Reader anti-collision method based on graph coloring and dual-channel transmission

Abstract

The invention provides a reader anti-collision method (GC-DCT) based on graph coloring and dual-channel transmission, working periods are allocated for readers by using a DSatur graph coloring algorithm, and when a reader with a certain color works, other readers communicate through a control channel, so that the readers without collision constraint also participate in a working state. Compared with a DiCa algorithm and an NFRA algorithm, the method has certain advantages in stability and throughput in a reader high-density environment.

Description

Technical Field

[0001] The invention belongs to the technical field of wireless radio frequency identification, and in particular relates to a reader anti-collision method based on graph coloring and dual-channel transmission. Background Art

[0002] As an advanced automatic identification technology, RFID technology has been widely used in various fields in recent years. In logistics management, it has the advantages of large reading range, ability to identify multiple tags at the same time, and the ability to store more information (see references [1]-[3]), which greatly improves the efficiency of warehouse management and transportation management.

[0003] When the signal ranges of adjacent readers overlap, the reader cannot read the tag normally, which affects the efficiency and reliability of the system (see reference [4]). Therefore, how to effectively solve the reader collision problem has become one of the keys to improving the performance of the RFID system.

[0004] At present, the reader anti-collision algorithm is mainly divided into methods based on scheduling time slots, carrier monitoring and adjusting transmission power. The NFRA algorithm proposed in reference [5] uses a server to assign work to readers. Each reader generates a 1-N random number. The reader compares its own random number with the OC value of the server's subscription command. If the two are the same, it sends a beacon to detect whether there is a collision. The reader that does not detect a collision starts working. The DiCa algorithm proposed in reference [6] adopts a dual-channel mechanism. The reader sends and receives signals on the control channel to detect whether there are other readers working in the same time and area. If a collision signal is detected, the reader will reselect the working time. Reference [7] proposes a hybrid anti-collision protocol that combines frequency division multiple access and time division multiple access. This protocol effectively reduces interference between readers by optimizing time slots and frequency resource management. Reference [8] proposes a reader anti-collision algorithm based on distributed machine learning. This method collects the working parameters of each reader and then uses a distributed learning algorithm to analyze and optimize them in real time, thereby dynamically adjusting the reader's working frequency and time slot to minimize signal conflicts.

[0005] Reference [9] proposed a transmission power adjustment algorithm that dynamically adjusts the power of the reader by monitoring the status of surrounding readers, allowing more readers to work in the same time slot, thereby improving the system throughput. Summary of the invention

[0006] In order to solve the collision problem between multiple readers, the present invention provides a reader anti-collision method based on graph coloring and dual-channel transmission (GC-DCT). The DSatur graph coloring algorithm is used to allocate working time periods for readers. When a reader of a certain color is working, the other readers will communicate through the control channel so that readers without collision constraints can also participate in the working state. Compared with the DiCa algorithm and the NFRA algorithm, the present invention has certain advantages in stability and throughput in a high-density reader environment.

[0007] The present invention adopts the following technical solution:

[0008] The reader network is considered as an undirected graph G = (V, E) (see references

[10] -

[11] ), where V is a finite set of vertices in G (a finite set of readers); E is a finite set of edges connecting two different vertices (readers) in V;<i,j> Represents an undirected edge between vertex i and vertex j. Each vertex represents a reader, and the edges in graph G are collision constraints between readers. Readers that have collision constraints with a certain reader are called neighbor readers.

[0009] A reader anti-collision method based on graph coloring and dual-channel transmission, comprising:

[0010] Step 1: First use the DSatur algorithm to color the reader network. The readers are divided into several colors, and readers of each color are assigned to different working time periods.

[0011] Step 2: Select a reader of a certain color to enter the working state. If a reader in the working state receives a request signal in the control channel, it sends a busy signal to tell the neighboring readers that it is working.

[0012] Step 3: The remaining readers send request signals on the control channel to indicate that they want to participate in the work state.

[0013] Step 4: If a busy signal is received, it means that a neighbor reader is working, and the reader enters the rest state and no longer sends request signals during the current working time slice.

[0014] Step 5: If no busy signal is received, it means that no neighbor reader is working. If no request signal is received at this time, no neighbor reader is "competing" with itself, and it can participate in the working state; but if a request signal is received, it means that there are idle readers who also "compete" with itself.

[0015] Step 6: If a request signal is received, a random delay is used to avoid conflicts, otherwise multiple idle readers with collision constraints will participate in the working state at the same time. After the delay is over, send the request signal again.

[0016] The DSatur algorithm is a classic graph coloring algorithm that aims to effectively assign the minimum number of colors to a given graph so that adjacent vertices have different colors. The core idea of ​​the DSatur algorithm is to select the next vertex to be colored based on the saturation of the vertex (that is, the number of adjacent colored vertices) to minimize the number of colors used.

[0017] In graph theory, the number of colors required to color the vertices of any simple graph does not exceed the maximum degree of the graph plus 1. For a graph G, let its maximum degree be Δ(G), then the graph can be colored with at most Δ(G)+1 colors. Each reader is included in the color set color_i, i<=n, n is the total number of colors of the graph after coloring, and from the above, we can see that n ≤ Δ(G)+1, stipulates that the size of i is the size of the number of colors.

[0018] The specific steps of using the DSatur algorithm to color graph G are as follows:

[0019] Step 1: Initialize the color of each vertex in G to empty, and initialize several empty color sets color1-color_n.

[0020] Step 2: Select the vertex with the highest degree and color it as color 1, and add it to the set color_1. If there are multiple vertices with the highest degree, select one randomly.

[0021] Step 3: Calculate the saturation of the uncolored vertex, that is, the number of vertices associated with this vertex that have been colored.

[0022] Step 4: Select the vertex with the highest saturation, assign it a color that is the smallest and not used by its neighboring vertices, and add it to the corresponding set. If there are multiple vertices with the same saturation, select the vertex with the largest degree; if the degrees are also the same, select one randomly.

[0023] Step 5: Repeat steps 3 and 4 until all vertices are colored.

[0024] Beneficial effects of the present invention:

[0025] The present invention proposes a reader anti-collision method based on graph coloring and dual-channel transmission, aiming to improve the throughput of the reader network. The present invention divides the reader network into multiple working groups. When each group of readers is working, the remaining readers without collision constraints are allowed to participate in the working state as much as possible, thereby improving the throughput of the reader network. The invention can effectively solve the collision problem of the reader network and significantly improve the throughput and stability of the system. Simulation results show that the advantages of the present invention will be more obvious in a high-density reader environment. Therefore, the present invention is more suitable for use in a dense reader environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a flowchart of the algorithm steps of the present invention;

[0027] Figure 2 It is the flow chart of DSatur algorithm;

[0028] Figure 3 This is an example diagram of the DSatur algorithm;

[0029] Figure 4 is the system throughput comparison curve;

[0030] Figure 5 is the system variance comparison curve. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] Using the DSatur algorithm to color the reader network, if the readers are colored in sequence, all readers outside the color of the work group will enter a resting state, which will waste the resources of readers that have no collision constraints with the work group. Figure 3 In the figure, when the blue reader enters the working state, readers A and F are in the resting state, and there is no collision constraint between these readers and the blue reader.

[0033] In order to solve this resource waste problem, a dual-channel mechanism is introduced (see reference

[12] ), which divides the communication channel into a data channel and a control channel. The data channel is used for communication between readers and tags; the control channel is used for communication between readers. The communication range of the control channel must be large enough to ensure that any two readers with collision constraints can communicate through the control channel. The communication of the two channels will not interfere with each other. Any readers that would interfere with each other on the data channel can communicate on the control channel. For example, Figure 3 In this example, although A and B may interfere with each other on the data channel, they can communicate on the control channel.

[0034] When a reader of a certain color is working, in order for other readers without collision constraints to participate, they need to communicate on the control channel. The remaining readers send request signals on the control channel to participate in the "competition". If a reader in the working group receives a request signal, it will send a busy signal. The reader that receives the busy signal means that it has a collision constraint with some readers in the working group, so it will no longer participate in the "competition" in this round; the reader that does not receive the busy signal means that it has no collision constraint with any reader in the working group, but these readers cannot all enter the working state at the same time, because there may be collision constraints between these readers, so it is necessary to adopt a random delay method (see reference

[13] ) to avoid this situation.

[0035] like Figure 1 As shown, a reader anti-collision method based on graph coloring and dual-channel transmission of the present invention comprises:

[0036] Step 1: First use the DSatur algorithm to color the reader network. The readers are divided into several colors, and readers of each color are assigned to different working time periods.

[0037] Step 2: Select a reader of a certain color to enter the working state. If a reader in the working state receives a request signal in the control channel, it sends a busy signal to tell the neighboring readers that it is working.

[0038] Step 3: The remaining readers send request signals on the control channel to indicate that they want to participate in the work state.

[0039] Step 4: If a busy signal is received, it means that a neighbor reader is working, and the reader enters the rest state and no longer sends request signals during the current working time slice.

[0040] Step 5: If no busy signal is received, it means that no neighbor reader is working. If no request signal is received at this time, no neighbor reader is "competing" with itself, and it can participate in the working state; but if a request signal is received, it means that there are idle readers who also "compete" with itself;

[0041] Step 6: If a request signal is received, a random delay is used to avoid conflicts, otherwise multiple idle readers with collision constraints will participate in the working state at the same time. After the delay is over, send the request signal again.

[0042] For example, in Figure 3 In the example, when the blue reader is the working group, the five readers B, D, G, H, and I all have collision constraints with the blue reader, so they no longer participate in the "competition" in this round. At this time, only three readers A, C, and F are left. These three readers have received the request signal, so they enter the random delay. Assuming that A ends the delay first and sends the request signal, it can enter the working state when it finds that there is no reader competing with it; after C ends the delay and sends the request signal, it will receive the busy signal sent by A, so it enters the resting state; similarly, F can also enter the working state after the delay ends. The working order of the entire reader network is: {A, D, F, H}, {B, C, G, I}, {A, F, E, J}.

[0043] The DSatur algorithm is a classic graph coloring algorithm that aims to effectively assign the minimum number of colors to a given graph so that adjacent vertices have different colors. The core idea of ​​the DSatur algorithm is to select the next vertex to be colored based on the saturation of the vertex (that is, the number of adjacent colored vertices) to minimize the number of colors used.

[0044] In graph theory, the number of colors required to color the vertices of any simple graph does not exceed the maximum degree of the graph plus 1. For a graph G, let its maximum degree be Δ(G), then the graph can be colored with at most Δ(G)+1 colors. Each reader is included in the color set color_i, i<=n, n is the total number of colors of the graph after coloring, and from the above, we can see that n≤Δ(G)+1, and the size of i is the size of the number of colors.

[0045] like Figure 2 As shown, the specific steps of using the DSatur algorithm to color the graph G are as follows:

[0046] Step 1: Initialize the color of each vertex in G to empty, and initialize several empty color sets color1-color_n.

[0047] Step 2: Select the vertex with the highest degree and color it as color 1, and add it to the set color_1. If there are multiple vertices with the highest degree, select one randomly.

[0048] Step 3: Calculate the saturation of the uncolored vertex, that is, the number of vertices associated with this vertex that have been colored.

[0049] Step 4: Select the vertex with the highest saturation, assign it a color that is the smallest and not used by its neighboring vertices, and add it to the corresponding set. If there are multiple vertices with the same saturation, select the vertex with the largest degree; if the degrees are also the same, select one randomly.

[0050] Step 5: Repeat steps 3 and 4 until all vertices are colored.

[0051] For example, in Figure 3 In the figure, the degree of the graph is 4, Δ(G) = 4, so the value of n is 5, that is, initialize several sets color_1-color_5. Assume that color_1 is yellow, color_2 is red, and color_3 is blue. After coloring with the DSatur algorithm, the color set obtained is: color_1 = {A, D, F, H}, color_2 = {B, C, G, I}, color_3 = {E, J}, color_4 = {}, color_5 = {}. Since color_4 and color_5 are empty sets, they are discarded.

[0052] Simulation test:

[0053] In order to evaluate the performance of the proposed reader anti-collision algorithm, a series of simulation experiments were carried out, and the efficiency and stability of the algorithm were measured by two key indicators: system throughput and variance.

[0054] work i represents the number of times the i-th reader works; time represents the simulation time; Represents the average number of times each reader works, that is,

[0055]

[0056] The system throughput is defined as the number of reader operations per second, i.e.

[0057]

[0058] The variance of the system is defined as:

[0059]

[0060] The simulation experiment is set up as follows: the readers are randomly distributed in a 30×30m warehouse. In order to test the performance of the reader network under different densities, the number of readers is set to 100, 150, 200, 250, 300, 350, 400, 450, 500 respectively; the recognition radius of the reader is 3m; the communication cycle between the reader and the tag (data channel) is 0.46s; the communication cycle between readers and readers (control channel) is 0.01s. By adjusting the number of readers, the layout of readers and tags and other parameters, the system throughput and variance under different conditions are observed and recorded.

[0061] System throughput is an important indicator to measure the number of tags that the algorithm successfully identifies per unit time. The throughput of the proposed algorithm is compared with that of DiCa and NFRA algorithms. The experimental results are as follows: Figure 4 As shown. Under any density of readers, the throughput of the proposed algorithm is significantly higher than that of the other two algorithms, and as the number of readers increases, the throughput advantage of the proposed algorithm becomes more obvious. Specifically, when the number of readers is 300, the throughput of the proposed algorithm is about 22.13% higher than that of the DiCa algorithm and about 56.84% higher than that of the NFRA algorithm; when the number of readers is 500, the throughput of the proposed algorithm is about 35.63% higher than that of the DiCa algorithm and about 59.56% higher than that of the NFRA algorithm; on average, it is about 26.96% higher than the DiCa algorithm and about 53.15% higher than the NFRA algorithm. It can be seen that the proposed algorithm greatly increases the throughput of the system by adjusting more readers of non-group colors to the working state.

[0062] In addition to throughput, we also calculated the system variance to evaluate the stability of the algorithm under different conditions. The smaller the variance, the smaller the difference in the number of readers working, that is, the algorithm has better stability. Figure 5 As shown in the figure. As can be seen from the figure, with the increase in the number of readers, the system variance of the GC-DCT algorithm gradually decreases, and begins to be lower than the other two algorithms when the number of readers reaches 400. In particular, when the number of readers is 500, the system variance of the algorithm in this paper is significantly lower than that of the other two algorithms, indicating that in high-density environments, the GC-DCT algorithm exhibits better stability. Although the system variance of the algorithm in this paper is higher than that of other algorithms in low-density environments, as the density increases, the GC-DCT algorithm can better adapt to complex reader environments and maintain a lower variance. In actual RFID system deployments, high-density environments are common challenges. Due to the low variance advantage of the algorithm in this paper in high-density reader environments, it has more practical application value.

[0063] In summary, the proposed algorithm outperforms the DiCa algorithm and the NFRA algorithm in terms of system throughput at various reader densities; in terms of system variance, it also outperforms the DiCa algorithm and the NFRA algorithm at higher density. In the case of high reader density, the proposed algorithm can not only significantly improve the system throughput, but also maintain a low variance, indicating that it is more suitable for applications in high-density environments.

[0064] References

[0065] [1]XIAOLIN JIA,QUANYUAN FENG,LISHAN YU.Stability Analysis of anEfficient Anti-Collision Protocol for RFID Tag Identification[J].IEEETransactions on Communications,2012,60(8):2285-2294.

[0066] [2]XIAOLIN JIA,BOLIC,MIODRAG,et al.An Efficient Dynamic Anti-Collision Protocol for Mobile RFID Tags Identification[J].IEEE communicationsletters:A publication of the IEEE Communications Society,2019,23(4):620-623.

[0067] [3]XIAOLIN JIA,QUANYUAN FENG,TAIHUA FAN.RFID Technology and ItsApplications in Internet of Things(IOT)[C].2012 2nd International Conference on Consumer Electronics.2012:1282-1285.

[0068] [4]B.Cao,S.Yang,J.Zhao,et al.Using parallel particle swarmoptimization for RFID reader-to-reader anti-collision[C].2018 13th IEEEConference on Industrial Electronics and Applications(ICIEA).2018:1250-1254.

[0069] [5]JUN-BONG EOM,SOON-BIN YIM,TAE-JIN LEE.An Efficient ReaderAnticollision Algorithmin Dense RFID Networks With Mobile RFID Readers[J].IEEE Transactions on Industrial Electronics,2009,56(7):2326-2336.

[0070] [6]Distributed Tag Access with Collision-Avoidance among Mobile RFIDReaders[C]. / / 2009International Conference on Computational Science andEngineering(CSE 2009).2009:621-626.

[0071] [7]Ouadou M,Mafamane R,Minaoui K.A Hybrid Anti-Collision ProtocolBased on Frequency Division Multiple Access(FDMA)and Time Division MultipleAccess(TDMA)for Radio Frequency Identification(RFID)Readers[J].Network,2024,4(2):217-236.

[0072] [8]Mafamane R, Ouadou M, Sahbani H, et al.DMLAR: Distributed MachineLearning-Based Anti-Collision Algorithm for RFID Readers in the Internet ofThings[J]. Computers, 2022, 11(7):123-135.

[0073] [9] Xu Yafeng, Cui Yinghua. Power control anti-collision algorithm based on reader status[J]. Computer Engineering and Design, 2019, 40(3): 607-610.

[0074]

[10] Liu Xiaonan, Liu Zhengyu, Xie Haoshan, et al. Solving graph coloring problem based on Grover algorithm[J]. Computer Science, 2023, 50(6): 351-357.

[0075]

[11] Xu Yafeng, Cui Yinghua. Anti-collision algorithm for RFID reader based on graph theory[J]. Journal of Computer Applications, 2017, 37(8): 2163-2167.

[0076]

[12] Dupin N.Matheuristic variants of DSATUR for the vertex coloringproblem[C] / / Metaheuristics International Conference.Cham:Springer NatureSwitzerland,2024:96-111.

[0077]

[13] Wei Xin. Research on anti-collision algorithm for RFID tags and readers[D]. Chengdu: University of Electronic Science and Technology of China, 2009.

[0078] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A reader anti-collision method based on graph coloring and dual-channel transmission, characterized in that ,include: Step 1: First, use the DSatur algorithm to color the reader network. The readers are divided into several colors, and readers of each color are assigned to different working time periods. Step 2: Select a reader of a certain color to enter the working state. If a reader in the working state receives a request signal in the control channel, it will send a busy signal to tell the neighboring readers that it is working; Step 3: The remaining readers send a request signal on the control channel to indicate that they want to participate in the work state; Step 4: If a busy signal is received, it means that a neighbor reader is working, and the reader enters the rest state and no longer sends request signals during the current working time slice; Step 5: If no busy signal is received, it means that no neighbor reader is working. If no request signal is received at this time, no neighbor reader is "competing" with itself, and it can participate in the working state; but if a request signal is received, it means that there are idle readers who also "compete" with itself; Step 6: If a request signal is received, a random delay is used to avoid conflicts, otherwise multiple idle readers with collision constraints will participate in the working state at the same time. After the delay is over, send the request signal again.

2. The DStaur algorithm according to claim 1, characterized in that The DSatur algorithm is a classic graph coloring algorithm that aims to effectively assign the minimum number of colors to a given graph so that adjacent vertices have different colors. The core idea of ​​the DSatur algorithm is to select the next vertex to be colored based on the saturation of the vertex (that is, the number of adjacent colored vertices) to minimize the number of colors used.

3. The reader anti-collision method based on graph coloring and dual-channel transmission according to claim 1 is characterized in that: In graph theory, the number of colors required for vertex coloring of any simple graph does not exceed the maximum degree of the graph plus 1. For a graph G, let its maximum degree be Δ(G), then the graph can be colored with at most Δ(G)+1 colors. Each reader is included in the color set color_i, i<=n, n is the total number of colors of the graph after coloring, n<=Δ(G)+1, and the size of i is the size of the number of colors.

4. The specific process of coloring the graph G according to the DSatur algorithm described in claim 1 is as follows: Step 1: Initialize the color of each vertex in G to empty, and initialize several empty color sets color1-color_n; Step 2: Select the vertex with the highest degree and color it as color 1, and add it to the set color_1. If there are multiple vertices with the highest degree, select one randomly; Step 3: Calculate the saturation of the uncolored vertex, that is, the number of vertices associated with this vertex that have been colored. ; Step 4: Select the vertex with the highest saturation, assign it a minimum color that is not used by its neighboring vertices, and add it to the corresponding set. If there are multiple vertices with the same saturation, select the vertex with the largest degree. If the degrees are also the same, select one randomly. Step 5: Repeat steps 3 and 4 until all vertices are colored.

5. The reader anti-collision method based on graph coloring and dual-channel transmission according to claim 1, characterized in that: The communication range of the control channel must be large enough to ensure that any two readers with collision constraints can communicate through the control channel, and the communications of the data channel and the control channel will not interfere with each other.

Images