Efficient checking system and method based on RFID technology

Abstract

The invention relates to the technical field of RFID checking, and particularly discloses an efficient checking system and method based on the RFID technology, and the method comprises the steps: dividing an RFID checking region into a plurality of logic blocks, carrying out the RFID tag scanning of each logic block through an RFID reader-writer, obtaining a preliminary feedback signal of each logic block, and then carrying out the RFID tag scanning of each logic block, and a data processing technology based on deep learning is introduced to carry out global correlation analysis on the preliminary feedback signal of each logic block, on the basis of the global correlation analysis, the transmitting power of the RFID reader-writer in each logic block is dynamically adjusted to carry out scanning again, and an inventory inventory list is generated according to a scanning result. Through the mode, the parameter configuration of the RFID reader-writer can be adaptively optimized, so that the reading rate and the accuracy of the RFID tag are improved, the phenomena of missed reading and excessive reading of the tag are reduced, and the accuracy and the efficiency of inventory checking are further improved.

Description

Technical Field

[0001] The present application relates to the field of RFID inventory technology, and more specifically, to an efficient inventory system and method based on RFID technology. Background Art

[0002] In many fields such as logistics warehousing, retail management, and asset tracking, inventory counting is a key link to ensure that the number of items matches the records. Traditional inventory counting methods rely on manual verification one by one, which is not only time-consuming and labor-intensive, but also susceptible to human errors, resulting in low inventory efficiency and difficulty in ensuring accuracy. With the rapid development of information technology, especially the widespread application of wireless radio frequency identification (RFID) technology, automated and intelligent inventory counting solutions have gradually become the goal pursued by the industry.

[0003] RFID (Radio Frequency Identification) technology allows contactless automatic identification of RFID tags attached to items through radio wave communication, greatly improving the speed and accuracy of data collection. However, although RFID technology provides the advantage of fast batch reading, it still faces many challenges in practical applications, especially in complex and changing environments (such as dense shelves, near metal obstacles), the propagation of RFID signals may be disturbed, resulting in a decrease in tag reading rate, the so-called "tag missed reading" or "multiple reading" phenomenon, which directly affects the accuracy and efficiency of inventory counting.

[0004] Therefore, an optimized and efficient inventory counting system and method based on RFID technology are expected. Summary of the invention

[0005] The present application provides an efficient inventory counting system and method based on RFID technology, which can adaptively optimize the parameter configuration of RFID readers, thereby improving the reading rate and accuracy of RFID tags, reducing tag missed reading and multiple reading phenomena, and further improving the accuracy and efficiency of inventory counting.

[0006] In the first aspect, an efficient inventory method based on RFID technology is provided, comprising: Step 1: Divide the RFID inventory area into multiple logical blocks; Step 2: Use an RFID reader to perform a preliminary scan on the RFID tags of the objects to be counted in each of the multiple logical blocks to obtain multiple logical block initial feedback signals, wherein the logical block initial feedback signals include transmission power, number of tags successfully read, signal strength and reading range; Step 3: Based on the preliminary feedback signals of the multiple logical blocks, dynamically adjust the transmission power of the RFID reader when rescanning the RFID tags of the objects to be counted in the respective logical blocks to obtain multiple optimized feedback signals of the logical blocks; Step 4: Generate an inventory count list based on the optimization feedback signals of the multiple logic blocks.

[0007] Second, an efficient inventory counting system based on RFID technology is provided, including: A logical block division module is used to divide the RFID inventory area into multiple logical blocks; A preliminary scanning module, used to use an RFID reader to perform preliminary scanning on the RFID tags of the objects to be counted in each of the multiple logical blocks to obtain multiple logical block preliminary feedback signals, wherein the logical block initial feedback signals include transmission power, number of tags successfully read, signal strength and reading range; A re-scanning module, configured to dynamically adjust the transmission power of the RFID reader when re-scanning the RFID tags of the objects to be counted in the respective logical blocks based on the preliminary feedback signals of the plurality of logical blocks, so as to obtain the optimized feedback signals of the plurality of logical blocks; The inventory count list generating module is used to generate an inventory count list based on the optimization feedback signals of the multiple logic blocks.

[0008] The present application provides an efficient inventory counting system and method based on RFID technology, which divides the RFID inventory counting area into multiple logical blocks, uses an RFID reader to scan the RFID tags of each logical block to obtain the preliminary feedback signal of each logical block, and then introduces a data processing technology based on deep learning to perform a global correlation analysis on the preliminary feedback signals of each logical block to reveal the essential propagation characteristics of the signal sent by the RFID reader in the inventory counting area under the current setting, and then based on this, according to the offset degree of the preliminary feedback signal of each logical block relative to the essential propagation characteristics of the signal, dynamically adjust the transmission power of the RFID reader in each logical block to scan again, and generate an inventory counting list based on the scanning results. In this way, the parameter configuration of the RFID reader can be adaptively optimized, thereby improving the reading rate and accuracy of the RFID tag, reducing the phenomenon of missed reading and multiple reading of tags, and thus improving the accuracy and efficiency of inventory counting. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0010] Figure 1 This is a schematic flow chart of an efficient inventory method based on RFID technology according to an embodiment of the present application.

[0011] Figure 2 This is a schematic flowchart of step 3 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0012] Figure 3 This is a data flow diagram of step 3 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0013] Figure 4 This is a schematic flowchart of step 33 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0014] Figure 5 This is a schematic flowchart of step 331 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0015] Figure 6 This is a schematic flowchart of step 332 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0016] Figure 7 This is a schematic flowchart of step 34 in the efficient inventory method based on RFID technology in an embodiment of the present application.

[0017] Figure 8 This is a schematic block diagram of an efficient inventory counting system based on RFID technology according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0019] In response to the above technical problems, the technical concept of this application is: divide the RFID inventory area into multiple logical blocks, use the RFID reader to scan the RFID tags of each logical block to obtain the preliminary feedback signal of each logical block, then introduce the data processing technology based on deep learning to perform global correlation analysis on the preliminary feedback signals of each logical block to reveal the essential propagation characteristics of the signal sent by the RFID reader in the inventory area under the current setting, and then based on this, according to the offset degree of the preliminary feedback signal of each logical block relative to the essential propagation characteristics of the signal, dynamically adjust the transmission power of the RFID reader in each logical block to scan again, and generate an inventory count list based on the scanning results. In this way, the parameter configuration of the RFID reader can be adaptively optimized, thereby improving the reading rate and accuracy of the RFID tag, reducing the phenomenon of missed reading and multiple reading of tags, and thus improving the accuracy and efficiency of inventory counting.

[0020] Based on this, in the technical solution of this application, if Figure 1 As shown, the efficient inventory method based on RFID technology includes: step 1: dividing the RFID inventory area into multiple logical blocks; step 2: using an RFID reader to perform a preliminary scan on the RFID tags of the objects to be inventoried in each of the multiple logical blocks to obtain multiple logical block preliminary feedback signals, wherein the initial feedback signal of the logical block includes the transmission power, the number of tags successfully read, the signal strength and the reading range; step 3: based on the multiple logical block preliminary feedback signals, dynamically adjusting the transmission power of the RFID reader when re-scanning the RFID tags of the objects to be inventoried in each of the logical blocks to obtain multiple logical block optimized feedback signals; step 4: generating an inventory count list based on the multiple logical block optimized feedback signals.

[0021] Exemplarily, in step 1, the RFID inventory area is divided into multiple logical blocks. It should be understood that in large-scale RFID inventory scenarios, such as large warehouses, logistics centers or industrial production workshops, the objects to be counted are often distributed in a vast space, and their layout and storage methods are diverse, and different logical blocks may have different environmental characteristics (such as object stacking density, spatial obstacle distribution, electromagnetic interference level, etc.). Therefore, this application helps to decompose complex inventory tasks into multiple relatively independent subtasks by dividing the inventory area into multiple logical blocks (such as by shelves, by categories of stored items, by physical partitions of space, etc.), more effectively adjust the inventory strategy according to the characteristics of each area, and perform refined management and optimization operations on the inventory of different areas to adapt to different inventory needs, thereby improving the operability and scalability of the inventory and the ability to solve local problems in a targeted manner.

[0022] In a specific embodiment, when dividing the logical blocks, it is first necessary to conduct a comprehensive assessment of the entire inventory area to identify which areas may cause RFID signal attenuation or reflection due to the above reasons, thereby affecting tag reading. Based on this assessment, each logical block can be defined according to the location and number of the shelf, the type and attribute of the stored items (for example, metal products may shield the RFID signal), and physical partitions (such as walls or other fixed structures). Doing so not only helps to decompose complex inventory tasks into smaller and more manageable parts, but also ensures that the environmental conditions within each logical block are relatively consistent, thereby optimizing the operating parameters of the RFID reader, such as transmission power and frequency selection, to meet the needs of specific areas. In addition, considering the flexibility in actual operation, the size of the logical block can also be flexibly adjusted according to the specific requirements of the inventory task. For those areas where the tags are densely packed or prone to reading problems, smaller logical blocks can be set for more detailed management and adjustment; while for areas where the tags are sparsely distributed and the reading conditions are good, the range of the logical block can be appropriately expanded to reduce the number of unnecessary partitions and simplify the inventory process. At the same time, as the inventory work progresses, the division of logical blocks can be dynamically adjusted according to the results of the preliminary scan. That is, if it is found that there are many tags that are not read correctly in certain areas, the area can be further subdivided in subsequent rounds. Otherwise, adjacent small blocks can be merged to form larger blocks, so as to achieve the best inventory effect.

[0023] Exemplarily, in step 2, the RFID reader is used to perform a preliminary scan on the RFID tags of the objects to be counted in each of the multiple logical blocks to obtain multiple logical block preliminary feedback signals, wherein the initial feedback signals of the logical blocks include the transmission power, the number of tags successfully read, the signal strength and the reading range. It should be understood that the RFID reader is used to transmit radio frequency signals to each logical block for preliminary scanning. When the signal covers the objects to be counted with RFID tags in the logical block, the tags will reflect or respond to the signals. The RFID reader receives these reflected or response signals and analyzes them to extract the transmission power, the number of tags successfully read, the signal strength and the reading range information of the logical block, thereby generating preliminary feedback signals for each logical block. Among them, the transmission power refers to the power of the RFID reader when transmitting the signal, which determines the propagation distance and coverage of the RFID signal; the number of tags successfully read reflects the total number of RFID tags successfully read in the logical block, which is an important indicator for evaluating the performance of the RFID reader in this area; the signal strength indicates the strength of the reflection or response signal received by the RFID reader, which is affected by many factors, such as distance, obstacles, environmental noise, etc.; the reading range refers to the maximum distance or area within the logical block where the RFID reader can effectively read the RFID tag, which represents the effective spatial range of the current signal coverage. These information together constitute a comprehensive description of the performance of the RFID inventory process in each logical block, providing a basis for subsequent data processing and parameter optimization.

[0024] Exemplarily, in step 3, based on the preliminary feedback signals of the multiple logical blocks, the transmission power of the RFID reader / writer when rescanning the RFID tags of the objects to be counted in the respective logical blocks is dynamically adjusted to obtain multiple optimized feedback signals of the logical blocks. It should be understood that, considering the differences in the environments of different logical blocks and the objects to be counted, a single transmission power may not be able to meet the inventory requirements of all areas. Therefore, the present application further analyzes the preliminary feedback signals of the multiple logical blocks to identify the signal propagation characteristics and tag reading effects of each logical block, and then dynamically adjusts the transmission power of the RFID reader / writer in the logical block to adapt to its unique environment and item distribution, thereby improving the accuracy and efficiency of the inventory. For example, in areas where the signal is weak or the tag reading is difficult, the transmission power is appropriately increased to enhance the signal coverage and reading capability. In areas where the signal is strong and the tag reading effect is good, the transmission power can be appropriately reduced to reduce electromagnetic interference and energy consumption, thereby avoiding energy waste and signal interference caused by excessively strong signals in some areas due to uniform transmission power, or incomplete tag reading caused by excessively weak signals in some areas.

[0025] In one embodiment, Figure 2and Figure 3 As shown, the step 3 includes: step 31: performing structured coding on each logic block preliminary feedback signal of the multiple logic block preliminary feedback signals to obtain multiple logic block preliminary feedback signal embedded coding vectors; step 32: extracting the first logic block preliminary feedback signal embedded coding vector corresponding to the first logic block from the multiple logic block preliminary feedback signal embedded coding vectors; step 33: inputting the multiple logic block preliminary feedback signal embedded coding vectors into the feedback signal intrinsic feature extractor to obtain the feedback signal intrinsic feature coding vector; step 34: generating a transmission power dynamic adjustment value based on the feature offset between the first logic block preliminary feedback signal embedded coding vector and the feedback signal intrinsic feature coding vector; step 35: adjusting the transmission power of the RFID reader when scanning the first logic block based on the transmission power dynamic adjustment value.

[0026] Exemplarily, in the step 31, each of the multiple logic block preliminary feedback signals is structured encoded to obtain multiple logic block preliminary feedback signal embedded coding vectors. It should be understood that considering that the logic block preliminary feedback signal contains multiple different types of data, such as transmission power, the number of tags successfully read, signal strength and reading range, it has different dimensions and value ranges, and it is difficult to be directly used for fusion analysis between multidimensional data. Therefore, the present application adopts a structured coding method to input each logic block preliminary feedback signal into a multi-layer perceptron model, so as to utilize the powerful nonlinear mapping ability of the multi-layer perceptron model, and to transmit and integrate information layer by layer through neurons and activation functions in the hidden layer, so as to map the multi-dimensional information contained in each logic block preliminary feedback signal to a low-dimensional embedding space, and encode it into a common embedding vector to obtain multiple logic block preliminary feedback signal embedded coding vectors. In this way, the association learning between the multi-dimensional data in the original feedback signal can be realized, so that the embedded coding vector of the preliminary feedback signal of each logic block can accurately reflect the RFID signal propagation characteristics and tag reading effect of the corresponding logic block, providing a unified and comparable data basis for subsequent parameter optimization configuration.

[0027] In one embodiment, each of the multiple logic block preliminary feedback signals is structured encoded to obtain multiple logic block preliminary feedback signal embedded coding vectors, including: inputting each of the multiple logic block preliminary feedback signals into an embedded encoder based on a multi-layer perceptron to obtain the multiple logic block preliminary feedback signal embedded coding vectors.

[0028] Exemplarily, in step 32, the first logic block preliminary feedback signal embedding coding vector corresponding to the first logic block is extracted from the plurality of logic block preliminary feedback signal embedding coding vectors. It should be understood that for each divided logic block, the corresponding logic block preliminary feedback signal embedding coding vector is extracted as basic data for subsequent targeted analysis and adjustment. Here, taking the first logic block as an example, by extracting the first logic block preliminary feedback signal embedding coding vector corresponding to the first logic block, the transmit power of the logic block can be specifically analyzed and optimized.

[0029] Exemplarily, in step 33, the multiple logic block preliminary feedback signals are embedded in the coding vector and input into the feedback signal intrinsic feature extractor to obtain the feedback signal intrinsic feature coding vector. It should be understood that, considering the layout, environment and distribution of the objects to be counted in the current inventory area, the signal sent by the RFID reader will be affected by many factors, such as the obstruction of space obstacles, the interference of the electromagnetic environment and the stacking density of objects, resulting in significant differences in the propagation characteristics of the RFID signal between different logic blocks. Therefore, in order to effectively compensate for this difference, the present application further performs signal intrinsic feature extraction on the multiple logic block preliminary feedback signal embedded in the coding vector to extract more essential and representative feature information, revealing the inherent propagation and attenuation law of the signal sent by the RFID reader in the overall area, thereby providing a stable reference benchmark for the transmission power adjustment of each local logic block.

[0030] In one embodiment, Figure 4 As shown, in the step 33, the multiple logic block preliminary feedback signals are embedded in the coding vectors and input into the feedback signal intrinsic feature extractor to obtain the feedback signal intrinsic feature coding vector, including: step 331: performing feature complementarity analysis based on the hub feature on the multiple logic block preliminary feedback signals embedded in the coding vectors to obtain multiple logic block preliminary feedback signals-hub feature complementary information embedded coding vectors; step 332: performing significance dynamic modulation aggregation coding on the multiple logic block preliminary feedback signals-hub feature complementary information embedded coding vectors to obtain the feedback signal intrinsic feature coding vector.

[0031] In one embodiment, Figure 5 As shown, in the step 331, the feature complementary analysis based on the hub feature is performed on the multiple logic block preliminary feedback signal embedded coding vectors to obtain multiple logic block preliminary feedback signal-hub feature complementary information embedded coding vectors, including: step 3311: inputting the multiple logic block preliminary feedback signal embedded coding vectors into the hub feature extraction network to obtain the logic block preliminary feedback signal hub feature coding vector. Specifically, the process can be expressed by the formula: in, represents the hub feature extraction network, represents the embedding coding vector of the preliminary feedback signals of the plurality of logic blocks, , , and respectively represent the first, second, and third logic blocks in which the preliminary feedback signals are embedded in the coding vector. and The initial feedback signal of the logic block is embedded in the coding vector. The number of encoding vectors embedded in the preliminary feedback signal of the logic block, and Respectively represent the weight parameter matrix and bias term of the hub feature extraction network, represents the hub feature relevance score transformation vector of the hub feature extraction network, Indicates the The corresponding hub feature relevance scoring factor, represents matrix multiplication, represents the normalized exponential function, Indicates the The corresponding normalized hub feature relevance score factor, Represents the hub feature encoding vector of the preliminary feedback signal of the logic block.

[0032] That is, the present application first constructs a hub feature extraction network based on a neural network architecture, and utilizes the powerful feature extraction capability of the neural network to perform hub feature correlation analysis on the multiple logic block preliminary feedback signal embedded coding vectors to learn the potential correlation between the features of the preliminary feedback signals of each logic block, thereby identifying and extracting representative feature representations in the multiple logic block preliminary feedback signal embedded coding vectors, namely, the logic block preliminary feedback signal hub feature coding vector, as a key descriptor of the overall propagation characteristics of the RFID signal in the inventory area.

[0033] Step 3312: extracting complementary information of each logic block preliminary feedback signal embedded coding vector relative to the logic block preliminary feedback signal hub feature coding vector in the plurality of logic block preliminary feedback signal embedded coding vectors to obtain the plurality of logic block preliminary feedback signal-hub feature complementary information embedded coding vectors. Specifically, the process can be expressed as follows: in, represents the Sigmoid activation function, represents the initial feedback signal embedded in the standardized logic block encoding vector, Represents the hub feature encoding vector of the initial feedback signal of the standardized logic block, and represents different weight matrices, represents point convolutional coding, represents the preliminary feedback signal of the logic block - the hub feature differential feature encoding vector, represents the difference operation, To take the absolute value, Indicates the The corresponding logic block initially feeds back the signal-hub feature complementary information embedded in the coding vector.

[0034] That is, in order to avoid losing too much detailed information of local logical blocks during the hub feature extraction process, the present application further integrates and associates the unique features of the preliminary feedback signals of each logical block with the hub feature coding vector of the preliminary feedback signal of the logical block as the core, and introduces a dynamic compensation mechanism to compensate and repair the lost detailed information of the logical block, thereby achieving comprehensive coverage and accurate description of the signal propagation characteristics of each logical block. In this process, first, by calculating the complementary information of the embedded coding vector of the preliminary feedback signal of each logical block relative to the hub feature coding vector of the preliminary feedback signal of the logical block, the characteristic differences and unique contributions of the signal propagation characteristics of each logical block relative to the global core information of the signal propagation characteristics of the inventory area are captured, and multiple logical block preliminary feedback signals-hub feature complementary information embedded coding vectors are generated, so as to restore the characteristic components of the embedded coding vector of the preliminary feedback signal of each logical block that were not considered and absorbed during the hub feature extraction process.

[0035] In one embodiment, Figure 6 As shown, in the step 332, the multiple logic block preliminary feedback signals-hub feature complementary information embedded coding vectors are subjected to significant dynamic modulation aggregation coding to obtain the feedback signal intrinsic feature coding vector, including: step 3321: each logic block preliminary feedback signal-hub feature complementary information embedded coding vector in the multiple logic block preliminary feedback signals-hub feature complementary information embedded coding vectors is input into the complementary information significant identification module based on the attention mechanism to obtain the attention weights of the multiple logic block preliminary feedback signal complementary information. Specifically, the process can be expressed by the formula: in, express The corresponding significant identification factor is and They represent the weight parameter matrix and bias term of the complementary information saliency identification module, represents the complementary information saliency score conversion vector of the complementary information saliency identification module, represents the exponential function operation with e as the base, express The corresponding logic block initially feeds back the complementary information attention weights.

[0036] That is, since the complementary information of the preliminary feedback signal features of each logical block relative to the global core information of the inventory area signal propagation may be a beneficial supplement to it, or it may be redundant or interfered with. Therefore, the present application further uses the attention mechanism to evaluate the importance of the preliminary feedback signal-hub feature complementary information embedded coding vector of each logical block, so as to dynamically adjust the weight distribution of the preliminary feedback signal-hub feature complementary information embedded coding vector of each logical block in the information aggregation process.

[0037] Step 3322: Based on the attention weights of the complementary information of the preliminary feedback signals of the multiple logic blocks, the preliminary feedback signals of the multiple logic blocks-hub feature complementary information embedded coding vectors are subjected to attention modulation to obtain multiple significant modulated preliminary feedback signals of the logic blocks-hub feature complementary information embedded coding vectors. Specifically, the process can be expressed by the formula: in, , , and Respectively , , and The corresponding logic block initially feeds back the signal-hub feature complementary information and embeds it into the coding vector. , , and Respectively , , and The corresponding logic block initially feeds back the complementary information attention weight of the signal, Represents the preliminary feedback signal of multiple significant modulation logic blocks-hub feature complementary information embedded in the coding vector.

[0038] That is, based on the calculated attention weights of the complementary information of the preliminary feedback signals of multiple logical blocks, the attention degree of the multiple logical block preliminary feedback signals-hub feature complementary information embedded coding vectors is modulated, and through feature enhancement or suppression, the modulated multiple logical block preliminary feedback signals-hub feature complementary information embedded coding vectors can better reflect the key role and contribution of the preliminary feedback signal features of each logical block in the description of the global core features of signal propagation in the inventory area.

[0039] Step 3323: Fusing the logic block preliminary feedback signal hub feature coding vector and the plurality of significant modulation logic block preliminary feedback signal-hub feature complementary information embedding coding vectors to obtain the feedback signal intrinsic feature coding vector. Specifically, the process can be expressed as follows: in, represents a cascade function, Represents the intrinsic feature encoding vector of the feedback signal.

[0040] That is, through cascade operation, the multiple significantly modulated logic block preliminary feedback signal-hub feature complementary information embedding coding vectors and the logic block preliminary feedback signal hub feature coding vectors are integrated to comprehensively consider the global core information of the signal propagation characteristics of the RFID inventory area and the local detail characteristics of the signal propagation characteristics of each logic block, thereby improving the accuracy and comprehensiveness of the feature expression, thereby achieving accurate description and modeling of the essential propagation characteristics of the signal in the RFID inventory area, ensuring that it can serve as a reliable reference in the subsequent transmission power optimization configuration process.

[0041] In one embodiment, Figure 7 As shown, in the step 34, based on the feature offset between the embedded coding vector of the preliminary feedback signal of the first logic block and the intrinsic feature coding vector of the feedback signal, a dynamic adjustment value of the transmission power is generated, including: step 341: performing feature offset measurement on the preliminary feedback signal of the first logic block and the intrinsic feature coding vector of the feedback signal to obtain the semantic offset coding vector of the first logic block feedback signal; step 342: inputting the semantic offset coding vector of the first logic block feedback signal into the decoder-based transmission power dynamic adjuster to obtain the dynamic adjustment value of the transmission power.

[0042] Exemplarily, in step 341, the feature offset measurement is performed on the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain the first logic block feedback signal semantic offset coding vector. In one embodiment, the feature offset measurement is performed on the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain the first logic block feedback signal semantic offset coding vector, including: calculating the position difference vector between the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain the first logic block feedback signal semantic offset coding vector. It should be understood that after the above processing, the feedback signal intrinsic feature coding vector represents the intrinsic characteristics of the signal propagation in the RFID inventory area. Therefore, in order to quantitatively evaluate the signal performance difference of the first logic block relative to the overall inventory area, the present application further calculates the position difference vector between the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to reveal the offset direction and degree of difference between the first logic block and the overall inventory area in signal propagation characteristics, and obtain the first logic block feedback signal semantic offset coding vector. It should be understood that each element of the semantic offset coding vector of the feedback signal of the first logic block represents the offset of the first logic block at the corresponding position relative to the intrinsic characteristics of the overall signal propagation of the inventory area. A positive value indicates that the signal propagation effect is better than the overall level, and a negative value indicates that the signal propagation effect is worse than the overall level. Therefore, based on the semantic offset coding vector of the feedback signal of the first logic block, the signal performance of the logic block in the RFID inventory process can be intuitively understood, providing targeted guidance for subsequent transmission power adjustments. For example, a logic block located in an area with dense metal shelves may have a much lower signal strength than other logic blocks in open areas, and its offset is large. The transmission power can be adjusted according to its offset degree, and the shielding effect of the metal can be overcome by significantly increasing the transmission power.

[0043] Exemplarily, in step 342, the semantic offset coding vector of the feedback signal of the first logical block is input into the transmit power dynamic adjuster based on the decoder to obtain the transmit power dynamic adjustment value. Specifically, the decoder is used to parse the offset information in the semantic offset coding vector of the feedback signal of the first logical block to identify the strength distribution of the signal propagation of the logical block during the RFID inventory process and its difference from the intrinsic characteristics of the signal propagation of the inventory area, and based on the mapping relationship between the transmit power and the offset information learned in the training process, convert the parsed offset information into the transmit power dynamic adjustment value for the first logical block.

[0044] Here, since the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector respectively represent the structured coding features of the first logic block preliminary feedback signal and the full sample domain distribution intrinsic features of the preliminary feedback signals of all logic blocks, when calculating the position difference vector between the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain the first logic block feedback signal semantic offset coding vector, the feedback signal semantic feature population attributes corresponding to different sample numbers will have fine-grained differential interactive fairness differences, thereby affecting the feature distribution interactive inclusiveness of the first logic block feedback signal semantic offset coding vector, so that the feature distribution of the first logic block feedback signal semantic offset coding vector has a feature manifold edge reinforcement structure, which reduces the accuracy of the decoding result obtained by the decoder-based transmit power dynamic adjuster. In a preferred example, in the process of passing the first logic block feedback signal semantic offset coding vector through a decoder-based transmit power dynamic adjuster to obtain a transmit power dynamic adjustment value, the first logic block feedback signal semantic offset coding vector is modulated by a feature manifold edge architecture, and the process includes: A random permutation operator is used to perform dynamic topological reorganization on the first logic block feedback signal semantic offset encoding vector to generate a logic block feedback signal semantic offset feature dynamic reorganization encoding tensor, expressed as ,in, represents the semantic offset coding vector of the feedback signal of the first logic block, is a random permutation operator, represents matrix multiplication, Represents the dynamic reorganization encoding tensor of the semantic offset feature of the feedback signal of the logic block; The outer product tensor generation operation of the semantic offset coding vector of the feedback signal of the first logic block is used to construct a complete linear field coding tensor, which is expressed as: ,in, represents the transposed vector of the semantic offset encoding vector of the first logic block feedback signal, represents a complete linear field encoding tensor; The dynamic reorganized coding tensor of the semantic offset feature of the logic block feedback signal is subjected to field projection with the complete linear field coding tensor to generate a scalar associated distribution heterogeneous coding tensor of the semantic offset feature of the logic block feedback signal, which is expressed as: ,in, Represents the logic block feedback signal semantic offset feature scalar associated distribution heterogeneous coding tensor; A topological slicing analysis is performed on the conjugate tensor of the logical block feedback signal semantic offset feature scalar associated distribution heterogeneous coding tensor and the first logical block feedback signal semantic offset coding vector to obtain a logical block feedback signal semantic offset feature heterogeneous topological tensor, which is expressed as: ,in, Before retaining The truncated decomposition of singular values, Represents the semantic offset feature heterogeneous topological tensor of the feedback signal of the logic block; The first logic block feedback signal semantic offset coding vector is field-fused with the logic block feedback signal semantic offset feature heterogeneous topology tensor to output an optimized first logic block feedback signal semantic offset coding vector.

[0045] Here, the semantic offset coding vector of the feedback signal of the first logic block is dynamically topologically reorganized by a random permutation operator, and embedded into a complete linear field generated by an outer product tensor, so as to realize the analyzable measurement of the scalar correlation distribution of the feature set in the heterogeneous topological structure. Based on this, the heterogeneous topological tensor is constructed by using the conjugate tensor transformation, and the reduced-dimensional manifold of the feature space is reconstructed in the field in combination with the topological slice analysis, so as to avoid the compression effect of the edge enhancement architecture on the probability of feature representation, thereby improving the accuracy of the dynamic adjustment value of the transmit power obtained by the decoder-based transmit power dynamic adjuster.

[0046] Exemplarily, in step 35, based on the dynamic adjustment value of the transmission power, the transmission power of the RFID reader when scanning the first logic block is adjusted. That is, based on the dynamic adjustment value of the transmission power, the transmission power of the RFID reader when scanning the first logic block is adjusted, and the first logic block is scanned again using the adjusted transmission power to obtain the optimized feedback signal of the logic block. By performing the above steps for each logic block, the optimized feedback signal of each logic block can be obtained in turn.

[0047] Exemplarily, in step 4, an inventory count list is generated based on the optimization feedback signals of the multiple logic blocks. It should be understood that by parsing the optimized feedback signals of each logic block, the item-related information therein, that is, the information stored in the RFID tag of the item, is extracted to determine the detailed information of the item, such as item number, name, specification, etc., and the item information of all logic blocks is summarized into a unified table or database to generate the final inventory count list. In this way, the actual situation of the current inventory can be accurately and clearly reflected, helping enterprises to quickly perform operations such as inventory verification, replenishment decisions, and logistics arrangements, thereby improving the efficiency of enterprise operations and the scientific nature of decision-making. At the same time, it provides data for subsequent inventory count history records, which is convenient for long-term tracking and analysis of inventory changes.

[0048] In summary, according to the embodiment of the present application, an efficient inventory method based on RFID technology is explained, which divides the RFID inventory area into multiple logical blocks, uses an RFID reader to scan the RFID tags of each logical block to obtain the preliminary feedback signal of each logical block, and then introduces a data processing technology based on deep learning to perform a global correlation analysis on the preliminary feedback signals of each logical block to reveal the essential propagation characteristics of the signal sent by the RFID reader in the inventory area under the current setting, and then based on this, according to the offset degree of the preliminary feedback signal of each logical block relative to the essential propagation characteristics of the signal, dynamically adjust the transmission power of the RFID reader in each logical block to scan again, and generate an inventory count list based on the scanning results. In this way, the parameter configuration of the RFID reader can be adaptively optimized, thereby improving the reading rate and accuracy of the RFID tag, reducing the phenomenon of missed reading and multiple reading of tags, and thus improving the accuracy and efficiency of inventory counting.

[0049] Figure 8 Schematic block diagram of an efficient inventory counting system based on RFID technology according to an embodiment of the present application. Figure 8 As shown, the efficient inventory counting system 100 based on RFID technology includes: a logical block division module 110, which is used to divide the RFID inventory counting area into multiple logical blocks; a preliminary scanning module 120, which is used to use an RFID reader to perform preliminary scanning on the RFID tags of the objects to be counted in each logical block of the multiple logical blocks to obtain multiple logical block preliminary feedback signals, wherein the initial feedback signal of the logical block includes the transmission power, the number of tags successfully read, the signal strength and the reading range; a re-scanning module 130, which is used to dynamically adjust the transmission power of the RFID reader when re-scanning the RFID tags of the objects to be counted in each logical block based on the multiple logical block preliminary feedback signals to obtain multiple logical block optimized feedback signals; an inventory count list generation module 140, which is used to generate an inventory count list based on the multiple logical block optimized feedback signals.

[0050] In one embodiment, the re-scanning module includes: a preliminary feedback signal embedding coding unit, used to perform structured coding on each logic block preliminary feedback signal of the multiple logic block preliminary feedback signals to obtain multiple logic block preliminary feedback signal embedding coding vectors; a first logic block vector extraction unit, used to extract a first logic block preliminary feedback signal embedding coding vector corresponding to the first logic block from the multiple logic block preliminary feedback signal embedding coding vectors; a feedback signal intrinsic feature coding unit, used to input the multiple logic block preliminary feedback signal embedding coding vectors into a feedback signal intrinsic feature extractor to obtain a feedback signal intrinsic feature coding vector; a transmission power dynamic adjustment value generation unit, used to generate a transmission power dynamic adjustment value based on the feature offset between the first logic block preliminary feedback signal embedding coding vector and the feedback signal intrinsic feature coding vector; a transmission power adjustment unit, used to adjust the transmission power of the RFID reader when scanning the first logic block based on the transmission power dynamic adjustment value.

[0051] Here, those skilled in the art will appreciate that the specific operations of the various modules and units in the above-mentioned efficient inventory counting system based on RFID technology have been described in detail above. Figures 1 to 7 The description of the efficient inventory method based on RFID technology has been introduced in detail, and therefore, its repeated description will be omitted.

[0052] An embodiment of the present application further provides a computer program product, which includes a computer program code. When the computer program code runs on a computer, the computer implements the methods in the above embodiments of the present application.

[0053] An embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are executed on a computer, the computer implements the methods in the above embodiments of the present application.

[0054] An embodiment of the present application also provides a chip, including a circuit, for executing the methods in the above embodiments of the present application.

[0055] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0056] In the description of the embodiments of the present application, unless otherwise specified, " / " means or, for example, A / B can mean A or B; "and / or" in this article is a kind of association relationship that describes associated objects, indicating that three relationships can exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In this application, "at least one" means one or more, and "multiple" means two or more. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0057] The prefixes such as "first" and "second" used in the embodiments of the present application are only used to distinguish different description objects, and have no limiting effect on the position, order, priority, quantity or content of the described objects. The use of prefixes such as ordinal numbers used to distinguish description objects in the embodiments of the present application does not constitute a limitation on the described objects. For the statement of the described objects, please refer to the description in the context of the claims or embodiments, and the use of such prefixes should not constitute an unnecessary limitation.

[0058] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0059] In the various embodiments of the present application, unless otherwise specified or logically conflicting, the terms and / or descriptions between the various embodiments are consistent and can be referenced to each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0060] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0061] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0062] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present 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. An efficient inventory counting method based on RFID technology, characterized in that: include: Step 1: Divide the RFID inventory area into multiple logical blocks; Step 2: Use an RFID reader to perform a preliminary scan on the RFID tags of the objects to be counted in each of the multiple logical blocks to obtain multiple logical block initial feedback signals, wherein the logical block initial feedback signals include transmission power, number of tags successfully read, signal strength and reading range; Step 3: Based on the preliminary feedback signals of the multiple logical blocks, dynamically adjust the transmission power of the RFID reader when rescanning the RFID tags of the objects to be counted in the respective logical blocks to obtain multiple optimized feedback signals of the logical blocks; Step 4: Generate an inventory count list based on the optimization feedback signals of the multiple logic blocks.

2. The efficient inventory counting method based on RFID technology according to claim 1 is characterized in that: The step 3 comprises: Performing structured coding on each of the plurality of logic block preliminary feedback signals to obtain a plurality of logic block preliminary feedback signal embedding coding vectors; Extracting a first logic block preliminary feedback signal embedding code vector corresponding to a first logic block from the plurality of logic block preliminary feedback signal embedding code vectors; Embedding the preliminary feedback signals of the plurality of logic blocks into a coding vector and inputting it into a feedback signal intrinsic feature extractor to obtain a feedback signal intrinsic feature coding vector; Generate a dynamic adjustment value of transmit power based on a feature offset between a preliminary feedback signal embedding coding vector of the first logic block and an intrinsic feature coding vector of the feedback signal; Based on the dynamic adjustment value of the transmission power, the transmission power of the RFID reader when scanning the first logic block is adjusted.

3. The efficient inventory counting method based on RFID technology according to claim 2 is characterized in that: Performing structured coding on each of the plurality of logic block preliminary feedback signals to obtain a plurality of logic block preliminary feedback signal embedding coding vectors, including: Each logic block preliminary feedback signal of the plurality of logic block preliminary feedback signals is input into an embedding encoder based on a multi-layer perceptron to obtain an embedding coding vector of the plurality of logic block preliminary feedback signals.

4. The efficient inventory counting method based on RFID technology according to claim 3 is characterized in that: Embedding the preliminary feedback signals of the plurality of logic blocks into a coding vector and inputting the coding vector into a feedback signal intrinsic feature extractor to obtain a feedback signal intrinsic feature coding vector, comprising: Performing feature complementation analysis based on the hub feature on the plurality of logic block preliminary feedback signal embedded coding vectors to obtain a plurality of logic block preliminary feedback signal-hub feature complementary information embedded coding vectors; The preliminary feedback signals of the multiple logic blocks-hub feature complementary information embedded coding vectors are subjected to significant dynamic modulation aggregation coding to obtain the feedback signal intrinsic feature coding vector.

5. The efficient inventory counting method based on RFID technology according to claim 4 is characterized in that: Performing feature complementation analysis based on the hub feature on the plurality of logic block preliminary feedback signal embedded coding vectors to obtain a plurality of logic block preliminary feedback signal-hub feature complementary information embedded coding vectors, including: Embedding the plurality of logic block preliminary feedback signals into encoding vectors and inputting them into a pivot feature extraction network to obtain a logic block preliminary feedback signal pivot feature encoding vector; The complementary information of each logic block preliminary feedback signal embedded coding vector relative to the logic block preliminary feedback signal hub feature coding vector is extracted to obtain the multiple logic block preliminary feedback signal-hub feature complementary information embedded coding vectors.

6. The efficient inventory counting method based on RFID technology according to claim 5 is characterized in that: The preliminary feedback signals of the plurality of logic blocks and the hub feature complementary information embedding coding vectors are subjected to significant dynamic modulation aggregation coding to obtain the feedback signal intrinsic feature coding vector, including: Inputting each logic block preliminary feedback signal-hub feature complementary information embedding coding vector in the plurality of logic block preliminary feedback signal-hub feature complementary information embedding coding vectors into a complementary information significant identification module based on an attention mechanism to obtain a plurality of logic block preliminary feedback signal complementary information attention weights; Based on the attention weights of the complementary information of the preliminary feedback signals of the multiple logic blocks, the preliminary feedback signals of the multiple logic blocks-hub feature complementary information embedded coding vectors are subjected to attention modulation to obtain multiple significant modulated preliminary feedback signals of the logic blocks-hub feature complementary information embedded coding vectors; The feedback signal intrinsic feature coding vector is obtained by fusing the logic block preliminary feedback signal hub feature coding vector and the plurality of significantly modulated logic block preliminary feedback signal-hub feature complementary information embedding coding vectors.

7. The efficient inventory counting method based on RFID technology according to claim 6 is characterized in that: Generating a dynamic adjustment value of transmit power based on a feature offset between a preliminary feedback signal embedding coding vector of the first logic block and an intrinsic feature coding vector of the feedback signal, including: Performing feature offset measurement on the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain a first logic block feedback signal semantic offset coding vector; The first logic block feedback signal semantic offset coding vector is input into a decoder-based transmit power dynamic adjuster to obtain the transmit power dynamic adjustment value.

8. The efficient inventory counting method based on RFID technology according to claim 7 is characterized in that: Performing feature offset measurement on the first logic block preliminary feedback signal and the feedback signal intrinsic feature coding vector to obtain a first logic block feedback signal semantic offset coding vector, including: A position difference vector between the first logic block preliminary feedback signal and the feedback signal intrinsic feature encoding vector is calculated to obtain the first logic block feedback signal semantic offset encoding vector.

9. An efficient inventory counting system based on RFID technology, characterized in that: include: A logical block division module is used to divide the RFID inventory area into multiple logical blocks; A preliminary scanning module, used to use an RFID reader to perform preliminary scanning on the RFID tags of the objects to be counted in each of the multiple logical blocks to obtain multiple logical block preliminary feedback signals, wherein the logical block initial feedback signals include transmission power, number of tags successfully read, signal strength and reading range; A re-scanning module, configured to dynamically adjust the transmission power of the RFID reader when re-scanning the RFID tags of the objects to be counted in the respective logical blocks based on the preliminary feedback signals of the plurality of logical blocks, so as to obtain the optimized feedback signals of the plurality of logical blocks; The inventory count list generating module is used to generate an inventory count list based on the optimization feedback signals of the multiple logic blocks.

10. The efficient inventory counting system based on RFID technology according to claim 9, characterized in that: The re-scanning module comprises: A preliminary feedback signal embedding coding unit, configured to perform structured coding on each of the plurality of logic block preliminary feedback signals to obtain a plurality of logic block preliminary feedback signal embedding coding vectors; A first logic block vector extraction unit is used to extract a first logic block preliminary feedback signal embedding code vector corresponding to a first logic block from the plurality of logic block preliminary feedback signal embedding code vectors; A feedback signal intrinsic feature encoding unit, used for embedding the preliminary feedback signals of the plurality of logic blocks into a coding vector and inputting it into a feedback signal intrinsic feature extractor to obtain a feedback signal intrinsic feature coding vector; A transmit power dynamic adjustment value generating unit, configured to generate a transmit power dynamic adjustment value based on a feature offset between an embedded coding vector of a preliminary feedback signal of the first logic block and an intrinsic feature coding vector of the feedback signal; The transmission power adjustment unit is used to adjust the transmission power of the RFID reader when scanning the first logic block based on the transmission power dynamic adjustment value.

Images