VarianceAnd according to the variance obtainedThe average obtained with step (1-3)Sentence Whether disconnected user is moved.

The passive type two real-time indoor orientation method of anchor point the most according to claim 2, it is characterised in that in step (1-4), WhenDuring more than a threshold value, it is judged that user starts mobile, wherein said threshold value is measurement gained in quiet environment 2 times of average.

The passive type two real-time indoor orientation method of anchor point the most according to claim 1, it is characterised in that step (2) is concrete Including following sub-step:

(2-1) the static CFR of acquisition subcarrier:

$H_s(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}{\mathrm{\Σ}}_{k\∈P_s}{a}_k(f,t)e^{-j2{\mathrm{\πf\τ}}_k\left(t\right)}$

Wherein H_s(f, t) centered by frequency be the channel frequency response that the subcarrier of f obtains in the measurement of time t, f represents subcarrier Centre frequency, P_sRepresent the propagation path set that the signal propagation path length of subcarrier does not changes, a_k(f t) represents The initial phase of kth bar propagation path and the decay of amplitude, e^-j2πΔftRepresent the phase drift that frequency shift (FS) causes, Represent the phase drift on kth bar propagation path, τ_kT () represents the propagation delay on kth bar propagation path；

(2-2) the dynamic CFR of acquisition subcarrier:

$H_d(f,t)={\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t)e^{-j2{\mathrm{\πd}}_k\left(t\right)/\mathrm{\λ}}$

Wherein P_dRepresent the propagation path set that the signal propagation path length of subcarrier changes, d_kT () represents through human body The kth bars diffusion path length of reflection.

(2-3) dynamic CFR that the static CFR obtained according to step (1) and step (2) obtain obtains total CFR:

$H(f,t)=e^{-j2\mathrm{\π}\mathrm{\Δ}ft}\left(H_s\right(f,t)+{\mathrm{\Σ}}_{k\∈P_d}{a}_k(f,t\left)e^{-j\frac{2{\mathrm{\πd}}_k\left(t\right)}{\mathrm{\λ}}}\right);$

(2-4) rate of change of signal propagation path is determined according to the frequency response of subcarrier；

(2-5) determine that user moves the signal propagation path length that causes according to the rate of change of the diffusion path length determined Change direction.

The passive type two real-time indoor orientation method of anchor point the most according to claim 4, it is characterised in that step (2-4) has Body is,

First, to H (f, t) delivery square:

$\begin{array}{c}{\left|H(f,t)\right|}^2=\underset{k\∈P_d}{\Σ}2\left|H_s\left(f\right)a_k(f,t)\right|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|a_k(f,t)a_l(f,t)\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}{\left|a_k(f,t)\right|}^2+H_s{\left(f\right)}^2\end{array}$

Wherein k, l represent different propagation paths respectively, and λ represents wavelength, θ_skRepresent that propagation path s's and propagation path k is initial Phase place sum, θ_klRepresent the initial phase sum of propagation path k and propagation path l；

Secondly, the power of CFR is done wavelet transformation, to obtain a series of frequency values: choose amplitude in a series of frequency values Maximum frequency values f_max, and calculate v_k=f_max× λ, wherein v_kRepresent the change of the propagation path of the signal reflected by user's body Change speed.

The passive type two real-time indoor orientation method of anchor point the most according to claim 5, it is characterised in that step (2-5) has Body is,

First initial phase and the decay of amplitude of kth bar propagation path are obtainedWherein G_rAnd G_tBe receive and Launching the signal power gain of antenna, λ is the wavelength launching signal, d_kIt is the propagation distance on kth bar propagation path, Γ_iFor instead Penetrate the factor；

Secondly, this decay is substituted in the power expression of step (2-4):

$\begin{array}{c}{\left|H(f,t)\right|}^2=\underset{k\∈P_d}{\Σ}2\left|H_s\left(f\right)\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}\right|\cos (\frac{2{\mathrm{\πv}}_{k}t}{\mathrm{\λ}}+\frac{2{\mathrm{\πd}}_k\left(0\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{sk})+\\ \underset{k\≠l}{\underset{k,l\∈P_d}{\Σ}}2\left|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}\frac{G_t{G}_r{\mathrm{\Γ}}_l{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_l\right)}^2}\right|\cos (\frac{2\mathrm{\π}(v_k-v_l)t}{\mathrm{\λ}}+\frac{2\mathrm{\π}\left(d_k\right(0)-d_l(0\left)\right)}{\mathrm{\λ}}+{\mathrm{\θ}}_{kl})+\\ \underset{k\∈P_d}{\Σ}{\left|\frac{G_t{G}_r{\mathrm{\Γ}}_k{\mathrm{\λ}}^2}{{\left(4{\mathrm{\πd}}_k\right)}^2}\right|}^2+H_s{\left(f\right)}^2\end{array}$

Finally, determine that the amplitude of CSI power signal is to reduce or increase, when the amplitude of CSI power signal reduces, then signal The change direction of propagation path be augment direction, when the amplitude of CSI power signal increases, then the change of the propagation path of signal Change direction for reducing direction.

The passive type two real-time indoor orientation method of anchor point the most according to claim 6, it is characterised in that step (3) includes Following sub-step:

(3-1) determine through user anti-according to the rate of change of signal propagation path and the change direction of signal propagation path length The diffusion path length of the signal penetrated:

(3-2) length of the signal propagation path determined according to position and the step (3-1) of two anchor points determines the position of user.

The passive type two real-time indoor orientation method of anchor point the most according to claim 7, it is characterised in that step (3-2) has Body includes:

(3-2-1) using supporting to a wifi transmitter and wifi receiver use of putting together as the first anchor point, one 2nd wifi receiver is as the second anchor point, using the position coordinates at user place as a point on elliptic wheel profile, with One anchor point and the second anchor point as two oval focuses, build an ellipse, in this ellipse, user to two focus away from It is length d of signal propagation path from sum_k(t)；

(3-2-2) according to said method and use the anchor point 1 of wifi transmitter and a wifi receiver position as circle The heart, constructing a radius is d_kT the circle of ()/2, one of two intersection points between this circle and above-mentioned ellipse are user place Position；

(3-2-3) according to determining the change direction of signal propagation path length and combining above-mentioned two intersection points determined, use is determined The position that family eventually arrives at；

The passive type two real-time indoor orientation method of anchor point the most according to claim 8, it is characterised in that step (3-2-3) Specifically, if user arrives an intersection point above, then its change direction is to reduce, if user arrives a following intersection point, Then its change direction should be to increase.

10. the passive type two real-time indoor locating system of anchor point based on channel condition information, it is characterised in that including:

According to this CSI information, first module, for persistently detecting the CSI information of subcarrier, and judges whether this user there occurs Mobile, if it occur that movement, then proceed to the second module, otherwise continue detection CSI information；

Second module, the change speed of the signal propagation path length for being caused owing to user moves by this CSI acquisition of information Rate and owing to user moves the change direction of the signal propagation path length caused；

3rd module, the initial position before moving according to the position between two anchor points and user, and use the second module The rate of change of the signal propagation path length obtained and the change direction of signal propagation path length obtain user and are currently located Position.