CN107689153A - A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal - Google Patents
A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal Download PDFInfo
- Publication number
- CN107689153A CN107689153A CN201710736698.7A CN201710736698A CN107689153A CN 107689153 A CN107689153 A CN 107689153A CN 201710736698 A CN201710736698 A CN 201710736698A CN 107689153 A CN107689153 A CN 107689153A
- Authority
- CN
- China
- Prior art keywords
- mrow
- data
- extension set
- network
- mtd
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
Abstract
The invention discloses a kind of traffic cross-sectional flow prediction algorithm based on WIFI signal, prediction device systems are deployed with along traffic route, the prediction device systems include multiple sub-networks, each sub-network includes a main frame and several extension sets, the extension set is based on WIFI agreements and passes through wireless passive perceptual model, gather the broadcast data packet that the mobile device based on WIFI signal agreement is sent in surrounding environment, and screen the wherein packet with mobile terminal device id information and retrieved, main frame is uploaded to after stamping extension set label, the data being collected into are carried out unified storage and stamp time tag by main frame, and it is uploaded in data server and stores, and by being analyzed data to obtain road section flow velocity assessment prediction result.The present invention realizes the data mining and analysis to mobile terminal gathered data, can be applied to the detection of traffic cross-sectional flow.
Description
Technical field
The invention belongs to development of Mobile Internet technology, more particularly to the improvement of macro-traffic information monitoring algorithm.
Background technology
Traffic flow data is the important information source of traffic operation dispatching and command system, can be command scheduling, the magnitude of traffic flow
Control and traffic guidance provide decision-making foundation.Existing Traffic flow detecting technology has a variety of, can be divided into contact according to mounting means
Formula detection mode and non-contact detection mode.Wherein contact measurement technology includes piezoelectricity, pressure pipe detection and loop coil
Detection.The major defect of this technology is that vehicle causes the service life of detector shorter rolling for road, is detected laying
, it is necessary to suspend traffic, destroy road surface during device, therefore more difficult, use cost height is gone along with sb. to guard him in installation.Non-contact detection technology master
To be wave frequency detection and video detection.Wave frequency detection is divided into microwave, ultrasonic wave and infrared etc. three kinds.Non-contact detection device can lead to
Support installation is crossed, easy to maintenance, service life length, its major defect is easily to be influenceed by weather and outdoor conditions, environment be present and fits
The problems such as answering property is not strong, volume of transmitted data is big, Detection accuracy is not high and cost is higher.
With the rapid development of China's highway network, freeway traffic flow detection application demand increases severely.In highway network
In, traffic flow information is equally important, and by flow information, highway network administrative department can understand the reality in each section in real time
When vehicle fleet size information, there is provided intuitively road network vehicle load amount, accurate data is provided for the scheduling and integrated planning of road network.
But highway network has that some are special, such as highway power supply inconvenience, information transfer difficulty, with
And fail to lay all kinds of detectors etc. in advance in process of construction, concentrated type monitoring can not be accomplished and managed, it is necessary to existing detection
Device is further designed and improved.
The content of the invention
The problem of existing for prior art, the invention provides a kind of traffic cross-sectional flow prediction based on WIFI signal
Algorithm, the macro-traffic of the invention based on the unique ID of portable mobile terminal, gathered data timestamp and detection device positional information
Infomation detection algorithm, data mining and analysis to mobile terminal gathered data are realized, can be applied to the pre- of traffic cross-sectional flow
Survey.
In order to solve the above technical problems, present invention employs following technical scheme：
A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal, prediction device systems are deployed with along traffic route,
The prediction device systems include multiple sub-networks, and each sub-network includes a main frame and several extension sets, the extension set base
In WIFI agreements by wireless passive perceptual model, gather what the mobile device based on WIFI signal agreement in surrounding environment was sent
Broadcast data packet, and screen the wherein packet with mobile terminal device id information and retrieved, stamp on after extension set label
Main frame is reached, the data being collected into are carried out unified storage and stamp time tag by main frame, and are uploaded in data server and are deposited
Storage, and by being analyzed data to obtain road section flow velocity assessment prediction result.
Further, distance is d between the adjacent extension set, and the signal covering radius of single extension set is r, and d ＞ 2r.
Further, the road section flow velocity assessment prediction comprises the following steps：
Step 1：The data D of complete period is gathered by extension set,Wherein, DijRepresent j-th
No. i-th extension set data of subnet, N represent the number of subnet, and M represents the extension set number in j-th of subnet；
Step 2：Data are cut into slices and extract section S and complete period (T- Δs t) data D ' to be analyzeds,DisThe data collected by the extension set that numbering is i under s subnets；
Step 3：Spatial match is carried out to the sub-network of deployment and corresponding actual section S, obtain sub-network main frame with it is corresponding
The extension set deployment scenario list of the number information in section and the sub-network；
Step 4：The data of each extension set collection are ranked up according to mobile terminal device ID number, establish data matrix
{Tower(i,s),t}；
Step 5：The ID data lists of foundation are classified by the number for appearing in different pieces of information matrix：For in the time
In section Δ t, what mobile terminal ID only occurred in the single extension set of sub-network, data corresponding to mobile terminal ID are individually extracted into
The follow-up efficiency analysis of row；For in period Δ t, mobile terminal ID occurs in sub-network two and above extension set, should
Data are directly as valid data corresponding to mobile terminal ID
Step 6：For in period Δ t, mobile terminal ID is whole to the movement only in the single extension set appearance of sub-network
Data corresponding to the ID of end, which are individually extracted, carries out follow-up efficiency analysis：1st, when within the Δ t times, mobile terminal ID is at single point
Repeat in the data matrix TOWER of machine, and there is case above in multiple mobile terminal ID, mark corresponding road section is congestion shape
Condition, and such ID data is designated as valid data；2nd, when within the Δ t times, not finding mobile terminal ID in single extension set number
According to repeating in matrix TOWER, then travel through whether the sub-network before and after the sub-network corresponding road section S identical ID number occurs,
If do not occurred, using the ID data as noise data processing, if occurred in other sub-networks, identify into significant figure
According to list；
Step 7：Repeat step 4~6 finishes until data processing；
Step 8：For non-congestion situation, after being screened according to data above, the number of mobile terminal ID number association is re-established
According to matrix { Tower ' (i, s), t }；
Step 9：The relation between time t and Tower in data matrix { Tower ' (i, s), t }, data are entered
One step is divided into bi-directional data matrix DLAnd DR, and approximation computation is carried out to it；
Step 10：Establish flow velocity prediction neural network：
(jth work song network covers the traffic flow speed in section to v in the Δ t periods after T+ Δs t) the expression T moment；(j) it is the
The effective ID of mobile terminal of the direction of j work song networks total amount, m, n represent the extension set numbering in the host subnet network of place,At the time of representing that ID-k occurs in sub-network Tower (j) on m-th of extension set, a (k) represents weight,A (j) and B (j) represents weight regulatory factor；T (j) represents j-th of Tower mileage time average；F
() is approximate fits function；I is the sum of extension set in the subnet；
Step-up error amountIt is C to set self feed back iterative steps, sets stopping criterion for iterationAccording to the error amount of outputBy the value for adjusting weight A (j) and B (j) so that (T+ Δs t) approaches calculation to v
Method training sample S, final network output weight A (j) and B (j) value are as allocative decision.
Further, the S of Algorithm for Training sample described in step 10 is the data that coil checker or radar detector obtain,
And it is used as effective reference unit.
Beneficial effect：The invention provides one kind to be based on the unique ID of portable mobile terminal, gathered data timestamp and detection
The macro-traffic infomation detection algorithm of device location information, realizes the data mining and analysis to mobile terminal gathered data, is
Magnitude of traffic flow detector and detecting system based on WIFI signal provide depth excavation and the magnitude of traffic flow based on the type data
Detection algorithm is realized, has been filled up application blank of the type data in terms of Vehicle Detection, has been promoted the development in wisdom traffic field.
Brief description of the drawings
Fig. 1 is that the road of the traffic throughput monitor system of the present invention based on WIFI signal disposes schematic diagram.
Fig. 2 is the schematic flow sheet of the traffic cross-sectional flow prediction algorithm of the present invention based on WIFI signal.
Fig. 3 is the noise reduction screening of valid data in the traffic cross-sectional flow prediction algorithm of the present invention based on WIFI signal
Schematic flow sheet；
Fig. 4 is that data approximate fits approach very in the traffic cross-sectional flow prediction algorithm of the present invention based on WIFI signal
The process schematic of real value；
Fig. 5 is the data and the comparison diagram of coil True Data of road detection system collection of the present invention based on WIF signals.
Fig. 6 is the enlarged drawing of encircled portion in Fig. 5.
Fig. 7 is the comparison diagram of inventive algorithm output result cross-sectional flow
Embodiment
Below in conjunction with the accompanying drawings and with specific embodiment, the present invention is furture elucidated.It should be understood that these embodiments are only used for
The bright present invention rather than limitation the scope of the present invention, after the present invention has been read, those skilled in the art are to the present invention's
The modification of the various equivalent form of values falls within the application appended claims limited range.
As shown in figure 1, in inventive network, comprehensive reference current moment target road section and approach way changes in flow rate, move
Dynamic Termination ID unit mileage hourage change, inputs historical data, by fitting function, realizes T+ time Δt traffic flow speeds
Prediction v (T+ Δ t), and Algorithm for Training is carried out dependent on referential S, feedback Rule of judgment is v/S, can be according to precision of prediction needs
The ratio domain value range is adjusted, when v/S income values are in domain value range, data fitting operations are completed in expression, formed with
Imitate cross-sectional flow prediction data.Specifically：
Predict device systems deployment way：In Fig. 1, Tower (j) represents j-th of sub-network host computer in disposed road network,
Tower (i, j) represents i-th of extension set in j-th of subnet.Each sub-network includes a main frame and several extension sets, extension set
Quantity can suitably increase and decrease according to road network condition, sub-network maximum coverage range 2Km, single extension set estimation range maximum radius
250m, user can adjust single extension set estimation range according to actual road conditions condition, between extension set distance d settings only need to be more than twice
Extension set estimation range, equally can according to actual road conditions condition carry out flexible modulation.Sub-network deployment density can be according to actual friendship
Logical environmental management demand is disposed.
Predict device data acquisition flow：Single pre- measurement equipment is extension set by wireless passive perceptual model, i.e., using TI
CC3XXXX family chips detector, by based on wifi agreements collection from mobile terminal device at random to surrounding environment send out
WIFI broadcast data bags are sent, and screens the wherein packet with equipment id information and is retrieved.Stamp on after extension set label
Main frame is reached, the data being collected into are carried out unified storage and stamp time tag by main frame, and are uploaded in data server
Row storage, is waited to be analyzed.Inventive algorithm principle process is specifically described：
First, the data D of complete period is gathered by extension set,Wherein, DijTable
Show No. i-th extension set data of j-th of subnet, N represents the number of subnet, and M represents the extension set number in j-th of subnet；
Step 2：Data are cut into slices and extract section S and complete period (T- Δs t) data D ' to be analyzeds,
Step 3：Spatial match is carried out to the sub-network of deployment and corresponding actual section S, obtain sub-network main frame with it is corresponding
The extension set deployment scenario list of the number information in section and the sub-network；
Step 4：The data of each extension set collection are ranked up according to mobile terminal device ID number, establish data matrix
{Tower(i,s),t}；
Step 5：The ID data lists of foundation are classified by the number for appearing in different pieces of information matrix：For in the time
In section Δ t, what mobile terminal ID only occurred in the single extension set of sub-network, data corresponding to mobile terminal ID are individually extracted into
The follow-up efficiency analysis of row；For in period Δ t, mobile terminal ID occurs in sub-network two and above extension set, should
Data are directly as valid data corresponding to mobile terminal ID
Step 6：For in period Δ t, mobile terminal ID is whole to the movement only in the single extension set appearance of sub-network
Data corresponding to the ID of end, which are individually extracted, carries out follow-up efficiency analysis：1st, when within the Δ t times, mobile terminal ID is at single point
Repeat in the data matrix TOWER of machine, mark corresponding road section is congestion, and the ID data are designated as into valid data；
2nd, repeat when within the Δ t times, not finding mobile terminal ID in single extension set data matrix TOWER, then traversal should
Whether the sub-network before and after sub-network corresponding road section S there is identical ID number, if do not occurred, using the ID data as noise
Data processing, if occurred in other sub-networks, identify into valid data list；
Step 7：Repeat step 4~6 finishes until data processing；
Step 8：For non-congestion situation, after being screened according to data above, the number of mobile terminal ID number association is re-established
According to matrix { Tower ' (i, s), t }；
Step 9：The relation between time t and Tower in data matrix { Tower ' (i, s), t }, data are entered
One step is divided into bi-directional data matrix DLAnd DR, and approximation computation is carried out to it；
Step 10：Flow velocity prediction neural network is established, as shown in Figure 4：
ID_X represents effective ID data under corresponding Tower；
The mileage time of each ID under the conditions of the corresponding Tower of T_X expressions；
Σ (j) represents the summation of ID quantity；
T (j) represents that the mileage time under corresponding Tower takes average；
A (j), B (j) are weight regulatory factor；
F () represents approximate fits function；
Z represents the feedback factor under self study；
Bias reference system S represents effective reference unit, such as coil checker data, radar detector data, uses
In Algorithm for Training；(T+ Δs t) represents to predict the traffic flow speed that the Tower (j) under following T+ time Δts covers section v.Specifically
Algorithm is as follows：
In formula：
(jth work song network covers the traffic flow speed in section to v in the Δ t periods after T+ Δs t) the expression T moment；(j) it is the
The effective ID of mobile terminal of the direction of j work song networks total amount, m, n represent the extension set numbering in the host subnet network of place,At the time of representing that ID-k occurs in sub-network Tower (j) on m-th of extension set, a (k) represents weight,A (j) and B (j) represents weight regulatory factor；T (j) represents j-th of Tower mileage time average.
Step-up error amountIt is C to set self feed back iterative steps, sets stopping criterion for iterationAccording to the error amount of outputBy the value for adjusting weight A (j) and B (j) so that (T+ Δs t) approaches calculation to v
Method training sample S, final network output weight A (j) and B (j) value are as allocative decision.Algorithm for Training sample described in step 10
This S is the data that coil checker or radar detector obtain, and is used as effective reference unit.
Tested below with the supreme extra large G42 highways section in Beijing, in road both sides, the installation and deployment present invention hands over
Through-flow amount detection systems.As shown in figure 5, the data and coil of the road detection system collection for being the present invention based on WIF signals are true
The comparison diagram of real data.Fig. 6 is the enlarged drawing of encircled portion in Fig. 5.It was found from Fig. 5 and 6, screened by inventive algorithm noise reduction
No. 31 machine data afterwards have high consistency, excellent performance compared with coil True Data (correction data), it was demonstrated that the present invention
Feasibility and data accuracy.
Fig. 7 is the comparison diagram of this algorithm output result cross-sectional flow, it was found from figure, output result and corresponding road section to it is corresponding when
Groove circle flow speed data is highly consistent.
Claims (4)
1. a kind of traffic cross-sectional flow prediction algorithm based on WIFI signal, prediction device systems, institute are deployed with along traffic route
Stating prediction device systems includes multiple sub-networks, and each sub-network includes a main frame and several extension sets, and the extension set is based on
WIFI agreements by wireless passive perceptual model, gather the mobile device based on WIFI signal agreement in surrounding environment send it is wide
Unicast packets, and screen the wherein packet with mobile terminal device id information and retrieved, uploaded after stamping extension set label
To main frame, the data being collected into are carried out unified storage and stamp time tag by main frame, and are uploaded in data server and are stored,
And by being analyzed data to obtain road section flow velocity assessment prediction result.
2. the traffic cross-sectional flow prediction algorithm based on WIFI signal according to claim 1, it is characterised in that：It is described adjacent
Distance is d between extension set, and the signal covering radius of single extension set is r, and d ＞ 2r.
3. the traffic cross-sectional flow prediction algorithm based on WIFI signal according to claim 2, it is characterised in that：The road
Cross-sectional flow assessment prediction comprises the following steps：
Step 1：The data D of complete period is gathered by extension set,Wherein, DijRepresent j-th of subnet
No. i-th extension set data, N represent the number of subnet, and M represents the extension set number in j-th of subnet；
Step 2：Data are cut into slices and extract section S and complete period (T- Δs t) data D to be analyzeds',DisThe data collected by the extension set that numbering is i under s subnets；
Step 3：Spatial match is carried out with corresponding actual section S to the sub-network of deployment, obtains sub-network main frame and corresponding road section
Number information and the sub-network extension set deployment scenario list；
Step 4：The data of each extension set collection are ranked up according to mobile terminal device ID number, establish data matrix { Tower
(i,s),t}；
Step 5：The ID data lists of foundation are classified by the number for appearing in different pieces of information matrix：For in period Δ
In t, what mobile terminal ID only occurred in the single extension set of sub-network, after individually extracting progress to data corresponding to mobile terminal ID
Continuous efficiency analysis；For in period Δ t, mobile terminal ID occurs in sub-network two and above extension set, the movement
Data are directly as valid data corresponding to Termination ID
Step 6：For in period Δ t, what mobile terminal ID only occurred in the single extension set of sub-network, to mobile terminal ID
Corresponding data, which are individually extracted, carries out follow-up efficiency analysis：1st, when within the Δ t times, mobile terminal ID is in single extension set
Repeat in data matrix TOWER, and there is case above in multiple mobile terminal ID, mark corresponding road section is congestion,
And such ID data is designated as valid data；2nd, when within the Δ t times, not finding mobile terminal ID in single extension set data square
Repeat in battle array TOWER, then travel through whether the sub-network before and after the sub-network corresponding road section S identical ID number occurs, if
Do not occur, then using the ID data as noise data processing, if occurred in other sub-networks, identify and arranged into valid data
Table；
Step 7：Repeat step 4~6 finishes until data processing；
Step 8：For non-congestion situation, after being screened according to data above, the data square of mobile terminal ID number association is re-established
Battle array { Tower ' (i, s), t }；
Step 9：The relation between time t and Tower in data matrix { Tower ' (i, s), t }, data are further
It is divided into bi-directional data matrix DLAnd DR, and approximation computation is carried out to it；
Step 10：Establish flow velocity prediction neural network：
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mi>v</mi>
<mrow>
<mo>(</mo>
<mi>T</mi>
<mo>+</mo>
<mi>&Delta;</mi>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>F</mi>
<mrow>
<mo>(</mo>
<mo>&CenterDot;</mo>
<mo>)</mo>
</mrow>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mi>N</mi>
</mrow>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mi>M</mi>
</mrow>
</munderover>
<mi>&Sigma;</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>A</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>A</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>-</mo>
<mi>N</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<mi>A</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<mi>A</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>+</mo>
<mi>M</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>1</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<munderover>
<mi>&Sigma;</mi>
<mrow>
<mi>j</mi>
<mo>-</mo>
<mi>N</mi>
</mrow>
<mrow>
<mi>j</mi>
<mo>+</mo>
<mi>M</mi>
</mrow>
</munderover>
<mi>&Sigma;</mi>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>B</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
<mtd>
<mrow>
<mi>B</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>-</mo>
<mi>N</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<mi>B</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<mi>B</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>+</mo>
<mi>M</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mn>1</mn>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mi>&Sigma;</mi>
<mi>j</mi>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>x</mi>
</munderover>
<mi>I</mi>
<mi>D</mi>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mtable>
<mtr>
<mtd>
<mrow>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>k</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>x</mi>
</munderover>
<mfrac>
<mrow>
<mfrac>
<mrow>
<mo>|</mo>
<msubsup>
<mi>t</mi>
<mi>k</mi>
<mrow>
<mi>T</mi>
<mi>o</mi>
<mi>w</mi>
<mi>e</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<mi>m</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</msubsup>
<mo>-</mo>
<msubsup>
<mi>t</mi>
<mi>k</mi>
<mrow>
<mi>T</mi>
<mi>o</mi>
<mi>w</mi>
<mi>e</mi>
<mi>r</mi>
<mrow>
<mo>(</mo>
<mi>n</mi>
<mo>,</mo>
<mi>j</mi>
<mo>)</mo>
</mrow>
</mrow>
</msubsup>
<mo>|</mo>
</mrow>
<mrow>
<mo>|</mo>
<mi>m</mi>
<mo>-</mo>
<mi>n</mi>
<mo>|</mo>
</mrow>
</mfrac>
<mo>&times;</mo>
<mi>i</mi>
<mo>&times;</mo>
<mi>a</mi>
<mrow>
<mo>(</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</mrow>
<mi>x</mi>
</mfrac>
</mrow>
</mtd>
<mtd>
<mrow>
<mo>(</mo>
<mn>0</mn>
<mo><</mo>
<mi>m</mi>
<mo>,</mo>
<mi>n</mi>
<mo><</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
</mtd>
</mtr>
</mtable>
</mtd>
</mtr>
</mtable>
</mfenced>
(jth work song network covers the traffic flow speed in section to v in the Δ t periods after T+ Δs t) the expression T moment；(j) it is jth number
The effective ID of mobile terminal of the direction of sub-network total amount, m, n represent the extension set numbering in the host subnet network of place,At the time of representing that ID-k occurs in sub-network Tower (j) on m-th of extension set, a (k) represents weight,A (j) and B (j) represents weight regulatory factor；T (j) represents j-th of Tower mileage time average；F
() is approximate fits function；I is the sum of extension set in the subnet；
Step-up error amountIt is C to set self feed back iterative steps, sets stopping criterion for iteration
According to the error amount of outputBy the value for adjusting weight A (j) and B (j) so that and v (T+ Δ t) approximate algorithm training sample S, most
Whole network output weight A (j) and B (j) value are as allocative decision.
4. the traffic cross-sectional flow prediction algorithm based on WIFI signal according to claim 1, it is characterised in that：In step 10
The Algorithm for Training sample S is the data that coil checker or radar detector obtain, and is used as effective reference unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710736698.7A CN107689153B (en) | 2017-08-24 | 2017-08-24 | A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710736698.7A CN107689153B (en) | 2017-08-24 | 2017-08-24 | A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107689153A true CN107689153A (en) | 2018-02-13 |
| CN107689153B CN107689153B (en) | 2018-11-30 |
Family
ID=61154945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710736698.7A Active CN107689153B (en) | 2017-08-24 | 2017-08-24 | A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107689153B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115148020A (en) * | 2022-06-13 | 2022-10-04 | 中国标准化研究院 | Monitoring system and method based on traffic flow in unit time of expressway |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2516479A (en) * | 2013-07-24 | 2015-01-28 | Shane Gregory Dunny | A system for managing vehicular traffic flow within a road network |
| CN106097731A (en) * | 2016-08-16 | 2016-11-09 | 寿光明 | Traffic flow detector based on WIFI signal and detecting system |
| CN106251646A (en) * | 2016-08-16 | 2016-12-21 | 寿光明 | Traffic flow detection system based on WIFI signal and detection method |
| CN106920388A (en) * | 2015-12-24 | 2017-07-04 | 北京奇虎科技有限公司 | A kind of highway monitoring system and control method |
-
2017
- 2017-08-24 CN CN201710736698.7A patent/CN107689153B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2516479A (en) * | 2013-07-24 | 2015-01-28 | Shane Gregory Dunny | A system for managing vehicular traffic flow within a road network |
| CN106920388A (en) * | 2015-12-24 | 2017-07-04 | 北京奇虎科技有限公司 | A kind of highway monitoring system and control method |
| CN106097731A (en) * | 2016-08-16 | 2016-11-09 | 寿光明 | Traffic flow detector based on WIFI signal and detecting system |
| CN106251646A (en) * | 2016-08-16 | 2016-12-21 | 寿光明 | Traffic flow detection system based on WIFI signal and detection method |
Non-Patent Citations (1)
| Title |
|---|
| 李珺: "基于Wi-Fi Direct的道路交通状态信息采集方法", 《公路》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115148020A (en) * | 2022-06-13 | 2022-10-04 | 中国标准化研究院 | Monitoring system and method based on traffic flow in unit time of expressway |
| CN115148020B (en) * | 2022-06-13 | 2023-06-02 | 中国标准化研究院 | Monitoring system and method based on traffic flow in unit time of expressway |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107689153B (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Demissie et al. | Intelligent road traffic status detection system through cellular networks handover information: An exploratory study | |
| CN106205114B (en) | A kind of Freeway Conditions information real time acquiring method based on data fusion | |
| CN107564281B (en) | A kind of macroscopical wagon flow volume forecasting algorithm based on WIFI signal | |
| CN101379536B (en) | Intelligent real-time distributed traffic sampling and navigation system | |
| CN107564283B (en) | A kind of macroscopical wagon flow flow detection algorithm based on WIFI signal | |
| Peixoto et al. | A traffic data clustering framework based on fog computing for VANETs | |
| CN102083204A (en) | Positioning and tracking system method of active nodes in linear environment | |
| CN108109423A (en) | Underground parking intelligent navigation method and system based on WiFi indoor positionings | |
| CN107507419A (en) | A kind of magnitude of traffic flow detector based on WIFI signal | |
| CN102062866A (en) | Method and device for calculating travelling speed between two geographic positions | |
| CN110213710A (en) | A kind of high-performance indoor orientation method, indoor locating system based on random forest | |
| CN202134048U (en) | Scenic area visitor distribution statistical system | |
| CN103152697A (en) | Method for realizing automatic floor positioning by using intelligent mobile phone Wi-Fi (Wireless Fidelity) function | |
| CN111583651A (en) | Road tunnel traffic situation sensing system and method based on radar | |
| CN108616812A (en) | Positioning of mobile equipment and tracing system based on deep learning and its application method | |
| CN107564284B (en) | A kind of traffic based on WIFI signal detection is passed unimpeded grade forecast system | |
| Nandury et al. | Strategies to handle big data for traffic management in smart cities | |
| CN107689153A (en) | A kind of traffic cross-sectional flow prediction algorithm based on WIFI signal | |
| CN109800903A (en) | A kind of profit route planning method based on taxi track data | |
| Cheng et al. | The optimal sampling period of a fingerprint positioning algorithm for vehicle speed estimation | |
| CN107529664B (en) | A kind of traffic based on WIFI signal is passed unimpeded grade detection system | |
| CN107564282B (en) | A kind of traffic cross-sectional flow detection method based on WIFI signal | |
| CN112298293B (en) | System and method for acquiring station passenger behavior trajectory parameters based on 5G | |
| Madhani et al. | Collaborative sensing using uncontrolled mobile devices | |
| CN105608890B (en) | A kind of personnel's trip parametric statistical methods based on mobile phone signal data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 211135 2, B unit 300, Zhihui Road, Kirin science and Technology Innovation Park, Jiangning District, Nanjing, Jiangsu. Applicant after: Nanjing tea, non krypton Mdt InfoTech Ltd Address before: 210000 room 529, science and technology base, 12 Hengguang Road, Nanjing economic and Technological Development Zone, Nanjing, Jiangsu Applicant before: Nanjing tea, non krypton Mdt InfoTech Ltd |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |