Positioning method and device, storage medium and computer program product

Abstract

The invention provides a positioning method and device, a storage medium and a computer program product, and the method comprises the steps: obtaining first tag feature information corresponding to reference tag information and second tag feature information corresponding to to-be-positioned tag information, the reference tag information being tag information deployed in a to-be-positioned area in advance; and determining position information of the to-be-positioned tag information in the to-be-positioned area based on the first tag feature information and the second tag feature information.

Description

Technical Field

[0001] The present application relates to the field of positioning technology, and in particular to a positioning method, device, storage medium and computer program product. Background Art

[0002] Passive IoT has been widely used in the field of smart retail. Radio Frequency Identification (RFID) tags are attached to all products, and handheld devices are used to improve work efficiency in warehousing, storage, checkout and other links.

[0003] However, in actual applications, methods for locating RFID tag information based on passive Internet of Things, such as the three-dimensional positioning method of ultra-high frequency radio frequency identification (UHF RFID) tags based on multiple base stations, will cause inaccurate RFID tag information positioning due to obstructions or interferences such as metal shelves and walls on site. Summary of the invention

[0004] In view of this, the present application hopes to provide a positioning method, device, storage medium and computer program product, which can improve the accuracy of RFID tag information positioning.

[0005] The technical solution of this application is implemented as follows:

[0006] In a first aspect, the present application provides a positioning method, the method comprising:

[0007] Acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located;

[0008] Based on the first tag feature information and the second tag feature information, position information of the tag information to be located in the area to be located is determined.

[0009] In a second aspect, the present application provides a positioning device, the positioning device comprising:

[0010] An acquiring unit, configured to acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located;

[0011] The determination unit is used to determine the position information of the tag information to be located in the area to be located based on the first tag characteristic information and the second tag characteristic information.

[0012] In a third aspect, the present application provides a positioning device, comprising: a processor and a memory; the processor implements the above positioning method when executing a running program stored in the memory.

[0013] In a fourth aspect, the present application provides a storage medium having a computer program stored thereon, which implements the above-mentioned positioning method when executed by a processor.

[0014] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the above-mentioned positioning method when executed by a processor.

[0015] The present application provides a positioning method, device, storage medium and computer program product, the method comprising: obtaining first label feature information corresponding to reference label information, and second label feature information corresponding to label information to be positioned, wherein the reference label information is label information pre-deployed in the area to be positioned; based on the first label feature information and the second label feature information, determining the position information of the label information to be positioned in the area to be positioned. The above implementation scheme is adopted, by deploying reference label information in the area to be positioned, because the reference label information is used as a reference, the position of the reference label information is known when deployed, therefore, by obtaining the first label feature information of the reference label information, and obtaining the second label feature information of the label information to be positioned, because the first label feature information corresponds to the reference label information of the deployed known position, the first label feature information and the second label feature information are combined, and the position information of the label information to be positioned corresponding to the second label feature information can be further judged according to the first label feature information, so that the position information of the label to be positioned obtained is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a positioning method provided in an embodiment of the present application Figure 1 ;

[0017] Figure 2 A schematic diagram of a reference tag deployment and antenna coverage provided in an embodiment of the present application;

[0018] Figure 3 A schematic diagram of a method for concurrently executing a non-interference list determined by antennas according to an embodiment of the present application;

[0019] Figure 4 A schematic diagram of reference tag information deployment corresponding to a positioning method provided in an embodiment of the present application;

[0020] Figure 5 This is a schematic diagram of the random forest algorithm training process in the embodiment of the present application;

[0021] Figure 6 A schematic diagram of tag feature data of reference tag information corresponding to different antenna sides provided in an embodiment of the present application;

[0022] Figure 7 A schematic diagram of a positioning method provided in an embodiment of the present application Figure 2 ;

[0023] Figure 8 A schematic diagram of a tag feature information flow for collecting reference tag information provided in an embodiment of the present application;

[0024] Fig. 9 A schematic diagram of reference label information deployment for shelf-level positioning provided by an embodiment of the present application;

[0025] Fig.10 A schematic diagram of characteristic data of reference tag information collected according to an embodiment of the present application;

[0026] Fig.11 A schematic diagram of a test result provided in an embodiment of the present application;

[0027] Fig.12 A schematic diagram of reference tag information deployment for antenna-level positioning provided in an embodiment of the present application;

[0028] Fig.13 A schematic diagram of a flow chart of an antenna-level positioning method provided in this application;

[0029] Fig.14 A schematic diagram of the structure of a positioning device provided in an embodiment of the present application Figure 1 ;

[0030] Fig.15 A schematic diagram of the structure of a positioning device provided in an embodiment of the present application Figure 2 . DETAILED DESCRIPTION

[0031] In order to enable a more detailed understanding of the features and technical contents of the embodiments of the present application, the technical solution of the present application is further elaborated in detail below in combination with the drawings and specific embodiments of the specification. The attached drawings are for reference only and are not used to limit the embodiments of the present application.

[0032] Unless otherwise defined, all technical and scientific terms used in the embodiments of the present application have the same meaning as those commonly understood by those skilled in the art to which the present application belongs. The terms used in the embodiments of the present application are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0033] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict. It should also be noted that the terms "first / second / third" involved in the embodiments of the present application are only used to distinguish similar objects and do not represent a specific ordering of the objects. It is understandable that "first / second / third" may be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0034] Passive Internet of Things has been widely used in the field of smart retail. For example, some merchants in the clothing industry have fully launched passive Internet of Things. All products are affixed with RFID tags, and handheld devices are used to improve work efficiency in the warehousing, storage, and checkout links.

[0035] However, with the further deepening of the digital transformation of the retail industry, upgrading management methods is crucial to the retail industry, and the demand for automated inventory based on passive IoT has increased significantly. For example, some merchants have clearly proposed the need to upgrade automated display verification methods (such as regional positioning), change the way of flight inspection in related technologies, and reduce management costs; some merchants have clearly proposed the need to upgrade the store's automated positioning capabilities to achieve rapid clothing search and differentiated consumer guidance. With RFID tags already adopted, how to achieve regional positioning through passive IoT networks in stores with multiple interference environments and multiple metal shelves has become the core demand of store management.

[0036] In the related art, the industry has proposed methods such as three-dimensional positioning methods of UHF RFID tags based on multiple base stations, fixed position identification and positioning methods such as checkpoints, etc., based on passive Internet of Things positioning methods, but none of them are suitable for store management scenarios. For example, in the three-dimensional positioning method of UHF RFID tags based on multiple base stations, a ranging method based on the Phase Difference of Arrival (PDOA) is proposed to locate the three-dimensional position of the target tag when the tag height is known or unknown. The positioning system includes a reader and three or more positioning base stations. When the tag height is unknown, a single base station uses three, four, six or more antennas. When the tag height is known, a single base station uses two antennas, and the same number of antennas are placed on each base station. The distance between the two antennas is set to be less than half the wavelength of the RF signal, and the PDOA ranging method at the same RF signal frequency is used to obtain the incident angle of the tag and the distance to the base station. The position of the tag to be measured is obtained by using a triangulation positioning method using multiple base stations.

[0037] However, this solution requires at least two or more receiving base stations to receive the RFID tag signal to locate the RFID tag. In actual applications, due to the presence of metal shelves, walls and other obstructions or interferences in retail stores, the multipath environment causes inaccurate RFID tag positioning. At the same time, due to the economic requirements of system deployment, it is difficult to meet the requirements that all tags are recognized by two or more deployed antennas.

[0038] Therefore, in response to the above problems, a high-reliability positioning method for passive Internet of Things in multi-interference environment and multipath environment is proposed in the embodiment of the present application. It can be applied in the case where a single or multiple antennas can be identified, and high-reliability positioning can be achieved by adding low-cost reference tags. This solution can achieve low-cost, high-reliability indoor meter-level positioning in complex environments, meeting the needs of retail, industry and other fields for indoor positioning.

[0039] The detailed scheme of the embodiment of this application is as follows:

[0040] To solve the above problems, the present application provides a positioning method, such as Figure 1 As shown, the method may include:

[0041] S101: Acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located.

[0042] The reference tag information is tag information pre-deployed in the area to be located.

[0043] In the embodiment of the present application, the reference tag information may be some RFID tags with known locations pre-deployed in the area to be located.

[0044] In the embodiment of the present application, the tag information to be located may be an RFID tag that needs to be located.

[0045] In the embodiment of the present application, the first tag characteristic information includes one or more of the following: first received signal strength indication (RSSI) information, a first phase, a first frequency point, and a first inventory count.

[0046] In the embodiment of the present application, the second tag characteristic information includes one or more of the following: second RSSI information, second phase, second frequency point, and second inventory times.

[0047] In the embodiment of the present application, the first tag characteristic information and the second tag characteristic information correspond to each other. For example, the first tag characteristic information is RSSI information, and the second tag characteristic information is also RSSI information.

[0048] In an embodiment of the present application, when deploying a passive IoT system, reference tags are first deployed in the area to be positioned (such as a shelf-level area). For example, in a certain clothing industry scenario, it is necessary to locate the RFID tags on the left and right sides of the shelf. Therefore, m reference tags can be deployed on the left and right sides of each shelf respectively.

[0049] It should be noted that in actual applications, whether to actually deploy reference tags can be determined based on actual applications. For example, in a fixed unmanned environment, if only the positioning target (i.e., the tag information to be positioned) changes and the surrounding environment does not change, after collecting the first tag feature information of the deployed reference tags, the reference tags may not be deployed in actual applications.

[0050] In the embodiment of the present application, the reference tag information is pre-deployed in the area to be located, so the regional location of the reference tag information deployed in the environment can be acquired in advance, that is, the location of the reference tag information is known.

[0051] In the embodiment of the present application, after the environment deployment is completed, that is, after the network devices and reference tags for measurement are deployed, the environment measurement can be performed.

[0052] In the embodiment of the present application, obtaining the first tag feature information corresponding to the reference tag information and the second tag feature information of the tag information to be located can be implemented in the following manner:

[0053] Acquire positioning requirement information or positioning level information corresponding to the positioning requirement information; determine an inventory mode of the tag information to be located based on the positioning requirement information or the positioning level information; and collect the first tag feature information and the second tag feature information based on the inventory mode.

[0054] In the embodiment of the present application, the positioning requirement information is such as the acquisition of small-scale reference tag information.

[0055] In the embodiment of the present application, the positioning level information is such as area level positioning, shelf left and right level positioning or meter level positioning.

[0056] It should be noted that in order to save system resources and to meet the needs of different scenarios, the area level, antenna level or more detailed levels, such as shelf level, shelf left and right level, meter level, etc., are configured. Different inventory requirements require the system to obtain data corresponding to different levels. In this way, targeted acquisition can not only reduce computing resources, but also reduce the waste of resources caused by data transmission.

[0057] In the embodiment of the present application, for the reference tag information, two modes can be used simultaneously to perform N rounds of inventory to obtain the first tag feature information, or some of the two modes can be selected to perform N rounds of inventory to obtain the first tag feature information. For different business needs, the inventory mode selection scheme can be as follows:

[0058] (1) The same as the actual business execution mode. If the actual business adopts polling inventory, the polling inventory mode (i.e. mode 1) is selected.

[0059] (2) For a small number of reference tags (e.g., a few or dozens) and a small space, such as less than 10 nodes or antennas, a polling inventory mode (i.e., mode 1) can be used.

[0060] (3) For low positioning requirements, such as regional level, all antennas can be counted concurrently (i.e., Mode 2).

[0061] (4) For situations with high positioning requirements or multiple positioning service combinations or large spaces during actual service execution, Mode 1+Mode 2 (executed N times each) can be used.

[0062] In the embodiment of the present application, the actual inventory mode corresponding to the reference tag can be determined from the above-mentioned inventory modes according to actual needs.

[0063] In the embodiment of the present application, if the inventory mode is a concurrent inventory mode, that is, all antennas are inventoried concurrently, it is necessary to obtain a non-interference list between antennas, so as to perform a concurrent inventory based on the non-interference list.

[0064] In the embodiment of the present application, based on the inventory mode, when collecting the first tag feature information and the second tag feature information, the following methods can be specifically used:

[0065] In the case where the inventory mode includes a concurrent inventory mode, a non-interference list is determined based on the reference tag information; and based on the non-interference list, the first tag feature information and the second tag feature information are collected.

[0066] In the embodiment of the present application, the do not interfere list is determined based on the reference tag information, which can be implemented in the following manner:

[0067] The non-interference list is determined based on the inventory information of whether the reference tag information is stored in the inventory; or, the non-interference list is determined based on a first comparison result of whether the reference tag information is stored in the inventory information and whether the signal quality parameter information of the reference tag information is greater than a first threshold value; or, the non-interference list is determined based on a second comparison result of whether the reference tag information is stored in the inventory information and whether the number of times the reference tag information is stored in a preset time is greater than a second threshold value.

[0068] In the embodiment of the present application, the signal quality parameter information may be an RSSI value.

[0069] In the embodiment of the present application, the specific values ​​of the first threshold and the second threshold can be selected according to actual conditions, and are not specifically limited in the embodiment of the present application.

[0070] In the embodiment of the present application, since the location of the reference tag information is known, the surrounding tag information is increased one by one in parallel to obtain the information on whether the network node interferes with the information. After the parallelization is completed, a non-interference list is obtained. The main method is as follows:

[0071] by Figure 2 For example, if the circle is the antenna coverage area (depending on the antenna selection, the coverage area is different), Figure 2 With antenna 1 as the center and non-interference as the goal, the three methods are shown in Table 1 below:

[0072] Table 1

[0073]

[0074]

[0075] In the embodiment of the present application, Figure 3 When performing concurrency, the specific concurrency logic is as follows:

[0076] If there are n antennas on site, the interference information is recorded in turn from 2 antennas to n antennas. For example, if there are 4 antennas, numbered 1, 2, 3, and 4, the combination selected for 2 antennas is (12, 13, 14, 23, 34), the combination selected for 3 antennas is (123, 124, 134, 234), and the combination selected for 4 antennas is (1234).

[0077] During concurrency, records are made according to the non-interference concurrency logic in Table 2 above, and the record combinations that meet the non-interference and the record combinations that do not meet the non-interference are determined, and the determination of whether the antenna concurrency is completed is continued. For example, when the goal is that all tags can be retrieved, when 2 antennas are concurrent, if the 12th antenna in the combination cannot retrieve all the reference tag information deployed on site, it is recorded as an interference combination. At this time, the antenna has not completed concurrency, and it continues to the inventory stage, with the 13th antenna in the combination concurrently. If the reference tag information can be retrieved, it is recorded as a non-interference combination, and the concurrency of the 14th antenna in the combination is polled, and the subsequent situations are deduced by analogy.

[0078] In the embodiment of the present application, different positioning requirements correspond to different collected data, as shown in the following Table 2 for example:

[0079] Table 2

[0080]

[0081] In the embodiment of the present application, for the tag information to be located, in the process of actually acquiring the second characteristic tag information corresponding to the tag information to be located, facing the actual needs of different services, different scheduling methods are used to acquire the second tag characteristic information of the tag information to be located, which can be summarized as shown in the following Table 3:

[0082] Table 3

[0083]

[0084]

[0085] In the embodiment of the present application, it can be seen from the above that the first tag characteristic information and the second tag characteristic information can be obtained by polling or the like. If the inventory mode is a concurrent inventory, it is necessary to further determine the non-interference list and refer to the non-interference list to collect the first tag characteristic information and the second tag characteristic information.

[0086] It should be noted that the collected first tag feature information and the collected second tag feature information correspond to each other.

[0087] S102: Determine position information of the tag information to be located in the area to be located based on the first tag feature information and the second tag feature information.

[0088] In the embodiments of the present application, different methods can be used to locate the tag to be located according to different positioning requirements, as follows:

[0089] When the number of reference tag information is one or more, and the number of corresponding first tag feature information is one or more, wherein the reference tag information and the first tag feature information are in a one-to-one relationship, the position information of the tag information to be located in the area to be located is determined based on the first tag feature information and the second tag feature information, which can be implemented in the following manner:

[0090] Based on the second tag feature information and one or more first tag feature information, determine the distance between each reference tag information and the tag information to be located, wherein different reference tag information is used to identify different positions; determine a minimum distance from the distances; and based on the minimum distance, determine the location information of the tag information to be located in the area to be located.

[0091] In the embodiment of the present application, when the interference environment is determined, the tag information to be located can be located by performing partition feature analysis based on the features of the tag information to be located.

[0092] In the embodiment of the present application, the first tag characteristic information may be RSSI information, and the second tag characteristic information may also be RSSI information.

[0093] In the embodiment of the present application, if the number of reference tag information is one, then the RSSI information is one; if the number of reference tag information is multiple, then the RSSI information is multiple.

[0094] In the embodiments of the present application, Figure 4 As shown, assuming that the area to be located is Figure 4 The five areas in the middle circle are position 1, position 2, position 3, position 4 and position 5. For antenna 1 and antenna 2, the collected features (such as reference tag information feature RSSI, phase feature, etc.) can be compared with the tag feature information of the tag information to be located to calculate the difference. The tag information to be located will eventually fall into the area with the smallest difference.

[0095] Specifically, for the above-mentioned regional-level positioning, at least the RSSI feature of the measured tag information is used. The RSSI feature can use the characteristic RSSI (such as the RSSI with the most recognition times) or the average RSSI. The target characteristic information of the tag information to be located must be aligned with the characteristic of the reference tag information. In the above-mentioned RFID system, there are 2 antennas (i=1, 2), 5 areas (w=1, 2...5), m tag information to be located, and c reference tag information. The signal strength of the jth tag information to be located on the i-th antenna is RSSI ij， i=1,2;j=1,...,m. Similarly, the signal strength of the rth reference tag information at the i-th antenna is RSSI ir， i=1,2;r=1,...,c。D ijr The Euclidean distance between each tag information to be located and the reference tag information is calculated as shown in the following formula (1):

[0096]

[0097] The location of the tag information to be located is the location area with the smallest RSSI difference from the reference tag information of the set area, which can be expressed as the following formula (2):

[0098] P(target label) = min(D ijr )(2)

[0099] It should be noted that the target tag is the tag information to be located.

[0100] In the embodiment of the present application, the above formula (2) can be understood as determining the area where the reference tag information corresponding to the smallest distance information is located as the position information of the tag information to be located in the area to be located.

[0101] It should be noted that, for the same antenna, reference tag information with different feature points needs to be selected for positioning.

[0102] In an embodiment of the present application, based on the first tag feature information and the second tag feature information, to determine the position information of the tag to be located in the area to be located, the following method may also be adopted:

[0103] Use the first tag feature information for model training to obtain a positioning model; input the second tag feature information into the positioning model to obtain the position information of the tag to be located corresponding to the second tag feature information in the area to be located.

[0104] In an embodiment of the present application, when using the first tag feature information of the collected reference tag information for model training, the random forest algorithm can be used for training the positioning model (other training models can also be used), and the specific training process is as follows:

[0105] (1) For the collected first tag feature information, randomly divide 80% of the tag data for training, and the remaining 20% of the tag data for testing (the model can be further self-optimized through a feedback function based on the test results).

[0106] (2) Based on the divided training set, construct a random forest model composed of N decision trees, as Figure 5 shown:

[0107] ① The training set contains a total of M feature information (that is, the first tag feature information includes M types), randomly select N (N < M) feature information from them as the feature subset, and select K times in total, that is, construct K feature subsets;

[0108] ② One feature subset corresponds to one decision tree. When the tree node splits, calculate the Gini coefficient of each feature in the subset, as shown in the following formula (3), where k is the number of categories, and p i is the proportion of the number of samples of category i in all samples. Formula (3) is as follows:

[0109]

[0110] It should be noted that the higher the Gini coefficient, the greater the influence of the feature information on the classification result, and the best feature information can be selected as the basis for node splitting.

[0111] ③ Each decision tree completes J splits to form a random forest model.

[0112] (3) Based on the above test set and the trained random forest model, conduct a positioning test:

[0113] ① Randomly select N (M < 11) feature information from the test set to construct K feature subsets.

[0114] ② For each label data, K feature subsets are input into K decision trees respectively, and each decision tree outputs the region prediction value corresponding to the label;

[0115] ③ Vote based on the results output by each tree, and output the final prediction result by majority voting.

[0116] In the embodiment of the present application, the optimal model parameters are determined through the above-mentioned training and testing of the random forest model to obtain a positioning model.

[0117] In the embodiment of the present application, by inputting the collected second characteristic tag information of the tag information to be located into the positioning model, the area position corresponding to the tag information to be located can be obtained through the positioning model.

[0118] It can be understood that a positioning method provided by an embodiment of the present application deploys reference tag information in an area to be positioned. Since the reference tag information is used as a reference, the position of the reference tag information is known during deployment. Therefore, by obtaining the first tag feature information of the reference tag information and obtaining the second tag feature information of the tag information to be positioned, since the first tag feature information corresponds to the reference tag information of the deployed known position, the obtained first tag feature information and the second tag feature information are combined, and the position information of the tag information to be positioned corresponding to the second tag feature information can be further judged based on the first tag feature information. In this way, the position information of the tag to be positioned obtained is more accurate.

[0119] Based on the above embodiment, in the embodiment of the present application, if the positioning is performed for the antenna level, at least the RSSI of the reference tag information and the antenna inventory times feature are used for positioning, the following formula (4) can be used to determine the position of the tag information to be located, and the formula (4) is as follows:

[0120] P(target tag) = M*P_ANT(max(count_ANT i)) + N*P_ANT(min(D ijr ))(4)

[0121] Where, M and N are weights, P_ANT(max(count_ANT i)) is the number of times the tag information to be located (i.e., the target tag) is received by antenna i the most, and P_ANT(min(D ijr )) is the location area where the RSSI difference between the tag information to be located and the antenna level reference tag information is the smallest.

[0122] It should be noted that the location of the tag information to be located is based on actual needs, and different weights (i.e., M, N) are taken for the inventory times and RSSI difference values ​​of different antennas as needed. In addition to RSSI, phase features can also be introduced, and different RSSI (average value, characteristic value) can be used for multiple calculations and verifications. The location of the tag information to be located is the location of the antenna with the most significant calculated features.

[0123] Based on the above embodiment, in the embodiment of the present application, if positioning is performed for a finer area, at least the RSSI and phase characteristics of the reference tag information are used, and the position of the tag information to be located can be determined by using the following formula (5), which is as follows:

[0124] P(target label) = M*P_ANT(min(D ijr ))±

[0125]

[0126] Among them, RSSI can be calculated multiple times using the average value and characteristic value, and the phase value can use the frame head phase, frame tail phase, initial phase, arrival phase, etc.; the antenna inventory times feature can be introduced as needed, and the target tag (i.e., the tag information to be located) position is the position of the antenna with the most significant calculated feature.

[0127] In an embodiment of the present application, in the case of multiple antennas, fusion judgment can be performed according to the above relationship:

[0128] (1) Based on distance judgment, that is, a comprehensive judgment is made based on the approximate position of the tag under each antenna:

[0129] Label i: nth farthest from antenna 1...; kth farthest from antenna j. Based on the above features, a positioning area can be delineated.

[0130] (2) Based on range judgment, that is, comprehensive judgment is made based on the distance from the range area:

[0131] Tag i: distance range m is close, distance range n is far. The positioning area can be delineated based on the above features.

[0132] Based on the above embodiments, the overall implementation idea of ​​the embodiments of the present application is as follows:

[0133] In a multi-interference, multi-path environment, based on the advantages of long-distance identification and networking coverage of the passive Internet of Things, tags at different locations show different characteristics when viewed from the antenna side, such as Figure 6 As shown in the figure, it can be seen that the tag information seen from antenna i and antenna j shows different characteristics. Through the flexible scheduling of multi-nodes and multi-antennas of the passive IoT system, combined with the characteristics of the reference tag information, highly reliable meter-level positioning can be achieved.

[0134] By using the technical solution in the embodiments of the present application, it is only necessary to complete the characteristic measurement of environmental characteristics and reference tag information at different locations in the initial stage of passive IoT network deployment, and then the tag positioning of the tag to be positioned can be achieved subsequently without repeating the measurement process.

[0135] In an embodiment of the present application, the implementation method mainly includes the following steps:

[0136] (1) In actual scenarios, the system measures the characteristics of the environment and the currently placed reference tag information.

[0137] (2) Obtain feature combination information of all reference tag information.

[0138] (3) Based on the divided positions, mark the positions of different reference label information (the automatic labeling mode can be used by importing labels).

[0139] (4) Train the positioning model.

[0140] (5) When there is a positioning requirement for the tag information to be positioned, the passive system nodes are scheduled to execute the service based on the service requirements.

[0141] (6) Based on the positioning model constructed in (4), the location result of the tag information to be located is output.

[0142] The overall interaction process in the RFID system corresponding to the above process is as follows Figure 7 As shown, specifically including:

[0143] Environmental measurement: Measurement control instructions and business instructions (such as inventory) can be sent to the passive system distribution node through the passive system management node. The passive system distribution node sends the business instructions to the tag or terminal for inventory. The tag or terminal reports the tag data (such as the first tag feature data of the reference tag) to the passive system management node.

[0144] When there is a positioning requirement: a service instruction (including positioning, positioning requirement, list of tags to be positioned, etc.) is sent to the passive system management node through the service platform.

[0145] The passive system management node selects a scheduling strategy and sends a service instruction to the passive system distribution node. The passive system distribution node sends the received service instruction (such as inventory) to the tag or terminal, obtains the tag data (such as the tag feature information of the tag information to be located) from the tag or terminal, and forwards the tag data to the passive system management node; or, the passive system distribution node issues a service instruction (such as inventory including selection) to the tag or terminal, the passive system distribution node obtains the tag data from the tag or terminal, and forwards it to the passive system management node.

[0146] Finally, the label data obtained from environmental measurements is used for model training to obtain a positioning model. The positioning model is further used to locate the information of the tag to be located, that is, the passive system management node adopts a positioning algorithm and finally outputs the positioning result to the business platform.

[0147] In the embodiment of the present application, since the interference environment has multipath refraction, reflection, diffraction, etc., the environmental data (such as the tag feature data of the reference tag) can be recorded by the environmental measurement part. The specific collection process during the above environmental measurement can also be adopted. Figure 8 To express, Figure 8 The specific process includes issuing positioning requirements, selecting inventory mode, obtaining label feature data of reference labels, calculating features, etc.

[0148] It should be noted that the above Figure 8 The specific implementation of each process can be found in the above embodiments and will not be described in detail here.

[0149] The embodiments of the present application are applicable to positioning scenarios without dynamic real-time interference, including store display verification (daily judgment of whether clothing is misplaced), unmanned large warehouse positioning, etc. Taking the store display verification scenario as an example, some clothing merchants have 5,000 stores and some have 3,000 stores. Taking warehouses as an example, the national warehouse area exceeds 400 million square meters, the demand for unmanned operation is strong, and the number of super-large logistics parks exceeds 3,000. Therefore, this embodiment has high universality and a large market space.

[0150] Based on the above embodiments, examples corresponding to different positioning requirements are also given in the embodiments of the present application:

[0151] Example 1: Shelf-level positioning.

[0152] like Fig. 9 As shown, reference tag information is deployed in the positioning division area. In the example of this embodiment, three shelves are divided into left and right, with a total of six positioning areas. Three reference tag information is deployed in each positioning area, so that the positioning of the left and right columns of the shelf can be achieved.

[0153] (1) Sending positioning requirements: The business platform sends positioning requirements to the passive IoT management node. In this implementation example, the positioning requirement is shelf-level.

[0154] It should be noted that the positioning requirements can be selected according to different actual requirements.

[0155] (2) Environmental measurement: The passive IoT management node controls the passive IoT distribution node to perform environmental measurement and specifies the measurement time t. Since the deployed reference tag information is known, the system selects the reference tag information through the select command and collects the corresponding tag feature information. The collected tag feature information includes: RSSI (mandatory), initial phase, arrival phase, frequency, time, and inventory count. The characteristic RSSI, characteristic initial phase, and characteristic arrival phase can be further extracted.

[0156] (3) Obtaining the interference list: Obtaining the interference list with the goal of ensuring that all reference tag information can be retrieved.

[0157] (4) System scheduling execution: When the tag information to be located needs to be located, antenna 1 and antenna 2 can poll or execute the service concurrently based on the obtained interference list.

[0158] (5) Fusion positioning:

[0159] Assuming the time is 1 minute, the collected information is as follows: Fig.10 As shown, Fig.10 Only part of the collected data is listed.

[0160] The specific positioning method is as follows:

[0161] (1) Randomly divide 80% of the data collected above into reference label information for training to form a training set;

[0162] (2) Based on the training set, a random forest model consisting of N decision trees is constructed. Specifically:

[0163] ① The training set contains 11 features (i.e. Fig.10 ), randomly select m (m<11) features as feature subsets, select N times in total, that is, construct N feature subsets; where m can be set from 1 to 10, and the model universality is further enhanced by training 10 times.

[0164] ② One feature subset corresponds to one decision tree. When the tree node is split, the Gini coefficient of each feature in the subset is calculated as shown in the following formula (6), where k is the number of categories and p is i is the proportion of category i samples to all samples,

[0165]

[0166] It should be noted that the higher the Gini coefficient, the greater the influence of the feature on the classification result, and the best feature can be selected as the basis for node splitting;

[0167] Based on the above collected data, the Gini coefficients corresponding to the features are as follows: number of arrivals: 0.1367; arrival phase frame head: 0.1653; arrival phase frame tail: 0.3955; RSSI: 0.0928; frequency: 0.0013; timestamp: 0.0008; feature RSSI: 0.1426; average RSSI: 0.1377.

[0168] It can be seen that in this test environment, the influence on the classification results is as follows: frame tail phase > frame head phase > characteristic RSSI > average RSSI > disk arrival times > RSSI > frequency > timestamp.

[0169] ③Therefore, a feature weight configuration model based on the Gini coefficient is introduced. The model normalizes the Gini coefficient and attaches weight values ​​to different features.

[0170] ④ Each decision tree completes m splits to form a random forest model.

[0171] (3) Positioning of newly added tag information (such as replacing some tag information on the shelf): Randomly and evenly replace the tag information on the shelf on site, use the model for positioning, and compare the difference between the actual position of the newly added tag information and the position predicted by the positioning model. The output result is: 100% accurate positioning can be achieved for randomly and evenly replaced tags. The results are as follows: Fig.11 shown.

[0172] In this way, by using the trained positioning model, when the second tag feature data of the tag information to be positioned is input, accurate positioning of the tag information to be positioned can be achieved.

[0173] Example 2: Antenna-level positioning.

[0174] like Fig.12 As shown, reference tag information is deployed in the positioning division area. In the example of the embodiment of the present application, it is divided into 4 small areas, and 2 reference tag information is deployed in each positioning area, which can achieve antenna-level positioning (positions 1 and 2 belong to antenna 1, and positions 3 and 4 belong to antenna 2). It can also achieve position-level positioning in a directional search scenario according to needs.

[0175] The interactive process of this example implementation can be as follows Fig.13 As shown, the specific steps include:

[0176] (1) Sending positioning requirements: The service platform sends positioning requirements to the passive IoT management node, which may include antenna level and location level in this implementation example.

[0177] (2) Environmental measurement: The passive IoT management node controls the distributed nodes to perform environmental measurement and can send measurement control instructions to the passive distributed nodes, including the specified measurement time t. Since the deployed measurement tags are known, the system selects the reference tag information through the select instruction, that is, the passive system management node sends the select instruction to the passive system distribution node, and the passive system distribution node sends the select instruction to the tag or terminal to collect the corresponding tag feature information. The tag feature information corresponding to each Electronic Product Code (EPC) obtained from the tag or terminal includes: RSSI, frequency, phase, inventory count, etc.

[0178] (3) Obtaining the interference list: (e.g., using method 2 in the interference list acquisition method) The goal is to ensure that all reference tag information can be obtained and the receiving RSSI is greater than -90dBm. Assume that there is interference between antenna 1 and antenna 2, so only polling is possible.

[0179] (4) Scheduling and execution: When positioning is required, a service instruction is sent to the passive system management node through the service platform to select a scheduling method. Based on the interference list, antenna 1 and antenna 2 can execute the service in a polling manner. For example, a query instruction is sent to the passive system distribution node through the passive system management node, and the passive system distribution node obtains the tag feature information of the tag to be located from the tag or terminal through the query instruction to further realize antenna-level positioning. Alternatively, based on location-level positioning, a select instruction can be sent to the passive system distribution node through the passive system management node, and the passive system distribution node obtains the tag feature information corresponding to the specified target EPC.

[0180] (5) Fusion positioning:

[0181] Assuming the time is 1 minute, the collected information is shown in Table 4 below:

[0182] Table 4

[0183]

[0184]

[0185] It should be noted that: 1. The frequency point may not be fixed. In the case of frequency hopping, the frequency point information needs to be listed one by one. 2. The phase is not needed for antenna-level positioning, but if more precise positioning is required, it needs to be listed one by one together with the frequency point information.

[0186] Taking the information of a tag to be located as an example, if the collected information is as shown in the following Table 5:

[0187] Table 5

[0188]

[0189] The positioning calculation process is as follows:

[0190] Antenna 1 difference = ((Tag A Antenna 1 RSSI - Location 1 Antenna 1 RSSI) 2 + (Tag A Antenna 1 RSSI - Location 2 Antenna 1 RSSI) 2 )^(1 / 2)

[0191] Antenna 2 difference: ((Tag A Antenna 2 RSSI - Location 3 Antenna 2 RSSI) 2 +(Tag A Antenna 2 RSSI - Location 4 Antenna 2 RSSI) 2 )^(1 / 2)

[0192] Based on the above formula, we can calculate:

[0193] Antenna 1 difference = [(69-67) 2 +(69-69) 2 +(69-75) 2 +(69-78) 2 ]^(1 / 2)=11

[0194] Antenna 2 difference = [(75-81) 2 +(75-82) 2 +(75-84) 2 +(75-83) 2 ]^(1 / 2)=13.6

[0195] Therefore, the difference of antenna 1 is less than the difference of antenna 2, and the tag information A to be located is under antenna 1.

[0196] Based on the above embodiments, the embodiments of the present application have the following technical advantages:

[0197] 1. Expand the passive IoT positioning capabilities and improve positioning accuracy in complex interference environments.

[0198] 2. Meet the low-cost positioning needs based on passive Internet of Things in different scenarios.

[0199] 3. It can enhance the system’s differentiated service capabilities.

[0200] Based on the above embodiments, it can be seen that the embodiments of the present application mainly include the following points:

[0201] 1. In the area where positioning is required, feature data of the environment and reference tag information are obtained. The reference tag is the tag used in the real scene. The signaling side mainly includes the newly added measurement control instructions and carried parameters, and the business side includes the positioning demand division method, inventory mode selection based on demand, and the process of obtaining the interference list and the non-interference list.

[0202] 2. System node scheduling based on different positioning business needs, including polling inventory based on the last recorded final location as the starting node, concurrent inventory based on non-interference list information, and partition inventory based on non-interference list information.

[0203] 3. Tag feature information based on reference tag information obtained based on different positioning service requirements, including RSSI, initial phase, arrival phase, frequency, time, inventory times, characteristic RSSI, characteristic initial phase, characteristic arrival phase, etc.

[0204] 4. A method for constructing a positioning model based on an optimized random forest, including configuring feature weights based on the Gini coefficient to avoid the mutual exclusion problem between original features, introducing a self-feedback mechanism, training with 80% of the data, testing based on the remaining 20% ​​of the data, and self-optimizing the feature weights based on the position verification of the test set annotations.

[0205] Based on the above embodiment, in another embodiment of the present application, a positioning device 1 is provided, such as Fig.14 As shown, the positioning device 1 includes:

[0206] The acquisition unit 10 is used to acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located.

[0207] The determination unit 11 is used to determine the position information of the tag information to be located in the area to be located based on the first tag feature information and the second tag feature information.

[0208] In one embodiment, the number of reference tag information is one or more, the number of corresponding first tag feature information is one or more, and the reference tag information and the first tag feature information are in a one-to-one relationship.

[0209] In one embodiment, the determining unit 11 is further used to determine the distance between each reference tag information and the tag information to be located based on the second tag feature information and one or more first tag feature information, wherein different reference tag information is used to identify different locations;

[0210] The determining unit 11 is further configured to determine a minimum distance from the distances.

[0211] The determination unit 11 is further configured to determine the position information of the tag information to be located in the area to be located based on a minimum distance.

[0212] In one embodiment, the positioning device may further include: a training unit.

[0213] The training unit is used to perform model training using the first label feature information to obtain a positioning model.

[0214] The training unit is further used to input the second label feature information into the positioning model to obtain the position information of the label information to be positioned corresponding to the second label feature information in the area to be positioned.

[0215] In one embodiment, the acquiring unit 10 is further configured to acquire the positioning requirement information or the positioning level information corresponding to the positioning requirement information.

[0216] The determining unit 11 is further configured to determine an inventory mode of the reference tag information and the tag information to be located based on the positioning requirement information or the positioning level information.

[0217] The acquisition unit 10 is further used to collect the first tag feature information and the second tag feature information based on the inventory mode.

[0218] In one embodiment, the inventory mode includes a concurrent inventory mode.

[0219] In one embodiment, the determining unit 11 is further configured to determine the do not interfere list based on the reference tag information when the inventory mode includes the concurrent inventory mode.

[0220] The acquisition unit 10 is further configured to collect the first tag feature information and the second tag feature information based on the non-interference list.

[0221] In one embodiment, the determination unit 11 is further used to determine the non-interference list based on the inventory information of whether the reference tag information is stored in the inventory; or, determine the non-interference list based on a first comparison result of whether the reference tag information is stored in the inventory information and whether the signal quality parameter information of the reference tag information is greater than a first threshold; or, determine the non-interference list based on a second comparison result of whether the reference tag information is stored in the inventory information and whether the number of times the reference tag information is stored in a preset time is greater than a second threshold.

[0222] In one embodiment, the first tag characteristic information includes one or more of the following: first RSSI information, a first phase, a first frequency point, and a first inventory count.

[0223] In one embodiment, the second tag characteristic information includes one or more of the following: second RSSI information, second phase, second frequency point, and second inventory times.

[0224] An embodiment of the present application provides a positioning device, which obtains first label feature information corresponding to reference label information and second label feature information corresponding to label information to be located, wherein the reference label information is label information pre-deployed in the area to be located; based on the first label feature information and the second label feature information, the position information of the label information to be located in the area to be located is determined. It can be seen that a positioning device proposed in an embodiment of the present application deploys reference label information in the area to be located. Since the reference label information is used as a reference, the position of the reference label information is known during deployment. Therefore, by obtaining the first label feature information of the reference label information and obtaining the second label feature information of the label information to be located, since the first label feature information corresponds to the reference label information of the deployed known position, the obtained first label feature information and the second label feature information are combined, and the position information of the label information to be located corresponding to the second label feature information can be further judged according to the first label feature information. In this way, the position information of the label to be located obtained is more accurate.

[0225] Fig.15 This is a schematic diagram of the structure of a positioning device 1 provided in an embodiment of the present application. In practical applications, based on the same disclosed concept of the above embodiments, Fig.15 As shown, the positioning device 1 of the embodiment of the present application includes a processor 12 , a memory 13 and a communication bus 14 .

[0226] In the specific implementation process, the acquisition unit 10, the determination unit 11, and the training unit can be implemented by a processor 12 located on the positioning device 1, and the processor 12 can be at least one of an application-specific integrated circuit (ASIC), a digital signal processor (DSP), a digital signal processing image processing device (DSPD), a programmable logic image processing device (PLD), a field programmable gate array (FPGA), a CPU, a controller, a microcontroller, and a microprocessor. It can be understood that for different devices, the electronic device used to implement the above-mentioned processor function can also be other, and the embodiments of the present application do not make specific limitations.

[0227] In the embodiment of the present application, the communication bus 14 is used to realize the connection and communication between the processor 12 and the memory 13; when the processor 12 executes the running program stored in the memory 13, the following positioning method is implemented:

[0228] Obtain first tag feature information corresponding to the reference tag information and second tag feature information corresponding to the tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located; determine the location information of the tag information to be located in the area to be located based on the first tag feature information and the second tag feature information

[0229] In one embodiment, the number of reference tag information is one or more, the number of corresponding first tag feature information is one or more, and the reference tag information and the first tag feature information are in a one-to-one relationship.

[0230] In one embodiment, the processor 12 is further used to determine the distance between each reference tag information and the tag information to be located based on the second tag feature information and one or more first tag feature information, wherein different reference tag information is used to identify different positions; determine a minimum distance from the distances; and determine the position information of the tag information to be located in the area to be located based on the minimum distance.

[0231] In one embodiment, the processor 12 is further used to perform model training using the first label feature information to obtain a positioning model; and input the second label feature information into the positioning model to obtain position information of the label information to be located corresponding to the second label feature information in the area to be located.

[0232] In one embodiment, the processor 12 is further used to obtain positioning requirement information or positioning level information corresponding to the positioning requirement information; determine an inventory mode of reference tag information and tag information to be located based on the positioning requirement information or the positioning level information; and collect first tag feature information and second tag feature information based on the inventory mode.

[0233] In one embodiment, the inventory mode includes a concurrent inventory mode.

[0234] In one embodiment, the processor 12 is further configured to determine a non-interference list based on reference tag information when the inventory mode is a concurrent inventory mode; and collect the first tag feature information and the second tag feature information based on the non-interference list.

[0235] In one embodiment, the processor 12 is further configured to determine a non-interference list based on the inventory information of whether the reference tag information is stored in the inventory; or, determine the non-interference list based on a first comparison result of whether the reference tag information is stored in the inventory information and whether the signal quality parameter information of the reference tag information is greater than a first threshold; or, determine the non-interference list based on a second comparison result of whether the reference tag information is stored in the inventory information and whether the number of times the reference tag information is stored in a preset time is greater than a second threshold.

[0236] In one embodiment, the first tag characteristic information includes one or more of the following: first RSSI information, a first phase, a first frequency point, and a first inventory count.

[0237] In one embodiment, the second tag characteristic information includes one or more of the following: second RSSI information, second phase, second frequency point, and second inventory times.

[0238] Based on the above embodiments, the embodiments of the present application provide a storage medium on which a computer program is stored. The above computer-readable storage medium stores one or more programs. The above one or more programs can be executed by one or more processors and applied to a positioning device. The computer program implements the positioning method as described above.

[0239] Based on the above embodiments, an embodiment of the present application provides a computer program product, including a computer program. The computer program can be executed by one or more processors and applied to a positioning device. The computer program implements the positioning method as described above.

[0240] It should be noted that in the embodiments of the present application, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0241] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment method can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the relevant technology can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD), including a number of instructions for an image display device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the method described in each embodiment of the embodiment of the present application.

[0242] The above is only a specific implementation of the embodiment of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A positioning method, characterized in that: The method comprises: Acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located; Based on the first tag feature information and the second tag feature information, position information of the tag information to be located in the area to be located is determined.

2. The method according to claim 1, characterized in that The number of the reference tag information is one or more, the number of the corresponding first tag feature information is one or more, the reference tag information and the first tag feature information are in a one-to-one relationship, and the determining, based on the first tag feature information and the second tag feature information, the position information of the tag information to be located in the area to be located includes: Based on the second tag feature information and one or more first tag feature information, determining the distance between each reference tag information and the tag information to be located, wherein different reference tag information is used to identify different locations; determining a smallest distance from among the distances; Based on the minimum distance, position information of the tag information to be located in the area to be located is determined.

3. The method according to claim 1, characterized in that The determining, based on the first tag feature information and the second tag feature information, the position information of the tag information to be located in the area to be located includes: Using the first label feature information to perform model training to obtain a positioning model; The second tag feature information is input into the positioning model to obtain the position information of the tag information to be positioned corresponding to the second tag feature information in the area to be positioned.

4. The method according to claim 1, characterized in that: The step of acquiring first tag feature information corresponding to the reference tag information and second tag feature information of the tag information to be located includes: Acquire positioning requirement information or positioning level information corresponding to the positioning requirement information; Determining an inventory mode of the reference tag information and the tag information to be located based on the positioning requirement information or the positioning level information; Based on the inventory mode, the first tag feature information and the second tag feature information are collected.

5. The method according to claim 4, characterized in that The inventory mode includes a concurrent inventory mode, and the collecting of the first tag feature information and the second tag feature information based on the inventory mode includes: In a case where the inventory mode includes a concurrent inventory mode, determining a do not interfere list based on the reference tag information; Based on the non-interference list, the first tag feature information and the second tag feature information are collected.

6. The method according to claim 5, characterized in that The step of determining the do not interfere list based on the reference tag information includes: determining the do not interfere list based on the inventory information of whether the reference tag information is inventoried; Alternatively, the do not interfere list is determined based on a first comparison result of whether the reference tag information is inventoried and whether the signal quality parameter information of the reference tag information is greater than a first threshold; Alternatively, the do not interfere list is determined based on a second comparison result of inventory information of whether the reference tag information is inventoried and whether the number of times the reference tag information is inventoried within a preset time is greater than a second threshold.

7. The method according to claim 1, characterized in that The first tag characteristic information includes one or more of the following: first received signal strength indication RSSI information, a first phase, a first frequency point, and a first inventory count.

8. The method according to claim 1, characterized in that The second tag characteristic information includes one or more of the following: second RSSI information, second phase, second frequency point, and second inventory times.

9. A positioning device, characterized in that: The positioning device comprises: An acquiring unit, configured to acquire first tag feature information corresponding to reference tag information and second tag feature information corresponding to tag information to be located, wherein the reference tag information is tag information pre-deployed in the area to be located; A determination unit is used to determine the position information of the tag information to be located in the area to be located based on the first tag feature information and the second tag feature information.

10. A positioning device, characterized in that: The positioning device comprises: a processor and a memory; when the processor executes the running program stored in the memory, the method according to any one of claims 1 to 8 is implemented.

11. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 8.

Images