Method and device for detecting position change of electronic price tag

Abstract

The invention discloses a position change detection method and device for an electronic price tag, and the method comprises the steps: obtaining a plurality of base stations capable of receiving signals of the electronic price tag, and constructing a positioning data buffer queue for the electronic price tag by each base station according to the received signals; according to the positioning data buffer queue of each base station for the electronic price tag, judging whether each base station participates in the movement judgment of the electronic price tag; for each base station participating in movement judgment, the base station participating in movement judgment is divided into two positioning data cache sub-queues for the positioning data cache queue of the electronic price tag, and whether the electronic price tag moves or not is judged by comparing the two positioning data cache sub-queues to obtain a judgment result; and judging whether the electronic price tag moves or not according to judgment results of all the base stations participating in movement judgment. According to the invention, the position change of the electronic price tag can be effectively detected.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method and apparatus for detecting changes in the position of electronic price tags. Background Technology

[0002] This section is intended to provide background or context for the embodiments of the invention set forth in the claims. The description herein is not an admission that it is prior art simply because it is included in this section.

[0003] Product location information is crucial for the smartification and digitalization of supermarkets. However, in real-world supermarket scenarios, factors such as display strategies and seasonal changes can cause changes in the location of products, and consequently, the location of corresponding electronic shelf labels. Therefore, it is necessary to monitor the location of electronic shelf labels to determine if changes have occurred.

[0004] Existing technologies have several significant shortcomings in addressing changes in the position of goods and electronic shelf labels. From the perspective of position change monitoring, current technologies largely rely on manual inspections or fixed sensor networks to track the location of electronic shelf labels. Manual inspections are not only labor-intensive but also prone to missed detections and misjudgments due to limitations in inspection frequency and staff negligence. While fixed sensor networks can improve monitoring efficiency to some extent, their deployment costs are high, and signals are easily affected by factors such as obstruction from shelves and interference from metal products within supermarkets. Furthermore, they often create blind spots in densely populated areas, making it difficult to accurately capture subtle movements of electronic shelf labels.

[0005] In summary, a high-precision electronic price tag position change detection solution is currently needed. Summary of the Invention

[0006] This invention provides a method for detecting position changes in electronic price tags, which can effectively detect position changes in electronic price tags, including:

[0007] Multiple base stations capable of receiving signals from electronic price tags are obtained, and each base station constructs a location data cache queue for the electronic price tag based on the received signal;

[0008] Based on the location data cache queue of each base station for the electronic price tag, determine whether each base station participates in the movement determination of the electronic price tag;

[0009] For each base station participating in the motion determination, the location data cache queue of the electronic price tag of the base station participating in the motion determination is divided into two location data cache sub-queues. By comparing the two location data cache sub-queues, it is determined whether the electronic price tag has moved and the determination result is obtained.

[0010] Based on the judgment results of all base stations involved in the movement determination, it is determined whether the electronic price tag has been moved.

[0011] This invention provides an electronic price tag position change detection device, which can effectively detect position changes of electronic price tags, including:

[0012] The base station acquisition module is used to acquire multiple base stations that can receive signals from the electronic price tag. Each base station constructs a location data cache queue for the electronic price tag based on the received signal.

[0013] The mobile detection base station determination module is used to determine whether each base station participates in the mobile detection of the electronic price tag based on the location data cache queue of each base station for the electronic price tag.

[0014] The first motion determination module is used to divide the location data cache queue of the electronic price tag of each base station participating in motion determination into two location data cache sub-queues, and determine whether the electronic price tag has moved by comparing the two location data cache sub-queues to obtain the determination result.

[0015] The second mobility determination module is used to determine whether the electronic price tag has moved based on the determination results of all base stations participating in the mobility determination.

[0016] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the above-described method for detecting changes in the position of electronic price tags.

[0017] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for detecting changes in the position of electronic price tags.

[0018] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method for detecting changes in the position of electronic price tags.

[0019] In this embodiment of the invention, multiple base stations capable of receiving electronic price tag signals are obtained. Utilizing the signal coverage and positioning data of these multiple base stations, information about the electronic price tag can be acquired from different angles. Compared to positioning with a single base station, this reduces positioning errors caused by factors such as signal obstruction and multipath effects, improving the accuracy and reliability of positioning. For example, in a large shopping mall, multiple base stations can more comprehensively cover the area where the electronic price tag is located, avoiding positioning deviations caused by a single base station due to excessive distance or signal obstruction. The system determines whether a base station participates in motion detection based on its positioning data cache queue for the electronic price tag, and divides and compares the positioning data cache queues of base stations participating in motion detection. This method fully utilizes the historical positioning data stored by the base stations, analyzing data changes over different time periods to determine whether the electronic price tag has moved. This reduces the impact of random factors on the motion detection results and enhances the reliability of motion detection. For instance, when the electronic price tag is subjected to brief signal interference, comparing data from multiple time periods can prevent the interference from being mistaken for movement of the electronic price tag. Instead of relying on complex hardware or high-precision sensors to directly measure the movement of electronic price tags, this system analyzes and processes signal data received from existing base stations to determine location and movement, reducing hardware costs and complexity. Furthermore, utilizing historical data in a cached queue eliminates the need for extensive real-time computations, further reducing system computing resource requirements and lowering operating costs. It can quickly determine whether an electronic price tag has moved based on signals received from the base station and cached location data, meeting the need for real-time monitoring of price tag position changes. It adapts better to complex environments, accurately determining the movement of electronic price tags and ensuring stability and reliability under varying conditions. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0021] Figure 1 This is a flowchart of the electronic price tag position change detection method in an embodiment of the present invention;

[0022] Figure 2 This is a flowchart illustrating the detection of electronic price tag position changes based on multi-base station collaboration in an embodiment of the present invention.

[0023] Figure 3 This is a flowchart illustrating the position update of the electronic price tag in an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram of the channel acquisition point positioning data and acquisition point location information in an embodiment of the present invention;

[0025] Figure 5 This is the architecture of the CNN network in the embodiments of the present invention;

[0026] Figure 6 This is a schematic diagram illustrating the trajectory tracking of the shopping cart and shopping basket in an embodiment of the present invention;

[0027] Figure 7 This is a heat map of the shopping cart generated in an embodiment of the present invention;

[0028] Figure 8 This is a flowchart illustrating the detection of electronic price tag position changes based on the final positioning index value in an embodiment of the present invention.

[0029] Figure 9 This is a schematic diagram of the electronic price tag position change detection device in an embodiment of the present invention;

[0030] Figure 10 This is a schematic diagram of a computer device in an embodiment of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0032] Figure 1 The flowchart of the electronic price tag position change detection method in this embodiment of the invention includes:

[0033] Step 101: Obtain multiple base stations that can receive signals from electronic price tags. Each base station constructs a location data cache queue for the electronic price tags based on the received signals.

[0034] Step 102: Based on the location data cache queue of each base station for the electronic price tag, determine whether each base station participates in the movement determination of the electronic price tag;

[0035] Step 103: For each base station participating in the motion determination, the location data cache queue of the electronic price tag of the base station participating in the motion determination is divided into two location data cache sub-queues. By comparing the two location data cache sub-queues, it is determined whether the electronic price tag has moved and the determination result is obtained.

[0036] Step 104: Based on the judgment results of all base stations involved in the movement judgment, determine whether the electronic price tag has moved.

[0037] In this embodiment of the invention, multiple base stations capable of receiving electronic price tag signals are obtained. Utilizing the signal coverage and positioning data of these multiple base stations, information about the electronic price tag can be acquired from different angles. Compared to positioning with a single base station, this reduces positioning errors caused by factors such as signal obstruction and multipath effects, improving the accuracy and reliability of positioning. For example, in a large shopping mall, multiple base stations can more comprehensively cover the area where the electronic price tag is located, avoiding positioning deviations caused by a single base station due to excessive distance or signal obstruction. The system determines whether a base station participates in motion detection based on its positioning data cache queue for the electronic price tag, and divides and compares the positioning data cache queues of base stations participating in motion detection. This method fully utilizes the historical positioning data stored by the base stations, analyzing data changes over different time periods to determine whether the electronic price tag has moved. This reduces the impact of random factors on the motion detection results and enhances the reliability of motion detection. For instance, when the electronic price tag is subjected to brief signal interference, comparing data from multiple time periods can prevent the interference from being mistaken for movement of the electronic price tag. Instead of relying on complex hardware or high-precision sensors to directly measure the movement of electronic price tags, this system analyzes and processes signal data received from existing base stations to determine location and movement, reducing hardware costs and complexity. Furthermore, utilizing historical data in a cached queue eliminates the need for extensive real-time computations, further reducing system computing resource requirements and lowering operating costs. It can quickly determine whether an electronic price tag has moved based on signals received from the base station and cached location data, meeting the need for real-time monitoring of price tag position changes. It adapts better to complex environments, accurately determining the movement of electronic price tags and ensuring stability and reliability under varying conditions.

[0038] The execution subject of the method proposed in this embodiment of the invention is a backend server, but it can also be other devices with data processing capabilities, which are not limited here.

[0039] In step 101, multiple base stations capable of receiving signals from electronic price tags are obtained, and each base station constructs a location data cache queue for the electronic price tags based on the received signals;

[0040] For a specific electronic shelf label, the signal can be an AOA (Angle of Arrival) signal. This AOA signal is received by multiple base stations. Each base station calculates the AOA signal to obtain positioning data and caches a positioning data queue of a certain length. Therefore, joint motion determination can be performed using the positioning results from multiple base stations. In this embodiment of the invention, the method is applied to a backend server. The multiple base stations capable of receiving the electronic shelf label's signal need to send their location data cache queues to the backend server. The method of this embodiment can be executed for each electronic shelf label.

[0041] In one embodiment, each base station stores the location data of the electronic price tag within a first preset time period in its location data cache queue.

[0042] The location data cache queue pre-stores historical data for a preset duration (e.g., 5 minutes) to avoid the need for real-time acquisition and processing of raw signals for each movement determination. For example, in warehouse management, when multiple base stations process a large amount of price tag data simultaneously, the caching mechanism can reduce system load and ensure real-time performance. Multiple base stations are required to participate in movement determination. By comparing the determination results of different base stations (e.g., at least two base stations trigger movement determination simultaneously), misjudgments caused by multipath effects or temporary interference from a single base station can be filtered out. For example, in a large shopping mall, a base station may lose signal due to customer obstruction, but other base stations can still receive signals normally; a majority voting mechanism can prevent misjudgments.

[0043] In one embodiment, after obtaining signals from multiple base stations capable of receiving electronic price tags, the method further includes:

[0044] Based on the signal quality index value of the electronic price tag, a first preset number of base stations are selected from multiple base stations that can receive the signal of the electronic price tag;

[0045] Based on the location data cache queue for each base station regarding the electronic price tag, determine whether each base station participates in the movement determination of the electronic price tag, including:

[0046] Based on the location data cache queue of each base station in the first preset number of base stations selected, it is determined whether each base station in the first preset number of base stations participates in the movement judgment of the electronic price tag.

[0047] Signal quality metrics include, but are not limited to, RSSI or SNR. When selecting multiple base stations capable of receiving electronic price tags, choose those with higher RSSI or SNR values. The selected base stations can be sorted in descending or ascending order, and the base stations with the highest signal quality metrics should be selected. For descending order, select the first few base stations; for ascending order, select the last few. Subsequent analysis only needs to be performed on the first preset number of selected base stations.

[0048] In step 102, based on the location data cache queue of each base station for the electronic price tag, it is determined whether each base station participates in the movement determination of the electronic price tag;

[0049] In one embodiment, determining whether each base station participates in the movement determination of the electronic price tag based on the location data cache queue of each base station includes:

[0050] For each base station, select the location data within the second preset time period from the location data cache queue for the electronic price tag of that base station;

[0051] Determine whether the number of location data points within the second preset time period of the base station is greater than the set threshold.

[0052] If so, determine whether the base station is involved in the movement of the electronic price tag;

[0053] Otherwise, it is determined that the base station is not involved in the movement determination of the electronic price tag.

[0054] Specifically, for each base station, location data within a second preset duration (which can be represented as the past T_past hours) is selected from the location data cache queue for the electronic price tag. The second preset duration is less than the first preset duration. It is then determined whether the number of location data is greater than a set threshold. If it is greater than the threshold, the base station participates in the movement determination of the electronic price tag; otherwise, it does not participate in the movement determination of the electronic price tag. Therefore, the number of base stations participating in the movement determination does not exceed the total number of all base stations, thus improving the efficiency of movement determination.

[0055] Based on the location data cache queue of each base station in the first preset number of base stations selected for the electronic price tag, when determining whether each base station in the first preset number of base stations should participate in the movement judgment of the electronic price tag, the selected first preset number of N base stations can be traversed first. For any one of the N base stations, the movement judgment of the electronic price tag can be performed. In this way, the number of base stations participating in the movement judgment does not exceed N, thus improving the efficiency of the movement judgment.

[0056] In this embodiment of the invention, the positioning data includes a positioning timestamp and a positioning index value, wherein the positioning index value is location information, angle information, or distance information.

[0057] In step 103, for each base station participating in the motion determination, the location data cache queue of the electronic price tag of the base station participating in the motion determination is divided into two location data cache sub-queues. By comparing the two location data cache sub-queues, it is determined whether the electronic price tag has moved and the determination result is obtained.

[0058] In one embodiment, for each base station participating in the motion determination, the location data cache queue of the electronic price tag for that base station is divided into two location data cache sub-queues. By comparing the two location data cache sub-queues, it is determined whether the electronic price tag has moved, and the determination result is obtained, including:

[0059] For each base station participating in motion determination, the location data cache queue for the electronic price tag will be used as the location timestamp obtained by deducing the first offset time backward from the current time.

[0060] Based on the time point of movement determination, the location data cache queue is divided into two location data cache sub-queues;

[0061] Calculate the ratio of the number of location data in the two location data cache sub-queues. If the ratio exceeds the ratio threshold, determine that the judgment result corresponding to the base station participating in the movement judgment is that the electronic price tag has moved, and record the movement judgment time point.

[0062] If the ratio does not exceed the ratio threshold, the front and back segments of the location data cache queue are trimmed according to the trimming ratio to obtain a trimmed queue. From back to front, each location data in the trimmed queue is used as a split point to divide the location data cache queue. Each split generates two location data cache sub-queues, and the average location index value of the two location data cache sub-queues is calculated. When the difference between the average location index values ​​of the two location data cache sub-queues exceeds the first difference threshold, the splitting stops, and the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has moved. The time point when the base station obtains the location data corresponding to this split is taken as the motion judgment time point. Otherwise, the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has not moved.

[0063] In the above embodiment, for the base station participating in the movement determination of the price tag, for its location data within the past T_past hours, the location timestamp of the first offset time length (generally equal to half the location data cache queue time length, which can be expressed as T_offset hours) is used as the movement determination time point. Based on the movement determination time point, the selected location data is divided into two parts, forming two location data cache sub-queues. The ratio of the number of location data in the two location data cache sub-queues is calculated. If the ratio exceeds a threshold, the electronic price tag is determined to have moved on that base station, and the movement determination time point is recorded; otherwise, it is processed according to the trimming ratio. The front and back segments of the location data cache queue are trimmed to obtain a trimmed queue. From back to front, each location data in the trimmed queue is used as a split point to divide the location data cache queue. Each split generates two location data cache sub-queues, and the average location index value of the two location data cache sub-queues is calculated. When the difference between the average location index values ​​of the two location data cache sub-queues exceeds a first difference threshold, the splitting stops, and the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has moved. The time point when the base station obtains the location data corresponding to this split is taken as the motion judgment time point. Otherwise, the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has not moved.

[0064] By calculating the ratio of location data in two location data cache sub-queues, sudden movement of electronic shelf labels can be quickly identified. For example, if an electronic shelf label is suddenly picked up or moved quickly, the signal quality index value received by a base station for that label may fluctuate significantly due to distance changes, causing a sudden decrease or surge in data volume within a certain period. In this case, the ratio of location data exceeding a threshold (e.g., 70%) can directly determine movement, avoiding subsequent complex calculations on invalid data. This design is particularly suitable for detecting rapid movement of goods in scenarios such as warehousing and retail. If the ratio does not trigger movement determination, further calculation is required. According to the pruning ratio, the first and last sections of the location data cache queue are pruned (e.g., the first 10% and the last 10%) to obtain a pruned queue. From the back to the front, each location data in the pruned queue is used as a dividing point to divide the location data cache queue. Each division generates two location data cache sub-queues, and the difference in the average location index value of the two sub-queues (such as the difference in average position information, average angle information, or average distance information) is calculated. For example, assuming electronic shelf labels move slowly, the change in the amount of location data in the queue may not be obvious, but the difference in the average position information between the two segments will exceed a preset threshold (e.g., 0.5 meters). By calculating the average difference in statistics, the position change can be quantified, effectively identifying minute movements and avoiding missed detections. The hierarchical judgment mechanism combines the characteristics of sudden data changes and gradual position changes, enabling it to capture both sudden movements (such as goods being quickly removed) and slow movements (such as shelf labels being gradually adjusted). The proportional threshold can be dynamically adjusted according to the base station deployment environment (such as densely packed shelving areas or open aisles). For example, the proportional threshold can be relaxed in areas where the signal is easily obstructed to avoid misjudging movement due to temporary signal loss. The difference threshold is dynamically optimized using historical data or machine learning models. For example, in promotional areas where electronic shelf labels are frequently adjusted, the difference threshold can be increased to reduce false alarms. This adaptive capability allows the system to maintain stable performance in complex environments (such as warehouses and shopping malls).

[0065] According to the pruning ratio, the first and last sections of the location data cache queue are pruned (e.g., the first 10% and the last 10%) to obtain a pruned queue. Each location data point in the pruned queue is then used as a dividing point to segment the location data cache queue from back to front. The purpose of pruning the first and last sections of the location data cache queue to obtain the pruned queue, and then using each location data point in the pruned queue as a dividing point to segment the location data cache queue, is to ensure that the minimum number of location data points in each of the two resulting location data cache sub-queues is greater than a certain set value (e.g., the minimum number of location data points in each location data cache sub-queue must be greater than 10% of the total number of location data points in the location data cache queue). This is to avoid misjudging the electronic price tag movement due to distortion of one or a few consecutive location data points. When the location data cache queue is segmented from back to front using each location data point in the pruned queue as a dividing point, and the difference in the average location index values ​​of the two sub-queues formed by each segmentation does not exceed a first difference threshold, then the judgment result corresponding to the base station participating in the movement judgment is determined to be that the electronic price tag has not moved.

[0066] It should be noted that the first offset duration can also be set based on the specific positioning situation, and there are no restrictions here.

[0067] In addition to the method described above for determining whether an electronic shelf label has moved (referred to as the first method), this invention also proposes a second method for determining whether an electronic shelf label has moved, including:

[0068] According to the pruning ratio, the front and back segments of the location data cache queue are pruned to obtain a pruned queue. From back to front, each location data in the pruned queue is used as a split point to divide the location data cache queue. Each split generates two location data cache sub-queues, and the average location index value of the two location data cache sub-queues is calculated. When the difference between the average location index values ​​of the two location data cache sub-queues exceeds the first difference threshold, the splitting stops, and the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has moved. The time point when the base station obtains the location data corresponding to this split is taken as the motion judgment time point. Otherwise, the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has not moved.

[0069] It should be noted that in the second method mentioned above, the pruning ratio can be 0, that is, no pruning is required, and each location data in the location data cache queue is directly used as a split point to divide the location data cache.

[0070] In one embodiment, the positioning data includes a positioning timestamp and a positioning index value, wherein the positioning index value is location information, angle information, or distance information;

[0071] When the positioning indicator value is location information, the average positioning indicator value is the average location information, the difference is the Euclidean distance, and the first difference threshold is the first distance threshold.

[0072] When the positioning indicator value is angle information, the average positioning indicator value is average angle information, the difference is angle difference, and the first difference threshold is the first angle difference threshold.

[0073] When the positioning indicator value is distance information, the average positioning indicator value is the average distance information, the difference is the distance difference, and the first difference threshold is the first distance difference threshold.

[0074] The above provides various application scenarios, which can be selected as needed. For example, for ranging schemes such as UWB and CS, the embodiments of this application can also be used for distance change and movement determination by multiple base stations, that is, when using distance information obtained by ranging, this scheme is still applicable.

[0075] In step 104, the electronic price tag is determined to have moved based on the judgment results of all base stations involved in the movement determination.

[0076] In one embodiment, determining whether an electronic price tag has moved based on the judgment results of all base stations participating in the movement determination includes:

[0077] The statistical judgment result is the first number of base stations participating in the movement judgment. If the first number is consistent with the second number of all base stations participating in the movement judgment, and the difference between the movement judgment time points recorded by all base stations participating in the movement judgment whose judgment result is that the electronic price tag has moved is within the difference threshold, it is determined that the electronic price tag has moved; otherwise, it is determined that the electronic price tag has not moved.

[0078] This mechanism requires all base stations involved in motion detection to determine movement (the first number equals the second number), effectively filtering out false alarms caused by a single base station due to multipath effects, temporary obstruction, or hardware failure. For example, in a warehouse environment, if a base station mistakenly determines movement due to signal fluctuations caused by shelf swaying, but other base stations do not trigger synchronously, the system will not misjudge the electronic price tag as moved. However, when the price tag is actually moved, all base stations will trigger synchronous detection due to signal changes caused by physical displacement, ensuring accuracy. This mechanism is similar to the majority consensus algorithm in a distributed system, forming collective intelligence through multi-base station collaboration. For example, in a smart supermarket, if two of the three base stations determine movement but at different times, the system will still consider it as no movement; however, when all base stations synchronously determine movement within milliseconds, it can confirm that the price tag has undergone physical displacement, avoiding false alarms caused by occasional interference.

[0079] In one embodiment, the method further includes:

[0080] The statistical judgment result is the first number of base stations involved in the movement judgment when the electronic price tag has moved;

[0081] If the second number of all base stations involved in the mobility determination is greater than the first number, the first difference threshold is reduced.

[0082] Base stations that participated in the movement judgment and whose judgment result was that the electronic price tag had not moved were designated as secondary movement judgment base stations.

[0083] For each secondary motion detection base station, determine whether the difference between the two average positioning index values ​​of the secondary motion detection base station exceeds the reduced first difference threshold. If so, determine that the judgment result corresponding to the secondary motion detection base station is that the electronic price tag has moved, and record the motion detection time point. Otherwise, determine that the judgment result corresponding to the secondary motion detection base station is that the electronic price tag has not moved.

[0084] Based on the judgment results of all base stations involved in the movement determination, it is determined whether the electronic price tag has been moved.

[0085] In this embodiment of the invention, when some base stations do not trigger motion determination (the second number > the first number), the system automatically reduces the first difference threshold (e.g., from 0.8 meters to 0.5 meters) to capture more subtle motion signals by lowering the detection threshold. If the second number is greater than the first number, the first difference threshold is reduced proportionally, which can be the first number / the second number, and the determination continues for these base stations that participated in the motion determination and whose determination results indicate that the electronic price tag has not moved. The determination process is similar to steps 103 and 104, and will not be repeated here.

[0086] Steps 101-104 above will be performed periodically. The principle behind this is a method for detecting changes in the position of electronic price tags using a multi-base station collaborative approach. Based on the above embodiments, Figure 2 This is a flowchart of the electronic price tag location change detection based on multiple base stations in an embodiment of the present invention. The flowchart takes the location index value as the location information as an example, which can track the location change of the electronic price tag in real time as much as possible. This method is a small-scale detection method. The flowchart is a comprehensive summary of steps 101-104 and all related steps, which will not be repeated here.

[0087] Figure 3 The flowchart for updating the position of the electronic price tag in this embodiment of the invention includes:

[0088] Step 301: Select the determined moving electronic price tag as the electronic price tag to be analyzed, and obtain the positioning data of the electronic price tag to be analyzed from multiple base stations that can receive the signal of the electronic price tag within a fourth preset time period.

[0089] Step 302: Calculate the average positioning index value of each base station for the electronic price tag to be analyzed based on the positioning data of the electronic price tag to be analyzed within the fourth preset time period.

[0090] Step 303: Based on the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag on each shelf in the fingerprint database, the electronic price tag to be analyzed is matched with each electronic price tag in the fingerprint database to obtain the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint database. The fingerprint database is used to store the fingerprints of multiple electronic price tags on multiple shelves. The fingerprint of each electronic price tag includes the dot location information of the electronic price tag, the average positioning index value of each base station among multiple base stations that can receive the signal of the electronic price tag for the electronic price tag, and the unique identifier of each base station.

[0091] Step 304: If the maximum matching degree of the electronic price tag to be analyzed is greater than the matching degree threshold, the dot position information of the electronic price tag in the fingerprint database corresponding to the maximum matching degree is used as the current position information of the electronic price tag to be analyzed.

[0092] In this embodiment of the invention, if the maximum value of the matching degree corresponding to the electronic price tag to be analyzed is greater than the matching degree threshold, the current location information of the electronic price tag to be analyzed is determined based on the location data of the electronic price tag to be analyzed.

[0093] This application employs a fingerprint matching method for updating the location of electronic shelf labels (ESLs). Typically, even if ESLs have some mobility on the shelf, the locations where they can be placed are almost fixed. Furthermore, the average location index values ​​of different ESLs at the same or adjacent locations (on the same shelf) across multiple base stations exhibit consistency and stability. In addition, because different base stations cover different areas, the location data distribution characteristics of labels on different shelves also show a certain degree of specificity. Simultaneously, the fingerprint matching method is not affected by absolute positioning accuracy; it primarily relies on the relative consistency of the average location index values.

[0094] The fingerprint database stores the average signal characteristics of each electronic shelf label on each shelf under historical scenarios from multiple base stations, transforming the movement behavior of electronic shelf labels into quantifiable signal characteristic differences. For example, in a smart supermarket, when an electronic shelf label moves from shelf A to shelf B, the distribution of signal quality index values ​​detected by its surrounding base stations will highly match the fingerprint database characteristics of shelf B, thus accurately identifying the location change. The fingerprint database is constructed through the fusion of multiple electronic shelf label data. This group feature model can effectively suppress individual differences of single devices (such as signal attenuation caused by battery power fluctuations), enabling the system to maintain stable accuracy in complex environments. When the maximum matching degree exceeds a threshold, the dot position information of the electronic shelf label in the fingerprint database corresponding to the maximum matching degree is directly used as the current location information of the electronic shelf label to be analyzed, achieving millisecond-level fast positioning; if the threshold is not met, the system switches to trilateration or triangulation algorithms based on the original positioning data to ensure that location information can still be output in scenarios with signal attenuation. For example, in a warehouse corner with severe signal obstruction, the system automatically lowers the threshold and estimates the location using the original positioning data to avoid missed detections.

[0095] In one embodiment, the fingerprint database is constructed using the following steps:

[0096] Based on the marking information when the electronic shelf labels on each shelf are put into use, obtain the binding relationship and marking location information of the electronic shelf labels on each shelf;

[0097] The electronic shelf labels that have not been moved within five preset time periods after being marked on each shelf are used as fingerprints to construct electronic shelf labels.

[0098] For each fingerprint-based electronic price tag, obtain the average positioning index value of each of the multiple base stations that can receive the signal of the fingerprint-based electronic price tag for that fingerprint-based electronic price tag;

[0099] Based on the average positioning index value of the electronic price tag built for each fingerprint by each base station, the fingerprint of each fingerprint is obtained for building the electronic price tag.

[0100] Add a timestamp to each fingerprint and use the dot position information of each fingerprint to build an electronic price tag as the key value of each fingerprint;

[0101] Build a fingerprint database using all fingerprints.

[0102] In the above embodiments, electronic shelf labels that have not moved within a fifth preset time period after being marked are selected as fingerprint construction objects, eliminating position drift interference caused by device movement. The fifth preset time period can be the same as or different from the fourth preset time period. This mechanism ensures that only stable and reliable position information is retained in the fingerprint database. An average positioning index value is calculated for each fingerprint-constructed electronic shelf label, and statistical methods are used to smooth the instantaneous fluctuations of signal quality index values ​​(such as Bluetooth RSSI). This processing method is similar to the strategy of using the mean RSSI to generate fingerprints in Wi-Fi positioning, which can control the positioning error within a smaller range. Using the marked position information as the fingerprint key establishes a direct mapping relationship between fingerprint features and physical location. Compared with traditional interpolation algorithms, this avoids position estimation deviations caused by environmental changes (such as the addition of obstacles), and is particularly suitable for supermarket environments with dense shelves and complex signal propagation.

[0103] In one embodiment, for each fingerprint-based electronic price tag, the average positioning index value of each of the multiple base stations capable of receiving signals from the fingerprint-based electronic price tag is obtained for that fingerprint-based electronic price tag, including:

[0104] Determine whether the variance of the positioning index value of the fingerprint-constructed electronic price tag on each base station that can receive the signal of the fingerprint-constructed electronic price tag is within a preset variance threshold. If so, use the positioning data of the fingerprint-constructed electronic price tag within a fifth preset time period on the base station that can receive the signal of the fingerprint-constructed electronic price tag as the positioning data for fingerprint construction.

[0105] Based on the fingerprint construction location data of each fingerprint-based electronic price tag, determine the average location index value for each fingerprint-based electronic price tag.

[0106] In one embodiment, before determining whether the variance of the location index value of the fingerprint-constructed electronic price tag on each base station capable of receiving the signal of the fingerprint-constructed electronic price tag is within a preset variance threshold, the method further includes:

[0107] For each fingerprint-based electronic price tag, a second preset number of base stations are selected from multiple base stations that can receive the signal from the fingerprint-based electronic price tag.

[0108] Each location data point on each base station capable of receiving the signal from the fingerprint-constructed electronic price tag is determined to have a variance within a preset variance threshold. If so, the location data of the fingerprint-constructed electronic price tag within a fifth preset time period on each base station capable of receiving the signal from the fingerprint-constructed electronic price tag is used as the location data for fingerprint construction, including:

[0109] Determine whether the variance of the location index value of the fingerprint-constructed electronic price tag on each selected base station is within a preset variance threshold. If so, use the location data of the fingerprint-constructed electronic price tag on the selected base station within a fifth preset time period as the location data for fingerprint construction.

[0110] In the above embodiments, during the fingerprint database construction phase, based on the initial electronic shelf label's location information (the true value of the electronic shelf label's location information) when it goes online, the specific shelf label information on each shelf is obtained, thus binding the electronic shelf label to the shelf and establishing the binding relationship. Then, the original location data within a stable time period (e.g., nighttime, and this time period is close to the total shelf labeling time) for a fifth preset duration (T2 hours, configurable) is retrieved. Data from the T2 hours after labeling can be selected, or location data from T2 / 2 hours before and after the labeling time can be selected. Note that electronic shelf labels that successfully move within this time period are not included in the fingerprint database construction. Furthermore, the selected T2-hour location data needs to be filtered. For each electronic shelf label, taking the location index as location information as an example, if the variance of the location information of the electronic shelf label on a certain base station among the second number of base stations selected by the index RSSI or PSNR meets a preset variance threshold, then the location data of that base station can be used as the location data for fingerprint construction of the electronic shelf label; otherwise, it is not used as the location data for fingerprint construction of the electronic shelf label.

[0111] Then, calculate the average value of the location information of the Jth price tag on the i-th shelf from the location data of the K-th base station. (i.e., the average positioning index value), therefore, the fingerprint information of each electronic price tag on the i-th shelf can be constructed (taking the positioning index value as location information as an example; the principle is the same when the positioning index value is angle information and distance information):

[0112]

[0113] in, For the i-th shelf, For the Jth electronic price tag, Let J be the average location information of the J-th electronic shelf label on the i-th shelf, located on the K-th base station. It can be seen that multiple fingerprints exist on a single shelf, and each fingerprint on each shelf corresponds to the location information of each electronic shelf label on that shelf across multiple base stations capable of receiving its signal (the average location index value of multiple base stations corresponding to one electronic shelf label on one shelf is one fingerprint unit). Furthermore, to prevent shelf movement in real-world scenarios, the fingerprint key value is set to the dot-mapping location information of the electronic shelf label, i.e., its specific location information in the actual environment, represented as... The purpose of adding a timestamp to each fingerprint is to delete fingerprints older than a certain threshold if too many fingerprints are found on a particular shelf during subsequent fingerprint updates. A fingerprint database can be constructed based on the correspondence between fully marked shelves and price tags, represented as follows:

[0114]

[0115] in, Let be the timestamp of the i-th fingerprint.

[0116] In future fingerprint database maintenance, whenever a new electronic price tag is added online, causing the fingerprint information to be reported, the above method can be used to obtain the corresponding new fingerprint and directly add it to the fingerprint database.

[0117] To prevent excessive additions, the fingerprints of new electronic shelf labels are matched with those of existing electronic shelf labels on the shelf before addition. If the matching degree exceeds a certain threshold, the addition is abandoned to prevent redundant additions. Furthermore, if the matching degree is less than another threshold, the newly added fingerprint is considered to be erroneous or abnormal data, and such data is not added to the fingerprint database and is abandoned.

[0118] In one embodiment, before calculating the average positioning index value of each base station for the electronic price tag to be analyzed based on the positioning data of the electronic price tag to be analyzed within a fourth preset time period, the method further includes:

[0119] From multiple base stations that can receive signals from the electronic price tag to be analyzed, a second preset number of base stations are selected, and the location data of the selected base stations for the electronic price tag to be analyzed is obtained.

[0120] Based on the location data of the electronic price tag to be analyzed within the fourth preset time period, the average location index value of each base station for the electronic price tag to be analyzed is calculated, including:

[0121] Based on the location data of the selected base stations within the fourth preset time period, the average location index value of the selected base stations for the electronic price tag to be analyzed is calculated.

[0122] In the above embodiment, when selecting a second preset number of base stations from multiple base stations capable of receiving signals from the electronic price tag to be analyzed, the base station with the larger signal quality index value can be selected based on the RSSI or SNR index. Assuming that K base stations are selected, all the positioning data of the selected base stations for the electronic price tag to be analyzed can be represented as follows:

[0123]

[0124] Among them, electronic price tags The Kth base station All include Indicators, these indicators are used to screen base stations, Average location information is a type of average positioning index value.

[0125] In one embodiment, before matching the electronic shelf label to be analyzed with each electronic shelf label in the fingerprint database, the method further includes:

[0126] After obtaining multiple base stations that can receive signals from the electronic price tags to be analyzed, the base station with the highest quality index value of the signal received from the electronic price tags to be analyzed is selected from among the multiple base stations that can receive signals from the electronic price tags to be analyzed.

[0127] The average value of the location information of the selected base stations in the location data of the electronic price tag to be analyzed within the fourth preset time period is taken as the current location information of the electronic price tag to be analyzed.

[0128] The shelves within the preset range of the current location information are selected as the shelves to be analyzed;

[0129] The process involves matching the electronic shelf labels to be analyzed with each electronic shelf label in the fingerprint database, including:

[0130] Match the electronic shelf labels to be analyzed with each electronic shelf label on the shelf to be analyzed in the fingerprint database.

[0131] In the above embodiments, the main purpose is to screen shelves in order to improve matching efficiency.

[0132] In one embodiment, based on the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag on each shelf in the fingerprint database, the electronic price tag to be analyzed is matched with each electronic price tag in the fingerprint database to obtain the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint database, including:

[0133] The base station corresponding to each electronic price tag on each shelf in the fingerprint database is taken as the first base station set for each electronic price tag, and multiple base stations that can receive the signal from the electronic price tag to be analyzed are taken as the second base station set.

[0134] Electronic shelf labels whose first base station set in the fingerprint database contains the second base station set are used as electronic shelf labels to be matched;

[0135] For each electronic price tag pair formed by the electronic price tag to be matched and the electronic price tag to be analyzed, based on the unique identifier of the base station, find the difference between the average positioning index value of the same base station for the electronic price tag to be matched and the average positioning index value for the electronic price tag to be analyzed, and determine the matching degree of each electronic price tag pair based on the differences of all base stations.

[0136] In the above embodiments, electronic price tags need to be... Matching fingerprints with those in the fingerprint database is represented as follows (using the average location index value as an example):

[0137]

[0138] The fingerprint database contains fingerprints of multiple electronic price tags. Each electronic price tag's fingerprint includes the average positioning index values ​​of K base stations for that electronic price tag. These K base stations form the first set of base stations for the electronic price tag. The base stations corresponding to multiple electronic price tags to be analyzed (also K in the above formula) are used as the second set of base stations. If the first set of base stations contains the second set of base stations, then the electronic price tag in the fingerprint database can be used for matching and becomes the electronic price tag to be matched.

[0139] For each electronic price tag pair formed by the electronic price tag to be matched and the electronic price tag to be analyzed, for example, for and Each electronic price tag pair has K base stations, and the same base station, for example... Calculate the average position information of the two locations. and The Euclidean distance is used to determine... and The degree of matching.

[0140] In one embodiment, determining the matching degree of each electronic price tag pair based on the differences among all base stations includes:

[0141] The reciprocal of the sum of the differences between all base stations is used as the matching degree of each electronic price tag pair;

[0142] Alternatively, the reciprocal of the maximum value among all the differences between the base stations can be used as the matching degree of each electronic price tag pair.

[0143] If the maximum matching degree of the electronic price tag to be analyzed is greater than the matching degree threshold, the dot location information of the electronic price tag in the fingerprint database corresponding to the maximum matching degree is taken as the current location information of the electronic price tag to be analyzed. Otherwise, the current location information of the electronic price tag to be analyzed is determined based on the positioning data of the electronic price tag to be analyzed. Specifically, referring to the aforementioned steps, the base station with the highest quality index value of the signal of the electronic price tag to be analyzed is selected from multiple base stations that can receive the signal of the electronic price tag to be analyzed. The average value of the location information of the electronic price tag to be analyzed in the positioning data of the selected base station within the fourth preset time period is taken as the current location information of the electronic price tag to be analyzed.

[0144] Another method for determining the current location information of the electronic price tag to be analyzed is given below. This method is an artificial intelligence-based location update method, and the training data used by this method includes two types:

[0145] The first method is to obtain the fingerprint construction location data for fingerprint electronic price tags. The acquisition method is exactly the same as the fingerprint matching method. The fingerprint electronic price tag can be referred to as a collection point here. Since this collection point is the electronic price tag, there is no need to collect it again. If an electronic price tag moves, the fourth preset time corresponding to the electronic price tag needs to be shortened or otherwise adjusted to avoid the electronic price tag moving within the selected fourth preset time.

[0146] The second method involves setting up collection points in the aisles of the store. Each collection point continuously sends signals and collects location data of the aisle collection points for a fourth preset time period through a base station. The structure of the location data of the aisle collection points is consistent with the structure of the location data used for fingerprint construction, including location index values ​​at multiple time points and signal quality index values. When the store map changes, such as aisle changes, collection points can be planned on the new aisle. When the shelves change, all price tags on the changed shelves can be marked once.

[0147] By training a neural network with training data, a well-trained location prediction model is obtained. Subsequently, after obtaining the location data of the electronic price tag to be analyzed within the fourth preset time period, the data is input into the well-trained location prediction model to obtain the current location information of the electronic price tag to be analyzed.

[0148] In one embodiment, the method further includes:

[0149] The location data of the electronic price tag within the fourth preset time period is input into the location prediction model to obtain the current location information of the electronic price tag;

[0150] The training steps for a location prediction model include:

[0151] Obtain the fingerprint location data for each fingerprint used to construct the electronic price tag;

[0152] Obtain the location data of each collection point in the passageway within the store for a fourth preset time period. The structure of the location data of the passageway collection points is consistent with the location data used to construct the fingerprint.

[0153] The fingerprint construction data for each electronic price tag is generated using location data and dot location information, as well as channel collection point location data and collection point location information for each collection point, to form training data.

[0154] Windowing is applied to the training data to create an expanded training dataset;

[0155] Based on the expanded training dataset, the neural network is trained to obtain a well-trained location prediction model.

[0156] The electronic shelf label in the above embodiments can be any electronic shelf label that needs to determine its current location information. In the scenario of the present invention, the electronic shelf label can be the electronic shelf label to be analyzed, that is, the electronic shelf label that has been determined to have moved.

[0157] In an embodiment of the present invention, Figure 4 This is a schematic diagram of the channel collection point positioning data and collection point location information in an embodiment of the present invention. The collection point location information of each collection point is also obtained by marking points. The collection points can be smart shopping carts, smart shopping baskets, and aisle cleaning robots, etc. These collection points can obtain specific location information through internal devices (such as navigation devices) to form accurate collection point location information.

[0158] Figure 4 In the training data, the location information of the collection points, the positioning data (x_N, y_N) collected by each base station, RSSI_N, and SNR_N are all included. Windowing the training data (e.g., time window T1) aims to generate more data, thus improving the accuracy of model training. Base stations that do not receive a signal are padded with zeros. Furthermore, since the dimensions of coordinates, RSSI, and PSNR are inconsistent, normalization is necessary.

[0159] In this embodiment of the invention, the neural network can be a one-dimensional CNN network architecture. The CNN network architecture is a lightweight neural network used for classification. Figure 5 This is an example of the CNN network architecture in this embodiment of the invention. Other similar architectures can also be used, and the neural network can have other structures; no restrictions are placed here. During training, the expanded training dataset can be divided into an 8:2 ratio, corresponding to the training set and the test set respectively. During training, each data collection point can be sampled according to this ratio. During initial unified training, all data is preprocessed as described above and then fed into the neural network in batches of a certain size for training. The backpropagation loss function is used to further improve the network's feature learning. The number of iterations can be adjusted as needed, and a convergence detection strategy is added. The network stops training when convergence or the number of iterations reaches its maximum. Theoretically, during the training phase, the loss function will continuously decrease and converge to a smaller value, while the accuracy of the test set will continuously increase and converge to a relatively high value. When new data is available in the future, it will be organized according to the above method, and the location prediction model will be fine-tuned to be compatible with new classification labels. In the online inference phase, the location data of the electronic price tag to be analyzed within a fourth preset time period is obtained and input into the location prediction model to obtain the current location information of the electronic price tag to be analyzed.

[0160] Below is a specific example of an AI-based location update method. Taking a shopping cart as the data collection point, a corresponding positioning device is attached to the shopping cart, etc. This positioning device continuously sends signals. After receiving the signals, the base stations installed at an angle or on the ground in the store process the received signals to obtain AOA information (azimuth angle, elevation angle), RSSI index or SNR index to form channel collection point positioning data. Using the above channel collection point positioning data and collection point location information from multiple base stations for a fifth preset time period, and fingerprints on the original shelves to construct positioning data and dot location information, training data is formed. The training data is windowed to form an extended training dataset. Based on the extended training dataset, the neural network is trained to obtain a trained location prediction model.

[0161] Then, the location data of the electronic price tag to be analyzed within the fourth preset time period is input into the location prediction model to obtain the current location information of the electronic price tag to be analyzed.

[0162] The predicted data is input into the pre-trained AI model to predict its location and output the location.

[0163] Figure 6 This is a schematic diagram illustrating the trajectory tracking of shopping carts and baskets in an embodiment of the present invention. When shopping carts and baskets move within a physical store, their positions can be continuously output through continuous positioning. By plotting all the positioning positions of the corresponding positioning devices within a certain time period into a corresponding curve in chronological order, the trajectory of the shopping carts and baskets can be tracked. Furthermore, based on the location information of the assets of interest within the store over a period of time, a heatmap of the entire store can also be generated. Figure 7 This invention provides a heat map of the shopping cart in this embodiment, which identifies areas within the store that receive high and low customer attention, thus supporting adjustments to store displays. Furthermore, the location results can be used for inventory management of all assets and applications such as electronic fences.

[0164] This invention also proposes a method for detecting changes in the position of electronic price tags based on the final positioning index value. This method is a large-scale detection method.

[0165] Figure 8 This is a flowchart illustrating the detection of electronic price tag position changes based on the final positioning index value in an embodiment of the present invention. In one embodiment, the method further includes:

[0166] Step 801: After obtaining multiple base stations that can receive signals from electronic price tags, select the base station with the highest quality index value of the signal received from the multiple base stations that can receive signals from electronic price tags.

[0167] Step 802: Calculate the average positioning index value of the electronic price tag based on the positioning data cache queue of the selected base stations for the electronic price tag within the third preset time period.

[0168] Step 803: Calculate the difference between the average positioning index value and the average positioning index value of the last electronic price tag update;

[0169] Step 804: When the difference is greater than the second difference threshold, it is determined that the electronic price tag has moved.

[0170] In this embodiment of the invention, the selected base station stores the location data of the electronic price tag within a first preset time period in its location data cache queue.

[0171] In this embodiment of the invention, the positioning data includes a positioning timestamp and a positioning index value, wherein the positioning index value is location information, angle information, or distance information;

[0172] When the positioning index value is the location information, the average positioning index value is the average location information, the difference is the Euclidean distance, and the second difference threshold is the second distance threshold.

[0173] When the positioning indicator value is angle information, the average positioning indicator value is average angle information, the difference is angle difference, and the second difference threshold is the second angle difference threshold.

[0174] When the positioning indicator value is distance information, the average positioning indicator value is the average distance information, the difference is the distance difference, and the second difference threshold is the second distance difference threshold.

[0175] For an electronic shelf label, location calculation can be performed using a single base station or multiple base stations based on its signal (e.g., AOA signal). For a single base station, the optimal base station can be selected based on the maximum RSSI or SNR value to obtain the final output location index value of the electronic shelf label. Therefore, in this method for detecting changes in the location of electronic shelf labels, the location index value at the time of the last movement determination of the label can be recorded first, using the location information... For example, based on the location data cache queue at the current moment, the location information of the electronic price tag within the third preset time period (e.g., T hours) is retrieved, and the average value of the location information is calculated. This is used as the average position information, and then the Euclidean distance between the average position information and the position information at the time of the last movement judgment of the electronic price tag is calculated. When the European distance If the distance exceeds the second distance threshold, the price tag is determined to have moved.

[0176] The third preset duration is less than the first preset duration. After determining that the electronic price tag has moved through one of the two methods, the position of the electronic price tag needs to be updated. In this embodiment of the invention, the location is based on multi-base station location fingerprint matching.

[0177] In addition, if both methods are used simultaneously, the movement of the price tag can be detected as comprehensively as possible. If the first method determines that the electronic price tag has moved, and the second method determines that the electronic price tag has not moved, then the electronic price tag has been determined to have moved. In other words, if either method determines that the electronic price tag has moved, then the electronic price tag has been determined to have moved.

[0178] In one embodiment, the method further includes:

[0179] The mobile electronic price tag is identified as the electronic price tag to be analyzed, and the positioning data of the electronic price tag to be analyzed from multiple base stations that can receive the signal of the electronic price tag to be analyzed within a fourth preset time period is obtained.

[0180] Based on the location data of the electronic price tag to be analyzed from multiple base stations that can receive signals from the electronic price tag to be analyzed within a fourth preset time period, calculate the average location index value of each base station for the electronic price tag to be analyzed.

[0181] Based on the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag on each shelf in the fingerprint database, the electronic price tag to be analyzed is matched with each electronic price tag in the fingerprint database to obtain the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint database. The fingerprint database is used to store the fingerprints of multiple electronic price tags on multiple shelves. The fingerprint of each electronic price tag includes the dot location information of the electronic price tag, the average positioning index value of each base station among multiple base stations that can receive the signal of the electronic price tag for the electronic price tag, and the unique identifier of each base station.

[0182] If the maximum matching degree of the electronic price tag to be analyzed is greater than the matching degree threshold, the dot position information of the electronic price tag in the fingerprint database corresponding to the maximum matching degree will be used as the current position information of the electronic price tag to be analyzed.

[0183] In one embodiment, the fingerprint database is constructed using the following steps:

[0184] Based on the tracking information of each electronic shelf label on each shelf when it goes online, obtain the binding relationship and tracking position information of each electronic shelf label on each shelf;

[0185] The electronic shelf labels that have not been moved within five preset time periods after being marked on each shelf are used as fingerprints to construct electronic shelf labels.

[0186] For each fingerprint-based electronic price tag, obtain the average positioning index value of each of the multiple base stations that can receive the signal of the fingerprint-based electronic price tag for that fingerprint-based electronic price tag;

[0187] Based on the average positioning index value of the electronic price tag built for each fingerprint by each base station, the fingerprint of each fingerprint is obtained for building the electronic price tag.

[0188] Add a timestamp to each fingerprint and use the dot position information of each fingerprint to build an electronic price tag as the key value of each fingerprint;

[0189] Build a fingerprint database using all fingerprints.

[0190] In one embodiment, for each fingerprint-based electronic price tag, the average positioning index value of each of the multiple base stations capable of receiving signals from the fingerprint-based electronic price tag is obtained for that fingerprint-based electronic price tag, including:

[0191] For each fingerprint-based electronic price tag, a second preset number of base stations are selected from multiple base stations that can receive the signal from the fingerprint-based electronic price tag.

[0192] Determine whether the variance of the location index value of the fingerprint-constructed electronic price tag on each selected base station is within the preset variance threshold. If so, use the location data of the fingerprint-constructed electronic price tag on the selected base station within the fifth preset time period as the fingerprint construction location data.

[0193] Based on the fingerprint construction location data of each fingerprint-based electronic price tag, determine the average location index value for each fingerprint-based electronic price tag.

[0194] In one embodiment, before matching the electronic shelf label to be analyzed with each electronic shelf label in the fingerprint database, the method further includes:

[0195] After obtaining multiple base stations that can receive signals from the electronic price tags to be analyzed, the base station with the highest quality index value of the signal received from the electronic price tags to be analyzed is selected from among the multiple base stations that can receive signals from the electronic price tags to be analyzed.

[0196] The average value of the location information of the selected base stations in the location data of the electronic price tag to be analyzed within the fourth preset time period is taken as the current location information of the electronic price tag to be analyzed.

[0197] The shelves within the preset range of the current location information are selected as the shelves to be analyzed;

[0198] The process involves matching the electronic shelf labels to be analyzed with each electronic shelf label in the fingerprint database, including:

[0199] Match the electronic shelf labels to be analyzed with each electronic shelf label on the shelf to be analyzed in the fingerprint database.

[0200] In one embodiment, based on the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag on each shelf in the fingerprint database, the electronic price tag to be analyzed is matched with each electronic price tag in the fingerprint database to obtain the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint database, including:

[0201] The base station corresponding to each electronic price tag on each shelf in the fingerprint database is taken as the first base station set for each electronic price tag, and multiple base stations that can receive the signal from the electronic price tag to be analyzed are taken as the second base station set.

[0202] Electronic shelf labels whose first base station set in the fingerprint database contains the second base station set are used as electronic shelf labels to be matched;

[0203] For each electronic price tag pair formed by the electronic price tag to be matched and the electronic price tag to be analyzed, based on the unique identifier of the base station, find the difference between the average positioning index value of the same base station for the electronic price tag to be matched and the average positioning index value for the electronic price tag to be analyzed, and determine the matching degree of each electronic price tag pair based on the differences of all base stations.

[0204] In one embodiment, determining the matching degree of each electronic price tag pair based on the differences among all base stations includes:

[0205] The reciprocal of the sum of the differences between all base stations is used as the matching degree of each electronic price tag pair;

[0206] Alternatively, the reciprocal of the maximum value among all the differences between the base stations can be used as the matching degree of each electronic price tag pair.

[0207] In one embodiment, the method further includes:

[0208] The location data of the electronic price tag to be analyzed within the fourth preset time period is input into the location prediction model to obtain the current location information of the electronic price tag to be analyzed.

[0209] The training steps for a location prediction model include:

[0210] Obtain the fingerprint location data for each fingerprint used to construct the electronic price tag;

[0211] Obtain the location data of each collection point in the passageway within the store for a fourth preset time period. The structure of the location data of the passageway collection points is consistent with the location data used to construct the fingerprint.

[0212] The fingerprint construction data for each electronic price tag is generated using location data and dot location information, as well as channel collection point location data and collection point location information for each collection point, to form training data.

[0213] Windowing is applied to the training data to create an expanded training dataset;

[0214] Based on the expanded training dataset, the CNN network is trained to obtain a well-trained location prediction model.

[0215] This invention also proposes an electronic price tag position change detection device, the principle of which is similar to the electronic price tag position change detection method, and will not be described in detail here.

[0216] Figure 9 This is a schematic diagram of the structure of the electronic price tag position change detection device in an embodiment of the present invention, including:

[0217] The base station acquisition module 901 is used to acquire multiple base stations that can receive signals from electronic price tags. Each base station constructs a location data buffer queue for the electronic price tag based on the received signal.

[0218] The mobile detection base station determination module 902 is used to determine whether each base station participates in the mobile detection of the electronic price tag based on the location data cache queue of each base station for the electronic price tag.

[0219] The first motion determination module 903 is used to divide the location data cache queue of the electronic price tag of each base station participating in motion determination into two location data cache sub-queues, and determine whether the electronic price tag has moved by comparing the two location data cache sub-queues to obtain the determination result.

[0220] The second mobility determination module 904 is used to determine whether the electronic price tag has moved based on the determination results of all base stations participating in the mobility determination.

[0221] In one embodiment, the apparatus further includes a screening module for:

[0222] Based on the signal quality index value of the electronic price tag, a first preset number of base stations are selected from multiple base stations that can receive the signal of the electronic price tag;

[0223] The mobile detection base station determination module 902 is used to determine whether each base station in the first preset number of selected base stations participates in the mobile detection of the electronic price tag based on the location data cache queue of each base station for the electronic price tag in the first preset number of selected base stations.

[0224] In one embodiment, the mobile base station determination module 902 is used for:

[0225] For each base station, select the location data within the second preset time period from the location data cache queue for the electronic price tag of that base station;

[0226] Determine whether the number of location data points within the second preset time period of the base station is greater than the set threshold.

[0227] If so, determine whether the base station is involved in the movement of the electronic price tag;

[0228] Otherwise, it is determined that the base station is not involved in the movement determination of the electronic price tag.

[0229] In one embodiment, the first movement determination module 903 is used for:

[0230] For each base station participating in motion determination, the location data cache queue for the electronic price tag will be used as the location timestamp obtained by deducing the first offset time backward from the current time.

[0231] Based on the time point of movement determination, the location data cache queue is divided into two location data cache sub-queues;

[0232] Calculate the ratio of the number of location data in the two location data cache sub-queues. If the ratio exceeds the ratio threshold, determine that the judgment result corresponding to the base station participating in the movement judgment is that the electronic price tag has moved, and record the movement judgment time point.

[0233] If the ratio does not exceed the ratio threshold, the front and back segments of the location data cache queue are trimmed according to the trimming ratio to obtain a trimmed queue. From back to front, each location data in the trimmed queue is used as a split point to divide the location data cache queue. Each split generates two location data cache sub-queues, and the average location index value of the two location data cache sub-queues is calculated. When the difference between the average location index values ​​of the two location data cache sub-queues exceeds the first difference threshold, the splitting stops, and the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has moved. The time point when the base station obtains the location data corresponding to this split is taken as the motion judgment time point. Otherwise, the judgment result corresponding to the base station participating in the motion judgment is determined to be that the electronic price tag has not moved.

[0234] In one embodiment, the positioning data includes a positioning timestamp and a positioning index value, wherein the positioning index value is location information, angle information, or distance information;

[0235] When the positioning indicator value is location information, the average positioning indicator value is the average location information, the difference is the Euclidean distance, and the first difference threshold is the first distance threshold.

[0236] When the positioning indicator value is angle information, the average positioning indicator value is average angle information, the difference is angle difference, and the first difference threshold is the first angle difference threshold.

[0237] When the positioning indicator value is distance information, the average positioning indicator value is the average distance information, the difference is the distance difference, and the first difference threshold is the first distance difference threshold.

[0238] In one embodiment, the second movement determination module 904 is used for:

[0239] The statistical judgment result is the first number of base stations participating in the movement judgment. If the first number is consistent with the second number of all base stations participating in the movement judgment, and the difference between the movement judgment time points recorded by all base stations participating in the movement judgment whose judgment result is that the electronic price tag has moved is within the difference threshold, it is determined that the electronic price tag has moved; otherwise, it is determined that the electronic price tag has not moved.

[0240] In one embodiment, the second movement determination module 904 is further configured to:

[0241] The statistical judgment result is the first number of base stations involved in the movement judgment when the electronic price tag has moved;

[0242] If the second number of all base stations involved in the mobility determination is greater than the first number, the first difference threshold is reduced.

[0243] Base stations that participated in the movement judgment and whose judgment result was that the electronic price tag had not moved were designated as secondary movement judgment base stations.

[0244] For each secondary motion detection base station, determine whether the difference between the two average positioning index values ​​of the secondary motion detection base station exceeds the reduced first difference threshold. If so, determine that the judgment result corresponding to the secondary motion detection base station is that the electronic price tag has moved, and record the motion detection time point. Otherwise, determine that the judgment result corresponding to the secondary motion detection base station is that the electronic price tag has not moved.

[0245] Based on the judgment results of all base stations involved in the movement determination, it is determined whether the electronic price tag has been moved.

[0246] In one embodiment, the device further includes a third movement determination module, used for:

[0247] After obtaining multiple base stations that can receive electronic price tags, the base station with the highest quality index value of the signal received from the multiple base stations that can receive electronic price tags is selected.

[0248] The average positioning index value of the electronic price tag is calculated based on the positioning data cache queue of the selected base stations for the electronic price tag within a third preset time period.

[0249] Calculate the difference between the average positioning index value and the average positioning index value of the last electronic price tag update;

[0250] When the difference is greater than the second difference threshold, it is determined that the electronic price tag has moved.

[0251] In one embodiment, the device further includes a location update module for:

[0252] The mobile electronic price tag is identified as the electronic price tag to be analyzed, and the positioning data of the electronic price tag to be analyzed from multiple base stations that can receive the signal of the electronic price tag to be analyzed within a fourth preset time period is obtained.

[0253] Based on the location data of the electronic price tag to be analyzed within the fourth preset time period, calculate the average location index value of each base station for the electronic price tag to be analyzed.

[0254] Based on the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag on each shelf in the fingerprint database, the electronic price tag to be analyzed is matched with each electronic price tag in the fingerprint database to obtain the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint database. The fingerprint database is used to store the fingerprints of multiple electronic price tags on multiple shelves. The fingerprint of each electronic price tag includes the dot location information of the electronic price tag, the average positioning index value of each base station among multiple base stations that can receive the signal of the electronic price tag for the electronic price tag, and the unique identifier of each base station.

[0255] If the maximum matching degree of the electronic price tag to be analyzed is greater than the matching degree threshold, the dot position information of the electronic price tag in the fingerprint database corresponding to the maximum matching degree will be used as the current position information of the electronic price tag to be analyzed.

[0256] In one embodiment, the device further includes a fingerprint database construction module, used for:

[0257] Based on the tracking information of each electronic shelf label on each shelf when it goes online, obtain the binding relationship and tracking position information of each electronic shelf label on each shelf;

[0258] The electronic shelf labels that have not been moved within five preset time periods after being marked on each shelf are used as fingerprints to construct electronic shelf labels.

[0259] For each fingerprint-based electronic price tag, obtain the average positioning index value of each of the multiple base stations that can receive the signal of the fingerprint-based electronic price tag for that fingerprint-based electronic price tag;

[0260] Based on the average positioning index value of the electronic price tag built for each fingerprint by each base station, the fingerprint of each fingerprint is obtained for building the electronic price tag.

[0261] Add a timestamp to each fingerprint and use the dot position information of each fingerprint to build an electronic price tag as the key value of each fingerprint;

[0262] Build a fingerprint database using all fingerprints.

[0263] In one embodiment, the fingerprint database construction module is used for:

[0264] Determine whether the variance of the positioning index value of the fingerprint-constructed electronic price tag on each base station that can receive the signal of the fingerprint-constructed electronic price tag is within a preset variance threshold. If so, use the positioning data of the fingerprint-constructed electronic price tag within a fifth preset time period on the base station that can receive the signal of the fingerprint-constructed electronic price tag as the positioning data for fingerprint construction.

[0265] Based on the fingerprint construction location data of each fingerprint-based electronic price tag, determine the average location index value for each fingerprint-based electronic price tag.

[0266] In one embodiment, the fingerprint database construction module is used for:

[0267] After obtaining multiple base stations that can receive signals from the electronic price tags to be analyzed, the base station with the highest quality index value of the signal received from the electronic price tags to be analyzed is selected from among the multiple base stations that can receive signals from the electronic price tags to be analyzed.

[0268] The average value of the location information of the selected base stations in the location data of the electronic price tag to be analyzed within the fourth preset time period is taken as the current location information of the electronic price tag to be analyzed.

[0269] The shelves within the preset range of the current location information are selected as the shelves to be analyzed;

[0270] The process involves matching the electronic shelf labels to be analyzed with each electronic shelf label in the fingerprint database, including:

[0271] Match the electronic shelf labels to be analyzed with each electronic shelf label on the shelf to be analyzed in the fingerprint database.

[0272] In one embodiment, the location update module is used to:

[0273] The base station corresponding to each electronic price tag on each shelf in the fingerprint database is taken as the first base station set for each electronic price tag, and multiple base stations that can receive the signal from the electronic price tag to be analyzed are taken as the second base station set.

[0274] Electronic shelf labels whose first base station set in the fingerprint database contains the second base station set are used as electronic shelf labels to be matched;

[0275] For each electronic price tag pair formed by the electronic price tag to be matched and the electronic price tag to be analyzed, based on the unique identifier of the base station, find the difference between the average positioning index value of the same base station for the electronic price tag to be matched and the average positioning index value for the electronic price tag to be analyzed, and determine the matching degree of each electronic price tag pair based on the differences of all base stations.

[0276] In one embodiment, the location update module is used to:

[0277] The reciprocal of the sum of the differences between all base stations is used as the matching degree of each electronic price tag pair;

[0278] Alternatively, the reciprocal of the maximum value among all the differences between the base stations can be used as the matching degree of each electronic price tag pair.

[0279] In summary, the method and apparatus proposed in this invention obtain multiple base stations capable of receiving electronic price tag signals. Utilizing the signal coverage and positioning data of multiple base stations, information about the electronic price tag can be obtained from different angles. Compared to positioning with a single base station, this reduces positioning errors caused by factors such as signal obstruction and multipath effects, improving the accuracy and reliability of positioning. For example, in a large shopping mall, multiple base stations can more comprehensively cover the area where the electronic price tag is located, avoiding positioning deviations caused by a single base station due to excessive distance or signal obstruction. The method determines whether a base station participates in motion detection based on its positioning data cache queue for the electronic price tag, and divides and compares the positioning data cache queues of base stations participating in motion detection. This approach fully utilizes the historical positioning data stored by the base stations, analyzing data changes over different time periods to determine whether the electronic price tag has moved, reducing the impact of random factors on the motion detection results and enhancing the reliability of motion detection. For instance, when the electronic price tag is subjected to brief signal interference, comparing data from multiple time periods can prevent misjudging the interference as movement of the electronic price tag. Instead of relying on complex hardware or high-precision sensors to directly measure the movement of electronic price tags, this system analyzes and processes signal data received from existing base stations to determine location and movement, reducing hardware costs and complexity. Furthermore, utilizing historical data in a cached queue eliminates the need for extensive real-time computations, further reducing system computing resource requirements and lowering operating costs. It can quickly determine whether an electronic price tag has moved based on signals received from the base station and cached location data, meeting the need for real-time monitoring of price tag position changes. It adapts better to complex environments, accurately determining the movement of electronic price tags and ensuring stability and reliability under varying conditions.

[0280] This invention also provides a computer device. Figure 10 This is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 1000 includes a memory 1010, a processor 1020, and a computer program 1030 stored in the memory 1010 and executable on the processor 1020. When the processor 1020 executes the computer program 1030, it implements the above-mentioned method for detecting changes in the position of electronic price tags.

[0281] This invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for detecting changes in the position of electronic price tags.

[0282] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method for detecting changes in the position of electronic price tags.

[0283] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0284] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0285] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0286] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0287] The above specific embodiments further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of detecting a change in position of an electronic price tag, characterized by, The method comprises the following steps: a plurality of base stations capable of receiving signals of an electronic price tag are obtained, and each base station constructs a positioning data cache queue for the electronic price tag according to the received signals; whether each base station participates in the movement judgment of the electronic price tag is determined according to the positioning data cache queue of each base station for the electronic price tag; for each base station participating in the movement judgment, the positioning data cache queue of the base station participating in the movement judgment is divided into two positioning data cache sub-queues, whether the electronic price tag moves is determined by comparing the two positioning data cache sub-queues, and a determination result is obtained; whether the electronic price tag moves is determined according to the determination results of all the base stations participating in the movement judgment.

2. The method of claim 1, wherein, The positioning data cache queue of each base station for the electronic price tag stores the positioning data of the electronic price tag within a first preset time length.

3. The method of claim 1, wherein, After the plurality of base stations capable of receiving signals of the electronic price tag are obtained, the following steps are further included: a first preset number of base stations are selected from the plurality of base stations capable of receiving signals of the electronic price tag according to the quality index values of the signals of the electronic price tag; whether each base station participates in the movement judgment of the electronic price tag is determined according to the positioning data cache queue of each base station for the electronic price tag, which comprises the following steps: whether each base station of the selected first preset number of base stations participates in the movement judgment of the electronic price tag is determined according to the positioning data cache queue of each base station of the selected first preset number of base stations for the electronic price tag.

4. The method of claim 1, wherein, Whether each base station participates in the movement judgment of the electronic price tag is determined according to the positioning data cache queue of each base station for the electronic price tag, which comprises the following steps: for each base station, positioning data within a second preset time length is selected from the positioning data cache queue of the base station for the electronic price tag; whether the number of the positioning data within the second preset time length of the base station is greater than a set threshold value is determined; if yes, it is determined that the base station participates in the movement judgment of the electronic price tag; otherwise, it is determined that the base station does not participate in the movement judgment of the electronic price tag.

5. The method of claim 1, wherein, For each base station participating in the movement judgment, the positioning data cache queue of the base station participating in the movement judgment is divided into two positioning data cache sub-queues, whether the electronic price tag moves is determined by comparing the two positioning data cache sub-queues, and a determination result is obtained, which comprises the following steps: for the positioning data cache queue of each base station participating in the movement judgment for the electronic price tag, a positioning timestamp obtained by counting back a first offset time length from a current time point is taken as a movement judgment time point; the positioning data cache queue is divided into two positioning data cache sub-queues according to the movement judgment time point; the proportion of the number of positioning data of the two positioning data cache sub-queues is calculated, and if the proportion exceeds a proportion threshold value, it is determined that the determination result corresponding to the base station participating in the movement judgment is that the electronic price tag moves, and the movement judgment time point is recorded. If the ratio does not exceed the ratio threshold, the front and rear of the positioning data cache queue are cropped according to the cropping ratio to obtain a cropped queue, each positioning data in the cropped queue is taken as a segmentation point to segment the positioning data cache queue from back to front, each segmentation generates two positioning data cache sub-queues, and the average positioning index values of the two positioning data cache sub-queues are calculated. When the difference between the average positioning index values of the two positioning data cache sub-queues exceeds the first difference threshold, the segmentation is stopped, and it is determined that the judgment result corresponding to the base station participating in the movement judgment is that the electronic price tag moves, and the time point at which the base station obtains the positioning data corresponding to this segmentation is taken as the movement judgment time point. Otherwise, it is determined that the judgment result corresponding to the base station participating in the movement judgment is that the electronic price tag does not move.

6. The method of claim 5, wherein, The positioning data includes a positioning timestamp and a positioning index value, and the positioning index value is position information, angle information, or distance information. When the positioning index value is position information, the average positioning index value is average position information, the difference is the Euclidean distance, and the first difference threshold is the first distance threshold. When the positioning index value is angle information, the average positioning index value is average angle information, the difference is the angle difference, and the first difference threshold is the first angle difference threshold. When the positioning index value is distance information, the average positioning index value is average distance information, the difference is the distance difference, and the first difference threshold is the first distance difference threshold.

7. The method of claim 1, wherein, According to the judgment results of all base stations participating in the movement judgment, whether the electronic price tag moves is determined, including: The first number of base stations participating in the movement judgment whose judgment result is that the electronic price tag moves is counted. If the first number is consistent with the second number of all base stations participating in the movement judgment, and the difference between the movement judgment time points recorded by all base stations participating in the movement judgment whose judgment result is that the electronic price tag moves is within the difference threshold, it is determined that the electronic price tag moves. Otherwise, it is determined that the electronic price tag does not move.

8. The method of claim 5, wherein, Further including: The first number of base stations participating in the movement judgment whose judgment result is that the electronic price tag moves is counted. If the second number of all base stations participating in the movement judgment is greater than the first number, the first difference threshold is reduced. The base stations participating in the movement judgment whose judgment result is that the electronic price tag does not move are taken as secondary movement judgment base stations. For each secondary movement judgment base station, whether the difference between the two average positioning index values of the secondary movement judgment base station exceeds the reduced first difference threshold is determined. If yes, it is determined that the judgment result corresponding to the secondary movement judgment base station is that the electronic price tag moves, and the movement judgment time point is recorded. Otherwise, it is determined that the judgment result corresponding to the secondary movement judgment base station is that the electronic price tag does not move. According to the judgment results of all base stations participating in the movement judgment, whether the electronic price tag moves is determined.

9. The method of claim 1, wherein, Further including: After obtaining a plurality of base stations that can receive signals of the electronic price tag, the base station with the maximum quality index value of the signal of the electronic price tag is selected from the plurality of base stations that can receive signals of the electronic price tag. According to the positioning data of the base station for the electronic price tag in the positioning data cache queue within the third preset time length, the average positioning index value of the electronic price tag is calculated; The difference between the average positioning index value and the average positioning index value of the last electronic price tag update is calculated; When the difference is greater than the second difference threshold, it is determined that the electronic price tag has moved.

10. The method of claim 1, wherein, Also includes: The electronic price tag determined to move is taken as a to-be-analyzed electronic price tag, and the positioning data of the base station for the to-be-analyzed electronic price tag within a fourth preset time length is obtained; According to the positioning data of the to-be-analyzed electronic price tag within the fourth preset time length, the average positioning index value of each base station for the to-be-analyzed electronic price tag is calculated; According to the average positioning index value of each base station for the to-be-analyzed electronic price tag and the fingerprint of each electronic price tag of each shelf in the fingerprint library, the to-be-analyzed electronic price tag and each electronic price tag in the fingerprint library are matched to obtain the matching degree of the to-be-analyzed electronic price tag and each electronic price tag in the fingerprint library, and the fingerprint library is used to store the fingerprints of the electronic price tags of the plurality of shelves, the fingerprint of each electronic price tag includes the dotting position information of the electronic price tag, the average positioning index value of each base station for the electronic price tag and the unique identifier of each base station. If the maximum value of the matching degree corresponding to the to-be-analyzed electronic price tag is greater than the matching degree threshold, the dotting position information of the electronic price tag in the fingerprint library corresponding to the maximum value of the matching degree is taken as the current position information of the to-be-analyzed electronic price tag.

11. The method of claim 10, wherein, The fingerprint library is constructed by the following steps: Based on the dotting information when each electronic price tag on each shelf is online, the binding relationship and dotting position information of each electronic price tag on each shelf are obtained; The electronic price tag on each shelf that has not moved within a fifth preset time length after dotting is taken as a fingerprint construction electronic price tag; For each fingerprint construction electronic price tag, the average positioning index value of each base station for the fingerprint construction electronic price tag is obtained; According to the average positioning index value of each base station for each fingerprint construction electronic price tag, the fingerprint of each fingerprint construction electronic price tag is obtained; A timestamp is added to each fingerprint, and the dotting position information of each fingerprint construction electronic price tag is taken as the key value of each fingerprint; All fingerprints are used to construct a fingerprint library.

12. The method of claim 11, wherein, For each fingerprint construction electronic price tag, the average positioning index value of each base station for the fingerprint construction electronic price tag is obtained, including: Respectively, whether the variance of the positioning index value of the positioning data of the fingerprint construction electronic price tag on each base station that can receive the signal of the fingerprint construction electronic price tag is within the preset variance threshold, if yes, the positioning data of the fingerprint construction electronic price tag on the base station that can receive the signal of the fingerprint construction electronic price tag within the fifth preset time length is taken as the fingerprint construction positioning data; According to the fingerprint construction positioning data of each fingerprint construction electronic price tag, the average positioning index value of each fingerprint construction electronic price tag is determined.

13. The method of claim 10, wherein, Before matching the to-be-analyzed electronic price tag with each electronic price tag in the fingerprint library, it also includes: After obtaining the multiple base stations capable of receiving the signal of the electronic price tag to be analyzed, the base station receiving the signal of the electronic price tag to be analyzed with the maximum quality index value is screened out from the multiple base stations capable of receiving the signal of the electronic price tag to be analyzed; The average value of the position information of the screened base station in the positioning data of the electronic price tag to be analyzed within the fourth preset time length is taken as the current position information of the electronic price tag to be analyzed; The shelves within the preset range of the current position information are taken as the shelves to be analyzed; The matching between the electronic price tag to be analyzed and each electronic price tag in the fingerprint library includes: The matching between the electronic price tag to be analyzed and each electronic price tag on the shelf to be analyzed in the fingerprint library.

14. The method of claim 12, wherein, The matching between the electronic price tag to be analyzed and each electronic price tag in the fingerprint library is performed according to the average positioning index value of each base station for the electronic price tag to be analyzed and the fingerprint of each electronic price tag of each shelf in the fingerprint library, and the matching degree between the electronic price tag to be analyzed and each electronic price tag in the fingerprint library is obtained, including: The base stations corresponding to each electronic price tag of each shelf in the fingerprint library are taken as a first base station set of each electronic price tag, and the multiple base stations capable of receiving the signal of the electronic price tag to be analyzed are taken as a second base station set; The electronic price tag in the fingerprint library, of which the first base station set contains the second base station set, is taken as an electronic price tag to be matched; For each electronic price tag pair formed by each electronic price tag to be matched and the electronic price tag to be analyzed, the difference between the average positioning index value of the same base station for the electronic price tag to be matched and the average positioning index value of the same base station for the electronic price tag to be analyzed is found according to the unique identifier of the base station, and the matching degree of each electronic price tag pair is determined according to the difference values of all base stations.

15. The method of claim 14, wherein, The matching degree of each electronic price tag pair is determined according to the difference values of all base stations, including: The reciprocal of the accumulated difference values of all base stations is taken as the matching degree of each electronic price tag pair; Or, the reciprocal of the maximum value in the difference values of all base stations is taken as the matching degree of each electronic price tag pair.

16. An apparatus for detecting a change in position of an electronic price label, comprising: Including: The base station obtaining module is configured to obtain multiple base stations capable of receiving the signal of the electronic price tag, and each base station constructs a positioning data cache queue for the electronic price tag according to the received signal; The mobile judgment base station determining module is configured to determine whether each base station participates in the mobile judgment of the electronic price tag according to the positioning data cache queue of each base station for the electronic price tag; The first mobile judgment module is configured to, for each base station participating in the mobile judgment, divide the positioning data cache queue of the base station participating in the mobile judgment into two positioning data cache sub-queues, judge whether the electronic price tag moves by comparing the two positioning data cache sub-queues, and obtain a judgment result; The second mobile judgment module is configured to determine whether the electronic price tag moves according to the judgment results of all base stations participating in the mobile judgment.

17. A computer device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the computer program comprises computer program code configured to cause the processor to perform the method of any one of claims 1 to 16. The processor executes the computer program to realize the method in any one of claims 1 to 15.

18. A computer-readable storage medium, characterized in that, The computer readable storage medium stores a computer program, and the computer program is executed by the processor to realize the method in any one of claims 1 to 15.

19. A computer program product, characterised in that, The computer program product includes a computer program, and the computer program is executed by the processor to realize the method in any one of claims 1 to 15.

Images