A conflict resolution method based on DQRA-MIMO system

Abstract

The application discloses a conflict resolution method based on a DQRA-MIMO system, which comprises the following steps: step one: when a user accesses for the first time, an access request sequence is randomly selected in a time slot, a base station feeds back a time slot competition result, and the total number of users is estimated according to the time slot competition result; step two: based on the estimated total number of users, the resource utilization rate is maximized to determine the optimal resource quantity; step three: the number of users in a conflict queue and the number of groups are estimated, the number of users in each group is calculated, and the access time slot is dynamically grouped and adapted according to the expected group number; and step four: when subsequent users access, the users are combined in batches according to the estimated total number of users to reduce the conflict probability, and the users in the conflict queue are processed in cycles according to step three until the data transmission is completed. Through user number estimation, resource optimization configuration and dynamic grouping and adaptation of the conflict queue, the application solves the problems of access conflict and resource waste, and improves the transmission efficiency of the system.

Description

Technical Field

[0001] This invention relates to the field of mobile communication technology, and in particular to a conflict resolution method based on a DQRA-MIMO system. Background Technology

[0002] In recent years, with the development of IoT technology, machine-to-machine (M2M) communication services have become increasingly widespread. The demand for small data packet transmission from a large number of users has placed higher demands on the conflict resolution and resource utilization efficiency of wireless access systems. Traditional access systems often suffer from resource waste and low transmission efficiency when handling a large number of uncertain user accesses due to user-resource mismatch. For example, when users in a conflict queue access the system sequentially, resource allocation can become unreasonable after a cycle, reducing data transmission efficiency. While Dynamic Queuing Random Access (DQRA) technology can resolve access conflicts, it lacks mechanisms for user number estimation and dynamic resource adaptation. Multiple-input multiple-output (MIMO) technology can improve transmission capacity, but it needs to be matched with access conflict resolution capabilities. Existing DQRA-MIMO systems cannot adjust resource allocation in real time when the number of users changes dynamically, resulting in underutilization of MIMO antenna resources and low conflict resolution efficiency. Therefore, an efficient conflict resolution method is urgently needed to achieve user number estimation, optimized resource allocation, and dynamic packet adaptation to improve system resource utilization and transmission efficiency. Summary of the Invention

[0003] The purpose of this invention is to provide a conflict resolution method based on a DQRA-MIMO system. This invention solves access conflicts and resource waste by estimating the number of users, optimizing resource allocation, and dynamically adapting conflict queues, thereby improving system transmission efficiency.

[0004] The technical solution of this invention: a conflict resolution method based on a DQRA-MIMO system, comprising the following steps:

[0005] Step 1: When a user accesses the network for the first time, the system randomly selects a time slot to send an access request sequence. The base station then feeds back the time slot contention results and estimates the total number of users based on these results.

[0006] Step 2: Based on the estimated total number of users, maximize resource utilization to determine the optimal amount of resources;

[0007] Step 3: Estimate the number of users and groups in the conflict queue, calculate the number of users in each group, and dynamically group and adapt access time slots according to the expected number of groups;

[0008] Step 4: When subsequent users connect, they are grouped in batches according to the estimated total number of users to reduce the probability of conflict. Users in the conflict queue are processed in a loop according to Step 3 until the data transmission is completed.

[0009] In the above-mentioned conflict resolution method based on the DQRA-MIMO system, step one includes the time slot contention result as an idle, successful, or conflicted state.

[0010] In the aforementioned conflict resolution method based on the DQRA-MIMO system, step one, the method for estimating the total number of users is as follows:

[0011] The total number of users is obtained by minimizing the mean squared error between the actual competition results and the theoretical expected value:

[0012]

[0013] In the formula: C is the number of conflicting time slots, S is the number of successful time slots, E is the number of idle time slots, and C is the expected value of C. It is the expected value of S. It is the expected value of E;

[0014] in, K is the total number of time slots;

[0015] k represents the number of antennas; i represents the number of users; M represents the amount of resources;

[0016]

[0017] In the aforementioned conflict resolution method based on the DQRA-MIMO system, step two, the formula for calculating resource utilization is as follows:

[0018]

[0019] In the formula: N represents the total number of users, M represents the number of resources; P is the probability that the number of users in a single time slot does not exceed the number of antennas k. i represents the number of users.

[0020] The aforementioned conflict resolution method based on the DQRA-MIMO system solves the problem by traversing discrete resource values ​​or using a gradient descent algorithm to find the number of resources M that maximizes resource utilization U.

[0021] In the aforementioned conflict resolution method based on the DQRA-MIMO system, step three involves estimating the number of users in the conflict queue as follows:

[0022] n = N × (1 - P);

[0023] The estimated number of groups is calculated as follows:

[0024]

[0025] In the formula: n is the number of users in the collision queue; M represents the number of resources; and k is the number of antennas.

[0026] The number of users in each group is calculated as follows:

[0027]

[0028] In the aforementioned conflict resolution method based on the DQRA-MIMO system, step three involves dynamically grouping and adapting access time slots according to the expected number of groups. This involves dynamically dividing the users in the conflict queue into multiple groups from the head to the tail of the queue according to the estimated number of groups φ, with each group containing approximately [number of users]. They are then assigned to suitable AR time slots for reconnection in groups.

[0029] In the aforementioned conflict resolution method based on the DQRA-MIMO system, in step two, when using the gradient descent algorithm to solve for the optimal number of resources, the learning rate ranges from [0.001, 0.1].

[0030] In the aforementioned conflict resolution method based on the DQRA-MIMO system, in step four, when grouping users in batches according to the estimated total number of users, the fluctuation range of the number of users in each batch shall not exceed 10% of the estimated total number of users.

[0031] The aforementioned conflict resolution method based on the DQRA-MIMO system adjusts the resource quantity when the system detects that m consecutive time slots are all idle time slots. The adjustment amount is 1 / m of the current resource quantity.

[0032] In the aforementioned conflict resolution method based on the DQRA-MIMO system, in step three, if the number of users in the conflict queue is less than the expected number of groups, the remaining users are divided into a separate group for access.

[0033] The aforementioned conflict resolution method based on the DQRA-MIMO system, when a new user joins the system and the original estimated total number of users deviates from the actual total number of users by more than 20%, re-executes steps one to four to estimate the number of users, configure resources, and adapt user groups.

[0034] The aforementioned conflict resolution method based on the DQRA-MIMO system, when estimating the total number of users in step one, if the fluctuation range of the total number of users estimated multiple times exceeds a set threshold, then an adaptive adjustment mechanism is activated. By performing a weighted analysis on the recent competition results of multiple time slots, the weight of recent data is increased, and the total number of users is re-estimated to more accurately reflect the actual changes in the number of users.

[0035] Compared with existing technologies, this invention optimizes the conflict resolution module mechanism by estimating the total number of users based on a single user access scenario and configuring appropriate resources, thus obtaining the optimal resource combination for the expected number of users. Furthermore, it can estimate the number of users in the conflict queue, thereby dynamically grouping and adapting users to resources. It can also ensure that subsequent accesses have a user count close to or reaching the expected number, achieving efficient resource utilization probabilistically, enabling the system to support a large number of users. Experiments show that this invention improves resource utilization by an average of 25%–30%, effectively reducing resource waste; it reduces transmission latency by more than 50%, meeting real-time requirements; and it reduces the number of conflict resolution loops by 60%, accelerating user access speed. In summary, this invention provides a high-efficiency, low-latency solution for small data packet transmission for massive numbers of IoT users, possessing broad application prospects. Attached Figure Description

[0036] Figure 1 This is a flowchart illustrating the present invention;

[0037] Figure 2 This is a schematic diagram of the dynamic grouping and adaptation access phase for conflicting users;

[0038] Figure 3 This is the principle of the conflict resolution module;

[0039] Figure 4 This is a graph showing the relationship between resource quantity and resource utilization rate when the total number of simulated users is equal to 100. Detailed Implementation

[0040] The present invention will be further described below with reference to the accompanying drawings and embodiments, but this should not be construed as limiting the present invention.

[0041] Example: A conflict resolution method based on a DQRA-MIMO system. This invention estimates the number of users in a single user access result and then allocates the optimal amount of resources using an algorithm. Based on this, before subsequent user access, users can be grouped according to the estimated expected number of users to adapt to resource allocation. In addition, user groups in the conflict queue are also dynamically grouped and adapted according to the expected number of users before access, thereby adapting to resource allocation. The system repeats this process before users successfully transmit data. Throughout the process, compared to the original mechanism, optimizations have been made in the initial user access and conflict user re-access stages, and the iterative process improves the system's transmission efficiency. Figure 1 As shown, the entire process includes the following steps:

[0042] Step 1: When a user accesses the network for the first time, the system randomly selects a time slot to send an access request sequence. The base station then feeds back the time slot contention results and estimates the total number of users based on these results.

[0043] In this step, when a user accesses the network for the first time, a time slot is randomly selected. Users with data to send (active users) randomly choose an AR time slot (access request time slot) and send their AR sequence. Each AR time slot may have 0, 1, or more users sending their AR sequences. Users compete for space, and the base station feeds back the competition result in the time slot: Idle / Success / Collision (I / S / C). An I result indicates no user is sending an AR sequence in the time slot; an S result indicates one user is sending an AR sequence, thus successful transmission; and a C result indicates multiple users are sending their AR sequences, resulting in a collision. Users listen to the time slots sequentially. If the competition result in the AF time slot (access feedback time slot) corresponding to their AR time slot is "S", the user has successfully competed. If the competition result in the AF time slot corresponding to their AR time slot is "C", the user and their competitors enter the RQ queue to wait for collision resolution; a time slot with no user access is designated as "E". Let the total number of time slots be K. When there are k antennas and N users transmitting, the expected values ​​of C, S, and E are respectively:

[0044] K is the total number of time slots;

[0045] k represents the number of antennas; i represents the number of users; M represents the amount of resources;

[0046]

[0047] At this point, the original number of users can be estimated using the I / S / C results of the time slot contention, i.e., the total number of users can be obtained by minimizing the mean square error between the actual contention result and the theoretical expected value.

[0048]

[0049] In step one, if the fluctuation range of the estimated total number of users exceeds a set threshold (e.g., 10%) for multiple consecutive estimates (at least three times), an adaptive adjustment mechanism is activated. This mechanism re-estimates the total number of users by weighting the results of multiple recent time slot competitions, increasing the weight of recent data, and thus more accurately reflecting changes in the actual number of users.

[0050] Step Two: Based on the estimated total number of users, maximize resource utilization to determine the optimal resource quantity. In this step, let the number of users be N, the number of resources be M, and the number of antennas be k. In subsequent calculations, all discrete variables will be treated as continuous variables, and the expected value will be rounded to the nearest integer. Assuming that the number of users in each time slot is independent, the number of users in each channel can be approximately considered to follow a parameter N and ... binomial distribution:

[0051]

[0052] The utilization rate U is defined as the percentage of successful users out of the total number of users that the resources can support.

[0053]

[0054] Given the user number N and the number of antennas k, the system calculates the number of resources M that maximizes resource utilization U by iterating through discrete resource values ​​or using a gradient descent algorithm. When using gradient descent to find the optimal resource quantity, the learning rate ranges from [0.001, 0.1]. After determining the optimal resource quantity in step two, the system monitors the time slot status in real time. If m consecutive time slots are found to be idle, the resource quantity is adjusted by 1 / m of the current resource quantity. After adjustment, the system proceeds to step three.

[0055] Step 3: Estimate the number of users and groups in the conflict queue, calculate the number of users in each group, and dynamically group and adapt access time slots according to the expected number of groups; after one queue allocation, the number of users n = N × (1-P) and the number of user groups in the conflict queue (RQ) can be estimated. At this point, it is estimated that each group contains One user. Then proceed. Calculate the estimated number of dynamically grouped fit groups. For example... Figure 2 As shown, the dynamic grouping and adaptation of access time slots according to the expected number of groups involves dynamically dividing users in the conflict queue into multiple groups from the head to the tail of the queue according to the estimated number of groups φ, with each group having approximately [number of users]. The system then assigns users to suitable AR time slots for re-access, repeating this process until all users have entered the success queue (DQ). If the number of users in the conflict queue is less than the expected number of groups, the remaining users are grouped separately for access.

[0056] Step 4: When subsequent users are added, they will be grouped in batches according to the expected number of users N. The fluctuation range of the number of users in each batch shall not exceed 10% of the estimated total number of users to reduce the probability of conflict.

[0057] like Figure 3 As shown, during the initial user access phase, the system groups the total number of users to approximate the desired number N before allowing them to access the system in batches. If new users join the system and the original estimated total number of users deviates from the actual total number by more than 20%, steps one through four are re-executed to estimate the number of users, configure resources, and adapt user groups. Users entering the conflict queue are processed according to the procedure in step three.

[0058] Furthermore, corresponding experiments were conducted using the conflict resolution method of the present invention, as follows:

[0059] 1. Experimental setup

[0060] Table 1: Experimental Design Comparison of Optimized System and Original System

[0061] parameter Value illustrate Estimated number of users N 100,300,500,700,1000 Simulate user scenarios of different scales Resource quantity M 50,150,250,350,500 Covering a range of situations from insufficient to excessive resources. Number of antennas k 4 Fixed MIMO configuration

[0062] 2. Performance Indicators

[0063] Resource utilization rate U: The proportion of successful users to the total resource carrying capacity.

[0064] Transmission latency: The time from user access to completion of data transmission.

[0065] Conflict resolution time: The average number of loops required for a user in the conflict queue to complete access.

[0066] Figure 4 The graph shows the relationship between N=100 and resource utilization rate U in the simulation.

[0067] 3. Experimental Results

[0068] Table 2: Comparison of Resource Utilization Rates

[0069] Number of users N Traditional system U The U of the present invention Increase 100 68％ 85％ +17％ 300 52％ 78％ +26％ 500 41％ 70％ +29％ 700 35％ 65％ +30％ 1000 28％ 60％ +32％

[0070] As can be seen from Table 2, the present invention significantly improves resource utilization through dynamic resource allocation, and its advantages are more obvious when the number of users is large.

[0071] Table 3: Comparison of Transmission Delay (Unit: ms)

[0072] Number of users N Traditional system latency This invention delay Reduced latency 100 120 80 -33％ 300 280 150 -46％ 500 450 220 -51％ 700 620 290 -53％ 1000 850 380 -55％

[0073] As can be seen from Table 3, the user dynamic grouping adaptation mechanism reduces the number of collision retransmissions and significantly reduces transmission latency.

[0074] Table 4: Comparison of Conflict Resolution Time (Unit: Number of Loops)

[0075] Number of users N Traditional system This invention Efficiency Improvement 100 3.2 1.8 -44％ 300 5.5 2.5 -55％ 500 7.8 3.0 -62％ 700 9.2 3.5 -62％ 1000 12.1 4.0 -67％

[0076] As can be seen from Table 4, the conflict resolution efficiency of the present invention is significantly improved, and the user dynamic grouping adaptation strategy reduces redundant competition.

[0077] Through experimental comparative analysis, the optimization mechanism of this invention demonstrates outstanding performance in the following aspects:

[0078] (1) Resource utilization rate: On average, it increases by 25% to 30%, effectively reducing resource waste.

[0079] (2) Transmission efficiency: The latency is reduced by more than 50%, meeting the real-time requirements.

[0080] (3) Conflict resolution efficiency: The number of loops is reduced by 60%, which speeds up user access.

[0081] In summary, the conflict resolution mechanism designed in this invention can effectively estimate the number of users when they access the system and allocate resources by calculating the expected value of resource quantity. Given resources, this invention can adjust the initial number of users accessing the system to optimize resource utilization, reducing the number of cases where user contention results in I and C. Compared to previous systems where users accessed the system one by one from beginning to end, this invention dynamically groups and adapts users in the conflict queue, ensuring that the number reaches the expected value according to the estimated value, thus improving resource utilization. In general, this invention fundamentally improves resource utilization and reduces resource surplus by configuring resources and allocating users, while also accelerating the overall transmission process. This invention provides a high-efficiency, low-latency solution for small data packet transmission for massive numbers of IoT users and has broad application prospects.

Claims

1. A method for resolving collision based on DQRA-MIMO system, characterized in that: The method comprises the following steps: Step one: when a user accesses for the first time, randomly select a time slot to send an access request sequence, and the base station feeds back the time slot competition result, and the total number of users is estimated according to the time slot competition result; Step two: based on the estimated total number of users, the resource utilization rate is maximized to determine the optimal resource quantity; Step three: the number of users in the conflict queue and the number of groups are estimated, the number of users in each group is calculated, and the access time slot is dynamically grouped and adapted according to the expected number of groups; Step four: when subsequent users access, the users are combined in batches according to the estimated total number of users to reduce the conflict probability, and the users in the conflict queue are processed in a cycle according to step three until the data transmission is completed; In step one, the method for estimating the total number of users is: The total number of users is obtained by minimizing the mean square error of the actual competition result and the theoretical expected value: ； wherein: is the number of collision slots, is the number of success slots, is the number of idle slots, is the expected value of is the expected value of is the expected value of is the expected value of is the expected value of is the expected value of wherein , is the total number of slots; , is the number of antennas; denotes the number of users; denotes the number of resources; ； In step two, the calculation formula of the resource utilization rate is: ； In the formula: is the total number of users, represents the number of resources; is the probability that the number of users per time slot does not exceed the number of antennas, , , represents the number of users; In step three, the number of users in the conflict queue is estimated as follows: ； The number of groups is estimated and calculated as follows: ； In the formula: is the number of users in the conflict queue; represents the number of resources, is the number of antennas; The number of users in each group is calculated as follows: ； In step three, the dynamic grouping access time slot according to the desired group number is to group the users in the conflict queue according to the estimated group number From the head to the tail of the queue, dynamically divide into multiple groups, and each group has a number close to And distribute to the adaptive AR time slot for re-access.

2. The method for resolving collision based on DQRA-MIMO system according to claim 1, characterized in that: In step one, the time slot competition result includes idle, successful or conflict state.

3. The method for resolving collision based on DQRA-MIMO system according to claim 1, characterized in that: Solving by iterating over discrete resource values or gradient descent algorithm to maximize resource utilization Maximum number of resources value.

4. The method for resolving collision based on DQRA-MIMO system according to claim 1, characterized in that: In step two, when the gradient descent algorithm is used to solve the optimal resource quantity, the learning rate is in the range of [0.001, 0.1].

5. The method for resolving collision in DQRA-MIMO system according to claim 1, wherein: In step four, when the users are combined in batches according to the estimated total number of users, the fluctuation range of the number of users in each batch does not exceed 10% of the estimated total number of users.

6. The method for resolving collision in DQRA-MIMO system according to claim 1, wherein: When the system detects that m consecutive time slots are all idle time slots, the resource quantity is adjusted, and the adjustment amplitude is 1 / m of the current resource quantity.

7. The method for resolving collision in DQRA-MIMO system according to claim 1, wherein: In step three, if the number of users in the conflict queue is less than the expected number of groups, the remaining users are separately grouped for access.

8. The method for resolving collision in DQRA-MIMO system according to claim 1, wherein: When there is a new user in the system and the deviation between the original estimated total number of users and the actual total number of users exceeds 20%, steps one to four are re-executed for user number estimation, resource configuration and user grouping adaptation.

9. The method for resolving collision in DQRA-MIMO system according to claim 1, wherein: In step one, when the fluctuation range of the estimated total number of users exceeds the set threshold value for multiple times, an adaptive adjustment mechanism is started, the recent multiple time slot competition results are analyzed by weighting, the weight of recent data is increased, the total number of users is re-estimated, and the actual number of users is more accurately reflected.

Images