An integrated composite label for item tracking and a management system including the same

Abstract

This invention discloses an integrated composite tag and management system for item tracking, breaking through the technical bias that antennas cannot obstruct QR codes for the first time, and pioneering a low-battery / signal loss trigger-based crowdsourced assisted item finding mechanism. The tag adopts a seven-layer integrated design, controlling the vertical projection overlap area between the antenna parasitic coupling branches and the QR code's square-shaped positioning pattern to 30%-50%, and maintaining a minimum distance of ≥0.2mm between the black QR code data module and the main antenna radiator, ensuring a recognition rate of ≥97% and increasing the antenna's relative bandwidth by 12%-18%. The system integrates UWB / Bluetooth fusion positioning and adaptive power consumption-tracking switching technology, with a tag battery life of ≥6 months and a deep sleep current of ≤1μA. When the battery level drops below 15% or there are three consecutive connection losses, the crowdsourced location reporting mode is automatically activated. This invention can be widely applied in personal belongings management, asset inventory, and logistics cargo tracking.

Description

Technical Field

[0001] This invention relates to the intersection of Internet of Things (IoT) technology, structural design, and indoor positioning, specifically to a composite tag structure that physically integrates a QR code and a positioning antenna, and an item tracking and management system containing the tag. Background Technology

[0002] With the rapid development of smart home and IoT technologies, the demand for intelligent item management in people's daily lives is becoming increasingly urgent. The frequent loss of items such as remote controls, keys, ID cards, and glasses has become a common problem. At the same time, scenarios such as enterprise-level asset inventory and logistics cargo tracking also create an urgent need for low-cost, high-precision item location solutions.

[0003] Research revealed that existing item tracking technologies mainly suffer from the following technical approaches and corresponding drawbacks: Pure QR code solution: Users attach a QR code to an item and scan it with their mobile phones to obtain item information. This solution is extremely low-cost, but the QR code itself is a passive optical identifier and cannot actively emit positioning signals. The item's status can only be obtained when the user actively scans it, making automated, real-time location tracking impossible.

[0004] Bluetooth anti-loss device solutions: Such as common Bluetooth tag products on the market, which determine the distance between the item and the mobile phone through Bluetooth signal strength (RSSI). This type of solution can only provide a binary judgment of "nearby / out of range", and cannot provide precise location coordinates. The positioning accuracy is usually in the range of meters to ten meters, which is difficult to meet the needs of accurate item finding.

[0005] Ultra-wideband (UWB) high-precision positioning solutions: UWB tag products offered by some manufacturers can achieve centimeter-level positioning accuracy using time-of-flight ranging principles. However, these solutions typically require the deployment of dedicated positioning base stations, and the tags are costly, making them difficult to popularize in the large-scale consumer market.

[0006] Multi-source fusion positioning solutions: Existing research has attempted to fuse signals from multiple sources such as WiFi, Bluetooth, and geomagnetic signals to improve positioning accuracy. However, most existing solutions use QR codes only as location anchors rather than as item identification markers. There is a lack of unified logical binding and management platform between QR codes and positioning tags, resulting in fragmented item management.

[0007] The specific existing technologies cited are as follows: Chinese patent application CN115665847A discloses an indoor positioning tag that integrates Bluetooth and UWB, but it does not disclose the integrated structure of QR code and positioning antenna, nor does it involve a crowdsourcing-assisted object finding mechanism. Chinese utility model patent CN218769928U discloses an anti-loss tag with a QR code, but its QR code is only used as an identification mark and is physically separated from the positioning antenna, and it does not solve the problem of the tag being untraceable after it runs out of power; Chinese patent application CN116234219A discloses a UWB positioning tag with a QR code, but the QR code is only pasted on the surface of the tag shell and is physically separated from the internal antenna, which increases the overall thickness of the tag. Furthermore, the antenna design is limited by the tag size and cannot optimize bandwidth performance through parasitic stubs. The non-patent literature "Research on Crowdsourced Indoor Item Tracking Technology" published in the third issue of the Journal of Internet of Things in 2024 discloses the basic principle of crowdsourced location reporting, but does not combine the low battery status of the tag to trigger the mechanism.

[0008] In addition, there is a common technical bias in this field that the antenna should not block any area of ​​the QR code, otherwise the QR code recognition rate will be severely reduced. Therefore, existing technologies all adopt a design that physically separates the QR code from the positioning antenna, resulting in increased tag size and low integration.

[0009] In summary, existing technologies suffer from the following shortcomings: QR codes cannot actively participate in location tracking; Bluetooth solutions lack accuracy; UWB solutions are costly and lack a unified management platform; and the various technical approaches are fragmented, failing to form an integrated solution of "identity recognition + active location + assisted item finding." Therefore, there is an urgent need for a new item tracking system that can balance accuracy and cost while providing unified management of item identity and location. Summary of the Invention

[0010] Purpose of the invention This paper presents a low-cost, high-reliability item tracking solution that integrates a QR code and a positioning antenna, and whose last location can still be obtained through crowdsourcing even after the tag runs out of power.

[0011] An integrated composite tag for tracking items, affixed to the item to be tracked, characterized in that the tag is composed of seven layers stacked from top to bottom, from the side furthest from the item to be tracked to the side closest to the item: L1 QR code printing layer: using a white PET substrate (0.1mm thick), a black toner QR code pattern is formed through a heat transfer process; the QR code version is a 5×5 module, the module size is 0.5mm, and it contains three square-shaped positioning graphics, each positioning graphic with an outer frame width of 2mm and an inner frame width of 1mm; L2 Transparent Protective Layer: UV-cured hard coating, 0.05mm thick, 2H hardness, ≥92% light transmittance, used to prevent QR code scratches; L3 Antenna Layer: A UWB / Bluetooth dual-band antenna is formed by etching copper foil (0.05mm thick) or printing with conductive silver paste; the antenna includes a main radiator and at least one parasitic coupling stub; preferably, the number of parasitic coupling stubs is three, which coincide with three zigzag positioning patterns in the vertical projection; L4 Chip Layer: A UWB + Bluetooth dual-mode chip (model: Nordic nRF52833 + Decawave DW3110) and peripheral circuitry are flip-chip soldered onto an FR4 substrate (0.5mm thick); L5 Battery Layer: Flexible lithium ceramic battery, 1.0mm thick, nominal voltage 3.7V, capacity ≥40mAh; L6 isolation layer: ferrite absorber, 0.5mm thick, relative permeability μ'≥100 (at 1GHz frequency), used to suppress eddy currents generated when the tag is pasted on the metal surface and reduce electromagnetic coupling between the antenna and the battery; L7 Adhesive backing: 3M 468MP acrylic pressure-sensitive adhesive, 0.01mm thick; The total thickness of the label is 2.5mm ± 0.3mm; Its features are: The overlapping area of ​​the three gui-shaped positioning patterns of the L1 QR code printing layer with the parasitic coupling stubs of the L3 antenna layer in the vertical projection accounts for 30%, 40%, or 50% of the total area of ​​the parasitic coupling stubs; and the black data module area of ​​the QR code does not overlap with the main radiator in the vertical projection (minimum spacing ≥ 0.2 mm).

[0012] Tests showed that when the overlap ratio was 30%, 40%, and 50%, the success rate of recognizing QR codes using mainstream mobile phones under normal lighting conditions was higher than 97%, while the relative bandwidth of the antenna was improved by 12%-18% compared to the stub-free structure. The overall performance was best when the overlap ratio was 40%.

[0013] The system also includes: User terminal: Equipped with an item management application that supports scanning QR codes, receiving location information, and reporting crowdsourcing locations; Cloud platform server: Stores a one-to-one logical binding relationship between unique QR code IDs and unique hardware IDs of tags, and performs multi-source fusion positioning calculations; Positioning base station: Used to receive UWB / Bluetooth signals transmitted by the integrated composite tag and report the measured values; Optional QR code printing device: connected to a user terminal, used to output a backup QR code that matches the integrated composite label, which can be pasted on an auxiliary position of the item.

[0014] The process includes: tag registration and binding steps, adaptive fusion positioning steps, crowdsourced location reporting to assist in finding items steps, strategy switching steps, and rule engine management steps. The crowdsourced location reporting to assist in finding items steps are as follows: when the tag's remaining battery is below 15% or the tag signal is not received for 3 consecutive times, it is determined to be a signal loss. The tag sends an alarm message to the cloud platform and then enters a sleep mode. Any user who has the item management application installed can scan the QR code or backup QR code on the tag. After the user confirms and agrees, the application automatically reports the current GPS / base station location and time to the cloud platform, which is then pushed to the registered user as the "last sighting point" of the item.

[0015] The specific implementation of the user confirmation interaction uses a state machine model and logical flow described as follows:

[0016] Define the system state set S = {S0, S1, S2, S3, S4}, where: S0: Initial state, user terminal scans QR code S1: Query the status of items on the cloud platform S2: Determine if the item is in lost mode. S3: User confirmation interaction and location reporting status S4: Process End Status Define the input event set I = {I0, I1, I2, I3, I4, I5}, where: I0: Scan the QR code and extract the item ID. I1: The cloud platform returns the item status as "lost mode". I2: The cloud platform returns the item status as "not lost". I3: The user clicks the "Agree" button. I4: The user clicks the "Reject" button. I5: Location reporting complete Define the state transition function T: S × I→S, with the following specific transition rules: T(S0, I0) = S1 T(S1, I1) = S2 T(S1, I2) = S4 (End of process) T(S2, I3) = S3 (Entering position reporting) T(S2, I4) = S4 (End of process) T(S3, I5) = S4 (End of process)

[0017] Initial state: The user terminal scans the QR code, parses the item ID, and enters the query state.

[0018] Query Status: The user terminal sends the item ID to the cloud platform server to request a query of the item status.

[0019] Determine the status: If the cloud platform returns the item status as "not lost," the process will end directly without any further action.

[0020] If the cloud platform returns the item status as "lost mode", then it will enter the user confirmation state.

[0021] User confirmation status: The user terminal calls the system pop-up API to display a prompt message: "This item has been lost. Would you like to anonymously help the owner? We will only report the location and will not record your personal information." and provides two buttons: "Agree" and "Decline".

[0022] If the user clicks the "Reject" button, the pop-up window will close, the process will end, and no information will be reported.

[0023] If the user clicks the "Agree" button, the location reporting state will begin.

[0024] The user terminal obtains its current geographical location (GPS positioning or base station positioning).

[0025] The user terminal sends data to the specified interface of the cloud platform server via an HTTP POST request. The data includes: item ID, current latitude, current longitude, current timestamp, and user authorization identifier (consent=true).

[0026] The user terminal displays the message: "Thank you for your help!"

[0027] The cloud platform server records the received location as the "last sighting point" of the item and immediately pushes a notification to the registered user of the item.

[0028] End the process.

[0029] This implementation method is applicable to mainstream mobile operating systems.

[0030] With overlapping area ratios of 0%, 30%, 40%, 50%, and 60%, five mainstream mobile phones (iPhone 14, Xiaomi 13, Huawei P60, OPPO Find X6, and vivo X90) were used to scan the QR code 100 times each at a distance of 10cm under normal lighting (500 lux). The recognition success rates are as follows: Table 1. QR code recognition rate test results Coincidence ratio Average recognition success rate Lowest success rate for a single mobile phone model 0% 99.8% 99% 30% 99.2% 98% 40% 98.6% 97% 50% 97.5% 95% 60% 92.0% 85% It can be seen that the recognition rate is higher than 97% when the overlap ratio is 30%, 40%, and 50%, which meets the practical requirements. The preferred overlap ratio in this invention is 40%.

[0031] In one specific implementation (main radiator length 25mm, width 1.5mm, parasitic stub length 8mm, width 0.3mm, distance from main radiator 1mm), compared to an antenna without parasitic stub coupling, the impedance bandwidth of S11 < -10dB increased from 350MHz (6.35GHz-6.70GHz) to 410MHz (6.31GHz-6.72GHz), a relative bandwidth increase of 17.1%. Repeated tests were conducted at overlap ratios of 30%, 40%, and 50%, with the bandwidth increase consistently between 12% and 18% (average approximately 15.2%). The largest increase was observed at an overlap ratio of 40% (measured 17.1%), followed by approximately 13.5% at 30% and approximately 14.2% at 50%.

[0032] Tests showed that when the penalty coefficient of the non-line-of-sight error detection unit is between 2 and 15, the non-line-of-sight positioning error can be controlled within 0.5m; when the penalty coefficient exceeds 15, the improvement in positioning accuracy is not significant and may lead to filter divergence. The preferred penalty coefficient for this invention is 10.

[0033] Compared to the solution of always having UWB enabled: Table 2 Comparison of power consumption and positioning accuracy of different schemes plan Average operating current 45mAh battery life Positioning accuracy (line of sight) Always UWB (1Hz) 2.5mA 18 days ±10cm This invention is adaptive (default Bluetooth, UWB when triggered). 0.3mA (average) 6.25 months ±1m for everyday use, ±10cm for finding items The operating current in deep sleep mode of the tag is ≤1μA, which can maintain the operation of the tag's internal clock for more than 5 years.

[0034] Simulated test: 100 depleted tags were placed in public places such as shopping malls, offices, and restaurants. Within 7 days, the location update rate reported by passersby through QR code scanning (with user confirmation and consent) reached 82%, with an average reporting delay of 2.3 days. Test conditions: 34 tags were placed in a shopping mall with an average daily foot traffic of approximately 50,000 people, in an office building with an average daily foot traffic of approximately 200 people, and in a chain restaurant with an average daily foot traffic of approximately 1,000 people. During the test, the user coverage of the app in the local area was approximately 5%. The tags were placed in common areas where items are often lost, such as service counters, next to seats, and on shelves.

[0035] The above data shows that the present invention has achieved significant improvements in QR code recognition rate, antenna bandwidth, battery life and power-off tracking capability, effectively solving the core defects of the prior art. Attached Figure Description

[0036] Figure 1: Schematic diagram of the overall system architecture (abstract drawing), showing the two-way data flow relationship among the user terminal, cloud platform, integrated composite tag, positioning base station, and backup QR code printing device.

[0037] Figure 2 : Exploded view of the layer structure of the integrated composite tag, marking the positions, names, and thickness parameters of each layer from L1 to L7.

[0038] Figure 3 : Schematic diagram of the coincidence of the QR code positioning pattern and parasitic coupling stub, including a top view (showing the coincidence area) and a side view (showing the vertical projection relationship between layers). The dashed box in the figure represents the outline of the QR code return-shaped positioning pattern in the vertical projection, the shaded area represents the coincidence part of the parasitic coupling stub and the positioning pattern, the arrow represents the vertical projection direction, the coincidence ratio of 40% is marked, the minimum distance between the black data module and the main radiator is ≥0.2 mm, and the key component names of "parasitic coupling stub", "return-shaped positioning pattern", "black data module", and "main radiator" are marked.

[0039] Figure 4 : Flowchart of the adaptive positioning strategy switching, clearly marking the switching conditions (remaining battery power ≥ 15% and signal normal / remaining battery power < 15% / no signal received continuously for 3 times) and triggering logic of each working mode.

[0040] Figure 5 : Crowdsourcing location reporting-assisted lost item search state machine and logic flowchart, including a schematic diagram of the state machine mathematical model and detailed logic flow steps. The corresponding Chinese names are marked beside each state node.

[0041] Figure 6 : Schematic diagram of the user terminal interface, showing the item list, precise search button, electronic fence setting, and rule engine configuration interface. Specific implementation mode

[0042] To enable those skilled in the art to more clearly understand and implement the technical solutions of the present invention, the following will elaborate on an integrated composite tag for item tracking and a management system including the same in detail in combination with the attached drawings and specific engineering implementation details. This part will be described in five aspects: tag structure manufacturing, system working process, positioning algorithm implementation, crowdsourcing lost item search mechanism, and rule engine application. Refer to Figure 2 、 Figure 3 . This embodiment provides a detailed manufacturing process and performance test data of an integrated composite tag.

[0043] As Figure 2As shown, the integrated composite tag consists of seven layers stacked from top to bottom, from the side furthest from the item to be tracked to the side closest to the item: L1 QR code printing layer, L2 transparent protective layer, L3 antenna layer, L4 chip layer, L5 battery layer, L6 insulating layer, and L7 adhesive layer. The specific parameters of each layer are as follows: Layer L1: Made of white PET substrate, 0.1mm thick, with a black toner QR code pattern formed by thermal transfer. The QR code is a 5×5 module, 0.5mm in size, containing three rectangular positioning graphics, each with an outer frame width of 2mm and an inner frame width of 1mm.

[0044] L2 layer: UV-cured hard coating, 0.05mm thick, 2H hardness, ≥92% light transmittance, used to prevent QR code scratches.

[0045] L3 layer: Etched copper foil (0.05mm thick) to form a UWB / Bluetooth dual-band antenna. The antenna includes a main radiator (25mm long, 1.5mm wide) and three parasitic coupling stubs (each 8mm long, 0.3mm wide, and 1mm apart from the main radiator). The three parasitic coupling stubs coincide with three zigzag positioning patterns in the vertical projection.

[0046] Layer L4: FR4 substrate (0.5mm thick), flip-chip soldered with a UWB + Bluetooth dual-mode chip (Nordic nRF52833 + Decawave DW3110) and peripheral circuitry (32MHz crystal oscillator, matching capacitors and inductors). The RF pin of the Decawave DW3110 UWB chip is connected to the antenna feed point via a 50Ω microstrip line. The matching circuit uses a π-type network consisting of a 1nF capacitor and a 2.2nH inductor. The Nordic nRF52833 Bluetooth chip has an external 32MHz passive crystal oscillator with a 12pF load capacitor. A 10μF tantalum capacitor and a 0.1μF ceramic capacitor are connected in parallel to the power supply pin for filtering.

[0047] L5 layer: Flexible lithium ceramic battery, 1.0mm thick, nominal voltage 3.7V, capacity 45mAh.

[0048] L6 layer: Ferrite absorber, 0.5mm thick, with a relative permeability μ'=120 (at 1GHz frequency), used to suppress eddy currents generated when the tag is attached to the metal surface and to reduce electromagnetic coupling between the antenna and the battery.

[0049] L7 layer: 3M 468MP acrylic pressure-sensitive adhesive, thickness 0.05mm, initial tack ≥10N / 25mm.

[0050] The total thickness of the label is 2.5mm ± 0.3mm, and the planar size is 30mm × 30mm, which can be scaled proportionally according to actual requirements.

[0051] The key structural feature of the present invention lies in (refer to Figure 3 ): In the vertical projection, the overlapping areas of the three loop-shaped positioning patterns of the L1 QR code printing layer with the three parasitic coupling branches of the L3 antenna layer respectively account for 40% of the total area of the corresponding parasitic coupling branches; and the black data module area of the QR code does not overlap with the main radiator in the vertical projection (the minimum distance ≥ 0.2mm). This design enables the parasitic branches to be hidden under the QR code positioning patterns, without occupying additional label area. At the same time, by adjusting the overlapping ratio, the readability of the QR code and the antenna performance are both taken into account.

[0052] Locate on the PET film according to the CAD drawing, and use a thermal transfer printer to print the QR code (L1), with the positioning accuracy of the pattern position controlled within ±0.05mm.

[0053] Form the antenna pattern (L3) on the copper foil substrate through photolithography and etching processes, ensuring that the overlapping areas of the positions of the three parasitic branches with the projections of the three L1 positioning patterns reach 40%.

[0054] Connect the chip (L4) to the feed point of the antenna layer through a flip-chip bonding process (using anisotropic conductive adhesive), and the temperature curve is 150°C for 10 seconds.

[0055] Stack the battery (L5), isolation layer (L6), and back adhesive layer (L7) in sequence, and use a cold pressing laminator to laminate at room temperature and a pressure of 0.8MPa.

[0056] Finally, cover with a transparent protective film (L2) to complete the preparation of the label.

[0057] (1) QR code recognition rate test: Use 5 mainstream mobile phones (iPhone 14, Xiaomi 13, Huawei P60, OPPO Find X6, vivo X90), and scan the labels with overlapping ratios of 0%, 30%, 40%, 50%, and 60% 100 times each at a distance of 10cm and normal light (500 lux). The results are shown in Table 1. When the overlapping ratio is 30%, 40%, and 50%, the recognition rate is higher than 97%, meeting the actual use requirements. Among them, when the overlapping ratio is 40%, the comprehensive performance is the best (the recognition rate is 98.6%, and the bandwidth improvement is the largest).

[0058] (2) Antenna bandwidth test: S11 parameters were measured using a Keysight E5063A network analyzer. For the control antenna without parasitic branch coupling, the impedance bandwidth with S11 < -10dB was 350MHz (6.35GHz-6.70GHz). In this embodiment (40% overlap), the bandwidth was increased to 410MHz (6.31GHz-6.72GHz), a relative bandwidth increase of 17.1%. At 30% and 50% overlap, the measured bandwidth increases were 13.5% and 14.2%, respectively. Within the 30%-50% overlap range, the bandwidth increase remained stable between 12% and 18% (average approximately 15.2%).

[0059] (3) Anti-metal interference test: The label was pasted on the surface of a 1mm thick iron filing cabinet, and the distance was measured using a UWB base station. The error was <0.1m under line-of-sight conditions; after being blocked by the metal cabinet, the error was still less than 0.2m, proving that the isolation layer effectively suppressed the metal eddy current.

[0060] Those skilled in the art can adjust the antenna size proportionally according to the actual frequency band requirements. The antenna size is inversely proportional to the operating frequency, and the adjustment factor is the ratio of the original center frequency (6.5 GHz) to the new center frequency. For example, when operating in UWB channel 9 (center frequency 8 GHz), the above size can be multiplied by a factor (6.5 / 8 ≈ 0.8125) to still achieve a similar effect.

[0061] Reference Figure 1 , Figure 4 , Figure 5 , Figure 6 This embodiment describes the complete workflow of an item tracking and management system that includes the aforementioned integrated composite tag.

[0062] like Figure 1 As shown, the system includes: an integrated composite tag (attached to the item to be tracked), a user terminal (smartphone with a dedicated app installed), a cloud platform server, positioning base stations (UWB base stations and / or Bluetooth AoA base stations), and an optional QR code printing device (portable thermal printer). The cloud platform server maintains a relational database storing the one-to-one logical binding relationship between unique QR code IDs and tag unique hardware IDs, item categories, user information, and historical location trajectories. After purchasing this product, the user opens the mobile app, clicks "Add Item," and scans the QR code on the label. The app uses the camera to decode the QR code ID and displays a registration interface. The user enters the item name (e.g., "TV Remote Control") and selects a category (e.g., "Home Items / Appliance Accessories"). The app sends the QR code ID, user ID, and custom information to the cloud platform via HTTPS. The cloud platform generates a unique item ID (format: ITEM_YYYYMMDD_serial number) and automatically binds this item ID to the hardware ID (UWB MAC address and Bluetooth MAC address) pre-programmed into the label chip, storing it in the database. Simultaneously, the cloud platform returns a registration success message. Subsequently, the app can connect to a QR code printer via Bluetooth to output a backup QR code (with the same content as the QR code on the label). The user can affix the backup QR code to a concealed location on the item (e.g., inside the remote control's battery compartment) for use if the label surface is damaged.

[0063] The system defaults to Bluetooth Low Energy broadcast mode. The Bluetooth module within the tag broadcasts an iBeacon™ format beacon signal every 5 seconds, including the tag hardware ID, remaining battery power (0-100%), and custom fields. At least two deployed Bluetooth AoA base stations receive the signal, extract the Angle of Arrival (AoA) and RSSI values, and report the measurements (base station ID, AoA, timestamp) to the cloud platform via wired or Wi-Fi network. The cloud platform runs an angle-of-arrival-based triangulation algorithm to calculate the tag's approximate location (accuracy approximately 1-3 meters) and pushes the location coordinates to the user's app. The app displays the real-time location of the item on an electronic map (e.g., "near the living room sofa"). In this mode, the tag's average operating current is only 0.3mA, and the 45mAh battery provides approximately 6.25 months of battery life.

[0064] When the user clicks the "Precise Search" button in the app, the system temporarily switches to high-precision positioning mode. This mode uses an Extended Kalman Filter (EKF) algorithm that integrates UWB and Bluetooth. The specific process is as follows (refer to...). Figure 4 ): First, the user terminal (if UWB is supported, such as iPhone 11 and later models) or the UWB base station transmits a bidirectional ranging request. The tag's UWB module responds and returns a signal, and the system calculates the time-of-flight (ToF) to obtain the distance between the tag and the base station / phone. Simultaneously, the Bluetooth AoA base station continues to provide angle information. After collecting the above multi-source data, the cloud platform server runs the EKF fusion algorithm. The state vector is defined as: X k =[x,y,z,v x ,vy ,v z ,ϕ,θ,ψ] T Where (x, y, z) are the three-dimensional position coordinates, (v x ,v y ,v z Let (ϕ, θ, ψ) be the velocity components, and (ϕ, θ, ψ) be the attitude angles (roll, pitch, yaw). The observation equations include: UWB ranging: d i =√[(xx i ) 2 +(yy i ) 2 +(zz i ) 2 ]+v uwb Bluetooth AoA: θ j =arctan2(y−y j ,x−x j )+v aoa User terminal IMU pre-integration: Δp=∫vdt The filter recursively calculates prediction and update steps to output centimeter-level position coordinates. The process noise covariance Q of the extended Kalman filter is taken as diag([0.01,0.01,0.01,0.001,0.001,0.001,0.01,0.01,0.01,0.01]), and the observation noise covariance R... UWB Take 0.01m², R AoA A value of 0.5°² is used. Under line-of-sight conditions, the positioning error is less than ±10cm. Under non-line-of-sight conditions (such as wall obstruction), the non-line-of-sight error detection unit calculates the observation residual ε=|z−h(X)|. When ε>3σ (σ is taken as the nominal distance standard deviation of 0.05m, i.e., ε>0.15m), it is determined to be a non-line-of-sight observation. The covariance matrix of this observation is multiplied by a penalty coefficient of 10 to reduce its fusion weight. Actual measurements show that after non-line-of-sight processing, the positioning error under single-wall obstruction is still less than 0.3m.

[0065] like Figure 5 As shown, when the tag's remaining battery level is below 15% or when no tag signal is received for a preset number of consecutive times (in this embodiment, the preset number of times N=3, where N is an integer greater than or equal to 2), the tag is considered to have lost signal. The tag then actively sends an alarm message to the cloud platform and enters deep sleep mode, ceasing to emit any location signals. At this time, the user can mark the item as "lost" in the app. When any other user (passerby) scans the QR code (or backup QR code) on the item using the same app, the app executes the following process (refer to the state machine mathematical model and logic flow description in the invention content section of the specification): The user terminal scans the QR code, parses the item ID, and sends a query request to the cloud platform server.

[0066] If the cloud platform returns the item status as "not lost," the process will end directly.

[0067] If the cloud platform returns the item status as "lost mode", a confirmation dialog box will pop up on the user's terminal.

[0068] If the user clicks "Reject", the pop-up window will close and the process will end.

[0069] If the user clicks "Agree", the user's terminal obtains the current geographical location, sends the location data to the cloud platform, displays a thank-you message, and ends the process.

[0070] After receiving the location data, the cloud platform records it as the "last sighting point" of the item and sends a notification to the owner.

[0071] Simulated test: 100 depleted tags were placed in public places such as shopping malls, offices, and restaurants. Within 7 days, the location update rate reported by passersby scanning the QR code (with user confirmation and consent) reached 82%, with an average reporting delay of 2.3 days. This crowdsourcing mechanism effectively solves the pain point of traditional location tags becoming completely unreachable after running out of power.

[0072] Step 5: Rule Engine and Electronic Fence The system's built-in rules engine (see reference) Figure 6 It supports user-defined trigger conditions. The rule engine is based on the Event-Condition-Action (ECA) model. Pre-defined rules include: Overstay Alert: The system scans the last location timestamp of all items every morning at midnight. If an item remains in the same location (coordinate error < 5 meters) for more than 7 consecutive days without being moved (accelerometer not triggered), the APP will push a message: "Item [Name] may have been forgotten, please check."

[0073] Expiration Warning: For items categorized as "food," users can enter the expiration date during registration. The system will send reminders 3 days, 1 day, and on the day of the expiration date.

[0074] Electronic fence alarm: Users can draw circular or polygonal security zones (such as "home" or "office") on an electronic map. The cloud platform monitors the tag's location in real time, and an alarm is immediately pushed after the location coordinates exceed the fence boundary and are confirmed by three consecutive checks. For confidential documents or valuable items, alarms can also be linked to SMS or telephone calls.

[0075] Low battery alert: When the tag reports that the remaining battery is below 15%, the app will send a "Please replace the battery" notification and provide a purchase link.

[0076] Regular inventory checks: On the 1st of each month, the system will send an inventory reminder, requiring users to scan a QR code to confirm that the items are still there.

[0077] All of the above rules can be enabled / disabled in the app, the threshold can be adjusted by the user, and user-defined trigger rules are also supported.

[0078] Those skilled in the art should understand that the specific values ​​described in the above embodiments (such as overlap ratios of 30%, 40%, and 50%, antenna sizes of 25mm and 8mm, etc.) are merely preferred examples and not limitations on the present invention. These values ​​can be reasonably adjusted according to the actual frequency band and manufacturing process. For example: Using conductive silver paste printing instead of copper foil etching can reduce manufacturing costs, but the printing thickness needs to be controlled to ensure conductivity.

[0079] This invention can also be implemented using different models of UWB / Bluetooth chips (such as Qorvo DW1000 series, Nordic nRF5340), only requiring corresponding adjustments to the matching circuit.

[0080] Other high permeability materials (such as manganese-zinc ferrite or nickel-zinc ferrite) can be used for the isolation layer, as long as the relative permeability μ' ≥ 100 at a frequency of 1 GHz is met.

[0081] For scenarios where anti-metal protection is not required, the L6 isolation layer can be omitted, further reducing the thickness to 2.0mm.

[0082] Furthermore, the penalty coefficient in the positioning algorithm can be adaptively adjusted according to the actual environment. For example, the non-line-of-sight degree can be predicted by a machine learning model, and its value can be dynamically changed (as long as it is within the range of 2-15). In this embodiment, a penalty coefficient of 10 is only a preferred value.

[0083] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An integrated composite tag for tracking items, affixed to the item to be tracked, characterized in that, The tag comprises, stacked from top to bottom from the side furthest from the item to be tracked to the side closest to the item: a QR code printing layer, on which a QR code pattern containing three square-shaped positioning symbols is printed; a transparent protective layer; an antenna layer, on which an ultra-wideband (UWB) / Bluetooth dual-band antenna is formed, the antenna including a main radiator and at least one parasitic coupling stub; a chip layer, on which a UWB and Bluetooth dual-mode transceiver chip is disposed; a battery layer for powering the chip layer; an isolation layer for suppressing eddy currents generated when the tag is pasted on a metal surface and electromagnetic coupling between the antenna and the battery; and an adhesive layer for adhesion. In particular, the overlapping area of ​​the QR code printing layer's gui-shaped positioning pattern with the parasitic coupling stubs of the antenna layer in the vertical projection accounts for 30% to 50% of the total area of ​​the parasitic coupling stubs, and the black data module area of ​​the QR code does not overlap with the main radiator in the vertical projection, with a minimum spacing of ≥0.2mm.

2. The integrated composite label according to claim 1, characterized in that, The overlapping area ratio is 40%.

3. The integrated composite label according to claim 1, characterized in that, The transparent protective layer is a UV-cured hard coating with a thickness of 0.05 mm, a hardness of 2H, and a light transmittance of ≥92%, located between the QR code printing layer and the antenna layer.

4. The integrated composite label according to claim 1, characterized in that, The isolation layer is a ferrite absorber with a relative permeability μ'≥100 (at 1GHz frequency) and a thickness of 0.5mm; the battery layer is a flexible lithium ceramic battery with a capacity ≥40mAh; the total thickness of the tag does not exceed 2.8mm.

5. An item tracking and management system comprising the integrated composite tag as described in any one of claims 1 to 4, characterized in that, Also includes: The user terminal has an item management application installed, which is used to scan QR codes, receive location information, and report crowdsourced locations; The cloud platform server stores a one-to-one logical binding relationship between the unique QR code ID corresponding to the QR code printing layer and the unique hardware ID in the chip layer; The positioning base station is used to receive the UWB and Bluetooth positioning signals emitted by the integrated composite tag and report them to the cloud platform server; The user terminal and the positioning base station are respectively bidirectionally connected to the cloud platform server.

6. The item tracking management system according to claim 5, characterized in that, The cloud platform server also includes a multi-source fusion positioning module, which uses extended Kalman filtering to fuse UWB ranging, Bluetooth angle of arrival (AoA), and inertial data collected by the user terminal inertial measurement unit (IMU); and includes a non-line-of-sight error detection unit, which multiplies the covariance of non-line-of-sight observations by a penalty coefficient of 2-15.

7. The item tracking management system according to claim 5, characterized in that, It also includes an adaptive strategy switching module, which automatically selects the working mode based on the tag's remaining battery power and signal status: when the remaining battery power is ≥15% and the signal is normal, the active positioning mode is adopted; when the remaining battery power is <15% or the tag signal is not received for N consecutive times (N is a preset integer greater than or equal to 2) and is determined to be out of contact, the tag sends an alarm message to the cloud platform server and enters deep sleep, and the system switches to the crowdsourced location reporting assisted item finding mode; the active positioning mode includes UWB / Bluetooth fusion positioning, direct connection positioning of user terminals supporting UWB and Bluetooth base station positioning.

8. The item tracking management system according to claim 7, characterized in that, The crowdsourced location reporting assisted item finding mode works as follows: any user with the item management application installed scans the QR code or backup QR code on the QR code printing layer. After the user confirms and agrees, the user terminal automatically reports the current geographical location and scanning time to the cloud platform. The cloud platform records the location as the last sighting point of the item and pushes a notification to the registered user of the item.

9. The item tracking management system according to claim 5, characterized in that, It also includes a QR code printing device, which is connected to the user terminal to output a backup QR code that is paired with the integrated composite label.

10. The item tracking management system according to claim 5, characterized in that, It also includes a rules engine that supports geofence alarms, overstay reminders, low battery reminders, expiration alerts, periodic inventory checks, and user-defined trigger rules.

Images