A bluetooth signal based traffic analysis method and system

Abstract

The embodiment of the specification provides a Bluetooth signal-based traffic analysis method and system, the method comprising: dividing a region into a plurality of first areas and a plurality of second areas according to a first division method and a second division method respectively, wherein each second area in the plurality of second areas comprises a plurality of first areas; determining a first crowd in each first area in the plurality of first areas based on received data of a Bluetooth transceiver related to each first area; determining a second crowd in each second area in the plurality of second areas based on the first crowd in each first area of the plurality of first areas included in each second area; and determining a total crowd in the region based on the second crowd in each second area in the plurality of second areas.

Description

Technical Field

[0001] This manual relates to the field of pedestrian flow analysis, and in particular to a method and system for pedestrian flow analysis based on Bluetooth signals. Background Technology

[0002] Large commercial areas and tourist attractions have relatively high crowd densities, often leading to stampedes and other safety accidents due to overcrowding. Current methods for calculating crowd flow primarily rely on smart devices receiving Bluetooth or iBeacon signals. Based on these signals, spatiotemporal data is directly generated to determine crowd volume. However, the presence of obstructing buildings within these areas can prevent the attenuation of Bluetooth or iBeacon signals, resulting in low accuracy in crowd flow prediction.

[0003] Therefore, it is desirable to provide a traffic flow analysis method and system based on Bluetooth signals to improve the accuracy of traffic flow prediction. Summary of the Invention

[0004] This specification provides one or more embodiments of a Bluetooth signal-based traffic analysis method. The Bluetooth signal-based traffic analysis method includes: dividing a region into multiple first regions and multiple second regions according to a first division method and a second division method; wherein each of the multiple second regions includes multiple first regions; determining a first pedestrian flow in each of the multiple first regions based on received data from Bluetooth transceivers associated with each of the multiple first regions; determining a second pedestrian flow in each of the multiple second regions based on the first pedestrian flow in each of the multiple first regions included in each of the multiple second regions; and determining the total pedestrian flow in the region based on the second pedestrian flow in each of the multiple second regions.

[0005] This specification provides one or more embodiments of a Bluetooth-based traffic analysis system, including a segmentation module, a first pedestrian flow determination module, a second pedestrian flow determination module, and a total pedestrian flow determination module. The segmentation module is used to divide a region into multiple first regions and multiple second regions according to a first segmentation method and a second segmentation method, respectively. Each of the multiple second regions includes multiple first regions. The first pedestrian flow determination module is used to determine the first pedestrian flow of each of the multiple first regions based on received data from Bluetooth transceivers associated with each of the multiple first regions. The second pedestrian flow determination module is used to determine the second pedestrian flow of each of the multiple second regions based on the first pedestrian flow of each of the multiple first regions contained within each of the multiple second regions. The total pedestrian flow determination module is used to determine the total pedestrian flow of the region based on the second pedestrian flow of each of the multiple second regions.

[0006] This specification provides one or more embodiments of a Bluetooth-based traffic analysis device, including a processor for executing a Bluetooth-based traffic analysis method.

[0007] This specification provides one or more embodiments of a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes a traffic analysis method based on Bluetooth signals. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is a schematic diagram illustrating an application scenario of a Bluetooth-based traffic analysis system according to some embodiments of this specification;

[0010] Figure 2 This is an exemplary block diagram of a Bluetooth-based traffic analysis system according to some embodiments of this specification;

[0011] Figure 3 This is an exemplary flowchart of a traffic analysis method based on Bluetooth signals, as shown in some embodiments of this specification;

[0012] Figure 4 This is an exemplary flowchart illustrating the determination of the first pedestrian flow in each of a plurality of first areas according to some embodiments of this specification;

[0013] Figure 5This is an exemplary flowchart illustrating the determination of second pedestrian flow in each of a plurality of second areas according to some embodiments of this specification;

[0014] Figure 6 This is an exemplary flowchart illustrating a method for determining the total pedestrian traffic in an area according to some embodiments of this specification. Detailed Implementation

[0015] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0016] It should be understood that the terms “system,” “device,” “unit,” and / or “module” used herein are one way to distinguish different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0017] As indicated in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not specifically refer to the singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of expressly identified steps and elements, which do not constitute an exclusive list, and the method or apparatus may also include other steps or elements.

[0018] Flowcharts are used in this specification to illustrate the operations performed by the system according to embodiments of this specification. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0019] Figure 1 This is a schematic diagram illustrating an application scenario of traffic analysis based on Bluetooth signals, according to some embodiments of this specification.

[0020] Some embodiments of this specification can be applied to determine visitor flow in large commercial areas or tourist attractions. For example, a server can predict the total visitor flow in a tourist attraction based on received data from Bluetooth transceivers in various designated areas of the attraction.

[0021] like Figure 1As shown, the server 110 predicts the total number of visitors to the tourist area based on the received data from multiple terminal devices 120 (e.g., Bluetooth transceivers 120-1, 120-2, ..., 120-n) related to the various divisions of the tourist area.

[0022] In some embodiments, server 110 can be used to manage resources and process data and / or information from at least one component of the application scenario (e.g., terminal device 120 and storage 140) or external data sources (e.g., cloud data centers).

[0023] In some embodiments, server 110 may include processing device 112. In some embodiments, terminal device 120 may include processing device 112. Processing device 112 may process data and / or information obtained from terminal device 120. For example, processing device 112 may acquire received data from terminal device 120 related to a segmented area, and determine the pedestrian flow in the segmented area based on the received data of the segmented area, thereby determining the total pedestrian flow in the entire area.

[0024] In some embodiments, the terminal device 120 may be installed within an area for transmitting and receiving data. For example, the terminal device 120 may include multiple Bluetooth transceivers (e.g., Bluetooth transceiver 120-1, Bluetooth transceiver 120-2, ..., Bluetooth transceiver 120-n, etc.). In some embodiments, each Bluetooth transceiver may transmit Bluetooth signals to any other Bluetooth transceiver and receive Bluetooth signals transmitted by any other Bluetooth transceiver. In some embodiments, each Bluetooth transceiver may have a unique signal identifier, enabling the Bluetooth transceiver to distinguish signals from different transceivers.

[0025] In some embodiments, network 130 can facilitate the exchange of information and / or data. In some embodiments, one or more components of application scenario 100 (e.g., server 110, terminal device 120, and memory 140) can send information and / or data to other components of application scenario 100 via network 130. For example, server 110 can generate control commands and send them to at least one of a plurality of terminal devices 120 via network 130. As another example, terminal device 120 can send received data to server 110 via network 130.

[0026] In some embodiments, memory 140 may be used to store data and / or instructions. In some embodiments, memory 140 may be included in server 110, terminal device 120, and other possible application components. In some embodiments, memory 140 may store data obtained from server 110 and terminal device 120. In some embodiments, memory 140 may store data and / or instructions used by server 110 to perform or use in order to complete the exemplary methods described herein. In some embodiments, memory 140 may include mass storage, removable storage, read-write storage, read-only storage, etc., or any combination thereof. Exemplarily, mass storage may include disks, optical disks, etc. In some embodiments, memory 140 may be implemented on a cloud platform.

[0027] Figure 2 This is an exemplary block diagram of a Bluetooth-based traffic analysis system according to some embodiments of this specification.

[0028] In some embodiments, the Bluetooth-based traffic analysis system 200 may include a segmentation module 210, a first traffic flow determination module 220, a second traffic flow determination module 230, and a total traffic flow determination module 240.

[0029] In some embodiments, the partitioning module 210 can be used to divide the region into a plurality of first regions and a plurality of second regions according to a first partitioning method and a second partitioning method, respectively. Each of the plurality of second regions includes a plurality of first regions. For more information on this part, please see below. Figure 3 The description of step 310 in the document.

[0030] In some embodiments, the first pedestrian flow determination module 220 can be used to determine the first pedestrian flow in each of the multiple first regions based on the received data of the Bluetooth transceiver associated with each of the multiple first regions.

[0031] In some embodiments, the second pedestrian flow determination module 230 may be used to determine the second pedestrian flow of each of the plurality of second regions based on the first pedestrian flow of each of the plurality of first regions contained in each of the plurality of second regions.

[0032] In some embodiments, the total pedestrian flow determination module 240 can be used to determine the total pedestrian flow of a region based on the second pedestrian flow of each of the plurality of second regions.

[0033] In some embodiments, the first pedestrian flow determination module 220 may be further configured to compare the received data of the Bluetooth transceivers associated with each of the multiple first regions with the received data of a blank control group based on the transmit power of the Bluetooth transceivers associated with each of the multiple first regions, to determine the Bluetooth signal attenuation data of each of the multiple first regions. Then, based on the received data of the Bluetooth transceivers associated with each of the multiple first regions and the Bluetooth signal attenuation data of each of the multiple first regions, the first pedestrian flow in each of the multiple first regions is determined by a first prediction model. The first prediction model is a machine learning model. For more details on this part, please see below. Figure 4 And its description.

[0034] In some embodiments, the second pedestrian flow determination module 230 may be further configured to determine the second pedestrian flow of each of the plurality of second regions based on the first pedestrian flow of each of the plurality of first regions contained in each of the plurality of second regions and the spatial layout of each of the plurality of second regions. For more details on this section, please see below. Figure 5 And its description.

[0035] In some embodiments, the total pedestrian flow determination module 240 may be further configured to determine multiple second partitioning methods, and use the sum of the second pedestrian flows corresponding to each of the multiple second partitioning methods as a candidate value for the total pedestrian flow. Then, a weighted calculation is performed based on the multiple candidate values ​​for the total pedestrian flow corresponding to the multiple second partitioning methods to determine the total pedestrian flow. For more details on this part, please refer to the following text. Figure 6 And its description.

[0036] It should be understood that Figure 1 and Figure 2 The system and its modules shown can be implemented in various ways. For example, in some embodiments, Figure 2 The system shown can be implemented using a computer-readable storage medium.

[0037] It should be noted that the above description of the Bluetooth-based traffic analysis system and its modules is for convenience only and should not be construed as limiting this specification to the embodiments described. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the various modules or construct subsystems connected to other modules without departing from these principles. In some embodiments, Figure 2The first and second pedestrian flow determination modules disclosed herein can be different modules within the same system, or a single module can implement the functions of two or more of the aforementioned modules. For example, the modules can share a single storage module, or each module can have its own separate storage module. Such variations are all within the scope of protection of this specification.

[0038] Figure 3 This is an exemplary flowchart illustrating a Bluetooth signal-based traffic analysis method according to some embodiments of this specification. In some embodiments, process 300 may be executed by server 110. Figure 3 As shown, process 300 includes the following steps 310-340.

[0039] Step 310: Divide the region into multiple first regions and multiple second regions according to the first division method and the second division method respectively; wherein each of the multiple second regions includes multiple first regions. In some embodiments, step 310 may be performed by the division module 210.

[0040] A region can refer to a specific geographical space. For example, a region can include one or more of the following: any scenic area, any commercial district, etc.

[0041] The first region refers to a portion of the area defined by one of the boundaries of the Bluetooth transceivers placed within it. For example, the first region can be a portion of the area that includes at least two Bluetooth transceivers. The second region can be formed by merging multiple first regions, with a central Bluetooth transceiver as the center and multiple Bluetooth transceivers surrounding that central transceiver. For example, the second region can be formed by merging six first regions.

[0042] The first partitioning method refers to any feasible method of dividing a region into multiple first regions. For example, the first partitioning method can be the equal area partitioning method.

[0043] The second partitioning method refers to any feasible method for dividing a region into multiple second regions. In some embodiments, the first partitioning method differs from the second partitioning method. In some embodiments, the order in which the first and second partitioning methods are used to partition the region can be interchanged. For example, multiple first regions can be obtained first using the first partitioning method, and then multiple second regions can be obtained based on the obtained multiple first regions using the second partitioning method. Alternatively, multiple second regions can be obtained first using the second partitioning method, and then multiple first regions can be obtained based on the obtained multiple second regions using the first partitioning method.

[0044] In some embodiments, the second partitioning method may include a hexagonal honeycomb partitioning method. The hexagonal honeycomb partitioning method may refer to dividing each second region into seven hexagons. These seven equal-sized hexagons are arranged such that, with one hexagon as the center, any one side of the remaining six hexagons coincides with one side of the central hexagon.

[0045] In some embodiments, the first partitioning method may include dividing each first region into the sum of the outer hexagon and the central hexagon of each second region. In some embodiments, each second region may include six first regions.

[0046] In some embodiments, the partitioning module 210 can divide the regions one-to-one according to the first partitioning method and the second partitioning method to obtain multiple first regions and multiple second regions. In some embodiments, the order in which the first partitioning method and the second partitioning method are used to partition the regions can be interchanged.

[0047] In some embodiments, the partitioning module 210 may first obtain multiple second regions using a second partitioning method, and then obtain multiple first regions using a first partitioning method. For example, as shown below... Figure 6 As shown, the partitioning module 210 can partition multiple second regions using a second partitioning method (e.g., a hexagonal honeycomb partitioning method). Each second region can include hexagons A, B, C, D, E, F, and G. Then, the partitioning module 210 can use a first partitioning method to divide each first region into the sum of all the outer hexagons of each second region (e.g., hexagons A, B, C, D, E, or F) and one-sixth of the central hexagon (e.g., one-sixth of hexagon G), thereby partitioning multiple first regions.

[0048] Step 320: Based on the received data from the Bluetooth transceiver associated with each of the plurality of first regions, determine the first pedestrian flow in each of the plurality of first regions. In some embodiments, step 320 may be performed by the first pedestrian flow determination module 220.

[0049] The Bluetooth transceivers associated with the first area include Bluetooth transceivers placed within the first area and Bluetooth transceivers that transmit or receive Bluetooth signals through the first area.

[0050] The first pedestrian flow refers to the total number of people passing through a certain first area during a certain time period.

[0051] In some embodiments, the first pedestrian flow in each of the plurality of first areas can be determined manually.

[0052] In some embodiments, the first pedestrian flow determination module 220 may, based on the transmit power of the Bluetooth transceiver associated with each of the plurality of first areas, compare the received data of the Bluetooth transceiver associated with each of the plurality of first areas with the received data of a blank control group to determine the Bluetooth signal attenuation data for each of the plurality of first areas. Then, based on the received data of the Bluetooth transceiver associated with each of the plurality of first areas and the Bluetooth signal attenuation data for each of the plurality of first areas, a first pedestrian flow for each of the plurality of first areas is determined using a first prediction model, wherein the first prediction model is a machine learning model. For more details on this section, please see below. Figure 4 And its description.

[0053] Step 330: Based on the first pedestrian flow of each of the multiple first regions contained in each of the multiple second regions, determine the second pedestrian flow of each of the multiple second regions. In some embodiments, step 330 may be performed by the second pedestrian flow determination module 230.

[0054] The second pedestrian flow refers to the total number of people passing through a certain second area during a certain time period.

[0055] In some embodiments, the second pedestrian flow determination module 230 may use the sum of the first pedestrian flow in the six first areas as the second pedestrian flow.

[0056] In some embodiments, the second pedestrian flow determination module 230 can determine the second pedestrian flow in each of the plurality of second regions based on the first pedestrian flow in each of the plurality of first regions contained in each of the plurality of second regions and the spatial layout of each of the plurality of second regions. For a more detailed description of this part, please see below. Figure 5 And its related descriptions.

[0057] Step 340: Determine the total pedestrian flow of the area based on the second pedestrian flow in each of the plurality of second areas. In some embodiments, step 340 may be performed by the total pedestrian flow determination module 240.

[0058] Total pedestrian flow refers to the total number of people passing through the entire area during a certain period of time.

[0059] In some embodiments, the total pedestrian flow determination module 240 may use the sum of the second pedestrian flows in multiple second areas as the total pedestrian flow.

[0060] In some embodiments, the total pedestrian flow determination module 240 can determine multiple second division methods and use the sum of the second pedestrian flows corresponding to each of the multiple second division methods as a candidate value for the total pedestrian flow. Then, a weighted calculation is performed based on the multiple candidate values ​​for the total pedestrian flow corresponding to the multiple second division methods to determine the total pedestrian flow. For a more detailed description of this part, please see below. Figure 6 And its related descriptions.

[0061] In some embodiments of this specification, by collecting the received data of the Bluetooth transceiver in the first area, the first pedestrian flow in the first area is determined. Then, based on the first pedestrian flow, the relationship between the signal attenuation due to obstruction and the crowd in the first area can be obtained, thereby accurately determining the second pedestrian flow in the second area and obtaining the total pedestrian flow in the entire area, thereby improving the prediction accuracy of the total pedestrian flow in the entire area.

[0062] Figure 4 This is an exemplary schematic diagram illustrating the determination of the first pedestrian flow in each of a plurality of first areas according to some embodiments of this specification.

[0063] like Figure 4 As shown, the first pedestrian flow determination module can compare the received data 411 of the Bluetooth transceiver associated with each of the multiple first areas with the received data 412 of the blank control group to determine the Bluetooth signal attenuation data 422 of each first area.

[0064] The received data for the blank control group was obtained by measuring the first area under cleared conditions, based on the transmission power of the Bluetooth transceiver associated with the first area. Cleared conditions refer to zero pedestrian traffic within the first area while other conditions (such as object placement) remain unchanged. For example, the received data for the blank control group was obtained by measuring the cleared first area under conditions where the transmission power was the same as that of the Bluetooth transceiver associated with the first area.

[0065] Bluetooth signal attenuation data refers to the amount of Bluetooth signal loss due to obstruction by pedestrian traffic. In some embodiments, Bluetooth signal attenuation data can be the difference between the received data of the Bluetooth transceiver in the blank control group and the received data of the current Bluetooth transceiver.

[0066] In some embodiments, for each of the plurality of first regions, the first pedestrian flow 440 of the first region can be determined by a first prediction model 430 based on the received data 411 and Bluetooth signal attenuation data 422 of the Bluetooth transceiver associated with the first region.

[0067] A first prediction model processes received data and Bluetooth signal attenuation data from Bluetooth transceivers associated with each of the multiple first regions to determine the first pedestrian flow in each of the multiple first regions. In some embodiments, the inputs to the first prediction model include received data 411 from Bluetooth transceivers associated with each of the multiple first regions, the transmit power 421 of the Bluetooth transceivers, Bluetooth signal attenuation data 422, and the location distribution of the Bluetooth transceivers 423. In some embodiments, the output of the first prediction model is the first pedestrian flow 440 in each of the multiple first regions. In some embodiments, the first prediction model is a machine learning model, such as a deep neural network model, a convolutional neural network model, etc.

[0068] In some embodiments, a first prediction model can be trained based on a large amount of labeled training data. Specifically, the labeled training data is input into the initial first prediction model, and the parameters of the initial first prediction model are updated by various methods based on the labels and the output of the initial first prediction model. For example, training can be based on gradient descent, and training ends when a first preset condition is met. The first preset condition can be the convergence of the loss function. In some embodiments, the training data can be historical pedestrian flow monitoring data (including historical Bluetooth transceiver reception data, historical Bluetooth transceiver transmission power, historical Bluetooth signal attenuation data, and historical Bluetooth transceiver location distribution related to each of multiple first areas). In some embodiments, the labels can be historical actual pedestrian flow data for each of multiple first areas, such as historical actual pedestrian flow data for each of multiple first areas obtained through monitoring equipment.

[0069] In some embodiments, the first pedestrian flow in each of the multiple first regions can be determined at the current moment based on the received data and Bluetooth signal attenuation data of the Bluetooth transceivers in each of the multiple first regions corresponding to multiple consecutive time points before the current moment, using a second prediction model.

[0070] The second prediction model processes the received data and Bluetooth signal attenuation data of Bluetooth transceivers in each of multiple first regions corresponding to multiple consecutive time points before the current time to determine the first pedestrian flow in each of the multiple first regions. In some embodiments, the inputs of the second prediction model include the received data, transmit power, Bluetooth signal attenuation data, and Bluetooth transceiver location distribution of Bluetooth transceivers associated with each of the multiple first regions corresponding to multiple consecutive time points before the current time. In some embodiments, the output of the second prediction model is the first pedestrian flow in each of the multiple first regions. In some embodiments, the second prediction model is a machine learning model, such as a deep neural network model, a convolutional neural network model, etc.

[0071] In some embodiments, the second prediction model may include multiple first prediction models and a determining model. Each of the multiple first prediction models processes the received data, transmit power, Bluetooth signal attenuation data, and Bluetooth transceiver location distribution of Bluetooth transceivers associated with each of the multiple first regions corresponding to a given time point, thereby determining the first pedestrian flow in each of the multiple first regions corresponding to that time point. The determining model processes the first pedestrian flow in each of the multiple first regions corresponding to each time point output by each of the multiple first prediction models to determine the final first pedestrian flow in each of the multiple first regions.

[0072] In some embodiments, a second prediction model can be trained based on a large amount of labeled training data. Specifically, the labeled training data is input into an initial second prediction model, and the initial second prediction model is trained and its parameters are updated based on the labels and the output of the initial second prediction model using various methods. For example, training can be performed using gradient descent, and training ends when a second preset condition is met. The second preset condition can be the convergence of the loss function. In some embodiments, the training data can be historical pedestrian flow monitoring data corresponding to multiple consecutive historical time points (including historical Bluetooth transceiver reception data, historical Bluetooth transceiver transmission power, historical Bluetooth signal attenuation data, and historical Bluetooth transceiver location distribution related to each of multiple first regions corresponding to multiple consecutive historical time points). In some embodiments, the labels can be the historical actual pedestrian flow of each of multiple first regions corresponding to the next time point of multiple consecutive historical time points, such as the historical actual pedestrian flow of each of multiple first regions at the next time point obtained through monitoring equipment.

[0073] In some embodiments, the training of the second prediction model includes the joint training of multiple first prediction models and a deterministic model. Specifically, labeled training data is input into multiple initial first prediction models, and the initial deterministic model processes the outputs of the multiple initial first prediction models to generate its own output. Based on the labels and the outputs of the initial deterministic model, training is performed using various methods to update the parameters of the multiple initial first prediction models and the initial deterministic model. For example, training can be performed using gradient descent, and training ends when a pre-defined joint training condition is met. This pre-defined joint training condition can be the convergence of the loss function. In some embodiments, the training data can be historical pedestrian flow monitoring data corresponding to multiple consecutive historical time points (including historical Bluetooth transceiver reception data, historical Bluetooth transceiver transmission power, historical Bluetooth signal attenuation data, and historical Bluetooth transceiver location distribution corresponding to multiple consecutive historical time points in multiple first regions). In some embodiments, the labels can be the historical actual pedestrian flow in each of the multiple first regions corresponding to the next time point of the multiple consecutive historical time points, such as the historical actual pedestrian flow in each of the multiple first regions at the next time point obtained through monitoring equipment.

[0074] The second prediction model processes the received data and Bluetooth signal attenuation data of the Bluetooth transceivers in each of the multiple first regions corresponding to multiple consecutive time points before the current time. It can combine the data from multiple historical time points to predict the first pedestrian flow, making the predicted value of the first pedestrian flow more accurate.

[0075] In some embodiments, various measurement methods can be used to improve the prediction accuracy of the first pedestrian flow in the first area. For example, Bluetooth signals can be transmitted based on multiple different transmission powers, and Bluetooth signal attenuation data corresponding to each transmission power can be determined based on the blank control group and the received data of the Bluetooth transceiver corresponding to each transmission power. Then, the average value of the different first pedestrian flows determined by the different Bluetooth signal attenuation data can be taken to determine the predicted value of the first pedestrian flow in each of the multiple first areas. As another example, the roles of the Bluetooth transmitter and the Bluetooth receiver can be swapped (i.e., the original Bluetooth transmitter becomes the current Bluetooth receiver, and the original Bluetooth receiver becomes the current Bluetooth transmitter), Bluetooth signal attenuation data in two directions can be measured, and the average value of the two first pedestrian flows determined by the Bluetooth signal attenuation data in the two directions can be used as the predicted value of the first pedestrian flow in each of the multiple first areas.

[0076] By transmitting Bluetooth signals at multiple different transmission powers and using the blank control group and Bluetooth transceiver reception data corresponding to each transmission power, various measurement methods such as determining the Bluetooth signal attenuation data corresponding to each transmission power and swapping the roles of Bluetooth transmitters and receivers can improve the prediction accuracy of the first pedestrian flow in each of the multiple first areas, making the final prediction of the total pedestrian flow in the area more accurate.

[0077] Figure 5 This is an exemplary schematic diagram illustrating the determination of the second pedestrian flow in each of a plurality of second areas according to some embodiments of this specification.

[0078] In some embodiments, for each of a plurality of second regions, the second pedestrian flow of each of the plurality of second regions can be determined based on the first pedestrian flow of each of the plurality of first regions and the spatial layout of the second region.

[0079] The spatial layout of the second area refers to the spatial arrangement within the second area, such as the area occupied by obstructing buildings and the area occupied by open spaces.

[0080] In some embodiments, the sum of the first pedestrian flows of the plurality of first areas contained in a plurality of second areas may be corrected based on the spatial layout of each of the plurality of second areas to determine the second pedestrian flow of each of the plurality of second areas. In some embodiments, the correction includes determining whether to compensate for or deduplicate the second pedestrian flow.

[0081] In some embodiments, the obstruction of Bluetooth signals by buildings can prevent the flow of people behind the buildings from being reflected in the Bluetooth signal attenuation data (i.e., regardless of how many people are behind the building, the Bluetooth signal will still be blocked). Therefore, compensation can be made for the sum of the first flow of people in multiple first areas, and the compensation amount can be added to the sum of the first flow of people in multiple first areas to obtain the second flow of people in each of multiple second areas. For example, the average flow of people behind the obstructing building can be used as the compensation value, or the compensation value can be determined based on the area occupied by the obstructing building. The average flow of people behind the obstructing building can be determined based on statistics from historical monitoring equipment, and the area occupied by the obstructing building can be determined based on the building planning drawings of the area.

[0082] In some embodiments, the pedestrian flow along the common edge of any two adjacent first areas may be counted repeatedly. Therefore, correcting the sum of the first pedestrian flows of the multiple first areas contained in each of the multiple second areas also includes deduplicating the duplicated pedestrian flows along the common edge of any two adjacent first areas. For example, if the pedestrian flow along the common edge of two adjacent first areas is 10 people and is counted repeatedly, deduplication includes subtracting 10 people from the sum of the first pedestrian flows of the multiple first areas. Duplicate values ​​for other adjacent common edges are processed similarly. The pedestrian flow along the common edge of two adjacent first areas can be obtained based on monitoring equipment.

[0083] In some embodiments, for each second region, the second pedestrian flow 530 of the second region can also be determined based on the region knowledge graph 510 and the third prediction model 520.

[0084] A regional knowledge graph can refer to a graph related to a region. A regional knowledge graph can include multiple nodes and multiple edges.

[0085] In a regional knowledge graph, node types can include first-region nodes, second-region nodes, and Bluetooth transceiver nodes, etc. First-region nodes refer to nodes comprised of nodes in the first region. Second-region nodes refer to nodes comprised of nodes in the second region. Bluetooth transceiver nodes refer to nodes that use Bluetooth transceivers. For example, see... Figure 5 The regional knowledge graph can include a first regional node represented by "1", a second regional node represented by "2", and a Bluetooth transceiver node represented by "3". Nodes have node attributes. For example, the node attributes of the first regional node include the first pedestrian flow, the node attributes of the second regional node include spatial layout, and the node attributes of the Bluetooth transceiver node include the distribution location of the Bluetooth transceiver, transmission frequency, and received data.

[0086] Multiple nodes can be connected by edges, and the attributes of the edges can reflect the relationships between the nodes. When there is a dependency relationship or other preset association between two nodes, they are connected by an edge. In some embodiments, the edge connecting a first region node represented by "1" and a second region node represented by "2" indicates that the first region node belongs to the second region node. In some embodiments, the edge connecting a first region node represented by "1" and a Bluetooth transceiver node represented by "3" indicates that the Bluetooth transceiver node is a Bluetooth transceiver associated with the first region.

[0087] A third prediction model is used to process a regional knowledge graph to determine the second pedestrian flow in each of a plurality of second regions. In some embodiments, the input of the third prediction model includes a regional knowledge graph, and the output is the second pedestrian flow in each second region. In some embodiments, the third prediction model is a graph neural network model.

[0088] In some embodiments, a third prediction model can be trained based on a large amount of labeled training data. Specifically, the labeled training data is input into the initial third prediction model, and the parameters of the initial third prediction model are updated using various methods based on the labels and the output of the initial third prediction model. For example, training can be based on gradient descent, and training ends when a third preset condition is met. The third preset condition can be the convergence of the loss function. In some embodiments, the training data can be a knowledge graph of multiple historical regions. In some embodiments, the labels can be the historical actual pedestrian traffic of each second region, such as the historical actual pedestrian traffic of each of multiple second regions obtained through monitoring equipment.

[0089] In some embodiments, the confidence level of the second pedestrian flow prediction value output by the third prediction model can also be determined based on the second pedestrian flow at each historical moment in the second region. For example, a corresponding normal distribution curve of pedestrian flow can be fitted to each second region based on the pedestrian flow at each historical moment in the second region, and the confidence level of the second pedestrian flow prediction value output by the third prediction model for each second region can be determined based on the normal distribution curve of pedestrian flow. For example, the probability corresponding to the second pedestrian flow prediction value on the normal distribution curve of pedestrian flow can be used as the confidence level of the second pedestrian flow prediction value.

[0090] In some embodiments, when the confidence level of the second pedestrian flow prediction is less than a threshold, multiple first areas can be sampled to determine at least one sampled first area, and the real-time pedestrian flow in the at least one sampled first area can be determined using a monitoring device. If the real-time pedestrian flow in the first area differs significantly from the predicted value of the first pedestrian flow, the Bluetooth receiver retransmits the Bluetooth signal with a different transmission power to predict the first pedestrian flow and then re-predicts the second pedestrian flow. In some embodiments, the first pedestrian flow can be re-predicted using a first prediction model, and the second pedestrian flow can be re-predicted using a second prediction model. In some embodiments, if the real-time pedestrian flow in the first area differs significantly from the predicted value of the first pedestrian flow, the first prediction model can be trained based on the real-time pedestrian flow in the first area to update the parameters of the first prediction model and make the prediction of the first prediction model more accurate.

[0091] By determining the second pedestrian flow in the second region based on the regional knowledge graph and the third prediction model, and by combining the spatial layout information of the second region with the relationship between the first and second regions, a more accurate value of the second pedestrian flow can be predicted.

[0092] In some embodiments, multiple second partitioning methods can be determined, and the sum of the second pedestrian flows corresponding to each of the multiple second partitioning methods is used as a candidate value for total pedestrian flow; and a weighted calculation is performed based on multiple candidate values ​​for total pedestrian flow to determine the total pedestrian flow of a region. The candidate value for total pedestrian flow refers to the predicted total pedestrian flow of a candidate region. In some embodiments, the candidate value for total pedestrian flow is the sum of the second pedestrian flows corresponding to each of the multiple second partitioning methods. In some embodiments, each second partitioning method may correspond to a set of predicted values ​​for the second pedestrian flow of all second regions, thus there is a one-to-one correspondence between the second partitioning method and the candidate value for total pedestrian flow.

[0093] Multiple second partitioning methods can be any feasible method for dividing the second region using the hexagonal honeycomb partitioning method.

[0094] like Figure 6 As shown, the second division method can include at least two methods: division method A 610 and division method B 620. Both division method A 610 and division method B 620 use a hexagonal honeycomb division method to closely divide the second region. Each second region obtained by division includes 7 regular hexagonal regions, but the specific regular hexagonal regions included in each second region are different in the two division methods A and B.

[0095] like Figure 6 As shown, in partitioning method A 610, a second region is composed of hexagons A, B, C, D, E, F, and G. In partitioning method B 620, a second region is composed of regular hexagonal regions: hexagons D, E, F, G, H, I, and J. This demonstrates that different combinations of regular hexagonal regions can form different second partitioning methods. These different second partitioning methods can correspond to different Bluetooth transceiver location distributions. For example... Figure 6 As shown, the positions of the Bluetooth transceivers are represented by triangles. In partitioning method A, the Bluetooth transceiver can be configured at the center of hexagon G and on the sides of hexagons A, B, C, D, E, and F. In partitioning method B, the Bluetooth transceiver can be configured at the center of hexagon E and on the sides of hexagons D, F, G, H, I, and J.

[0096] In some embodiments, the weights for the weighted calculation of each second partitioning method are determined based on the sum of the confidence levels of the predicted second pedestrian traffic values ​​for multiple second regions corresponding to that second partitioning method. For example, if there are six second regions corresponding to partitioning method A, and the confidence levels of the predicted second pedestrian traffic values ​​for each second region are 0.8, 0.7, 0.6, 0.6, 0.8, and 0.9, respectively, then the sum of the confidence levels of the predicted second pedestrian traffic values ​​for multiple second regions corresponding to partitioning method A is 4.4. The weights corresponding to the sum of confidence levels are determined according to preset rules. In some embodiments, the larger the sum of confidence levels, the larger the weight. The confidence level of the predicted second pedestrian traffic value for each second region can be determined based on the historical pedestrian traffic of each second region. For methods of determining confidence levels, see [link to relevant documentation]. Figure 5 Related descriptions.

[0097] The total pedestrian flow of a region is determined by weighted calculation based on multiple candidate values ​​of total pedestrian flow. The weight of each second division method is determined by the sum of the confidence levels of the predicted second pedestrian flow values ​​of multiple second regions corresponding to that second division method, which can make the predicted total pedestrian flow of the region more accurate.

[0098] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0099] Furthermore, this specification uses specific terms to describe embodiments thereof. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "an embodiment," "one embodiment," or "an alternative embodiment" in different locations throughout this specification do not necessarily refer to the same embodiment. Moreover, certain features, structures, or characteristics in one or more embodiments of this specification can be appropriately combined.

[0100] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this specification are not intended to limit the order of the processes and methods described herein. Although various examples have been discussed in the foregoing disclosure of some embodiments of the invention that are currently considered useful, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the spirit and scope of the embodiments described herein. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely using software solutions, such as installing the described system on existing servers or mobile devices.

[0101] Similarly, it should be noted that, in order to simplify the description disclosed herein and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of embodiments in this specification may sometimes combine multiple features into a single embodiment, drawing, or description thereof. However, this method of disclosure does not imply that the subject matter of this specification requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of a single embodiment disclosed above.

[0102] In some embodiments, numbers describing the quantity of components and attributes are used. It should be understood that such numbers used in the description of embodiments are modified in some examples with the terms "approximately," "approximately," or "generally." Unless otherwise stated, "approximately," "approximately," or "generally" indicates that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the specification and claims are approximate values, which may be changed depending on the characteristics required by individual embodiments. In some embodiments, numerical parameters should take into account specified significant digits and employ a general method of digit reservation. Although the numerical ranges and parameters used to confirm their breadth of range in some embodiments of this specification are approximate values, in specific embodiments, such values ​​are set as precisely as feasible.

[0103] For each patent, patent application, patent application publication, and other material, such as articles, books, specifications, publications, and documents, referenced in this specification, the entire contents of which are incorporated herein by reference. This excludes historical application documents that are inconsistent with or conflict with the content of this specification, as well as documents that limit the broadest scope of the claims in this specification (currently or subsequently appended to this specification). It should be noted that in the event of any inconsistency or conflict between the descriptions, definitions, and / or terminology used in the supplementary materials to this specification and the content of this specification, the descriptions, definitions, and / or terminology used in this specification shall prevail.

[0104] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A traffic analysis method based on Bluetooth signals, characterized in that, The method includes: The region is divided into multiple first regions and multiple second regions according to a first division method and a second division method, respectively; wherein each second region includes multiple first regions; the second division method includes a hexagonal honeycomb division method; the first division method includes equal area division or dividing each first region into the sum of the entirety of an outer hexagon and one-sixth of the central hexagon of each second region; the first region is a partial region including at least two Bluetooth transceivers; the second region is formed by merging the multiple first regions, with a Bluetooth transceiver as the center, and multiple Bluetooth transceivers surrounding the central Bluetooth transceiver. Based on the transmit power of the Bluetooth transceiver associated with each first region, the received data of the Bluetooth transceiver associated with each first region is compared with the received data of the blank control group to determine the Bluetooth signal attenuation data of each first region; Based on the received data of the Bluetooth transceiver and the Bluetooth signal attenuation data of each first area corresponding to multiple consecutive time points before the current time, the first pedestrian flow of each first area is determined by a second prediction model. The second prediction model includes multiple first prediction models and a determining model. Each of the multiple first prediction models is used to determine the first pedestrian flow of each first area based on the received data of the Bluetooth transceiver and the Bluetooth signal attenuation data of each first area corresponding to a certain time point. The determining model is used to process the first pedestrian flow of each first area corresponding to each time point output by each first prediction model to determine the final first pedestrian flow of each first area. Based on a regional knowledge graph, a graph neural network model is used to determine the second pedestrian flow in each second region. The regional knowledge graph includes first region nodes, second region nodes, and Bluetooth transceiver nodes, as well as edges connecting multiple nodes. The attributes of the first region nodes include the first pedestrian flow; the attributes of the second region nodes include spatial layout; and the attributes of the Bluetooth transceiver nodes include the distribution location, transmission frequency, and received data of the Bluetooth transceivers. The spatial layout includes the area occupied by obstructing buildings and open spaces in the second region. Multiple second partitioning methods are determined, and the sum of the second pedestrian flow in each second region corresponding to each second partitioning method is used as a candidate value for the total pedestrian flow; and The total pedestrian flow of the region is determined by weighted calculation based on multiple candidate values ​​of total pedestrian flow corresponding to the various second division methods, wherein the weight of each second division method is determined based on the confidence level of the second pedestrian flow of the multiple second regions corresponding to the second division method.

2. A traffic analysis system based on Bluetooth signals, characterized in that, It includes a segmentation module, a first pedestrian flow determination module, a second pedestrian flow determination module, and a total pedestrian flow determination module; The partitioning module is used to divide the region into multiple first regions and multiple second regions according to a first partitioning method and a second partitioning method, respectively; wherein each second region includes multiple first regions; the second partitioning method includes a hexagonal honeycomb partitioning method; the first partitioning method includes equal area partitioning or partitioning each first region into the sum of the entirety of an outer hexagon and one-sixth of the central hexagon of each second region; the first region is a partial region including at least two Bluetooth transceivers; the second region is formed by merging the multiple first regions, with a Bluetooth transceiver as the center, and multiple Bluetooth transceivers surrounding the central Bluetooth transceiver. The first pedestrian flow determination module is used to compare the received data of the Bluetooth transceiver associated with each first area with the received data of the blank control group based on the transmission power of the Bluetooth transceiver associated with each first area, and determine the Bluetooth signal attenuation data of each first area; Based on the received data of the Bluetooth transceiver and the Bluetooth signal attenuation data of each first area corresponding to multiple consecutive time points before the current time, the first pedestrian flow of each first area is determined by a second prediction model. The second prediction model includes multiple first prediction models and a determining model. Each of the multiple first prediction models is used to determine the first pedestrian flow of each first area based on the received data of the Bluetooth transceiver and the Bluetooth signal attenuation data of each first area corresponding to a certain time point. The determining model is used to process the first pedestrian flow of each first area corresponding to each time point output by each first prediction model to determine the final first pedestrian flow of each first area. The second pedestrian flow determination module is used to determine the second pedestrian flow of each second region based on a regional knowledge graph and a graph neural network model. The regional knowledge graph includes first region nodes, second region nodes, Bluetooth transceiver nodes, and edges connecting multiple nodes. The attributes of the first region nodes include the first pedestrian flow, the attributes of the second region nodes include spatial layout, and the attributes of the Bluetooth transceiver nodes include the distribution location, transmission frequency, and received data of the Bluetooth transceivers. The spatial layout includes the area occupied by obstructing buildings and open spaces in the second region. The total pedestrian flow determination module is used to determine multiple second division methods, and to use the sum of the second pedestrian flow in each second region corresponding to each second division method as a candidate value for the total pedestrian flow; and The total pedestrian flow of the region is determined by weighted calculation based on multiple candidate values ​​of total pedestrian flow corresponding to the various second division methods, wherein the weight of each second division method is determined based on the confidence level of the second pedestrian flow of the multiple second regions corresponding to the second division method.

3. A traffic analysis device based on Bluetooth signals, characterized in that, The device includes at least one processor and at least one memory; The at least one memory is used to store computer instructions; The at least one processor is configured to execute at least a portion of the computer instructions to implement the Bluetooth signal-based traffic analysis method as described in claim 1.

4. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions from the storage medium, the computer executes the traffic analysis method based on Bluetooth signals as described in claim 1.

Images