An RFID tag quantity estimation method, device, terminal and medium

Abstract

The application discloses an RFID tag quantity estimation method and device, a terminal and a medium. The estimation method provided by the application comprises the following steps: collecting time-continuous signals received by a read-write device, extracting received signal strength features and response time delay features, constructing a basic credibility function to generate a trust vector, and determining the tag quantity through dynamic threshold judgment after fusing the two types of trust vectors. According to the scheme, the time difference features can compensate for the distortion of the strength features when the signals are affected by the multipath effect by combining the strength and time two-dimensional features. Then, the conflict evidence between the features is effectively processed based on the evidence theory fusion mechanism, the processing capacity of the conflict evidence is improved, and the trust degree distribution difference in different environments is adapted through the dynamic threshold judgment strategy. Through the above processing procedure, the RFID tag quantity estimation accuracy can be kept high even when there is channel fading or noise interference.

Description

Technical Field

[0001] This application relates to the field of Internet of Things (IoT) technology, and in particular to a method, apparatus, terminal, and medium for estimating the number of RFID tags. Background Technology

[0002] With the widespread application of IoT technology, Ultra-High Frequency (UHF) Radio Frequency Identification (RFID) technology, as a non-contact, multi-tag identification technology, is becoming an important support for intelligent sensing and information collection. However, when multiple tags respond to the reader's interrogation signal simultaneously, their returned response signals will overlap, causing channel collisions. This significantly reduces the success rate of identification and system throughput, and increases the reader's power consumption. To solve the collision problem, the industry currently widely adopts the Dynamic Framed Slotted ALOHA (DFSA) protocol. In this protocol, the system balances identification latency and tag retransmission counts by reasonably setting the length of the interrogation frame. Its core premise is the accurate estimation of the number of tags within the frame.

[0003] Current methods for estimating the number of tags mainly include: statistical estimation, physical layer response signal-based estimation, and machine learning-based estimation. However, most existing estimation methods rely on a certain physical layer feature and are sensitive to channel conditions. Under complex conditions such as tag response overlap, channel fading, or noise interference, there is a technical problem that the estimation accuracy will decrease significantly. Summary of the Invention

[0004] This application provides an RFID tag quantity estimation method, apparatus, terminal, and medium to solve the technical problem of low accuracy in existing RFID tag quantity estimation methods.

[0005] To address the aforementioned technical problems, the first aspect of this application provides a method for estimating the number of RFID tags, comprising:

[0006] The RFID tag-based reading and writing device collects the time-continuous signal received by the reading and writing device;

[0007] Based on the time-continuous signal, the characteristics of the received signal strength and the response delay of the time-continuous signal are extracted.

[0008] Based on the received signal strength characteristics and the response delay characteristics, and combined with a preset basic confidence function, the corresponding received signal strength confidence vector and delay confidence vector are obtained.

[0009] The received strength trust vector and the delay trust vector are fused using the Dempster-Shafer combination method to obtain a fused trust value;

[0010] The estimated number of RFID tags is determined based on the comparison between the fused trust value and the dynamic decision threshold.

[0011] Preferably, the formula for calculating the received strength trust vector is as follows:

[0012]

[0013]

[0014] In the formula, The received strength trust vector, For the k-th tag component in the received strength trust vector, The RSSI feature distribution for the number of tags k. The set of candidate label counts.

[0015] Preferably, the response delay feature is specifically:

[0016]

[0017]

[0018] In the formula, Let be the delay trust vector. For the k-th label component in the delay trust vector, Let k be the delayed feature distribution of the number of labels. The set of candidate label counts.

[0019] Preferably, the formula for calculating the fusion trust value is:

[0020]

[0021] In the formula, The fusion trust value corresponding to the number of tags k. For the k-th tag component in the received strength trust vector, For the k-th label component in the delay trust vector, The set of candidate label counts.

[0022] Preferably, determining the estimated number of RFID tags based on the comparison result between the fused trust value and the dynamic decision threshold includes:

[0023] Based on the comparison result between the fusion trust value and the dynamic decision threshold, if the fusion trust value is not less than the dynamic decision threshold, the estimated number of RFID tags is determined based on the number of tags corresponding to the fusion trust value. If the fusion trust value is less than the dynamic decision threshold, the search continues downward based on the number of tags corresponding to the fusion trust value until the minimum number of tags is found.

[0024] Preferably, the method for determining the dynamic decision threshold includes:

[0025] Based on preset tag sample data, the trust distribution of different tag numbers is statistically analyzed, and the dynamic decision threshold corresponding to each tag number is determined according to the trust distribution.

[0026] Preferably, the maximum number of tags in the candidate tag set is 5.

[0027] Meanwhile, a second aspect of this application provides an RFID tag quantity estimation device, comprising:

[0028] The tag reading and writing signal acquisition unit is used in RFID tag-based reading and writing devices to acquire the time-continuous signal received by the reading and writing devices.

[0029] The signal feature extraction unit is used to extract the features of the observed variables based on the time-continuous signal to obtain the received signal strength features and response delay features of the time-continuous signal.

[0030] The feature conversion unit is used to obtain the corresponding received signal strength trust vector and delay trust vector based on the received signal strength feature and the response delay feature, combined with a preset basic confidence function.

[0031] The trust vector fusion unit is used to fuse the received strength trust vector and the delay trust vector using the Dempster-Shafer combination method to obtain a fused trust value.

[0032] The threshold decision unit is used to determine the estimated number of RFID tags based on the comparison result between the fused trust value and the dynamic decision threshold.

[0033] A third aspect of this application provides an RFID tag quantity estimation terminal, comprising: a memory and a processor;

[0034] The memory is used to store program code that corresponds to the RFID tag quantity estimation method provided in the first aspect of this application.

[0035] The processor is used to read and execute the program code to implement the RFID tag quantity estimation method.

[0036] The fourth aspect of this application provides a computer-readable storage medium storing program code that is read and executed by a processor to implement the RFID tag quantity estimation method provided in the first aspect of this application.

[0037] As can be seen from the above technical solutions, this application has the following advantages:

[0038] This application provides a method for estimating the number of RFID tags based on the fusion of received signal strength and response delay features. The method includes acquiring a time-continuous signal received by a reader / writer, extracting received signal strength features and response delay features, constructing a basic confidence function to generate a trust vector, and fusing the two trust vectors to determine the number of tags through dynamic threshold decision. This scheme, by combining strength and time-dimensional features, compensates for the distortion of strength features when the signal is affected by multipath effects. Furthermore, it effectively handles conflicting evidence between features based on an evidence theory fusion mechanism, improving the ability to process conflicting evidence. Finally, the dynamic threshold decision strategy adapts to the differences in trust distribution under different environments. Through this processing flow, it achieves a high tag number estimation accuracy even in the presence of channel fading or noise interference. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a flowchart illustrating an embodiment of an RFID tag quantity estimation method provided in this application.

[0041] Figure 2 This is a schematic diagram of an embodiment of an RFID tag quantity estimation device provided in this application.

[0042] Figure 3 This is a schematic diagram of the structure of an RFID tag quantity estimation terminal embodiment provided in this application. Detailed Implementation

[0043] In existing technologies, UHF RFID technology faces the problem of signal collisions caused by simultaneous responses from multiple tags. The Dynamic Frame Slotted ALOHA protocol relies on accurate tag count estimation to optimize frame length, but existing methods are mostly based on a single physical layer feature. When channel fading or noise interference exists, a single feature is easily affected by the environment, leading to a significant decrease in estimation accuracy. For example, in densely deployed warehouse shelving scenarios, the reflection effect of metal shelves on electromagnetic waves exacerbates signal strength fluctuations, while multipath effects increase the differences in tag response times. This makes traditional estimation methods based on received signal strength prone to misjudgment, and most algorithms fail when the number of colliding tags exceeds three, indicating a low upper limit that cannot meet the needs of high-density tag scenarios in practical applications, such as logistics sorting, warehouse inventory, and large conference check-in.

[0044] In view of this, embodiments of this application provide an RFID tag quantity estimation method, apparatus, terminal, and medium to solve the technical problem of low accuracy in existing RFID tag quantity estimation methods.

[0045] To make the inventive objectives, features, and advantages of this application more apparent and understandable, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] First, a detailed description of an embodiment of an RFID tag quantity estimation method provided in this application is as follows:

[0047] Please see Figure 1 This application provides an RFID tag quantity estimation method, which includes:

[0048] Step 101: The RFID tag-based reading and writing device collects the time-continuous signal received by the reading and writing device;

[0049] Step 102: Extract the features of the observed variables based on the time-continuous signal to obtain the received signal strength features and response delay features of the time-continuous signal;

[0050] Step 103: Based on the received signal strength characteristics and response delay characteristics, and combined with the preset basic confidence function, obtain the corresponding received signal strength confidence vector and delay confidence vector;

[0051] Step 104: The received strength trust vector and the delay trust vector are fused using the Dempster-Shafer combination method to obtain the fused trust value;

[0052] Step 105: Determine the estimated number of RFID tags based on the comparison results between the fusion trust value and the dynamic decision threshold.

[0053] It should be noted that the scheme in this embodiment is based on a dual-feature fusion method for estimating the number of radio frequency identification tags, which includes collecting the time-continuous signal received by the reading and writing device, extracting the received signal strength features and response delay features, constructing a basic confidence function to generate a trust vector, and determining the number of tags by fusing the two types of trust vectors and making a dynamic threshold decision.

[0054] Among them, the time-continuous signal refers to the radio frequency signal waveform continuously received by the reader within a fixed time window, which can be acquired through high-speed sampling circuits. Taking an RFID tag system based on the EPC Gen2 protocol as an example, the slotted ALOHA mechanism requires tags to respond probabilistically within a specified time slot. Multiple tags may simultaneously select the same time slot, resulting in signal overlap, also known as "collision slots." This type of physical layer superimposed signal makes it difficult for traditional methods to accurately decode the tag ID, and it is also difficult to directly know how many tags responded to the time slot. Let the time-continuous signal received by the reader be:

[0055]

[0056] In the formula, This represents the number of tags in the current response. For the i-th tag, the Manchester or FM0 modulated signal. For the propagation delay of multipath components, denoted as the complex fading coefficient of the corresponding component. For the carrier frequency of the RFID system, It is Gaussian white noise, which includes thermal noise, adjacent interference, aliasing noise, etc.

[0057] In practice, the channel can be simplified to frequency-flat fading (common in short-range UHF scenarios), so the above equation can be approximately simplified to:

[0058]

[0059] Here, H represents the equivalent complex gain of the tag-reader channel. This model clearly expresses the theoretical basis of the multi-tag superposition response.

[0060] Received signal strength characteristics reflect the degree of signal energy attenuation, specifically represented by RSSI. Response delay characteristics characterize the difference in propagation time of tag response signals, which can be measured by detecting the time interval between the rising edges of adjacent signals. The basic confidence function is used to transform physical features into confidence in the number of candidate tags, and a Gaussian distribution model can be used to establish a probabilistic mapping relationship between features and the number of tags. The confidence vector consists of the confidence corresponding to each number of candidate tags, and its dimension is equal to the preset maximum number of tags. The Dempster-Shafer combination method eliminates evidence conflicts through orthogonalization and normalization processing; in practice, it uses a joint calculation of the confidence product and the conflict coefficient. The dynamic decision threshold is determined based on the confidence distribution of different tag numbers in historical data and can be set as the lower limit of the confidence interval of the corresponding tag number confidence distribution.

[0061] Specifically, the reader first acquires a radio frequency signal containing the superimposed responses of multiple tags, and then separates the time-continuous signal through a signal processing module. The mean signal intensity and the time difference between adjacent pulses are calculated to form observations characterizing the spatial attenuation and temporal differences of the signal. Next, based on these two observations, a preliminary confidence vector for the number of candidate tags is output. Then, based on Dempster-Shafer theory, the two confidence vectors are combined to eliminate evidence conflicts caused by environmental interference. Finally, by comparing the fused confidence value with a dynamic threshold, the maximum possible number of tags that meet the conditions is determined.

[0062] This method combines strength and time-based features, allowing the time difference feature to compensate for the distortion of the strength feature when the signal is affected by multipath effects. The evidence theory fusion mechanism effectively handles conflicting evidence between features, avoiding misjudgments caused by single feature anomalies. Through these technical solutions, the problem of unstable single feature estimation in complex environments is solved. The dual-feature fusion mechanism enhances the completeness of feature representation, the evidence theory combination rules improve the ability to handle conflicting evidence, and the dynamic threshold decision strategy adapts to the differences in trust distribution under different environments. This allows for maintaining high accuracy in tag number estimation even in the presence of channel fading or noise interference.

[0063] More specifically, after acquiring the time-continuous signal from the read / write device, the next step is to extract the features of the observed variables based on the time-continuous signal to obtain the received signal strength features and response delay features of the time-continuous signal.

[0064] Specifically, after obtaining the received signal strength features and response delay features during the signal feature extraction stage, these two features are mapped to corresponding trust vectors. The two trust vectors are then fused using the Dempster-Shafer combination rule. This involves multiplying the trust components for each candidate tag number k and normalizing the product of all candidate tag numbers. This fusion process eliminates misjudgments caused by single features under noise interference. For example, when the received signal strength fluctuates abnormally due to multipath effects, the supplementary evidence provided by the delay features maintains the stability of the overall estimation.

[0065] More specifically, this application utilizes two types of observational features—RSSI and response delay—to construct independent sources of belief. At the physical layer, when multiple tags transmit concurrently, there is a statistical correlation between the total signal strength measured at the receiver and the number of tags. Assuming that the transmission power of each tag is approximately equal, the RSSI can be modeled as:

[0066]

[0067] in, Noise power; For the number of tags; The average power per tag can be expressed in the following form:

[0068]

[0069] in, In order to receive signals, This is the length of the integration time window. This definition represents the average power of the received signal per unit time, commonly used to characterize signal strength, and its unit is 1 Ω. In practice, it is often calculated using a sliding window or periodic averaging method.

[0070] Due to Rayleigh fading and the distance difference between tags, the RSSI distributions overlap significantly. Therefore, it cannot be directly mapped to the tag count and must be introduced into belief theory modeling as a source of uncertain evidence. The propagation path from tag to reader involves distance and reflection differences. Even with simultaneous responses in the same time slot, the channel propagation delay will cause a slight misalignment of the signal start point. These delay differences are statistically modeled as follows:

[0071]

[0072] In the formula, The symbol interval (tens of microseconds) for tag modulation. The starting point of the i-th label is delayed.

[0073] The average delay between each pair of signals is statistically extracted, i.e., the degree of dispersion. The specific calculation formula is as follows: (This is used as a characteristic of response delay.)

[0074]

[0075] Furthermore, after obtaining the received signal strength characteristics and response delay characteristics, the corresponding received signal strength trust vector and delay trust vector are obtained by combining them with the preset basic confidence function.

[0076] It should be noted that this embodiment sets up a framework for tag number recognition:

[0077]

[0078] Define the basic confidence level generated by each observation source as a function:

[0079]

[0080] This scheme constructs a basic confidence function for each observed variable (RSSI, Delay) to obtain the corresponding confidence vector, as follows:

[0081] Receiver Strength Trust Vector: Treating RSSI features as evidence variables, we establish the RSSI distribution corresponding to each number of labels k in the training dataset. It can be fitted using kernel density estimation (KDE):

[0082]

[0083] In the formula, To receive the strength trust vector, To receive the k-th tag component in the strength trust vector, The RSSI feature distribution for the number of tags k. The set of candidate label counts.

[0084] Based on the above calculation formula, the set of length and number of candidate tags can be obtained. Corresponding trust vector .

[0085] Delay Trust Vector: A feature of the delay in observations Let its distribution under different numbers of labels be as follows: ,but

[0086]

[0087] In the formula, For the k-th label component in the time-delay trust vector, Let k be the delay feature distribution of the number of labels k.

[0088] Obtain the delayed trust vector , Both are single-point basic confidence functions, meaning that a specific confidence level is assigned to each number of tags k, which meets the requirements of ease of engineering implementation.

[0089] Next, the classic Dempster-Shafer combination is used to fuse the confidence levels of evidence from both sources, as shown in the following expression:

[0090]

[0091] The conflict coefficient K is:

[0092]

[0093] In this embodiment, all basic confidence functions are single-point sets (i.e., Therefore, the fusion rule is simplified to:

[0094]

[0095] This method ensures that the combined BBA still maintains a single-point form, has low computational complexity, and is suitable for embedded implementation.

[0096] This scheme maintains high estimation accuracy even under deteriorating channel conditions by fusing trust vectors from two independent physical features. For example, in scenarios where tag response signals partially overlap, a single feature may not accurately distinguish more than three tags, but the joint probability distribution of the two features enhances distinguishability. This technical solution addresses the insufficient anti-interference capability of existing methods that rely on a single feature. By fusing dual trust evidence of received signal strength and latency, the scheme effectively suppresses the impact of single feature anomalies on the estimation results when tag response signals collide or environmental noise is present, thereby improving the accuracy of tag number estimation under complex channel conditions.

[0097] Furthermore, this application also proposes a detailed process for determining the estimated number of RFID tags based on the comparison result between the fusion trust value and the dynamic decision threshold. Specifically, it includes: if the fusion trust value is not less than the dynamic decision threshold, then the estimated value is determined based on the corresponding number of tags; if the fusion trust value is less than the dynamic decision threshold, then the search continues downward based on the current number of tags until the minimum number of tags is reached.

[0098] Among them, the dynamic decision threshold refers to the adaptive discrimination threshold derived from the statistical analysis of the label sample data. Specifically, it can be generated by confidence interval calculation or probability density function fitting, and is used to dynamically adjust the search termination condition.

[0099] Specifically, when the fusion trust value exceeds the dynamic decision threshold, it indicates that the current number of labels has high credibility and can be directly output as the estimation result. If the fusion trust value does not reach the threshold, it means that the current number of labels is not credible enough, and the search needs to be performed downwards along the candidate label set, for example, starting from the maximum number of labels of 5 and decreasing gradually until the first number of labels that meets the threshold condition is found or the preset minimum number of labels is reached. During the search process, the dynamic decision threshold can be updated in real time based on the historical data distribution, for example, automatically lowering the threshold in a low signal-to-noise ratio environment to adapt to channel conditions.

[0100] For example, setting a dynamic threshold for the number of tags k. The decision logic is as follows:

[0101] If it exists , Generally, the maximum value is used as the initial value, satisfying:

[0102]

[0103] The estimated number of labels is: ;

[0104] If no such condition is met, continue searching downwards k-1 until the minimum number of labels is reached (i.e., ...). =1).

[0105] This solution combines dynamic thresholding with iterative search to automatically adjust the discrimination criteria in complex signal environments. For example, when tag responses overlap, multi-level verification avoids single-judgment errors, while a downward search strategy effectively suppresses error accumulation. Through these technical solutions, the problem of tag number estimation deviation caused by single-judgment in existing technologies is solved. In scenarios with channel fading or noise interference, the collaborative mechanism of credibility verification and iterative search reduces the probability of misjudgment, while ensuring that the number of searches within the candidate tag number range does not exceed a preset maximum value, significantly improving the estimation stability in dense tag environments.

[0106] Furthermore, the method for determining the dynamic decision threshold includes statistically analyzing the trust distribution of different number of tags based on preset tag sample data, so as to determine the dynamic decision threshold corresponding to each number of tags according to the trust distribution.

[0107] The tag sample data refers to the pre-collected signal feature data set corresponding to different numbers of tags. Specifically, it can be implemented using received signal strength and response delay feature sample data under different tag numbers in laboratory environments or real-world application scenarios, used to establish the mapping relationship between different tag numbers and signal features. The trust distribution refers to the probability distribution characteristics of the fused trust value corresponding to different tag numbers. Specifically, it can be achieved by statistically analyzing the frequency or probability density of the fused trust value corresponding to each tag number in the tag sample data, reflecting the distribution pattern of trust values ​​under different tag numbers. The dynamic decision threshold is a critical value that is adaptively adjusted according to the trust distribution of different tag numbers. Specifically, it can be determined using probability quantiles or statistical confidence intervals, for example, using the quantiles in the trust distribution where the cumulative probability reaches a preset confidence level as the threshold, or calculated using the following formula: ,in, Let k be the mean trust level of the number of tags. Let k be the standard deviation of the confidence level for the number of labels.

[0108] Specifically, when determining the dynamic decision threshold, it is first necessary to acquire tag sample data, such as pre-collecting received signal strength and response delay characteristics data for tags ranging from 1 to 5 in a laboratory environment. For each tag number, the corresponding fusion trust value is calculated and its distribution characteristics are statistically analyzed, for example, by obtaining the probability density function of the trust value through kernel density estimation. Based on the characteristics of the trust distribution, an appropriate threshold determination rule is selected, such as using the quantile where the cumulative probability in the trust distribution reaches 95% as the dynamic decision threshold corresponding to that tag number. In this way, each tag number corresponds to a statistically validated threshold, enabling the decision conditions to be dynamically adjusted according to the trust distribution of the current tag number during real-time estimation. By constructing a Basic BeliefAssignment (BBA) model for tag numbers k=0~5 and introducing a dynamic threshold discrimination mechanism, it is still possible to accurately estimate up to 5 tags even when SNR≥6 dB, thus improving system capacity.

[0109] This scheme, by statistically analyzing the trust distribution characteristics under different tag numbers, can dynamically adjust the decision threshold based on actual data, effectively solving the problem of poor adaptability of fixed thresholds under complex channel conditions, and improving the scientific rigor and reliability of the decision process. Through the above technical solution, this application can dynamically adjust the decision threshold according to the trust distribution characteristics under different tag numbers, making the tag number estimation process more closely match the changing patterns of actual signal characteristics, effectively reducing the impact of noise interference and channel fading on the decision results, and improving the accuracy and stability of tag number estimation in complex environments.

[0110] The above is a detailed description of an embodiment of an RFID tag quantity estimation method provided in this application. The following is a detailed description of an embodiment of an RFID tag quantity estimation device provided in this application.

[0111] Please see Figure 2 This application further provides an RFID tag quantity estimation device, comprising:

[0112] The tag reading and writing signal acquisition unit 201 is used in RFID tag-based reading and writing devices to acquire the time-continuous signal received by the reading and writing devices.

[0113] The signal feature extraction unit 202 is used to extract the features of the observed variables based on the time-continuous signal to obtain the received signal strength features and response delay features of the time-continuous signal.

[0114] The feature conversion unit 203 is used to obtain the corresponding received signal strength trust vector and delay trust vector based on the received signal strength characteristics and response delay characteristics, combined with a preset basic confidence function.

[0115] Trust vector fusion unit 204 is used to fuse the received strength trust vector and the delay trust vector through Dempster-Shafer combination to obtain a fused trust value;

[0116] The threshold decision unit 205 is used to determine the estimated number of RFID tags based on the comparison result between the fused trust value and the dynamic decision threshold.

[0117] This device integrates dual-modal feature extraction with evidence theory to construct a redundant decision mechanism, enabling compensation through another feature even when one feature is interfered with. Furthermore, existing devices often employ fixed threshold decisions, which are ill-suited to dynamically changing tag densities. This device introduces a dynamic threshold adjustment strategy based on sample statistics, automatically optimizing the classification boundary according to the actual environment. Through these technical solutions, this application solves the problem of insufficient anti-interference capability caused by reliance on a single feature in existing technologies, maintaining stable estimation accuracy even when tag responses overlap or channel conditions deteriorate. Simultaneously, the dynamic threshold decision mechanism reduces dependence on preset parameters, improving the device's adaptability to different application scenarios. This device can be directly integrated into the baseband processing module of an RFID reader, providing reliable tag quantity input for the dynamic frame slot ALOHA protocol, thereby optimizing the frame length adjustment strategy and improving system throughput.

[0118] like Figure 3As shown, this application further proposes an RFID tag quantity estimation terminal, including a memory and a processor; the memory 33 is used to store program code, which corresponds to the RFID tag quantity estimation method provided in the above embodiments; the processor 31 is used to read and execute the program code to implement the estimation method, wherein the memory 33 and the processor 31 can be connected through a communication bus 34.

[0119] Here, memory 33 refers to the hardware module used to store program code, which can be implemented using non-volatile storage media such as ROM or flash memory to ensure that the program code can be completely preserved after power failure. The processor is the arithmetic unit that executes the program code, which can be implemented using an embedded microprocessor or digital signal processing chip, performing signal feature extraction and trust vector fusion through high-speed computation. The program code is the set of instructions containing each step of the RFID tag quantity estimation method, which can be written in C or assembly language and compiled into an executable file to guide the processor to complete signal acquisition, feature extraction, and fusion decision according to preset logic.

[0120] Specifically, the RFID tag quantity estimation terminal stores pre-written program code in its memory. This code includes algorithmic implementations for steps such as signal acquisition, received signal strength feature extraction, response delay feature extraction, trust vector generation and fusion, and dynamic threshold decision. During operation, the processor sequentially executes the instructions in the program code, controlling the reader / writer to acquire continuously occurring signals, extract signal strength and response delay features, and generate received signal strength trust vectors and delay trust vectors based on a preset confidence function. Subsequently, the processor calls the Dempster-Shafer combination rule to fuse the two types of trust vectors, obtaining a fused trust value. Finally, the estimated tag quantity is determined by comparing the results with a dynamic threshold.

[0121] In some specific implementations, the memory capacity can be configured to meet the program code storage requirements, for example, by using a 256KB flash memory chip; the processor frequency can be set to 100MHz or higher, for example, by using an ARM Cortex-M7 core, to ensure real-time processing of multi-tag signals. The program code can be remotely updated via serial port or wireless communication module to adapt to parameter adjustment requirements in different scenarios.

[0122] Furthermore, this application also proposes a computer-readable storage medium storing program code, which is read and executed by a processor to implement an RFID tag quantity estimation method.

[0123] Computer-readable storage media refers to the physical carrier used to store program code, which can be implemented using flash memory, hard disk, or solid-state storage. Its function is to persistently store program code and ensure portability. Program code refers to a computer program containing executable instructions, which can be implemented as a binary file compiled from machine language or a high-level language. Its function is to guide the processor in executing the various steps of the RFID tag quantity estimation method. The processor is the arithmetic unit capable of reading and executing program code, which can be implemented using a central processing unit or an embedded microcontroller. Its function is to complete operations such as signal acquisition, feature extraction, trust vector fusion, and threshold determination by running the program code.

[0124] Specifically, when the program code is read and executed by the processor, it first acquires a time-continuous signal through a read / write device, then extracts the received signal strength features and response delay features, and generates corresponding trust vectors based on a basic trust function. Next, the two types of trust vectors are fused using the Dempster-Shafer combination rule to obtain a fused trust value. Finally, the estimated number of tags is determined based on the comparison between the fused trust value and the dynamic decision threshold. Throughout this process, the execution logic of the program code ensures the joint use of received signal strength and response delay features, as well as the compensation and fusion of redundant evidence.

[0125] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the terminals, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0126] In the several embodiments provided in this application, it should be understood that the disclosed terminals, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0127] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the application described herein can be implemented, for example, in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0128] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0129] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0130] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0131] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0132] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for estimating the number of RFID tags, characterized in that, include: The RFID tag-based reading and writing device collects the time-continuous signal received by the reading and writing device; Based on the time-continuous signal, the characteristics of the received signal strength and the response delay of the time-continuous signal are extracted. Based on the received signal strength characteristics and the response delay characteristics, and combined with a preset basic confidence function, the corresponding received signal strength confidence vector and delay confidence vector are obtained. The received strength trust vector and the delay trust vector are fused using the Dempster-Shafer combination method to obtain a fused trust value; Based on the comparison results between the fused trust value and the dynamic decision threshold, the estimated number of RFID tags is determined; The formula for calculating the received strength trust vector is as follows: In the formula, The received strength trust vector, For the k-th tag component in the received strength trust vector, The RSSI feature distribution for the number of tags k. The set of candidate label counts; The specific response delay characteristic is as follows: In the formula, Let be the delay trust vector. For the k-th label component in the delay trust vector, Let k be the delayed feature distribution of the number of labels. The set of candidate label counts; The formula for calculating the fusion trust value is: In the formula, The fusion trust value corresponding to the number of tags k. For the k-th tag component in the received strength trust vector, For the k-th label component in the delay trust vector, The set of candidate label counts.

2. The RFID tag quantity estimation method according to claim 1, characterized in that, The step of determining the estimated number of RFID tags based on the comparison result between the fused trust value and the dynamic decision threshold includes: Based on the comparison result between the fusion trust value and the dynamic decision threshold, if the fusion trust value is not less than the dynamic decision threshold, the estimated number of RFID tags is determined based on the number of tags corresponding to the fusion trust value. If the fusion trust value is less than the dynamic decision threshold, the search continues downward based on the number of tags corresponding to the fusion trust value until the minimum number of tags is found.

3. The RFID tag quantity estimation method according to claim 1, characterized in that, The dynamic decision threshold is determined in the following ways: Based on preset tag sample data, the trust distribution of different tag numbers is statistically analyzed, and the dynamic decision threshold corresponding to each tag number is determined according to the trust distribution.

4. The RFID tag quantity estimation method according to claim 1, characterized in that, The maximum number of tags in the candidate tag set is 5.

5. An RFID tag quantity estimation device, characterized in that, include: The tag reading and writing signal acquisition unit is used in RFID tag-based reading and writing devices to acquire the time-continuous signal received by the reading and writing devices. The signal feature extraction unit is used to extract the features of the observed variables based on the time-continuous signal to obtain the received signal strength features and response delay features of the time-continuous signal. The feature conversion unit is used to obtain the corresponding received signal strength trust vector and delay trust vector based on the received signal strength feature and the response delay feature, combined with a preset basic confidence function. The trust vector fusion unit is used to fuse the received strength trust vector and the delay trust vector using the Dempster-Shafer combination method to obtain a fused trust value. The threshold decision unit is used to determine the estimated number of RFID tags based on the comparison result between the fused trust value and the dynamic decision threshold. The formula for calculating the received strength trust vector is as follows: In the formula, The received strength trust vector, For the k-th tag component in the received strength trust vector, The RSSI feature distribution for the number of tags k. The set of candidate label counts; The specific response delay characteristic is as follows: In the formula, Let be the delay trust vector. For the k-th label component in the delay trust vector, Let k be the delayed feature distribution of the number of labels. The set of candidate label counts; The formula for calculating the fusion trust value is: In the formula, The fusion trust value corresponding to the number of tags k. For the k-th tag component in the received strength trust vector, For the k-th label component in the delay trust vector, The set of candidate label counts.

6. An RFID tag quantity estimation terminal, characterized in that, include: Memory and processor; The memory is used to store program code, which corresponds to the RFID tag quantity estimation method as described in any one of claims 1 to 4; The processor is used to read and execute the program code to implement the RFID tag quantity estimation method.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that is read and executed by a processor to implement the RFID tag quantity estimation method as described in any one of claims 1 to 4.

Images