RFID tag interference solving method based on density peak clustering algorithm

Abstract

An RFID tag interference solving method based on a density peak value clustering algorithm reduces signal interference between tags and improves reliability and identification precision of a system by dynamically adjusting priorities and time windows of RFID tags. The method comprises the following core steps: firstly, acquiring a label response signal through an RFID reader-writer; secondly, preprocessing the data, and removing noise and abnormal values; then, calculating the local density of the label signal, and identifying a density peak value region by using a clustering algorithm; then, reducing signal collision of tags in a high-density region by dynamically adjusting priorities and time windows of the tags in the region, and preferentially processing the tags; and finally, outputting a processed tag identification result. According to the invention, the problem of signal interference in a high-density RFID tag environment is obviously solved, the identification accuracy and the operation stability of the system are improved, and the method is suitable for various high-density tag application scenes.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless radio frequency identification, and in particular relates to a solution to RFID tag interference based on a density peak clustering algorithm. Background Art

[0002] With the rapid development of information technology, RFID (radio frequency identification) technology has been widely used in logistics, retail, manufacturing and other fields. The RFID system automatically identifies and tracks objects through radio frequency signals, enabling rapid acquisition and management of item information. However, in actual applications, as the number of RFID tags increases, the problem of signal interference between tags becomes increasingly prominent, seriously affecting the recognition accuracy and reliability of the system.

[0003] In a high-density tag environment, multiple tags respond to RFID reader signals at the same time, resulting in signal collision and interference. This signal interference not only reduces the tag recognition rate, but may also cause data loss and misreading, thus affecting the overall performance of the system. Traditional anti-collision algorithms, such as the ALOHA algorithm and the tree anti-collision algorithm, can alleviate the signal interference problem to a certain extent, but in a high-density tag environment, it is still difficult to completely solve the interference problem between tags.

[0004] In order to deal with the signal interference problem in high-density RFID tag environments, a variety of improved methods and algorithms have been proposed. For example, the tag scheduling method based on signal strength and time interval dynamically adjusts the tag response time to reduce signal collisions; the signal processing method based on machine learning analyzes the characteristics of the tag response signal to identify and filter interference signals. However, these methods still face problems such as high algorithm complexity and high computational cost in practical applications, making it difficult to be effectively promoted in large-scale application scenarios. Summary of the invention

[0005] In view of the above technical problems, this technical solution provides a solution to RFID tag interference based on a density peak clustering algorithm. By analyzing the density distribution of tag response signals, the peak area of ​​signal density is identified, and tags in high-density areas are prioritized and dynamically scheduled, thereby reducing signal collisions and improving the recognition accuracy and reliability of the system. The density peak clustering algorithm has the advantages of low computational complexity and high recognition accuracy, and can implement an effective signal interference solution in a high-density tag environment. It can effectively solve the above problems. The present invention is implemented through the following technical solutions:

[0006] A method for solving RFID tag interference based on density peak clustering algorithm reduces signal interference between tags by dynamically adjusting the priority and time window of RFID tags. The specific operation steps of the method include:

[0007] Step 1: Collect tag response signal data through RFID reader, including location information, signal strength, frequency, response time and other parameters;

[0008] Step 2: Preprocess the collected data to remove noise and outliers;

[0009] Step 3: Based on the preprocessed data, calculate the local density of each tag response signal;

[0010] Step 4: Use the density peak clustering algorithm to identify the peak area of ​​signal density; the specific operation method is:

[0011] Using the location information collected in step 1, density calculation based on location information: Use location information to calculate the proximity relationship of tags based on the distance threshold d c , determine the number of neighbor labels for each label; local density ρ i The calculation formula is:

[0012] ρ i =∑ j χ(d ij -d c )

[0013] In the formula, χ represents the indicator function, d ij is the distance between labels i and j, d c is the cutoff distance;

[0014] Density peak identification based on location information: Combine location information and local density ρi to calculate the minimum distance δ from each label to a label with higher density i :

[0015]

[0016] For the label with the largest density, define δ i =max j (d ij ); Using the position information, then, combined with ρ i and δ i Draw a decision diagram based on the value of and identify the density peak point, i.e., the cluster center;

[0017] Step 5: Assign different priorities to tags in the density peak area based on parameters such as density and signal strength;

[0018] Step 6: Dynamically adjust the response time window of each tag to avoid tags responding at the same time;

[0019] Step 7: Prioritize tags in high-density areas to reduce signal interference and collision, and output processed tag recognition results.

[0020] Furthermore, the specific operation method of step one is:

[0021] Use RFID readers to obtain the geographic location information of the tag, and calculate the specific location of the tag by measuring the distance from the tag to multiple readers;

[0022] The location information includes the X and Y coordinates of the tag in two-dimensional space or the X, Y and Z coordinates in three-dimensional space. The weighted average method is used to fuse the location information of multiple measurements to improve the positioning accuracy. The weighted average method uses the tag positions measured by multiple readers and writers, gives each position a weight, and obtains the final position through weighted average. The calculation formula of the weighted average method is:

[0023]

[0024] In the above formula, n is the total number of positions involved in the calculation, i is the index of the current i-th position data point, and the value range of i is from 1 to n.

[0025] Furthermore, the specific operation method of step 2 is: using a median filtering algorithm to process the collected RFID tag signal; the calculation steps of the median filtering are: selecting a sliding window of the signal, sorting all signal values ​​in the window, and taking the middle value as the filtering result.

[0026] Furthermore, the specific steps of identifying high-density areas in step 4 are: collecting the location information of the tag, including X, Y, and Z coordinates, through the RFID reader. Using this location information, calculate the local density of each tag. The calculation method of the local density is as follows:

[0027] Set a cutoff distance d c , define all labels within this distance range as neighbor labels. Calculate the number of neighbors of each label, that is, the label density around the label.

[0028] For each label, calculate its minimum distance to the label with higher density. Draw a decision graph with one axis as local density and the other axis as the minimum distance to the label with higher density. In the decision graph, the peak points (i.e. the center points of the high-density area) will appear as individuals far away from other points. Use these peak points to identify the areas with the highest signal density, i.e. the high-density areas.

[0029] After identifying high-density areas, different priorities are assigned to tags in these areas based on local density and signal strength. Tags in high-density areas will be assigned higher priorities for priority processing.

[0030] Furthermore, the specific operation method of step five is: for tags in the density peak area, different priorities are assigned according to density and signal strength parameters; tags in high-density areas have higher priorities and are read first; and priorities are dynamically adjusted in combination with historical response records of tags.

[0031] Furthermore, the response time window of each tag is dynamically adjusted as described in step 6, and dynamic priority adjustment and time window allocation are performed in combination with the location information, including dynamic priority adjustment and time window adjustment;

[0032] Dynamic priority adjustment: first assign priority based on the density and signal strength of the tag location. Tags in high-density areas have higher priority and are read first. Combined with location information, dynamically adjust tag priority to ensure the flexibility and adaptability of the system. The specific operation is as follows: first, use the RFID reader to obtain the geographic location information of each tag, including its X and Y coordinates in space. Based on this location information, calculate the distribution density of the tag in space. Then use the density peak clustering algorithm to analyze the location information of all tags and calculate the tag density around each tag. Combine density with signal strength to assign preliminary priority to each tag. Tags with high density and high signal strength will be assigned higher priority. Update the location information of the tag in real time according to the movement of the tag and changes in the environment. When changes in the tag density in certain areas are detected, the system will dynamically adjust the priority of the tags in that area.

[0033] Time window adjustment: Dynamically adjust the response time window of each tag to avoid tags responding at the same location and time, reducing signal collisions. High-priority tags are assigned shorter time windows to ensure fast reading and processing. High-priority tags are assigned shorter time windows to ensure fast reading, while low-priority tags are assigned longer time windows. Through this synchronous adjustment, signal collisions between tags can be further reduced, improving the reading efficiency and accuracy of the system.

[0034] Furthermore, the specific operation method of step seven is: read the tags in order according to the adjusted priority and time window, and read the high priority tags in batches; process the read tag data, identify and record the tag information, and output the processed tag identification results to ensure the accuracy and completeness of the data.

[0035] Beneficial Effects

[0036] The RFID tag interference solution based on the density peak clustering algorithm proposed in the present invention has the following beneficial effects compared with the prior art:

[0037] (1) The present invention uses a density peak clustering algorithm to identify high-density areas and give priority to the tags in the area, effectively reducing signal collisions between tags. The strategy of giving priority to tags in high-density areas ensures that tags in key areas can be accurately read, improving the recognition accuracy of the overall system. At the same time, the priority and response time window of the tag are dynamically adjusted according to the real-time collected data and the historical response records of the tag to ensure the stable operation of the system in different environments. By reducing signal interference and collisions, the reliability of the system operation is improved and the probability of misreading and missed reading is reduced.

[0038] (2) The density peak clustering algorithm in the present invention has low computational complexity and can run efficiently in large-scale application scenarios. The optimized data preprocessing and density calculation steps ensure that the system can still respond quickly when processing a large amount of labeled data, thereby improving the overall computational efficiency.

[0039] (3) The method of the present invention is applicable to scenarios that require efficient and accurate management of a large number of RFID tags, such as logistics centers, warehouse management, and retail industries. It has broad application value and market prospects and can be used in various practical application scenarios that require high-density identification of RFID tags, providing an efficient and reliable solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 It is a schematic diagram of the overall process of the present invention.

[0041] Figure 2 It is a flow chart of the data acquisition and preprocessing module in the present invention.

[0042] Figure 3 It is a schematic diagram of the density calculation and density peak identification process in the present invention.

[0043] Figure 4 This is the result graph of the cutoff distance and cluster center in the example.

[0044] Figure 5 This is the result of the density peak clustering algorithm.

[0045] Figure 6 It is the experimental result figure in the embodiment. DETAILED DESCRIPTION

[0046] The following will be combined with the accompanying drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Under the premise of not departing from the design concept of the present invention, various modifications and improvements made by ordinary persons in the art to the technical solutions of the present invention should all fall within the protection scope of the present invention.

[0047] Embodiment 1:

[0048] A solution to RFID tag interference based on density peak clustering algorithm. This embodiment describes how to use a method based on density peak clustering algorithm in a large logistics center to dynamically adjust the priority and time window of each RFID tag to reduce interference between tags, thereby improving the reliability and accuracy of the system. In this embodiment, the logistics center is divided into the following main areas: receiving area, storage area, picking area, packaging area, shipping area, and return area. The RFID tags in each area have different distribution densities and quantities. The specific implementation steps are as follows; Figure 1 As shown, the specific operation steps of the method include:

[0049] Step 1: Collect tag response signal data through RFID reader, including parameters such as location information, signal strength, frequency and response time.

[0050] Multiple RFID readers are installed in the logistics center to regularly collect response signal data from all RFID tags. The collected data includes parameters such as signal strength, frequency, response time and location information. Through multi-point reader positioning technology, the geographic location information of each tag is obtained and the coordinates of the tag in two-dimensional or three-dimensional space are recorded.

[0051] Use RFID readers to obtain the geographic location information of the tag, and calculate the specific location of the tag by measuring the distance from the tag to multiple readers;

[0052] The location information includes the X and Y coordinates of the tag in two-dimensional space or the X, Y and Z coordinates in three-dimensional space. The weighted average method is used to fuse the location information of multiple measurements to improve the positioning accuracy. The weighted average method uses the tag positions measured by multiple readers and writers, gives each position a weight, and obtains the final position through weighted average. The calculation formula of the weighted average method is:

[0053]

[0054] In the above formula, n is the total number of positions involved in the calculation, i is the index of the current i-th position data point, and the value range of i is from 1 to n.

[0055] Step 2: Preprocess the collected data to remove noise and outliers. The process of data collection and preprocessing module is as follows: Figure 2 The core steps are:

[0056] Use the median filter algorithm to process the collected RFID tag signals. Median filtering can effectively remove short-term pulse noise and retain the edge characteristics of the signal. The calculation steps of the median filter are: select a sliding window of a fixed size, usually an odd length N, to slide one by one in the signal sequence. The size of the window N can be adjusted according to actual needs. Sort all signal values ​​in the sliding window. Set the signal sequence, and during the window sliding process, take out N signal values ​​each time, and sort these N values ​​from small to large to obtain the sorted signal value sequence. Take the middle value of the sorted sequence as the filtering result, that is, the median x med =x ((N+1) / 2) . This median replaces the center value of the current window in the original signal sequence, thereby filtering out the noise. Move the sliding window one unit on the signal sequence in turn, and repeat the above steps until the entire signal sequence is processed. After each move, the new median replaces the original signal value at the center of the current window. Select a sliding window of the signal, sort all the signal values ​​in the window, and take the middle value as the filtering result.

[0057] Step 3: Based on the preprocessed data, calculate the local density of each tag response signal.

[0058] Step 4: Use the density peak clustering algorithm to identify the peak area of ​​signal density; the specific operation method is:

[0059] like Figure 3 As shown, using the location information collected in step 1, density calculation based on location information: using location information to calculate the proximity relationship of tags, based on the distance threshold d c , determine the number of neighbor labels for each label; local density ρ i The calculation formula is:

[0060] ρ i =Σ j χ(d ij -d c )

[0061] In the formula, χ represents the indicator function, d ij is the distance between labels i and j, d c is the cutoff distance.

[0062] Density peak identification based on location information: Combine location information and local density ρi to calculate the minimum distance δ from each label to a label with higher density i :

[0063]

[0064] For the label with the largest density, define δ i =max j(d ij ); Using the position information, then, combined with ρ i and δ i Draw a decision diagram based on the value of and identify the density peak point, which is the cluster center.

[0065] like Figure 4 As shown, dc: 6.16834688187 represents the cutoff distance in the density peak clustering algorithm, which is used to calculate the local density of each point. Specifically, only points with a distance less than dc are considered neighbors to calculate the local density. centers[636455319367770151] represents the index of the center point (cluster center) identified by the clustering algorithm. These indices correspond to the labels selected as the center of each category in the dataset. The cutoff distance dc determines the range of neighbors for calculating the local density, while the cluster center index shows the center point of each category selected by the algorithm.

[0066] like Figure 5 As shown in the figure, the RFID tag distribution and clustering results obtained based on the density peak clustering algorithm are shown. The points of different colors in the figure represent tags of different categories, and the black cross represents the cluster center of each category. There are six points of different colors in the figure, representing six different categories. These categories have obvious distribution differences in different areas of the figure. The label points of each category are distributed in a specific area, reflecting the distribution of label positions in different functional areas of the logistics center. The black cross represents the cluster center (i.e., the density peak point) of each category. The location of the cluster center is roughly at the center of each category label point, indicating that the clustering algorithm effectively finds the center point of each category.

[0067] Step 5: Assign different priorities to tags in the density peak area based on parameters such as density and signal strength; The specific operation method is: Assign different priorities to tags in the density peak area based on parameters such as density and signal strength. Tags in high-density areas have higher priorities and are read first. Combined with the historical response records of tags, the priorities are dynamically adjusted to ensure the flexibility and adaptability of the system.

[0068] Step 6: Dynamically adjust the response time window of each tag to prevent tags from responding at the same time; the specific operation method is:

[0069] Dynamically adjust the response time window of each tag to avoid tags responding at the same location and time, reducing signal collisions. High-priority tags are assigned shorter time windows to ensure fast reading and processing. High-priority tags are assigned shorter time windows to ensure fast reading, while low-priority tags are assigned longer time windows. Through this synchronous adjustment, signal collisions between tags can be further reduced, improving the reading efficiency and accuracy of the system.

[0070] Step 7: Prioritize tags in high-density areas to reduce signal interference and collision, and output the processed tag recognition results. The specific operation method is:

[0071] Read tags according to the adjusted priority and time window sequence. Process the read tag data, identify and record tag information, and output the processed tag identification results to ensure the accuracy and integrity of the data.

[0072] In order to verify the advantages of the above scheme, the applicant conducted experiments to verify the scheme. The process of experiment and verification is as follows:

[0073] In order to evaluate the performance of the algorithm, this embodiment uses three indicators: accuracy, average response time and number of interferences.

[0074] The experimental results are as follows Figure 6 As shown in the figure, according to the experimental results, the Density Peak Clustering (DPC) model has significantly improved the accuracy of the K-means model and the DBSCAN model, reaching 98%. This result shows that the DPC algorithm has the highest accuracy in RFID tag recognition and classification, and can effectively reduce errors and false alarms and improve the recognition accuracy of the system. The average response time of the DPC algorithm is 1.08 seconds, which is significantly better than K-means' 1.60 seconds and DBSCAN's 1.57 seconds. This means that the DPC algorithm can process tag data more quickly in practical applications, improving the system's response speed and user experience. The number of interferences of the DPC algorithm is only 59 times, while K-means and DBSCAN are 135 and 114 times respectively. The DPC algorithm performs well in reducing tag signal interference and significantly improves the reliability and stability of the system.

[0075] In summary, the density peak clustering algorithm (DPC) has demonstrated its excellent performance in dealing with RFID tag interference problems. DPC not only has obvious advantages in accuracy, but also can significantly reduce the average response time and the number of interferences. These superior performances make the DPC algorithm an ideal choice for solving RFID tag interference problems.

[0076] The above embodiments are only for illustrating the technical concept and features of the present invention, and their purpose is to enable people familiar with the technology to understand the content of the present invention and implement it accordingly, and they cannot be used to limit the protection scope of the present invention. Any equivalent transformation or modification made according to the spirit of the present invention should be included in the present invention.

Claims

1. A solution to RFID tag interference based on density peak clustering algorithm, characterized by: By dynamically adjusting the priority and time window of RFID tags, signal interference between tags is reduced. The specific operation steps of this method include: Step 1: Collect tag response signal data through RFID reader, including location information, signal strength, frequency, response time and other parameters; Step 2: Preprocess the collected data to remove noise and outliers; Step 3: Based on the preprocessed data, calculate the local density of each tag response signal; Step 4: Use the density peak clustering algorithm to identify the peak area of ​​signal density; the specific operation method is: Using the location information collected in step 1, density calculation based on location information: Use location information to calculate the proximity relationship of tags based on the distance threshold d c , determine the number of neighbor labels for each label; local density ρ i The calculation formula is: r i =S j x(d ij -d c ) In the formula, χ represents the indicator function, d ij is the distance between labels i and j, d c is the cutoff distance; Density peak identification based on location information: Combine location information and local density ρi to calculate the minimum distance δ from each label to a label with higher density i : For the label with the largest density, define δ i =max j (d ij ); using the position information, then, combined with ρ i and δ i Draw a decision diagram based on the value of and identify the density peak point, i.e., the cluster center; Step 5: Assign different priorities to tags in the density peak area based on parameters such as density and signal strength; Step 6: Dynamically adjust the response time window of each tag to avoid tags responding at the same time; Step 7: Prioritize tags in high-density areas to reduce signal interference and collision, and output processed tag recognition results.

2. The RFID tag interference solution based on the density peak clustering algorithm according to claim 1 is characterized by: The specific operation method of step one is: Use RFID readers to obtain the geographic location information of the tag, and calculate the specific location of the tag by measuring the distance from the tag to multiple readers; The location information includes the X and Y coordinates of the tag in two-dimensional space or the X, Y and Z coordinates in three-dimensional space. The weighted average method is used to fuse the location information of multiple measurements to improve the positioning accuracy. The weighted average method uses the tag positions measured by multiple readers, gives each position a weight, and obtains the final position through weighted average. The calculation formula of the weighted average method is: In the above formula, n is the total number of positions involved in the calculation, i is the index of the current i-th position data point, and the value range of i is from 1 to n.

3. The RFID tag interference solution based on the density peak clustering algorithm according to claim 1 is characterized by: The specific operation method of step 2 is: use the median filtering algorithm to process the collected RFID tag signal; the calculation steps of the median filtering are: select a sliding window of the signal, sort all the signal values ​​in the window, and take the middle value as the filtering result.

4. The RFID tag interference solution based on density peak clustering algorithm according to claim 1 is characterized by: The specific operation method of step five is: for tags in the density peak area, different priorities are assigned according to density and signal strength parameters; tags in high-density areas have higher priorities and are read first; and priorities are dynamically adjusted based on historical response records of tags.

5. The RFID tag interference solution based on density peak clustering algorithm according to claim 1 is characterized by: The step 6 dynamically adjusts the response time window of each tag, and performs dynamic priority adjustment and time window allocation in combination with the location information, including dynamic priority adjustment and time window adjustment; Dynamic priority adjustment: first assign priority based on the density and signal strength of the tag location. Tags in high-density areas have higher priority and are read first. Combined with location information, dynamically adjust tag priority to ensure the flexibility and adaptability of the system. The specific operation is as follows: first, use the RFID reader to obtain the geographic location information of each tag, including its X and Y coordinates in space. Based on this location information, calculate the distribution density of the tag in space. Then use the density peak clustering algorithm to analyze the location information of all tags and calculate the tag density around each tag. Combine density with signal strength to assign preliminary priority to each tag. Tags with high density and high signal strength will be assigned higher priority. According to the movement of tags and changes in the environment, the location information of tags is updated in real time. When changes in the tag density in certain areas are detected, the system will dynamically adjust the priority of tags in that area. Time window adjustment: Dynamically adjust the response time window of each tag to avoid tags responding at the same location and time, reducing signal collisions; high-priority tags are assigned shorter time windows to ensure fast reading and processing; high-priority tags are assigned shorter time windows to ensure fast reading, while low-priority tags are assigned longer time windows; through this synchronous adjustment, signal collisions between tags can be further reduced, improving the reading efficiency and accuracy of the system.

6. The RFID tag interference solution based on density peak clustering algorithm according to claim 1 is characterized by: The specific operation method of step seven is: read the tags in order according to the adjusted priority and time window, and read the high-priority tags in batches; process the read tag data, identify and record the tag information, and output the processed tag identification results to ensure the accuracy and completeness of the data.

Images