Sensor-enhanced RFID tag information collection method based on double polling

Abstract

The invention provides a sensor-enhanced RFID tag information collection method based on double polling. The method comprises the following steps: step 1, a reader constructs an original expected frame vector by mapping unqueried tags; 2, dividing the original expected frame vector into two segments, and generating a composite vector through AND operation; step 3, the reader screens out all double indexes as polling vectors; step 4, after the tag receives the polling vector, calculating a hash index value; 5, each label determines a reply time slot according to the Hash index Xz and the value of the length fi of the original expected frame vector OV of the label; and step 6, if the tag is queried, keeping silence, otherwise, keeping active and waiting for a next polling vector. According to the method, each polling vector can simultaneously query a pair of tags, so that the utilization rate of each polling vector is effectively improved, the communication overhead of a reader is reduced, and the time efficiency of querying the tags by the reader is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of radio frequency identification and the Internet of Things, and in particular to a sensor-enhanced RFID tag information collection method based on dual polling. Background Art

[0002] Radio Frequency Identification (RFID) systems typically consist of a backend server, one or more readers, and a large number of tags. The reader communicates with tags within its radio frequency transmission range via a simple wireless channel. The backend server stores the ID of each tag and exchanges information with the reader via wired or wireless communication. With the increasing deployment of sensor-enhanced RFID systems, RFID-based information collection is becoming increasingly important. Tag information collection aims to efficiently and accurately collect valuable information from tagged objects. Polling-based information collection effectively avoids response collisions between RFID tags and is widely used for accurate tag querying. In large-scale RFID systems, readers can automatically manage the RFID system by querying RFID tags and collecting their responses, saving users significant time and labor costs in practice. In some environmentally sensitive RFID applications, microsensors are often integrated into tags to enhance the RFID system's sensing capabilities. In these applications, readers can not only identify target objects but also collect information about tagged objects in real time by querying sensor-enhanced RFID tags. For example, in a large food warehouse, the surrounding environment, including air humidity and temperature, can easily affect the quality of food preservation. To ensure food preservation quality, sensitive environmental information needs to be monitored in real time. Therefore, an efficient information collection solution is crucial for obtaining target information from labeled objects.

[0003] Tag information collection aims to efficiently obtain accurate, targeted responses from RFID tags. To this end, avoiding conflicting responses between different RFID tags is crucial. The frame-slotted Aloha method allows each RFID tag to randomly select a time slot to reply to the reader, and is widely used to combat tag conflicts and improve communication efficiency between readers and tags. However, a reader can only successfully receive a tag's reply in a single time slot selected by a tag, and the reader cannot receive any useful information from conflicting or empty time slots. More importantly, although many existing methods attempt to alleviate the problem of wasted time slots, conflicting and empty time slots are inevitable due to the randomness of the slot index selected by the tag. To improve the efficiency of data transmission between readers and tags, avoiding wasted time slots is a very important solution. However, in the traditional polling mode, a polling vector can only query one tag at a time, which is inefficient.

[0004] Polling-based information collection schemes can completely avoid tag conflicts and useless empty time slots by querying tags one by one to collect information. In the conventional polling protocol (CPP), many researchers have devoted themselves to minimizing the length of polling vectors to reduce polling overhead. However, an obvious shortcoming of CPP is that the communication between readers and tags adopts a one-to-one mode. That is, each polling vector can only be used to query one tag at a time. Therefore, when querying all tags, the reader will send a polling vector equal to the number of tags to query all tags. As the number of tags increases, in large-scale RFID systems, it may take a long time for the reader to complete the information collection process. Therefore, reducing the number of polling vectors and improving the utilization of each polling vector are crucial to improving collection efficiency. However, current sensor-enhanced RFID tag information collection faces the following challenges: (1) The traditional one-to-one polling mode makes the polling vector utilization low, how to improve the utilization of the polling vector; (2) The traditional 96-bit ID as a polling vector makes the polling vector length too long, how to shorten the polling vector length.

[0005] The existing methods for collecting tag information in RFID systems are HPP and TPP (see Liu J, Xiao B, Liu X, et al. Efficient Polling-Based Information Collection in RFID Systems [J]. IEEE / ACM Transactions on Networking, 2019, 27(3): 948-961.). First, the Hash Polling Protocol (HPP) can calculate the index of tag selection in advance through hash mapping, and then filter out the single index in which only one tag is selected. After the reader broadcasts all single indexes, the tag responds according to the corresponding index. Then, the Tree-based Polling Protocol (TPP) maps all tags to leaf nodes by establishing a polling tree, thereby avoiding the repeated broadcast of prefixes in the HPP protocol and improving polling efficiency. However, the tag information collection process of the above-mentioned large-scale RFID system is low in efficiency due to the long average polling vector and low polling vector utilization.

[0006] Chinese patent application number CN201510163224.9 describes an efficient RFID-based information collection protocol. The protocol consists of three phases: a first phase extracts a tag sample, a second phase identifies the tags for information collection, and a third phase collects the information from the tags. The invention proposes an algorithm for collecting useful information from a large number of tags integrated with microsensors in a large RFID system. The algorithm first uses the well-known sampling theorem to select a group of tags. It then utilizes a Bloom filter to significantly shorten the tag identification process. Finally, it uses a time-slot zeroing method to collect tag information. This algorithm achieves efficient and low-cost data collection.

[0007] In the Chinese patent application with application number CN200780039223.7, a tag is disclosed that can switch between first and second operating modes and consumes less energy in the first mode. In the second mode, the tag receives a wireless signal having a continuous first and second portion. If the tag recognizes a repeating pattern in the wireless signal, the tag ignores the second portion of the signal. Otherwise, the tag receives the second portion. In another embodiment, in the second mode, the tag can receive a wireless signal having a continuous first, second, and third portion. The tag switches between the first and second modes at a first rate until it detects the first portion, and then switches at a higher rate until it detects the second portion, and then stays in the second mode and receives the third portion.

[0008] In the Chinese patent application with application number CN202011064015.6, a PCIE device, an SR-IOV-based ordered data packet transmission method and system are involved; the method includes: generating N independent DMA read requests; polling the TLP request packets in the N DMA read requests according to a preset polling rule, that is: in the current polling process, the TLP request packets in a DMA read request are collected one by one until all TLP request packets in a DMA read request are collected; then the polling process for the next DMA read request is performed; a sorting tag is assigned to each TLP request packet; the TLP request packet carrying the sorting tag is transmitted to the host at high speed, and the host forms a corresponding TLP completion packet, each TLP completion packet carries requester identification information and a sorting tag; uniformly receiving the TLP completion packets transmitted by the host at high speed through the PCIE core, and uniformly sorting each TLP completion packet according to the sorting tag of each TLP completion packet; and returning the uniformly sorted TLP completion packets in sequence according to the requester identification information carried by each TLP completion packet.

[0009] The above existing technologies are all significantly different from the present invention and fail to solve the technical problem we want to solve. Therefore, we have invented a new sensor-enhanced RFID tag information collection method based on dual polling. Summary of the Invention

[0010] The purpose of the present invention is to provide a sensor-enhanced RFID tag information collection method based on dual polling, which solves the problem of slow tag information collection process leading to low information collection efficiency in the prior art, can improve tag information collection efficiency and shorten collection time.

[0011] The object of the present invention can be achieved by the following technical measures: a sensor-enhanced RFID tag information collection method based on dual polling, the sensor-enhanced RFID tag information collection method based on dual polling comprising:

[0012] Step 1: The reader constructs an original expected frame vector by mapping the tags that have not been queried;

[0013] Step 2: Divide the original expected frame vector into two segments and generate a composite vector through an AND operation;

[0014] Step 3: The reader selects all double indices as polling vectors;

[0015] Step 4: After receiving the polling vector, the tag calculates the hash index value;

[0016] Step 5: Each tag is indexed by hash index X z and the length f of the original expected frame vector OV of the label i The value of determines the reply time slot;

[0017] Step 6: If the tag is queried, it remains silent, otherwise it remains active and waits for the next polling vector.

[0018] The purpose of the present invention can also be achieved by the following technical measures:

[0019] In step 1, multiple rounds of tag information collection are performed. Considering any round of information collection i (1≤i), the reader first collects all n tags that have not been queried before the current round. i The labels are mapped to a length f i To construct an original expected frame vector OV from the array of h i is the upper limit of the hash index length; when there is only one tag mapping, the bit in OV is set to "1", otherwise, the bit is set to "0"; that is, "1" represents the expected single time slot, and "0" represents the expected non-single time slot; therefore, OV is the set of the expected states of all time slots in the current collection round, and the reader can predict the time slot index selected by each tag.

[0020] It is characterized in that, in step 2, the reader divides the original expected frame vector OV into two segments of the same length, and the size of each segment is f i is the length of the original expected frame vector of the label; then, using these two fragments, a synthetic vector SV is generated through a bitwise AND operation.

[0021] In step 3, if the value of a bit in the composite vector SV is "1" after the AND operation, the index of the bit is defined as a double index; after that, the reader can obtain all the double indexes and use them as the polling vector for this round of query tags.

[0022] In step 3, considering any label information collection round i, the length of the original expected frame vector OV is f i , the number of tags that have not been queried before this round is n i ; Then, the probability that a time slot in OV is a desired single time slot is:

[0023]

[0024] According to formula (1), when f i =n i When the expected single slot ratio P s Can reach maximum value That is to say, when hour, Because h i is a positive integer, so if Then we can get P s1 Value:

[0025]

[0026] like Then we have:

[0027]

[0028] Therefore, we can get P s1 and P s2 The difference:

[0029]

[0030] Let equation (4) be 0, then:

[0031]

[0032] By analyzing formula (5), we can get Therefore, in order to make P s Maximize the value of h i The value of is determined by the following formula:

[0033]

[0034] Use n' i Represents the number of polling vectors in this round, then:

[0035]

[0036] Therefore, the expected number of tags that can be queried in this round is 2n' i ; After the end of round i, the number of tags that have not been queried before round (i+1) is:

[0037]

[0038] In step 3, in round i, the size of SV is Then the polling vector length is between 1 and bits; therefore, the expected length of this round of polling vector is:

[0039]

[0040] Then we have:

[0041]

[0042] Subtracting Equation (10) from Equation (9) yields the expected value of the polling vector length:

[0043]

[0044] In step 4, the reader broadcasts a parameter <f i ,R>Query command to start the information collection round; after issuing the Query command, the reader will broadcast the polling vector in turn; once a polling vector is sent, the reader will start two time slots to collect information of a pair of target tags; after receiving the parameter command and the polling vector, the tag z(1≤z≤n i ) by calculating H(ID,R)mod f i Get the original hash index X z , where H() is the hash function deployed in advance, ID is the tag ID, R is the random seed, and f i is the frame length.

[0045] In step 5, each label is z and f i The value of determines the reply slot:

[0046] (1) If X z Less than and is equal to the polling vector, then the tag will reply to the reader in the first of the following two time slots;

[0047] (2) If X z equal to or greater than and is equal to the polling vector, then the tag will reply to the reader in the second of the next two time slots.

[0048] In step 5, the execution time of the i-th round is composed of two parts: the polling time of the reader and the response time of the tag; then the expected execution time of the round is obtained:

[0049]

[0050] Among them, E(h i ) is the expected value of the polling vector length, n' i Indicates the number of polling vectors in this round, t rt is the time it takes for the reader to transmit 1 bit of data to the tag, t tr is the time it takes for a tag to transmit 1 bit of data to a reader, 4t rt It indicates the time required for the reader to broadcast a 4-bit QueryRep command to initialize a time slot. The total execution time of the BDP method is:

[0051]

[0052] Where c≥1 and n c ≠0, n c+1 =0, that is, the reader can query all tags in c rounds.

[0053] In step 6, once a tag is queried, it remains silent during the subsequent query process, while the other tags remain active, waiting for the next polling vector. The above query process is repeated round by round until all tags are queried. In the BDP method, the reader can query a pair of tags with the same dual index, which not only improves the utilization of each hash index but also reduces the average length of the polling vector.

[0054] The purpose of the present invention can also be achieved through the following technical measures: a sensor-enhanced RFID tag information collection system based on dual polling, which uses a sensor-enhanced RFID tag information collection method based on dual polling to collect RFID system tag information.

[0055] The present invention's sensor-enhanced dual-polling RFID tag information collection method comprises the following steps: The tag information collection process consists of multiple rounds. In any round, the reader first constructs an original expected frame vector by mapping unqueried tags. The reader then splits the original expected frame vector into two segments of equal length and combines these segments into a composite vector using a bitwise AND operation. The reader then selects the dual indexes whose values ​​are all "1" and uses them as the polling vector for the tag query in this round. The reader then broadcasts the polling vectors sequentially to count the system's tags, thereby collecting specific tag information. Because each polling vector can query a pair of tags simultaneously, each polling vector sent by the reader opens two time slots to wait for the corresponding tags to respond. Once a tag is queried, it remains silent for the remainder of the query process, while the remaining tags remain active, awaiting the next polling vector. This query process is repeated round after round until all tags have been queried. The present invention effectively improves polling vector utilization and shortens information collection time. Compared to the prior art, the present invention has the following advantages:

[0056] (1) The present invention enables each polling vector to query a pair of tags at the same time, effectively improving the utilization rate of each polling vector.

[0057] (2) Compared with the traditional 96-bit ID polling vector, the present invention can significantly reduce the length of the polling vector, reduce the communication overhead of the reader, and improve the time efficiency of the reader querying the tag. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of the flow of the sensor-enhanced RFID tag information collection method based on dual polling of the present invention;

[0059] Figure 2Schematic diagram of an embodiment of the sensor-enhanced RFID tag information collection method based on dual polling of the present invention;

[0060] Figure 3 Schematic diagram of an embodiment of the sensor-enhanced RFID tag information collection method based on dual polling of the present invention;

[0061] Figure 4 for A schematic diagram comparing the average length of the polling vector of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the change of the number of tags;

[0062] Figure 5 for A schematic diagram comparing the average length of the polling vector of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the number of tags;

[0063] Figure 6 for A schematic diagram comparing the average length of the polling vector of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the change of the number of tags;

[0064] Figure 7 for A schematic diagram comparing the average length of the polling vector of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the change of the number of tags;

[0065] Figure 8 for A schematic diagram comparing the time it takes to query a tag versus the length of the tag's response in a sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and an existing method;

[0066] Figure 9 for A schematic diagram comparing the time it takes to query a tag versus the length of the tag's response in a sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and an existing method;

[0067] Figure 10 for A schematic diagram comparing the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method as the total execution time varies with the number of tags;

[0068] Figure 11 for A schematic diagram comparing the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method as the total execution time varies with the number of tags;

[0069] Figure 12 for A schematic diagram comparing the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method as the total execution time varies with the number of tags;

[0070] Figure 13 for A schematic diagram comparing the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the change of the channel error rate;

[0071] Figure 14 for A schematic diagram comparing the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling according to an embodiment of the present invention and the existing method with the change of the channel error rate;

[0072] Figure 15 for FIG. 4 is a schematic diagram showing a comparison of the total execution time of the sensor-enhanced RFID tag information collection method based on dual polling in an embodiment of the present invention and the existing method as the channel error rate changes. DETAILED DESCRIPTION

[0073] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0074] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations and / or combinations thereof.

[0075] The sensor-enhanced RFID tag information collection method based on dual polling of the present invention adopts a large-scale group RFID system including a background server, a reader and a group of sensor-enhanced RFID tags T={t1, t2, t3, ..., t n}, where n is the number of labels.

[0076] The goal is to efficiently collect l (1≤l≤96) bits of information from all n known tags in the shortest possible time using a polling-based approach. Each tag has a unique 96-bit ID.

[0077] Figure 1 The figure is a flow chart of the sensor-enhanced RFID tag information collection method based on dual polling of the present invention; the sensor-enhanced RFID tag information collection method based on dual polling comprises the following steps:

[0078] Step 1: The reader first constructs an original expected frame vector. This method requires multiple rounds of tag information collection. Consider any round of information collection i (1≤i), the reader first collects all n tags that have not been queried before the current round. i The labels are mapped to a length f i To construct an original expected frame vector OV from the array of h i is the upper limit of the hash index length. A bit in the OV is set to "1" when there is only one tag mapping; otherwise, it is set to "0." In other words, "1" represents a desired single slot, and "0" represents a desired non-single slot. Therefore, the OV is the set of expected states for all slots in the current collection round, and the reader can predict the slot index selected by each tag.

[0079] Step 2, then, the reader splits the OV into two segments of equal length, each of size The two segments are then used to generate a resultant vector SV through a bitwise AND operation.

[0080] In step 3, if the value of a bit in the SV is "1" after the AND operation, the index of that bit is defined as a double index. The reader can then obtain all double indices and use them as the polling vector for this round of querying tags.

[0081] Step 4, then, the reader broadcasts a parameter <f i ,R> to start the information collection round. After issuing the Query command, the reader will broadcast the polling vector in turn. Once a polling vector is sent, the reader will start two time slots to collect information from a pair of target tags. After receiving the parameter command and the polling vector, the tag z (1≤z≤n i ) by calculating H(ID,R)mod f i Get the original hash index X z .

[0082] Step 5. Then, each label is z and f i The value of determines the reply slot:

[0083] (1) If X z Less than and is equal to the polling vector, then the tag will reply to the reader in the first of the next two time slots.

[0084] (2) If X z equal to or greater than and is equal to the polling vector, then the tag will reply to the reader in the second of the next two time slots.

[0085] In step 6, once a tag has been queried, it remains silent for the remainder of the query process, while the remaining tags remain active, waiting for the next poll vector. This query process is repeated round after round until all tags have been queried. In the BDP (Basic Double Polling-based protocol) method, the reader can query a pair of tags with the same double index, which not only improves the utilization of each hash index but also reduces the average length of the poll vector.

[0086] The following are several specific embodiments of the present invention:

[0087] Example 1

[0088] In the first embodiment of the present invention, the sensor-enhanced RFID tag information collection method based on dual polling includes:

[0089] (I) The reader first constructs an original expected frame vector. This method requires multiple rounds of tag information collection. Consider any round of information collection i (1≤i), the reader first passes all n i The labels are mapped to a length f i To construct an original expected frame vector OV from the array of h i is the upper limit of the hash index length. A bit in the OV is set to "1" when there is only one tag mapping; otherwise, it is set to "0." In other words, "1" represents a desired single slot, and "0" represents a desired non-single slot. Therefore, the OV is the set of expected states for all slots in the current collection round, and the reader can predict the slot index selected by each tag.

[0090] (2) The reader then divides the OV into two segments of equal length, each of which is The two segments are then used to generate a resultant vector SV through a bitwise AND operation.

[0091] (3) If the value of a bit in SV is "1" after the AND operation, the index of this bit is defined as a double index. After that, the reader can obtain all double indexes and use them as the polling vector for this round of querying tags.

[0092] Consider any label information collection round i, the length of the original expected frame vector OV is f i , the number of tags that have not been queried before this round is n i Then, the probability that a time slot in OV is a desired single time slot is:

[0093]

[0094] According to formula (1), when f i =n i When the expected single slot ratio P s Can reach maximum value That is to say, when hour, Because h i is a positive integer, so if Then we can get P s1 Value:

[0095]

[0096] like Then we have:

[0097]

[0098] Therefore, we can get P s1 and P s2 The difference:

[0099]

[0100] Let equation (4) be 0, then:

[0101]

[0102] By analyzing formula (5), we can get Therefore, in order to make P s Maximize the value of h i The value of is determined by the following formula:

[0103]

[0104] Use n' i Represents the number of polling vectors in this round, then:

[0105]

[0106] Therefore, the expected number of tags that can be queried in this round is 2n' i After the end of round i, the number of tags that have not been queried before round (i+1) is:

[0107]

[0108] In round i, the size of SV is Then the polling vector length is between 1 and Therefore, the expected length of this round of polling vector is:

[0109]

[0110] Then we have:

[0111]

[0112] Subtracting Equation (10) from Equation (9) yields the expected value of the polling vector length:

[0113]

[0114] (IV) Then, the reader broadcasts a message with a parameter of <f i ,R> to start the information collection round. After issuing the Query command, the reader will broadcast the polling vector in turn. Once a polling vector is sent, the reader will start two time slots to collect information from a pair of target tags. After receiving the parameter command and the polling vector, the tag z (1≤z≤n i ) by calculating H(ID,R)mod f i Get the original hash index X z .

[0115] (5) Then, each tag is z and f i The value of determines the reply slot:

[0116] (1) If X z Less than and is equal to the polling vector, then the tag will reply to the reader in the first of the next two time slots.

[0117] (2) If X z equal to or greater than and is equal to the polling vector, then the tag will reply to the reader in the second of the next two time slots.

[0118] (6) Once a tag is queried, it remains silent for the remainder of the query process, while the remaining tags remain active, waiting for the next poll vector. This query process is repeated round after round until all tags have been queried. In the BDP method, the reader can query a pair of tags with the same dual index, which not only improves the utilization of each hash index but also reduces the average length of the poll vector.

[0119] The execution time of round i consists of two parts: the polling time of the reader and the response time of the tag. Then we get the expected execution time of the round:

[0120]

[0121] Among them, t rt is the time it takes for the reader to transmit 1 bit of data to the tag, t tr is the time it takes for a tag to transmit 1 bit of data to a reader, 4t rt It indicates the time required for the reader to broadcast a 4-bit QueryRep command to initialize a time slot. The total execution time of the BDP method is:

[0122]

[0123] Where c≥1 and n c ≠0, n c+1 =0, that is, the reader can query all tags in c rounds.

[0124] Example 2

[0125] In a specific embodiment 2 of the present invention, see Figure 2 , is any round of query process of the BDP method, assuming that there are 12 tags that have not been read before this round, and the original frame length is 16. The reader first constructs OV by hashing the IDs of all tags. The bits mapped by tags 1, 2, 3, 6, 7, 9, 11, and 12 are set to "1", and the remaining bits are set to "0". The reader then divides the OV into two fragments and combines them into a synthetic vector SV through a bitwise AND operation. According to the bits with the value of "1" in SV, the reader obtains all double indexes as polling vectors, namely "12", "112", and "1102". The reader then broadcasts the first polling vector "1" to query the corresponding two tags. Since the hash result of tag 1 is 1, which is equal to the polling vector, tag 1 will reply its 1-bit information to the reader in the next first time slot. The hash value of tag 7 is 9. and Tag 7 will then reply to the reader in the second time slot. Similarly, tags 3, 9, 6, and 11 will reply to the reader in the corresponding time slots after the polling vectors "112" and "1102" are broadcast, respectively. These tags will remain silent for the remainder of the time, while the remaining tags will participate in the next round of information collection.

[0126] When the number of known tags in a large-scale RFID system changes from 1,000 to 10,000, as Figure 4 、 5 As shown in Figure 6, it can be seen that the average polling vector length of the HPP and BDP methods increases with the increase in the number of tags. This is because hash indexes are used to query tags in the HPP and BDP methods, and the hash index length will increase with the increase in the number of tags, and accordingly, the polling vector length will also increase. However, the average polling vector length of the TPP method remains almost unchanged with the increase in the number of tags. The polling vector of the TPP method is part of a binary tree branch. The channel error rate will affect the average polling vector length because when a channel error occurs, the reader needs to rebroadcast the entire polling vector, which will cause additional polling overhead. The polling vector length of the BDP method proposed in this invention is shorter than that of the HPP and TPP methods. Therefore, when the channel bit error rate is large, the method proposed in this chapter will perform better.

[0127] Figure 7 、 8 Figure 9 shows the execution time for querying an RFID tag under different channel error rates as the tag reply length increases. If the wireless channel is ideal, all methods query a tag in less than 1 millisecond when collecting 8 bits of information. When the reader collects 96 bits of information, the query time remains less than 4 milliseconds. In ideal wireless channels, the BDP method requires less time to query a tag than the HPP and TPP methods. This is primarily due to the BDP method's use of dual indexes to query a pair of tags, thereby improving the utilization of each polling vector. Notably, the BDP method achieves a shorter query time than the HPP and TPP methods when the channel error rate is 20%. When a time slot is corrupted, the IPP method requires rebroadcasting the entire polling vector to query the tag, which is very time-consuming. In contrast, the BDP method uses dual hash indexes to query a pair of tags. In this case, both the length and number of hash indexes are reduced, significantly improving collection efficiency. Therefore, when the channel error rate is above 10%, the BDP method exhibits superior performance compared to the HPP and TPP methods, indicating that BDP is capable of performing better in complex communication environments.

[0128] Under different channel error rates, as the number of tags increases, the execution time of all methods is as follows: Figure 10 、 11As shown in , 12, the BDP method is better than the HPP and TPP methods. Figure 11 In the study, when the channel error rate is high, the BDP method significantly outperforms the HPP and TPP methods, demonstrating that the BDP method has stronger anti-interference capabilities than the HPP and TPP methods. In the HPP and TPP methods, if a time slot is corrupted, the previously broadcast polling vector becomes useless for tag queries. However, the polling vector in the method proposed in this invention can query a pair of tags at a time, so the polling vector only becomes invalid when the next two time slots are corrupted. Therefore, the time overhead of transmitting all polling vectors can also be reduced. In other words, the more complex the environment, the better the effectiveness of the method proposed in this invention.

[0129] Figure 13 、 14 Figures 15 and 16 demonstrate the impact of channel errors on execution time. It can be seen that when the channel error rate is greater than 10%, the BDP method outperforms the other methods. As the channel error rate increases, the execution time of all methods also increases because they need to rebroadcast the entire polling vector to query a tag. However, the proposed method can mitigate this impact because each polling vector can be used to query a pair of tags.

[0130] From the above, it can be seen that the sensor-enhanced RFID tag information collection method based on dual polling provided by the present invention can greatly reduce the tag information collection time, improve the tag inventory efficiency, and effectively complete the tag information collection of large-scale RFID systems.

[0131] Example 3

[0132] In the specific embodiment 3 of the present invention, see Figure 3 , is any round of query process of the BDP method, assuming that there are 6 tags that have not been read before this round, and the original frame length is 12. The reader first constructs OV by hashing the IDs of all tags. All tag mapping bits are set to "1" and the remaining bits are set to "0". The reader then divides the OV into two fragments and combines them into a synthetic vector SV through bitwise AND operation. According to the bits with the value of "1" in SV, the reader obtains all double indexes as the polling vector, that is, "11". The reader then broadcasts the polling vector "11" to query the corresponding two tags. Since the hash result of tag 2 is 3, which is equal to the polling vector, tag 1 will reply its 1-bit information to the reader in the next first time slot. The hash value of tag 5 is 9. and Then tag 5 will reply to the reader in the next second time slot. Tags 2 and 5 will remain silent for the rest of the time, while the rest of the tags will participate in the next round of information collection process.

[0133] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art may modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features therein. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

[0134] Except for the technical features described in the specification, all other technical features are known technologies to those skilled in the art.

Claims

1. A sensor-enhanced RFID tag information collection method based on dual polling, characterized in that: The sensor-enhanced RFID tag information collection method based on dual polling includes: Step 1: The reader constructs an original expected frame vector by mapping the tags that have not been queried; Step 2: Divide the original expected frame vector into two segments and generate a composite vector through an AND operation; Step 3: The reader selects all double indices as polling vectors; Step 4: After receiving the polling vector, the tag calculates the hash index value; Step 5: Each tag is indexed by hash index X z and the length f of the original expected frame vector OV of the label i The value of determines the reply time slot; Step 6: If the tag is queried, it remains silent, otherwise it remains active and waits for the next polling vector.

2. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 1, multiple rounds of tag information collection are performed. Considering any round of information collection i (1≤i), the reader first collects all n tags that have not been queried before the current round. i The labels are mapped to a length f i To construct an original expected frame vector OV from the array of h i is the upper limit of the hash index length; When there is only one tag mapped, the bit in OV is set to "1", otherwise, the bit is set to "0"; that is, "1" represents the expected single time slot, and "0" represents the expected non-single time slot; therefore, OV is the collection of the expected states of all time slots in the current collection round, and the reader can predict the time slot index selected by each tag.

3. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 2, the reader divides the original expected frame vector OV into two segments of the same length, each of which is f i is the length of the original expected frame vector of the label; then, using these two fragments, a synthetic vector SV is generated through a bitwise AND operation.

4. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 3, if the value of a bit in the composite vector SV is "1" after the AND operation, the index of the bit is defined as a double index; then, the reader can obtain all the double indexes and use them as the polling vector for this round of querying tags.

5. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 4, characterized in that: In step 3, considering any label information collection round i, the length of the original expected frame vector OV is f i , the number of tags that have not been queried before this round is n i ; Then, the probability that a time slot in OV is a desired single time slot is: According to formula (1), when f i =n i When the expected single slot ratio P s Can reach maximum value That is to say, when hour, Because h i is a positive integer, so if Then we can get P s1 Value: like Then we have: Therefore, we can get P s1 and P s2 The difference: Let equation (4) be 0, then: By analyzing formula (5), we can get Therefore, in order to make P s Maximize the value of h i The value of is determined by the following formula: Use n′ i Represents the number of polling vectors in this round, then: Therefore, the expected number of tags that can be queried in this round is 2n′ i ; After the end of round i, the number of tags that have not been queried before round (i+1) is:

6. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 5, characterized in that: In step 3, in round i, the size of SV is Then the polling vector length is between 1 and bits; therefore, the expected length of this round of polling vector is: Then we have: Subtracting Equation (10) from Equation (9) yields the expected value of the polling vector length:

7. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 4, the reader broadcasts a parameter <f i ,R>Query command to start the information collection round; after issuing the Query command, the reader will broadcast the polling vector in turn; once a polling vector is sent, the reader will start two time slots to collect information of a pair of target tags; after receiving the parameter command and the polling vector, the tag z(1≤z≤n i ) by calculating H(ID,R)mod f i Get the original hash index X z , where H() is the hash function deployed in advance, ID is the tag ID, R is the random seed, and f i is the frame length.

8. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 5, each label is z and f i The value of determines the reply slot: (1) If X z Less than and is equal to the polling vector, then the tag will reply to the reader in the first of the following two time slots; (2) If X z equal to or greater than and is equal to the polling vector, then the tag will reply to the reader in the second of the next two time slots.

9. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 8, characterized in that: In step 5, the execution time of the i-th round is composed of two parts: the polling time of the reader and the response time of the tag; then the expected execution time of the round is obtained: Among them, E(h i ) is the expected value of the polling vector length, n′ i Indicates the number of polling vectors in this round, t rt is the time it takes for the reader to transmit 1 bit of data to the tag, t tr is the time it takes for a tag to transmit 1 bit of data to a reader, 4t rt It indicates the time required for the reader to broadcast a 4-bit QueryRep command to initialize a time slot. The total execution time of the BDP method is: Where c≥1 and n c ≠0, n c+1 =0, that is, the reader can query all tags in c rounds.

10. The sensor-enhanced RFID tag information collection method based on dual polling according to claim 1, characterized in that: In step 6, once a tag is queried, it remains silent during the subsequent query process, while the other tags remain active, waiting for the next polling vector. The above query process is repeated round by round until all tags are queried. In the BDP method, the reader can query a pair of tags with the same dual index, which not only improves the utilization of each hash index but also reduces the average length of the polling vector.

11. A sensor-enhanced RFID tag information collection system based on dual polling, characterized in that: The sensor-enhanced RFID tag information collection system based on dual polling adopts the sensor-enhanced RFID tag information collection method based on dual polling as described in any one of claims 1 to 10 to collect RFID system tag information.

Images