Passenger flow statistical method based on wireless communication system
Through backscatter communication and cluster analysis of wireless communication systems, accurate passenger flow statistics without contact and without perception are achieved, solving the problem of insufficient accuracy and convenience of passenger flow statistics in the existing technology, and providing efficient and accurate passenger flow data collection.
Patent Information
- Application Number
- CN202411325651.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-23
- Publication Date
- 2024-12-24
AI Technical Summary
The existing passenger flow statistics methods have shortcomings in terms of accuracy, cost, aesthetics and convenience, especially when used in commercial places, and are difficult to meet the needs, and are limited by the environment and equipment, making it difficult to achieve efficient and accurate passenger flow data collection.
The passenger flow statistics method based on the wireless communication system is adopted, and static clutter removal processing and clustering analysis are performed through the backscattered communication signals at the sending and receiving ends, real-time passenger flow is calculated, and accurate statistics without contact and perception are achieved.
In the case of mobile targets and free movement, the subscriber for the passenger flow statistics provides accurate local and global passenger flow statistics, which has privacy protection, avoids the number of missing and environmental interference problems in traditional methods, and improves the accuracy of statistics. and efficiency.
Smart Images
Figure CN119183147A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of communication perception technology, and in particular to a passenger flow counting method based on a wireless communication system. Background Art
[0002] Traditional passenger flow statistics methods mainly include: manual passenger flow statistics, infrared induction passenger flow statistics, three-roller gate passenger flow statistics, gravity induction passenger flow statistics, etc. Among them, the manual statistics method is to count the passenger flow manually. The accuracy of this method depends on the concentration of the statisticians, duration, uninterrupted, comprehensive statistics, cost and other aspects. Infrared induction passenger flow statistics equipment can be divided into: infrared counter-radiation mode, infrared reflection mode and other equipment. Its main implementation method is to cut off or block the infrared rays of the human body passing through the infrared sensing area to make it produce resistance changes or to determine the number of human bodies by detecting specific infrared rays emitted by the human body. This method has a relatively moderate cost. When people enter and exit the door freely, the system can automatically obtain passenger flow data. The equipment is small and the installation is beautiful. However, there are also shortcomings. When multiple people pass by at the same time, it is easy to miss the number. Due to its own technical reasons, the infrared method cannot judge the direction of customers' entry and exit very well, and can only count whether someone has passed by. Therefore, data collection affects the results of passenger flow analysis. The three-roller gate method mainly adopts a mechanical method. Customers need to pass through the rolling gate to enter the relevant places. The rolling gate rolls once, thereby recording a person in and out. This method is more accurate for data statistics, but because a tripod turnstile machine needs to be installed at the entrance and exit, it lacks aesthetics for shopping malls, and is not convenient enough, and cannot enter and exit quickly, so it is not suitable for applications in shopping malls and other places. Gravity sensing mainly installs a gravity sensing device on the floor to calculate the number of passengers when a person steps on it. However, due to the relatively high installation requirements, high cost, and its stability issues, it is rarely used by commercial users.
[0003] The above-mentioned traditional passenger flow counting methods are difficult to meet the needs of commercial users. Therefore, detection methods based on sensor technology have emerged, such as optical cameras and laser radar. For optical cameras, the first type of algorithm is implemented through motion area detection and tracking algorithms. Its basic principle is to extract the pixel area where motion occurs in the fixed scene video, and combine some prior knowledge of human body size judgment to count these areas. When the size of the area is similar to that of the person in the empirical knowledge, it is determined that someone has passed, thereby realizing the estimation of passenger flow. The second video technology is binocular passenger flow counting. According to the videos obtained by each of the two cameras, a 3D image is obtained through calculation and processing, including the height of the person. By detecting the video content between 1 and 2 meters in height, combined with empirical knowledge, the recognition and judgment of the person are formed. Laser radar relies on the TOF (Time Of Flight) time flight principle, continuously sends light pulses to the target, and then uses the sensor to receive the light returned from the object. The distance of the target object is obtained by detecting the round-trip flight time of these emitted and received light pulses. Among them, optical cameras are easily interfered by environmental factors, and they are overly dependent on the surface texture of the object being photographed. Although the TOF of lidar performs well in strong light and dark light environments, has low cost and fast response speed, it is greatly affected by natural light and infrared rays outdoors and cannot guarantee accuracy at long distances.
[0004] With the development of AI technology, passenger flow counting methods using AI technology are gradually being adopted and used by shopping malls and other places. Through face recognition technology, AI can quickly and accurately identify and track customers entering and leaving the venue. This technology not only improves the accuracy of passenger flow statistics, but also enables real-time monitoring, providing strong support for security prevention. The application of AI technology in passenger flow counting equipment provides us with more accurate and efficient data support, but it also requires the installation of dedicated video equipment and analysis equipment to support its functions. Summary of the invention
[0005] In view of the above-mentioned deficiencies in the prior art, the present invention provides a passenger flow counting method based on a wireless communication system.
[0006] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is: A passenger flow counting method based on a wireless communication system comprises the following steps: Using a transmitting end to transmit a wireless communication signal including a backscatter communication activation signal; Using a wireless communication signal including a backscatter communication activation signal at a receiving end, and in response to the backscatter communication activation signal, modulating the information to be sent onto the wireless communication signal to feed back a perception signal; The receiving end receives the perception signal and calculates the real-time passenger flow based on the perception signal.
[0007] Preferably, calculating the real-time passenger flow according to the perception signal includes: Perform static clutter removal on the sensing signal to generate echo point cloud data of the static target; Cluster the echo point cloud data of static targets to obtain the number of people in each cluster area; The number of people counted in each cluster area is merged to obtain the passenger flow statistics result.
[0008] Preferably, static clutter removal is performed on the sensing signal to generate echo point cloud data of the static target, including: A1. Perform fast Fourier transform on the sensing signal to obtain multiple point cloud data; A2. Identify static targets based on pulse signals in point cloud data; A3. Average the pulse signals of the static target to obtain a reference received pulse signal, then subtract the reference received pulse signal from each pulse signal to obtain the static target echo signal, and generate the echo point cloud data of the static target.
[0009] Preferably, identifying a static target according to a pulse signal in point cloud data includes: According to the point cloud data, three adjacent pulse signals are selected as a pulse group, the absolute difference of the pulse group is calculated, and it is determined whether the absolute difference of the pulse group is less than the threshold value; if so, the current pulse signal is marked as a static target; otherwise, the current pulse signal is marked as a dynamic target.
[0010] Preferably, calculating the absolute difference of the pulse groups comprises: Difference is made between the current pulse signal and the previous pulse signal to obtain a first pulse difference; Difference is made between the next pulse signal and the current pulse signal to obtain a second pulse difference; The absolute value of the difference between the first pulse difference and the second pulse difference is calculated to obtain the absolute difference of the pulse group.
[0011] Preferably, the echo point cloud data of the static target is clustered to obtain the number of people in each clustering area, including: B1. Divide the area to be counted into multiple sub-areas; B2. For each block area, specify the centroid of the block area, and calculate the number of cluster centers according to the number of point clouds in the block area; B3, randomly select the number of point clouds of cluster centers as the initial cluster centers, and classify each point cloud into each cluster according to the distance from each point cloud to the initial cluster center; B4. Recalculate the cluster center of each cluster, and select the point cloud with the smallest distance from the point cloud in each cluster to the cluster center as the new cluster center; B5. Determine whether the number of point clouds in each cluster is less than or equal to the first point cloud number threshold; if so, classify the cluster as an outlier, and delete the cluster center of the cluster from the cluster center set; otherwise, the cluster center set remains unchanged; B6. Calculate the distance between each cluster center in the cluster center set, select the two clusters with the smallest distance between the cluster centers, use the point cloud with the smallest distance from the cluster center of the relative cluster in the two clusters as the temporary center point, and determine whether the number of point clouds within the set radius of the temporary center point is greater than the second point cloud number threshold; if so, merge the two clusters, and delete the cluster centers of the two clusters from the cluster center set; otherwise, the cluster center set remains unchanged; B7. Recalculate the cluster center of each cluster, and start searching for each cluster from the point cloud farthest from the cluster center until the number of point clouds that do not meet the set radius is less than or equal to the second point cloud number threshold. Define the clustered point cloud as a new cluster, and take the remaining point clouds of the original cluster that are not included in the clustered point cloud as another new cluster. Recalculate the cluster centers of the two new clusters, add the two new cluster centers to the cluster center set, and delete the cluster center of the original cluster from the cluster center set. B8, repeat steps B6 to B7 until the cluster center is no longer updated or the number of merges reaches the set number; B9. Repeat steps B4 to B8 until the set maximum number of cycles is reached or the cluster center is no longer updated, and the number of cluster centers in each block area is used as the statistical number of people in each cluster area.
[0012] Preferably, the number of people counted in each cluster area is merged to obtain passenger flow statistics results, including: Calculate the distance from each cluster center to the centroid of each block area, and take the block area with the smallest distance as the block area to which the cluster center belongs. The cluster center is only recorded once in the statistics.
[0013] Preferably, the number of people counted in each cluster area is merged to obtain passenger flow statistics results, including: Calculate the proportion of the point cloud in each cluster in each block area, take the block area with the largest proportion as the block area to which the cluster center belongs, and delete the counting information or point cloud information recorded in other block areas.
[0014] Preferably, the receiving end includes: A remote wireless signal receiving device, or a near-end device for a customer traffic statistics subscriber.
[0015] As a preference, it also includes: The passenger flow statistics results are displayed on the subscription terminal.
[0016] The present invention has the following beneficial effects: The present invention designs a synaesthesia pattern, based on existing wireless signals and combined with backscatter communication signals, and can provide accurate local and global passenger flow statistics for passenger flow statistics subscribers without contact, perception, awareness, free movement, and privacy protection for mobile targets. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 A flow chart of a passenger flow counting method based on a wireless communication system; Figure 2 is a schematic diagram of an embodiment of a receiving end; Figure 3 A schematic diagram of another embodiment of a receiving end. DETAILED DESCRIPTION
[0018] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.
[0019] like Figure 1 As shown, a passenger flow counting method based on a wireless communication system provided by an embodiment of the present invention includes the following steps S1 to S3: S1. Using a transmitting end to send a wireless communication signal including a backscatter communication activation signal; S2, using the wireless communication signal including the backscatter communication activation signal at the receiving end, and in response to the backscatter communication activation signal, modulating the information to be sent on the wireless communication signal to feed back the perception signal; S3. Utilize the receiving end to receive the sensing signal and calculate the real-time passenger flow based on the sensing signal.
[0020] This embodiment integrates wireless communication and wireless perception. Both wireless communication and wireless perception are based on electromagnetic wave theory. They can not only parse the source information carried by the electromagnetic wave signal, but also parse the perception information of the wireless environment characteristics. This dual function of communication and perception makes ISAC (Integrated Sensing And Communication) possible. Unlike the signal modulation and transmission process of traditional active wireless communication, the device (or tag) supporting backscatter communication does not have the ability to generate carrier waves and cannot "actively" send signals to the outside. Instead, it modulates the information that the device needs to send on a third-party carrier signal to change the physical properties of the third-party carrier signal such as amplitude, phase or frequency, thereby realizing the modulation and transmission of its own information without the need for RF devices such as mixers and oscillators. Since BC (Backscatter Communication) can complete the transmission of perception data in a very low-power backscatter communication mode, it is considered to integrate ISAC with BC to realize the synaesthesia integration of backscatter equipment. The present invention can realize passenger flow statistics in existing communication system equipment by designing a synaesthesia pattern in combination with backscatter communication technology.
[0021] In an optional embodiment of the present invention, calculating the real-time passenger flow according to the sensing signal includes: Perform static clutter removal on the sensing signal to generate echo point cloud data of the static target; Cluster the echo point cloud data of static targets to obtain the number of people in each cluster area; The number of people counted in each cluster area is merged to obtain the passenger flow statistics result.
[0022] In this embodiment, based on the principle of static clutter removal method, traditional methods include zero-speed channel zeroing method, moving target display method and phasor mean cancellation algorithm, some of which are not friendly to micro-moving targets, and some algorithms can only suppress the DC component to a certain extent, but the overall suppression effect is not very obvious. In view of these problems, the present invention proposes a static clutter removal method, which uses the difference between pulses to mark static points first, and then subtracts the mean value from the static points, which can effectively distinguish static targets, micro-moving targets and dynamic targets.
[0023] This embodiment performs static clutter removal processing on the sensing signal to generate echo point cloud data of the static target, including: A1. Perform fast Fourier transform on the sensing signal to obtain multiple point cloud data; A2. Identify static targets based on pulse signals in point cloud data; A3. Average the pulse signals of the static target to obtain a reference received pulse signal, then subtract the reference received pulse signal from each pulse signal to obtain the static target echo signal, and generate the echo point cloud data of the static target.
[0024] The identification of static targets based on pulse signals in point cloud data includes: According to the point cloud data, three adjacent pulse signals are selected as a pulse group, the absolute difference of the pulse group is calculated, and it is determined whether the absolute difference of the pulse group is less than the threshold value; if so, the current pulse signal is marked as a static target; otherwise, the current pulse signal is marked as a dynamic target.
[0025] The calculation of the absolute difference of the pulse group includes: Difference is made between the current pulse signal and the previous pulse signal to obtain a first pulse difference; Difference is made between the next pulse signal and the current pulse signal to obtain a second pulse difference; The absolute value of the difference between the first pulse difference and the second pulse difference is calculated to obtain the absolute difference of the pulse group.
[0026] Exemplarily, this embodiment first performs a fast Fourier transform on the received data to obtain data of each point cloud; Then three adjacent pulses are grouped together. Among the three adjacent pulses, the difference between the two groups is taken. The difference between the second pulse and the first pulse is recorded as , the difference between the third pulse and the second pulse is recorded as , and then calculate the absolute value of the difference between the two groups , d is compared with the threshold value T. If d is less than T, it is marked as a static point, otherwise it is marked as a dynamic point; Finally, the received pulses are averaged to obtain the reference received pulse, and then the target echo signal can be obtained by subtracting the reference received pulse from each pulse to form a point cloud. , two-dimensional coordinates ( , ), except for those marked as dynamic points.
[0027] The present invention designs a static clutter removal method for effectively distinguishing static targets, micro-moving targets and moving targets, and ensuring the overall suppression effect of static clutter removal.
[0028] In this embodiment, based on the clustering method, the basic methods include K-means and DBSCAN (Density-Based Spatial Clustering of Applications with Noise), and improvements based on the basic methods, such as conditional merging, combination of two algorithms, clustering algorithm with prediction, and search area formed by using the characteristics of the mobile target itself, are used to improve the accuracy of clustering. The present invention adopts a clustering method that combines coarse search with fine search, and conditional merging and conditional splitting.
[0029] This embodiment clusters the echo point cloud data of the static target to obtain the number of people in each clustering area, including: B1. Divide the area to be counted into multiple sub-areas; B2. For each block area, specify the centroid of the block area, and calculate the number of outliers according to the number of point clouds in the block area; for example, for each block area, specify the centroid B of the block area, and preliminarily calculate the number of cluster centers according to the number of point clouds N and the constant A in the block area ,in is the proportion of non-outliers, then 1- is the proportion of outliers; B3, randomly select the number of point clouds of cluster centers as initial cluster centers, and classify each point cloud into each cluster according to the distance from each point cloud to the initial cluster center; illustratively, randomly select Point clouds are used as initial cluster centers ; Calculate each point cloud arrive The distance between the cluster centers is calculated based on the minimum distance principle. Classify to In the clusters; B4. Recalculate the cluster center of each cluster, and select the point cloud with the smallest distance from the point cloud in each cluster to the cluster center as the new cluster center; illustratively, recalculate each cluster center D, that is, calculate the mean of the two-dimensional coordinates of the point cloud in each cluster; calculate the distance from the point cloud in each cluster to the cluster center D, and select the point cloud with the smallest distance to point D in each cluster as the new cluster center E; B5. Determine whether the number of point clouds in each cluster is less than or equal to the first point cloud number threshold; if so, classify the cluster as an outlier and delete the cluster center of the cluster from the cluster center set; otherwise, the cluster center set remains unchanged; illustratively, for the number of point clouds in each cluster, if it is less than or equal to , then the point cloud in this class is classified as an outlier, and the number of cluster centers is adjusted to , and delete the outlier center from the cluster center E; B6. Calculate the distance between each cluster center in the cluster center set, select the two clusters with the smallest distance between the cluster centers, and use the point cloud with the smallest distance from the cluster center of the relative cluster in the two clusters as the temporary center point, and determine whether the number of point clouds within the set radius of the temporary center point is greater than the second point cloud number threshold; if so, merge the two clusters and delete the cluster centers of the two clusters from the cluster center set; otherwise, the cluster center set remains unchanged; illustratively, calculate the distance between the cluster centers in E, starting from the two clusters with the smallest distance. and ,turn up Zhongli The nearest point cloud z, if z is the temporary center point and there are m point clouds within the radius r, if , then merge these two clusters. , delete the corresponding two cluster centers in E; B7. Recalculate the cluster center of each cluster, start searching for each cluster from the point cloud farthest from the cluster center, cluster the point clouds that meet the clustering conditions, and search the clustered point clouds again until the number of point clouds that do not meet the set radius is less than or equal to the second point cloud number threshold. Then define the clustered point cloud as a new cluster, and take the remaining point clouds in the original cluster that are not included in the clustered point cloud as another new cluster. Re-cluster the two new clusters. New cluster centers are calculated, and two new cluster centers are added to the cluster center set, and the cluster center of the original cluster cluster is deleted from the cluster center set; for example, the merged cluster center is calculated, that is, the mean of the two-dimensional coordinates of the point cloud in the merged cluster is calculated, and the cluster center E is re-added; for all merged clusters, the search starts from the point cloud farthest from the cluster center, and the point clouds that meet the clustering conditions are clustered, and the clustered point clouds are searched again for the clustering conditions until they are not satisfied and the radius r is the radius, and , then define this cluster as a new cluster , the remaining part is another new cluster, the two separated clusters recalculate the cluster centers respectively, delete the original cluster center points from E, and re-add the two new cluster centers; B8, repeat steps B6 to B7 until the cluster center is no longer updated or the number of merges reaches the set number; B9. Repeat steps B4 to B8 until the set maximum number of cycles is reached or the cluster center is no longer updated, and the number of cluster centers in each block area is used as the statistical number of people in each cluster area.
[0030] The present invention designs a clustering method, which first adopts a coarse clustering method of randomly selecting center point clustering, and combines fine clustering methods such as condition-based merging and splitting operations for accurate passenger flow statistics in block areas.
[0031] In this embodiment, the embodiment transitions from regional statistics to overall statistics based on a soft merging or hard merging method.
[0032] The hard merging method is to merge the number of people counted in each cluster area to obtain the passenger flow statistics results, including: Calculate the distance from each cluster center to the centroid of each block area, and take the block area with the smallest distance as the block area to which the cluster center belongs. The cluster center is only recorded once in the statistics.
[0033] This embodiment addresses the problem that point clouds at the edge or overlapping areas of a block area may appear in adjacent cluster centers, and performs deduplication processing during overall statistics. For example, the right edge of a block area is usually a line segment, but a part of the point cloud cluster may appear to the left of the edge line segment, while another part of the point cloud cluster appears to the right of the edge line segment. The point cloud cluster is counted in the block area, and will also be counted in the block area to the right of the block area. It is okay to count the block area in this way, but in the end, it can only be counted once when the overall count is performed. The extra duplicate counts need to be processed to ensure the correctness of the final statistical results.
[0034] The soft merging method is to merge the number of people counted in each cluster area to obtain the passenger flow statistics results, including: Calculate the proportion of the point cloud in each cluster in each block area, take the block area with the largest proportion as the block area to which the cluster center belongs, and delete the counting information or point cloud information recorded in other block areas.
[0035] Choose one of the merging methods to calculate the combined number of people in the entire venue.
[0036] The block area merging method designed by the present invention is divided into soft merging and hard merging methods, which mainly solve the problem of repeated counting from block statistics to overall statistics, and a single merging method can be used to merge passenger flows in the overall area.
[0037] The present invention provides accurate local and global passenger flow statistics for passenger flow statistics subscribers under the condition that the mobile target is contactless, has no perception awareness, moves freely, and has privacy protection.
[0038] In an optional embodiment of the present invention, the receiving end of this embodiment includes: A remote wireless signal receiving device, such as Figure 2As shown; or the proximal device of the customer flow statistics subscriber, such as Figure 3 shown.
[0039] In an optional embodiment of the present invention, the embodiment further includes: The passenger flow statistics results are displayed on the subscription terminal.
[0040] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0041] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0042] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0043] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.
[0044] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.
Claims
1. A passenger flow counting method based on a wireless communication system, characterized in that: The following steps are involved: Using a transmitting end to transmit a wireless communication signal including a backscatter communication activation signal; Using a backscatter communication device to respond to a backscatter communication activation signal in a wireless communication signal, modulate the information to be sent on the wireless communication signal to feed back a perception signal; The receiving end receives the perception signal and calculates the real-time passenger flow based on the perception signal.
2. A passenger flow counting method based on a wireless communication system according to claim 1, characterized in that: Calculating real-time passenger flow based on sensing signals includes: Perform static clutter removal on the sensing signal to generate echo point cloud data of the static target; Cluster the echo point cloud data of static targets to obtain the number of people in each cluster area; The number of people counted in each cluster area is merged to obtain the passenger flow statistics result.
3. A passenger flow counting method based on a wireless communication system according to claim 2, characterized in that: Perform static clutter removal on the sensing signal to generate echo point cloud data of static targets, including: A1. Perform fast Fourier transform on the sensing signal to obtain multiple point cloud data; A2. Identify static targets based on pulse signals in point cloud data; A3. Average the pulse signals of the static target to obtain a reference received pulse signal, then subtract the reference received pulse signal from each pulse signal to obtain the static target echo signal, and generate the echo point cloud data of the static target.
4. The passenger flow counting method based on a wireless communication system according to claim 3, characterized in that: Identifying static targets based on pulse signals in point cloud data includes: According to the point cloud data, three adjacent pulse signals are selected as a pulse group, the absolute difference of the pulse group is calculated, and it is determined whether the absolute difference of the pulse group is less than the threshold value; if so, the current pulse signal is marked as a static target; otherwise, the current pulse signal is marked as a dynamic target.
5. The passenger flow counting method based on a wireless communication system according to claim 4, characterized in that: Calculating the absolute difference of a pulse group involves: Difference is made between the current pulse signal and the previous pulse signal to obtain a first pulse difference; Difference is made between the next pulse signal and the current pulse signal to obtain a second pulse difference; The absolute value of the difference between the first pulse difference and the second pulse difference is calculated to obtain the absolute difference of the pulse group.
6. The passenger flow counting method based on a wireless communication system according to claim 2, characterized in that: Cluster the echo point cloud data of static targets to obtain the number of people in each cluster area, including: B1. Divide the area to be counted into multiple sub-areas; B2. For each block area, specify the centroid of the block area, and calculate the number of cluster centers according to the number of point clouds in the block area; B3, randomly select the number of point clouds of cluster centers as the initial cluster centers, and classify each point cloud into each cluster according to the distance from each point cloud to the initial cluster center; B4. Recalculate the cluster center of each cluster, and select the point cloud with the smallest distance from the point cloud in each cluster to the cluster center as the new cluster center; B5. Determine whether the number of point clouds in each cluster is less than or equal to the first point cloud number threshold; if so, classify the cluster as an outlier, and delete the cluster center of the cluster from the cluster center set; otherwise, the cluster center set remains unchanged; B6. Calculate the distance between each cluster center in the cluster center set, select the two clusters with the smallest distance between the cluster centers, use the point cloud with the smallest distance from the cluster center of the relative cluster in the two clusters as the temporary center point, and determine whether the number of point clouds within the set radius of the temporary center point is greater than the second point cloud number threshold; if so, merge the two clusters, and delete the cluster centers of the two clusters from the cluster center set; otherwise, the cluster center set remains unchanged; B7. Recalculate the cluster center of each cluster, and start searching for each cluster from the point cloud farthest from the cluster center until the number of point clouds that do not meet the set radius is less than or equal to the second point cloud number threshold. Define the clustered point cloud as a new cluster, and take the remaining point clouds of the original cluster that are not included in the clustered point cloud as another new cluster. Recalculate the cluster centers of the two new clusters, add the two new cluster centers to the cluster center set, and delete the cluster center of the original cluster from the cluster center set. B8, repeat steps B6 to B7 until the cluster center is no longer updated or the number of merges reaches the set number; B9. Repeat steps B4 to B8 until the set maximum number of cycles is reached or the cluster center is no longer updated, and the number of cluster centers in each block area is used as the statistical number of people in each cluster area.
7. The passenger flow counting method based on a wireless communication system according to claim 2, characterized in that: The number of people counted in each cluster area is combined to obtain the passenger flow statistics results, including: Calculate the distance from each cluster center to the centroid of each block area, and take the block area with the smallest distance as the block area to which the cluster center belongs. The cluster center is only recorded once in the statistics.
8. The passenger flow counting method based on a wireless communication system according to claim 2, characterized in that: The number of people counted in each cluster area is combined to obtain the passenger flow statistics results, including: Calculate the proportion of the point cloud in each cluster in each block area, take the block area with the largest proportion as the block area to which the cluster center belongs, and delete the counting information or point cloud information recorded in other block areas.
9. The passenger flow counting method based on a wireless communication system according to claim 1, characterized in that: The receiving end includes: A remote wireless signal receiving device, or a near-end device for a customer traffic statistics subscriber.
10. The passenger flow counting method based on a wireless communication system according to claim 1, characterized in that: Also includes: The passenger flow statistics results are displayed on the subscription terminal.