CN106154222B - A kind of method of direction of travel using radio-frequency signal detection people - Google Patents
A kind of method of direction of travel using radio-frequency signal detection people Download PDFInfo
- Publication number
- CN106154222B CN106154222B CN201610445766.XA CN201610445766A CN106154222B CN 106154222 B CN106154222 B CN 106154222B CN 201610445766 A CN201610445766 A CN 201610445766A CN 106154222 B CN106154222 B CN 106154222B
- Authority
- CN
- China
- Prior art keywords
- people
- equipment
- travel
- state information
- radio
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
- G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
Abstract
The invention discloses a kind of methods of the direction of travel using radio RF signal detection people, place an apparatus for transmitting signal Tx and two signal receivers Rx1, Rx2 in known fixed position respectively；Rx1, Rx2 are as vertical as possible with the line of Tx respectively, do not stop between Tx Rx1 and Tx Rx2 equipment pair；Signal is modulated using OFDM, and the information of change in signal strength is obtained by measure channel state information；The variation of multi-frequency wireless channel is extracted by measuring radio RF signal, the direction of travel of people is calculated.People must not carry sensing equipment and could detect direction of travel and limited technical solution of the present invention in by cognitive method based on equipment, can utilize existing wireless channel CSI numerical value, cost-effective without carrying out hardware modifications on wireless transmitting-receiving equipments；Can have the advantages that convenient and practical, detection error is smaller, with high accuracy with real-time continuous operation.
Description
Technical field
The present invention relates to no device-aware technology more particularly to a kind of utilization radio RF, (Radio Frequency, are penetrated
Frequently the method for the direction of travel of signal detection people), the disturbance generated using the walking of people to radio RF signal, by measuring simultaneously
The variation of multi-frequency wireless channel is extracted, the direction of travel of people is calculated.
Background technology
The walking manner of indoor people is a kind of important situation, and support can be provided for many applications by perceiving direction of travel.Sentence
Whether the prior art of the direction of travel of disconnected people needs to carry certain equipment according to perceived people, is roughly divided into two classes：
Without device-aware and perception based on equipment.
Wherein, based on equipment perceive allows people to carry certain sensor, according to the difference of sensor, perceives direction of travel
Method also there are many.2014, the Nirupam Roy of University of Illinois (UIUC) were in document " I am a
It is proposed is a kind of in Smartphone and I can Tell my UsersWalking Direction " perceives room using mobile phone
The method WalkCompass of interior direction of travel.This method utilizes acceleration transducer present on mobile phone, it can be deduced that, people exists
In moving process, the acceleration information in the dimension of three, space.Acceleration information projects on horizontal plane that understand can be horizontal
Movement on face accelerates situation.WalkCompass differentiates that direction of travel is the process of two steps：1) Nirupam Roy are observed
To when people walks, step touches ground and generates a kind of special vibrations, and this vibrations are transmitted to entire body, and can be placed on
The mobile phone of various positions can perceive with it.WalkCompass is shaken as reference using this, forward scanning signal,
Specific sample is extracted when trunk movement accounts for walking main process.Signal is then repaiied according to the data of mobile phone gyroscope
Just, the influence of the variation of mobile phone posture when being walked with compensation.Finally, vector acceleration is projected in the plane of gravity,
It is averaged per several steps, to carry out the estimation of the local direction of travel relative to mobile phone；2) the local direction based on mobile phone is reflected
It is mapped on global geomagnetic coordinate system system, this is completed using the geomagnetic sensor (electronic compass) in mobile phone.Due to mobile phone magnetic
Induction compass is easily influenced by Metals in Environments object, directly larger using compass data error.WalkCompass utilizes walking
In multiple surveyors' compass data, to carry out triangulation location, the information of triangulation location can correct the data of compass.The one of walking
In step, mobile distance is shorter, and compass can be influenced in multiple measure by environment same degree.Therefore, WalkCompass
Utilize the location information repeatedly measured in the blind airmanship tracking moving process based on gyroscope.
Differentiate that the method for the moving direction of people can also be completed by continuous positioning based on equipment, i.e., estimate by various methods
The position indoors such as mobile phone or wearable device measures direction of travel using the variation of different time estimation out position.Side
To measurement accuracy depend on the precision of positioning and unit interval and can calculate the quantity of result.In the auxiliary of no environmental unit
Under, it only completes to position by handheld device, presently mainly passes through the matched method of information fingerprint or the method based on ranging/angle
To complete.In finger print matching method, certain signal meets two kinds of requirements in area to be tested：1) ga s safety degree.That is detection zone
The signal of different location is different in domain.2) time stability.I.e. within the regular hour (such as in some months), this signal
Otherness remains unchanged.Algorithm is using the existing sensor for certain signal in mobile phone, place is carried out in advance a large amount of
Then sampling is mapped with location information after feature extraction and training, forms database.This is the process of study.Make
The signal of acquisition by the record in similary processing and database is compared, and estimated location, generally uses more by the used time
Bayes method.The problem of method of fingerprint matching, is to need to carry out largely sampling in advance and training, and per secondary environment, equipment
Resampling and training are needed after placement location variation.
The existing location technology based on fingerprint using information such as Geomagnetic signal, wireless signals (GSM/WiFi RSSI), essence
Degree is about in room-level, far from the requirement for reaching estimation indoor moving direction.And essence can be obtained using physical layer CSI information
The thin positioning of some.5 pattern Shun in 2013 etc. construct finely in document " CSI-Based Indoor Localization "
Indoor positioning FILA (Fine-grained Indoor Localization) system, the system use the base of WiFi CSI information
In the positioning of fingerprint matching, in the case of single AP, mean accuracy is about 3 meters；The situation mean accuracy of two AP is about 1.5
Rice.
When method based on ranging works, the distance of the signal source relative to multiple known locations can be measured respectively, so
Method of geometry estimated location is utilized afterwards.The signal that distance measuring method utilizes reaches time ToF (TimeofFlight) information, distance etc.
Signal velocity is multiplied by ToF.It needs to arrange special equipment in room using sound ranging；Using wireless signals such as WiFi then
Existing network facility can be multiplexed.However, since radio is with light velocity propagation, the ranging based on radio needs accurate reach
The measurement of time, and accurate ToF measures the big bandwidth of needs.Under the 20MHz bandwidth of existing WiFi agreements, when discernmible
Between between be divided into 50 nanoseconds, about 15 meters of precision.It is about 15 meters of situation in precision since common room-sized is less than 15 meters
Under, the different path lengths in 15 meters cannot be distinguished in this method, therefore can not meet the requirements.
Angle A oA (Angleof Arrival) information estimated location is reached by measuring signal based on the method for angle.
The estimation of angle needs aerial array, i.e., each receiving terminal mutiple antennas.Antenna amount is more, and angular error is smaller.Has skill
Art is mostly based on super resolution MUSIC algorithms or its a certain mutation.By positioning signal relative to several known ginsengs
The angle of position is examined, the position of signal source can be calculated by method of geometry.
Bear outstanding person in 2014 et al. is in document (Phaser:Enabling Phased Array Signal Processing
OnCommodity WiFi Access Points) in record, using multiple AP WiFi signal angle of arrival information into
Row positioning can reach about 1 meter of mean accuracy.But the precision needs to obtain under 5 antennas of each AP outfits, and common
WiFi equipment is configured without so multiple antennas.2015, Xiong Jie was in document (ToneTrack:Leveraging Frequency-
Agile Radios forTime-Based Indoor Wireless Localization) in construct ToneTrack, should
System has used frequency hopping, is switched fast in multiple adjacent WiFi frequency ranges, to improve positioning accuracy.ToneTrack makes
With the WiFi AP of 6 special WARP hardware simulations, binding reaches 0.9 meter of average positioning accurate using under 3 20MHz channels
Degree.
ManikantaKotaru in 2015 et al. is improved MUSIC algorithms using WiFiCSI information, so as to reduce
Antenna amount (the SpotFi needed:Decimeter Level Localization Using WiFi).SpotFi is used
6 AP, each AP assemble 3 antennas, and positioning accuracy is about 0.4 meter, have reached the comparable of previous each AP 6 antennas of needs
As a result.In the case where using two AP, SpotFi reaches about 1.6 meters of precision.
Scanning result is stitched together, by scanning all channels of WiFi to improve by Deepak Vasisht et al.
Temporal resolution and range accuracy (Sub-Nanosecond Time of Flight on CommercialWi-Fi
Cards).This method only needs an AP, can reach 0.58 meter of positioning accurate in the case of there are LoS (Lineof Sight)
Degree, reaches 1.18 meters of positioning accuracies in NLoS (Non Lineof Sight).But this method is needed in AP ends and mobile phone
Synchronous vacations driving is held, full band scan is realized, can not directly be realized on commercial WiFi basis of software.
Direction of travel judgement based on no device-aware is that the influence of signal is detected using people, usually using wireless
Signal.It does not need to carry any equipment when people walks indoors.Most common detection method is to utilize doppler principle.Doppler
Radar is exactly typical example.But the realization of radar is expensive, is difficult to minimize, and is detected for indoor moving and unreal
Border.Come examinations it is another kind of implementation method using software radio, advantage is can to send with precise synchronization, receiving terminal
Signal, but the problem is that accurate clock is synchronous and sampling is with high costs, and needs to send and receive end with cable connection,
To provide unified clock.This high-precision clock, which synchronizes, to be difficult to set in two individual sending devices of low cost and reception
It is realized in standby radio frequency chip.It is existing at present to be carried out using radio frequency RF signals without equipment Passive Positioning and sense
The work known is less, and 5 pattern Shun in 2013 et al. construct Pilot systems (Pilot:Passive Device-free Indoor
LocalizationUsing Channel State Information), Pilot is first and WiFi physical layers CSI is utilized
Information carries out the work without device-aware and positioning, but is only capable of being spaced 3~4 meters of location reference point in coarse divided chamber.
In conclusion can reach about 1 meter of positioning accuracy in the ideal case currently based on the cognition technology of equipment, need
People's Portable device, and equipment needs and infrastructure coordinates to run aware software.And have using WiFi signal without equipment sense
The detection error for the location technology known is more than the detection error using perceptual positioning method based on equipment, reaches 3-4 meters, existing
Positioning accuracy without device-aware method is excessively coarse, and error is too big, it is difficult to be directly used in estimation direction of travel.Not yet go out at present
It is now able to directly realize the technology for measuring direction of travel using existing commercial WiFi chip.
Invention content
In order to overcome the above-mentioned deficiencies of the prior art, the present invention provides a kind of new without device-aware method, using wireless
The method of the direction of travel of electric RF signal detections people, the disturbance generated using the walking of people to radio RF signal, passes through measurement
And the variation of multi-frequency wireless channel is extracted, the direction of travel of people is calculated.The present invention facilitate it is easy-to-use, with real-time continuous detection,
And accuracy of detection is high.
The principle of the present invention is：The mobile of people can impact radio wave.This body meeting for influencing to show people
Together with reflection radio wave, the electric wave of reflection and the Signal averaging originally sent out from Tx-Rx (transmission-reception), formed new
Signal strength.When people moves indoors, the path length that the radio signal of body reflection is passed through constantly changes, and
The path length of LoS does not change.Then, the radio wave from two paths can be superimposed, and the effect of superposition takes
Certainly in the phase relation of the radio signal from two paths.If two path phases are identical (or close), from people's
Reflection signal can enhance the signal from LoS；If two path phases are on the contrary, the reflection signal from people can weaken comes from
The signal of LoS.This phenomenon can be described by Fresnel region model.Fresnel region model is using Tx and Rx as two focuses
The distance and the half wavelength more than LoS length of one focus point of point distance on concentration ellipse cluster, wherein innermost layer ellipse, between oval
There is no an intersection point, the distance and half wavelength just more than adjacent inner layer ellipse of two focuses of point distance on later outer layer ellipse.
On the difference ellipse of Fresnel region, signal forms wave crest and trough, is alternately present.Continuous moving can be worn people indoors as a result,
More Fresnel region causes the variation repeatedly of signal strength, generates the variation of similar sine wave sample.The effect of superposition depends on using
Radio wavelength and people's distance a pair of Tx-Rx distance.The intensity of wireless signal can use receiving terminal RSSI
(Received Signal StrengthIndicator) is portrayed.
The present invention perceives the direction of travel into the people in room using the wireless signal of OFDM modulation.Especially by interior
The radio signal sending device and receiving device of fixed position, are counted using the channel state information acquired on receiving device
The direction of travel of people is calculated and estimates, to overcome people that must carry sensing equipment this constraint.The radio signal of OFDM modulation
Subchannel containing multiple frequencies, difference on the frequency is generally spaced equal between subchannel.And when the person moves, it is different in different subchannels
The variation tendency of the radio signal strength of wavelength is simultaneously asynchronous.By measuring the priority that signal strength fluctuates in different subchannels
Relationship, it can be determined that go out mobile relationship of the people relative to a pair of of Tx-Rx.The signal strength for portraying each subchannel in OFDM is logical
Measure channel state information (ChannelStateInformation, CSI) is crossed come what is completed.CSI information is described each
OFDM subchannels, wireless signal from transmitting terminal reach receiving terminal when, the attenuation of intensity and the situation of change of phase.
Technical solution provided by the invention is as follows：
A kind of processing method of direction of travel using radio RF signal detection people, puts respectively in known fixed bit
Put an apparatus for transmitting signal Tx (such as：WiFi wireless routers or wireless access point AP), two signal receiver Rx (examples
Such as：PC PCs, smart television etc.), two Rx are as vertical as possible with the line of Tx respectively, and signal is modulated using OFDM, and
The information of change in signal strength can be obtained by measure channel state information (CSI information), the people got in does not need to take
With any equipment.The present invention claims not stopping between Tx-Rx, there are stronger LoS (Lineof Sight) paths；
Processing method using the direction of travel of radio RF signal detection people includes the following steps：
1) data acquisition and pretreatment
A) Tx emits radio signal, and CSI information is acquired at two Rx ends.Complete CSI information includes OFDM subchannels
Transmission characteristic information (single plural number description), acquisition when transmission antenna number and reception antenna number, acquisition moment
Timestamp.Tx sends signal and does not require uniformly at equal intervals, can be with the normal transmission of wireless data communication；
B) after collected CSI signal times sequence is taken absolute value square, complex values are become into real number；
C) to the collected non-homogeneous CSI signal times sequence interpolations containing timestamp, uniform CSI signal times are formed
Sequence
D) application of CSI signal times sequence is filtered, to remove noise and high-frequency fluctuation；
2) feature extraction
A) it is window to whole time series cutting；
B) to each window, compare the delay between OFDM different sub-carriers；
C) according to the relationship being delayed between subcarrier, determine whether people has occurred movement, identification movement relative to each pair of Tx-Rx
Belong to close to relationship or far from relationship；
D) to each window, estimation quantity period of waves, as displacement distance；
Specifically, it is time window to the cutting of uniform channel status information time series entirety；To each time window, lead to
Cross the delay for calculating and comparing between OFDM different sub-carriers, the distribution being delayed in detection a period of time, respectively as people relative to Tx-
Thus Rx1 equipment pair or the corresponding moving direction of Tx-Rx2 equipment centering single pair equipment identify people relative to Tx-Rx1 equipment pair
Or whether Tx-Rx2 equipment is to having occurred apart from the movement changed；And estimate number period of waves of corresponding each time window
Amount, as relative to the displacement distance to equipment；The movement that distance change has occurred refers to people with respect to Mr. Yu to equipment
To having occurred movement, and people changes apart from the Tx to equipment plus the sum of path length of people's distance Rx；At this point, one
The sum of value of symbol of delay distribution between the subcarrier in time window corresponding in equipment is more than set threshold value；
3) direction estimation：By the position of transmitting antenna note of Tx into coordinate origin, cartesian coordinate system is built；By two Rx's
Remember into two reference axis relative to origin position reception antenna position；
A) if people has occurred relative to two pairs of Tx-Rx equipment and (shifting of distance change has occurred apart from the movement changed
It is dynamic to refer to that people occurs significantly after movement has occurred relative to a pair of of Tx-Rx equipment with the sum of the distance of Tx and Rx equipment respectively
Change, show as on Fresnel region model, have significantly from outer layer/internal layer ellipse inner layer/elliptical movement of outer layer, as significantly
Closer or far from relationship), then utilize method of geometry calculate moving direction；Under this situation, the computational methods in direction are as follows：By one
Section the time in, people is projected to relative to the moving direction and distance of each pair of Tx-Rx on corresponding axis, moving direction it is separate/
It is relative to the positive and negative of corresponding axis close to relationship projection, displacement distance is the period of waves time of CSI waveforms in this time
Number obtains (by 2d)).After projecting two reference axis, the direction of the vector of synthesis is the direction moved indoors for people；
If b) people have relative to two pairs of Tx-Rx equipment and only have occurred on a pair of Tx-Rx distance change movement (for
It is apparent it is close/far from relationship), another pair Tx-Rx it is upper it is close/far from relationship unobvious, then utilize empirical method estimation movement
Direction.
Under this situation, the c of step 2) step testing result wherein at least a pair of of performance in two couples of Tx-Rx unobvious
(people transceiver is had occurred relative to this it is mobile after respectively with the sum of the distance approximately equal to transceiver, do not have
It substantially change, i.e. the movement is not that the movement that distance changes has occurred.It shows on Fresnel model, is equivalent to same
Moved on layer ellipse), then experience method of estimation is：The vector for projecting two reference axis is the Distribution Value of delay, projects two
After a reference axis, the direction of the vector of synthesis is the direction moved indoors for people.
If c) close on two couples of Tx-Rx/far from the equal unobvious of relationship, judge not move for people.
Beneficial effects of the present invention：
The present invention provides a kind of new without device-aware method, utilizes the side of the direction of travel of radio RF signal detection people
Method, the disturbance generated using the walking of people to radio RF signal are changed by measuring and extracting multi-frequency wireless channel, calculated
Obtain the direction of travel of people.
The present invention is calculated the walking so as to estimate people using the channel state information acquired on transmission or receiving device
Direction, people must not carry sensing equipment and could detect direction of travel and limited in by cognitive method based on equipment, Ke Yili
It is cost-effective without carrying out hardware modifications on wireless transmitting-receiving equipments with existing wireless channel CSI numerical value；Detect mistake simultaneously
Difference is smaller, and detection error, which has reached, is equivalent to the multiple sensors such as accelerometer, gyroscope based on equipment of being integrated with
The angle detecting precision of WalkCompass methods.This method real-time continuous can be run, and is not needed to acquire environmental signal in advance and is referred to
Line.People's one to two step of walking indoors can perceive and calculate moving direction, and technical solution of the present invention is better than existing method, has
The advantages of practicability is high, precision is higher.
Description of the drawings
Wireless device in Fig. 1 embodiment of the present invention places schematic diagram；
Wherein, dotted line represents detection zone；TX is radio transition equipment；RX1, RX2 are respectively radio receiving equipment 1
With radio receiving equipment 2.
The calculation process frame of the method for the direction of travel of radio RF signal detection people is utilized in Fig. 2 embodiment of the present invention
Figure.
Specific embodiment
Below in conjunction with the accompanying drawings, the present invention, the model of but do not limit the invention in any way are further described by embodiment
It encloses.
The present invention provides a kind of method of the direction of travel using radio RF signal detection people, using the walking of people to nothing
The disturbance that line electricity RF signals generate is changed by measuring and extracting multi-frequency wireless channel, the direction of travel of people is calculated.
The present invention when implementing, need to place in the room a known fixed location apparatus for transmitting signal Tx (such as：
WiFi wireless routers or wireless access point AP), the signal receiver Rx of two known fixed locations (such as：PC people's electricity
Brain, smart television etc.), and two Rx equipment are as vertical as possible with the line of Tx respectively, signal uses OFDM modulation techniques, and energy
Reach the information by measuring CSI acquisition of information change in signal strength.The people got in does not need to carry any equipment.The present invention
It is required that not stopping between Tx-Rx, there are stronger LoS paths.Wireless device in Fig. 1 embodiment of the present invention places signal
Figure；Wireless device in the present embodiment includes a radio transition equipment TX and two radio transition equipments RX1, RX2.
Wherein, dotted line represents detection zone.The method of the direction of travel of radio RF signal detection people is utilized in Fig. 2 embodiment of the present invention
Calculation process block diagram.The specific embodiment of the present invention is as follows：
1) channel state information CSI is obtained；
A) CSI information both can respectively be acquired in two wireless device receiving terminals, can also be in wireless device transmitting terminal collection
Middle acquisition.One group of CSI time series is generated per a pair of Tx-Rx equipment rooms, two groups of Tx-Rx generate altogether two groups of CSI time serieses.
B) collected information, which concentrates in together, is handled.If CSI information is acquired in two receiving terminals, need to lead to
Cross network be pooled to it is same processing node be further processed；If CSI information is acquired in transmitting terminal, can directly exist
Sending node processing.
C) acquisition of CSI information is that the process of communication is followed to synchronously complete.Generally before each communication transfer, one is measured
Secondary CSI numerical value.
D) the different subchannels of OFDM codings can generate respective CSI measured values, i.e.,：CSI measurements each time include different frequencies
The CSI numerical value of the OFDM subchannels of rate.
E) CSI was constantly sampled with the time, formed time series.
Then square f) take absolute value to each CSI numerical value,.The purpose to take absolute value is long to plural modulus.
G) to the direction of travel of people in accurate judgement room, CSI sample rates should be at least up to 300Hz；
2) it is aligned the time series of two receiving terminal CSI
A) collected CSI numerical value can include the timestamp that network interface card is formed, and be free of absolute time in timestamp, only contain
The counter that internal crystal oscillator generates, therefore the timestamp absolute value that generates of different receiving terminals and differ.This step needs will be each
Subtract each other before and after the timestamp of the CSI time serieses of receiving terminal, form relative time stamp.I.e.：After subtracting each other, timestamp represents each
The time interval that CSI is measured is represented with the quantity that crystal oscillator counts.
B) relative time stamp is multiplied by network interface card crystal oscillator frequency to get to the timestamp of relative time；
C) the relative time stamp generated two receiving terminals carries out maximum common subsequence matching algorithm, finds out public sub- sequence
It is listed in the initial position of two relative time stamps；
D) initial position numerical value is subtracted each other to obtain difference, according to the CSI time sequences of difference two receiving terminals of alignment of initial position
Row；
3) to CSI waveforms into row interpolation
A) timestamp of relative time that 2b steps generate is added up, obtains the CSI time serieses to clock since 0.
I.e.：The timestamp of the relative time of 2b represents the time interval that adjacent CSI twice is measured, after adding up, at the time of expression since 0
Value；
B) using Spline Spline Interpolation Methods to CSI timestamps into row interpolation, during obtaining the uniform CSI of time interval
Between sequence；
4) noise reduction is carried out to CSI waveforms
A) noise reduction, Savitzky-Golay are carried out to CSI time serieses waveform using Savitzky-Golay smoothing methods
The parameter selection of noise reduction is：Select 3 order polynomials, window size be comprising 1 CSI period of waves preferably；
B) it is smoother to obtain to two CSI time serieses respectively using Savitzky-Golay smoothing processings several times
Curve.
5) delay between subcarrier is calculated
A) using sliding window by CSI time series fragments, window size is set as 0.1 second；
B) in window, the CSI waveforms of multiple OFDM subchannels are standardized.The method of standardization is：It calculates each
CSI sequences in the OFDM subchannels are subtracted this and are averaged by average value of the OFDM subchannel CSI time serieses in the window
Value so that CSI waveforms become to do around 0 to be fluctuated up and down；
C) the signal cross-correlation between different sub-carrier is calculated, cross-correlation is taken to reach the offset at maximum value, as two
Delay between subcarrier；
6) estimate relative movement distance in a period of time
A) using sliding time window by CSI time series fragments, window size is set as 0.5 second, is overlapped between window
50%；
B) quantity period of waves of CSI in the time window is obtained using Fast Fourier Transform (FFT) FFT to each window；
C) algorithm is walked to each OFDM subcarriers application b, obtains quantity period of waves of each subcarrier in window, take sub- load
Wave period of waves quantity estimation of the average value as displacement distance；The displacement distance and the distance correspondence got along well spatially,
It is only used for calculating angle；
D) the sum of displacement distance that the displacement distance accumulated in a period of time is calculated for each child window removes windows overlay
Part；
7) distribution being delayed in detection a period of time, obtains moving direction of the people relative to a pair of of Tx-Rx equipment；
A) the result fragment for being calculated the delay of 5) step carrier wave using sliding window, window size are set as 1 second；
B) in window, 5) delay value that step calculates is taken into sign function sgn, i.e.,：To positive number, result 1, to negative, knot
Fruit is -1, to 0, result 0；
C) it by all delay values in 1 second window, is added after taking symbol, obtains the distribution of delay value；
D) setting threshold value is 1/4 of the quantity of effective delay value in 1 second window；
E) when the distribution of the obtained delay values of step 7c is more than the threshold value set by step 7d, people is relative to this to Tx-
The moving direction at Rx equipment lines midpoint is just, people is far from Tx-Rx equipment at this time；When the distribution for the delay value that step 7c is obtained is small
When the negative value of the threshold value set by step 7d, direction is negative, and people is close to Tx-Rx equipment at this time；When the delay that step 7c is obtained
When the distribution of value is between positive negative threshold value, moving direction unobvious, direction 0, it is believed that people is not mobile or connects along Tx-Rx
The parallel angle movement of line；
8) estimate the moving direction of people
A) position of transmitting antenna of Tx is remembered into coordinate origin, builds cartesian coordinate system；
B) the reception antenna position of two Rx is remembered into two reference axis relative to origin position；
C) if the testing result in step 7e all shows apparent on two couples of Tx-Rx, i.e.,：The direction delay value point of 7e steps
The absolute value of cloth is more than threshold value.Then the computational methods in direction are as follows：By people relative to the moving direction of each pair of Tx-Rx and apart from throwing
On shadow to corresponding axis, close/separate relationship of moving direction is the positive negative in direction, is mapped as relative to each coordinate
The positive and negative 7e of axis, fluctuation number 6d of the displacement distance for CSI waveforms in a period of time.After projecting two reference axis, synthesis
The direction of vector is the direction moved indoors for people；
If d) the e of step 7) unobvious of the testing result of step wherein at least a pair of of performance in two couples of Tx-Rx, i.e. 7e)
It is 0 to walk direction, then experience method of estimation is：By 7c) step two is projected directly at two phases to the delay Distribution value calculated on Tx-Rx
It answers in reference axis.After projecting two reference axis, the direction of the vector of synthesis is the direction moved indoors for people.
Below by example, the present invention will be further described.
It is universal due to indoor WiFi equipment, following embodiment we the direction of travel of people is detected using WiFi signal.
The WiFi routers of the support 802.11n agreements of omnidirectional antenna are carried in 5G frequency ranges by the use of configuration as signal transmitting terminal Tx, it is sharp
By the use of two carrying omnidirectional antennas, it is mounted with the MiniPC of Intel5300 wireless network cards as signal receiving end.802.11n and with
Agreement afterwards supports OFDM modulation.CSI data can be acquired using Linux CSItools, collected CSI includes 30 sons and carries
The information of wave.Idiographic flow using the direction of travel of the method provided by the present invention detection people is as follows：
1) transmitting antenna on router is placed perpendicular to the ground, position mark is coordinate origin Tx, is placed on room one
Angle；
2) transmitting antenna of two MiniPC is placed perpendicular to the ground, position mark is placed on room to receive Rx1 and Rx2
Between another two corners.The wire definition of Tx and Rx1 is x-axis, and the wire definition of Tx and Rx2 are y-axis, establish coordinate system；
3) the continuous acquisition CSI data on Rx1 and Rx2, and pass through WiFi network and be transmitted to a PC centralized processing；
4) suitable data communication is generated in WiFi router sides, to reach the CSI sample rates of about 300Hz；
5) CSI data predictions are carried out, i.e.,：To collected CSI time serieses interpolation, S-G filtering is then carried out.Interpolation
When, using spline spline methods, with reach 300Hz and more than uniform sampling.Savitzky-Golay filtering is using 3 times
Multinomial, window size are set as about CSI period of waves.About rounding after sample rate divided by 40, then ensures window
Size is odd number (requirement of S-G filtering algorithms).
6) split window calculates the delay between subcarrier.Window is set as 0.1 second, is slided in a manner of overlapping 50%.Delay
Computational methods be calculate two different sub-carriers between cross-correlation.It is delay value to take the displacement at cross-correlation maximum.
7) distribution of computation delay value.The statistics of distribution needs slightly longer window, such as：The window of 1 second is taken, then is pressed
According to 0.1 second window calculation, the delay value for being overlapped 50% shared 20.Take and add up after symbol, then the value range of accumulated value [-
20,20] between.Threshold value is set as 20/4=5.
8) judge direction.Being distributed between 20 and 5 for delay, is judged as positive direction；Delay be distributed in -5 with -20 it
Between, it is judged as negative direction；Delay is distributed between -5 and 5, and directional information is indefinite.
9) judge displacement distance, that is, count the fluctuation number of CSI in a period of time.For example, the time window of selection 1 second,
The fluctuation number of CSI is estimated by Fast Fourier Transform (FFT) FFT methods.By the CSI time serieses of 1 second after FFT transform, shape
The sequence of Cheng Xin, the location number of the maximum value in sequence are the fluctuation number estimated.
10) direction of two dimensional surface is synthesized.If directional information is all clear and definite on two couples of Tx-Rx, the vector of two-dimensional directional can
To project on each pair of Tx-Rx, i.e.,：It projects in x-axis and y-axis.The value of projection fluctuates number for the 9th step and is being multiplied by the 8th step just
Negative value.Such as：The direction for projecting to x-axis is negative, and it is 3 times to fluctuate number；The direction of y-axis is projected to as just, and fluctuates number
It is 4, then availability vector (- 3,4) represents mobile direction.
If 11) twice in Tx-Rx directions, there is a pair of directional information indefinite, then the vector of two-dimensional directional utilizes experience
Method estimates that the accumulated value replacement of the 7th step is respectively adopted in the projection of i.e. x/y axis.
12) it is the actual direction walked by the angle of vector that x/y axis represents by the way that arctan functions is taken to calculate.Example
Such as：If mobile vector is (- 3,4), then the angle moved is -53.13 degree.
It should be noted that the purpose for publicizing and implementing example is to help to further understand the present invention, but the skill of this field
Art personnel are appreciated that：It is not departing from the present invention and spirit and scope of the appended claims, various substitutions and modifications are all
It is possible.Therefore, the present invention should not be limited to embodiment disclosure of that, and the scope of protection of present invention is with claim
Subject to the range that book defines.
Claims (10)
1. a kind of method of direction of travel using radio-frequency signal detection people, it is characterized in that, respectively in known fixation
Place an apparatus for transmitting signal Tx and two signal receivers Rx1, Rx2 in position；Rx1, Rx2 to the greatest extent may be used respectively with the line of Tx
Can be vertical, do not stop between Tx-Rx1 equipment pair and Tx-Rx2 equipment pair；Signal is modulated using OFDM, by measuring channel shape
The direction of travel of people is calculated in the information of state information extraction multi-frequency OFDM subchannels wireless channel variation；Including walking as follows
Suddenly：
1) data acquisition is obtained and is pre-processed：Tx emits radio signal, as the time constantly acquires channel state information, is formed
Time series, it is preprocessed to obtain uniform channel status information time series；
2) feature extraction is carried out, including：
It is time window to the uniform channel status information time series entirety cutting；
To each time window, by calculating the delay compared between OFDM different sub-carriers, point being delayed in detection a period of time
Cloth respectively as people relative to Tx-Rx1 equipment pair or the corresponding moving direction of Tx-Rx2 equipment centering single pair equipment, is thus known
Whether others is relative to Tx-Rx1 equipment pair or Tx-Rx2 equipment to having occurred apart from the movement changed；
Quantity period of waves of corresponding each time window is estimated, as people relative to the displacement distance to equipment；
3) direction estimation is carried out, obtains the direction of travel of the moving direction of people, i.e. people；Including two kinds of situations：
A) when people relative to Tx-Rx1 and Tx-Rx2 two to the relationship of equipment be have occurred distance change it is mobile when, utilization is several
Where the moving direction of people is calculated in method；
B) when the movement that distance changes, and the shifting that people occurs relative to another pair equipment have occurred relative to one pair of which equipment by people
It is dynamic be not distance change it is mobile when, estimate to obtain the moving direction of people using empirical method；The empirical method is：By described in
The distribution of delay is as the vector for projecting two reference axis, after the Distribution Value of delay is projected two reference axis, synthesis
The direction of vector is the direction moved indoors for people.
2. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step 1)
The Tx transmittings radio signal is as wireless data communication normally sends radio signal；The channel state information includes
Transmission antenna number, reception antenna number and the timestamp at acquisition moment when the transmission characteristic information of OFDM subchannels, acquisition.
3. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step 1)
Data acquisition obtains channel state information and the process of communication is followed to synchronously complete；Specifically adopted in wireless device transmitting terminal concentration
Collect or acquired respectively in two wireless device receiving terminals；Collected information is concentrated in together and is pre-processed.
4. as claimed in claim 3 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, the letter
Channel state information sample rate is at least up to 300Hz.
5. as claimed in claim 3 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, work as channel
Status information is when two wireless device receiving terminal Rx are acquired respectively, need to be aligned two wireless device receiving terminal channel status letters
The time series of breath, the alignment include the following steps：
A) collected channel state information numerical value includes the timestamp that network interface card is formed, by the channel state information of each receiving terminal
Subtract each other before and after the timestamp of time series, form relative time stamp；The relative time stamp represents that each channel state information is surveyed
The time interval of amount is represented with the quantity that crystal oscillator counts；
B) relative time stamp is multiplied by network interface card crystal oscillator frequency, obtains the timestamp of relative time；
C) the relative time stamp generated two receiving terminals is found out common subsequence and is existed by maximum common subsequence matching algorithm
The initial position of two relative time stamps；
D) initial position numerical value is subtracted each other to obtain difference, further according to the channel state information of difference two receiving terminals of alignment of initial position
Time series.
6. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step 1)
The pretreatment of the channel state information includes：After collected channel state information time series is taken absolute value square；It is right
Channel state information time series forms uniform channel status information time series into row interpolation；To the channel state information time
Sequence removes noise and high-frequency fluctuation by filtering noise reduction.
7. as claimed in claim 6 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, specific profit
The uniform channel status of time interval is obtained into row interpolation to channel state information time series with Spline Spline Interpolation Methods
Information time sequence；Channel state information time series waveform drops in specifically used Savitzky-Golay smoothing methods
It makes an uproar.
8. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step 2)
It is described calculate a period of time in OFDM subcarriers delay distribution, obtain moving direction, identification people with respect to Mr. Yu's equipment to whether
The movement that distance changes has occurred；Specifically comprise the following steps：
A) setting sliding window is used by channel state information time series fragment；
B) multiple OFDM subchannels channel state information waveforms are standardized in sliding window；
C) the signal cross-correlation between different OFDM subcarriers is calculated, cross-correlation is taken to reach the offset at maximum value, as two
Delay value between subcarrier；
D) delay value between the c) different sub-carrier is carried out by fragment using sliding window；
E) delay value between the c) different sub-carrier is taken into sign function sgn in sliding window, obtains result 1, -1 or 0；
F) all results addeds that will e) obtain obtain the distribution of delay value；
G) threshold value is set according to the quantity of effective delay value in sliding window；
H) when the distribution of delay value is more than set threshold value, the movement far from equipment occurs, direction is just；When point of delay value
When cloth is less than negative threshold value, the movement close to equipment occurs, direction is negative；When delay value distribution between threshold value positive negative value
Between when, moved for no movement or along the equipment angle parallel to line；
I) as people relative to the equipment to having occurred movement, and people respectively with the apparatus for transmitting signal and signal of the equipment centering
When variation has occurred in the sum of the distance of receiving device, identification obtains the movement that people changes relative to the equipment to distance has occurred.
9. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step 2)
The estimation obtains the fluctuation number of each time window, as displacement distance, specifically comprises the following steps：
A) channel state information time series fragment is set between the sliding window size and sliding window using sliding window
Duplication；
B) channel state information wave in the sliding window is obtained using Fast Fourier Transform (FFT) FFT methods to each sliding window
Dynamic amount of cycles；
C) above-mentioned steps b) is performed to each OFDM subcarriers, obtains the fluctuation quantity of each subcarrier, take the fluctuation number of subcarrier
Estimated value of the average value of amount as displacement distance, the estimated value of the displacement distance are used to calculate the angle of direction of travel；
D) the sum of estimated value of displacement distance for calculating each sliding window removes the part of windows overlay, during as one section
The displacement distance of interior accumulation.
10. as described in claim 1 using the method for the direction of travel of radio-frequency signal detection people, it is characterized in that, step
3) it is described people is calculated using method of geometry moving direction be specially：
People in a period of time projects to accordingly the moving direction and displacement distance of equipment relative to Tx-Rx1 and Tx-Rx2 two
In reference axis, the direction of synthesized vector is the direction moved indoors for people；The people is relative to Tx-Rx1 and Tx-Rx2
Separate/close relationship projection of the moving direction of two pairs of equipment is relative to the positive and negative of corresponding axis；The displacement distance by
Number period of waves of channel state information waveform obtains in this time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610445766.XA CN106154222B (en) | 2016-06-20 | 2016-06-20 | A kind of method of direction of travel using radio-frequency signal detection people |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610445766.XA CN106154222B (en) | 2016-06-20 | 2016-06-20 | A kind of method of direction of travel using radio-frequency signal detection people |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106154222A CN106154222A (en) | 2016-11-23 |
| CN106154222B true CN106154222B (en) | 2018-06-12 |
Family
ID=57353561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610445766.XA Active CN106154222B (en) | 2016-06-20 | 2016-06-20 | A kind of method of direction of travel using radio-frequency signal detection people |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106154222B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106842155B (en) * | 2017-01-17 | 2020-02-07 | 北京工业大学 | Wireless fixed emission source positioning method based on spatial interpolation and cluster analysis |
| CN106802404B (en) * | 2017-01-18 | 2018-10-16 | 深圳大学 | A kind of human body indoor positioning automatic testing method and system |
| JP2018141763A (en) * | 2017-02-28 | 2018-09-13 | パナソニックＩｐマネジメント株式会社 | Detector and control system |
| CN107396311B (en) * | 2017-07-17 | 2020-04-10 | 北京大学 | Non-contact sensing positioning method |
| WO2019138130A1 (en) * | 2018-01-15 | 2019-07-18 | Sony Mobile Communications Inc. | Determining a position of a device with respect to another device |
| CN109194416B (en) * | 2018-08-22 | 2022-06-07 | 平安科技（深圳）有限公司 | Information transmission method and device, computer equipment and storage medium |
| CN109341679A (en) * | 2018-09-30 | 2019-02-15 | 华中科技大学 | A kind of smart machine air navigation aid and navigation system |
| CN109711251B (en) * | 2018-11-16 | 2023-04-07 | 天津大学 | Commercial Wi-Fi-based independent direction gait recognition method |
| TWI695595B (en) * | 2019-01-03 | 2020-06-01 | 仁寶電腦工業股份有限公司 | Object orientation system, object orientation method and electronic apparatus |
| CN110363947B (en) * | 2019-07-11 | 2021-11-23 | 京东方科技集团股份有限公司 | Method for judging illegal intrusion based on WIFI (Wireless Fidelity) reflected signal and security system |
| CN110738130A (en) * | 2019-09-21 | 2020-01-31 | 天津大学 | Gait recognition method with independent path based on Wi-Fi |
| CN112153736B (en) * | 2020-09-14 | 2022-07-26 | 南京邮电大学 | Personnel action identification and position estimation method based on channel state information |
| CN112380903B (en) * | 2020-10-14 | 2024-02-02 | 东北电力大学 | Human body activity recognition method based on WiFi-CSI signal enhancement |
| CN112926514A (en) * | 2021-03-26 | 2021-06-08 | 哈尔滨工业大学（威海） | Multi-target detection and tracking method, system, storage medium and application |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104812061A (en) * | 2015-03-24 | 2015-07-29 | 成都希盟泰克科技发展有限公司 | Indoor range finding and positioning method based on MIMO-OFDM channel state information |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2520057A4 (en) * | 2009-12-29 | 2017-03-15 | Thomson Licensing | Method and apparatus for channel estimation |
-
2016
- 2016-06-20 CN CN201610445766.XA patent/CN106154222B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104812061A (en) * | 2015-03-24 | 2015-07-29 | 成都希盟泰克科技发展有限公司 | Indoor range finding and positioning method based on MIMO-OFDM channel state information |
Non-Patent Citations (4)
| Title |
|---|
| Communicating Is Crowdsourcing：Wi-Fi indoor Localization with CSI-Based Speed Estimation;Zhiping Jiang等;《Journal of computer Science and Technology》;20130731;1-15 * |
| MonoPHY：Mono-Stream-based Device-free WLAN Localization via Physical Layer Information;Heba Abdel-Nasser等;《2013 IEEE Wireless Communications and Networking Conference(WCNC):SERVICES&APPLICATIONS》;20130430;4569-4574 * |
| Tool Release：Gathering 802.11n Traces with Channel State information;Daniel Halperin等;《ACM SIGCOMM Computer Communication Review》;20110131;第41卷(第1期);53 * |
| WiFall：Device-free Fall Detection by Wireless Networks;Chunmei Han等;《IEEE INFOCOM 2014》;20141231;271-279 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106154222A (en) | 2016-11-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106154222B (en) | A kind of method of direction of travel using radio-frequency signal detection people | |
| Qian et al. | Widar2. 0: Passive human tracking with a single Wi-Fi link | |
| Wu et al. | WiDir: Walking direction estimation using wireless signals | |
| Mariakakis et al. | Sail: Single access point-based indoor localization | |
| WO2018068753A1 (en) | Time difference of arrival (tdoa) measurement method and apparatus, tdoa control apparatus and method, and terminal | |
| CN104135305A (en) | Ranging and positioning system | |
| KR20170086524A (en) | Partially synchronized multilateration or trilateration method and system for positional finding using rf | |
| US20130141283A1 (en) | Method for locating a radio center and system for locating a radio center and data processing unit | |
| CN101730851A (en) | System and method for positioning | |
| Khyam et al. | Multiple access chirp-based ultrasonic positioning | |
| JP2010197050A (en) | Position estimating system | |
| KR101331833B1 (en) | Method for positioning using time difference of arrival | |
| US20160041251A1 (en) | Ultrasound-based location determination and inertial navigation with accuracy improvement in determining client device location | |
| CN110069134A (en) | A method of hand aerial mobile track is restored using radio-frequency signal | |
| Naeem et al. | Performance analysis of TDOA-based indoor positioning systems using visible LED lights | |
| Li et al. | An indoor location algorithm based on Kalman filter fusion of ultra-wide band and inertial measurement unit | |
| CN109934031A (en) | The method and system of differentiation LOS/NLOS based on RFID system | |
| CN106323450B (en) | Vibration monitoring method of large flexible structure based on Doppler radar | |
| CN105652239A (en) | Self-adaptive high-precision indoor positioning method and system | |
| Wang et al. | Through the wall detection and localization of autonomous mobile device in indoor scenario | |
| RU2579983C1 (en) | Method of transmitting radio signals | |
| CN111405658B (en) | Indoor positioning method based on fusion of sound wave positioning and Bluetooth ranging | |
| KR101108707B1 (en) | Locating system and signal receiver for the system | |
| RU2578750C1 (en) | Method of transmitting radio signals | |
| US11079463B2 (en) | Method and device for position determination |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |