A clustering active user detection system and method based on a particle swarm optimization algorithm

Abstract

The application discloses a cluster active user detection system and method based on a particle swarm optimization algorithm, and belongs to the technical field of active user detection methods; the application comprises the following steps: S1, a target function is designed to evaluate the quality of cluster division; different weight coefficients are introduced to adjust the target function to adapt to different scene requirements; S2, a large-scale fading coefficient variance constraint is introduced into the system optimization process to ensure that the channel quality between access points and users in the cluster is uniform; S3, based on a particle swarm optimization function module, the selection of the number of clusters is dynamically optimized through a particle swarm optimization algorithm, and the selection optimization of the number of clusters is further realized through an adaptability function calculation module and a large-scale fading coefficient variance calculation module; and S4, after optimization, the optimal number of clusters and the corresponding adaptability value are output through the particle swarm optimization algorithm. Compared with the prior art, the application is suitable for active user detection in a de-cell large-scale multiple-input multiple-output network, and effectively improves the detection precision and the calculation efficiency.

Description

Technical Field

[0001] This invention relates to the field of active user detection methods, and in particular to a clustered active user detection system and method based on particle swarm optimization algorithm. Background Technology

[0002] Adaptive cluster number selection schemes offer significant advantages in dynamically changing network environments such as mobile networks, wireless sensor networks, and massive machine-type communications due to their dynamic adjustment capabilities. This scheme can adjust the cluster number in real time based on factors such as network load, user distribution, and communication demands, thereby effectively improving resource utilization efficiency. In decellularized massive MIMO systems, the number of devices is large and their distribution is complex, making traditional static cluster number selection methods difficult to adapt to this complex and ever-changing environment. Adaptive cluster number selection schemes, however, can optimize the cluster number based on real-time network conditions and demands, significantly enhancing the system's scalability and reliability.

[0003] In the field of communications, adaptive cluster selection algorithms encompass genetic algorithms, ant colony optimization algorithms, simulated annealing algorithms, and reinforcement learning algorithms. For adaptive cluster selection using the K-means algorithm, current mainstream solutions include K-means++, which improves cluster center selection, and adaptive selection methods based on silhouette coefficients. K-means++ improves the initial cluster center selection strategy of the traditional K-means algorithm, effectively enhancing clustering performance and reducing the probability of the algorithm converging to a local optimum. The silhouette coefficient, a key indicator for evaluating the quality of clustering results, ranges from -1 to 1, comprehensively considering intra-cluster compactness and inter-cluster separation. Specifically, a silhouette coefficient closer to 1 indicates a more reasonable distribution of data points within clusters, with high intra-cluster similarity and large inter-cluster differences; a silhouette coefficient close to -1 suggests that data points may have been incorrectly assigned and should belong to other clusters. Cluster selection methods based on silhouette coefficients calculate the silhouette coefficients for different cluster numbers and select the cluster number that maximizes the silhouette coefficient, thus fully considering both intra-cluster compactness and inter-cluster separation.

[0004] However, in scenarios involving decellularized massive MIMO and massive machine-type communication, the optimized algorithms for cluster number adaptive selection and active user detection still have many problems:

[0005] 1. Static settings have significant limitations: Existing cluster number selection methods mostly rely on pre-set static cluster numbers. This approach has significant limitations in practical communication applications, especially in dynamic network environments. Due to frequent changes in the number of devices and user load in the network, fixed cluster number allocation and geographical location-based allocation methods are difficult to adapt to these dynamic fluctuations. Too many clusters lead to resource waste, while too few result in resource insufficiency. In ultra-dense deployment scenarios, this may cause access point resource overload or some clusters to be idle, severely impacting communication performance.

[0006] 2. Insufficient consideration of channel factors: In decellularized massive MIMO networks and massive machine-type communication scenarios, the channel quality differences between access points and users have a significant impact on system performance. Traditional clustering methods such as K-means fail to effectively consider large-scale fading coefficients and channel characteristics, resulting in inconsistent signal quality within clusters, which in turn affects cooperative detection performance.

[0007] 3. Insufficient dynamic response capability: Traditional cluster number selection algorithms cannot respond to network changes in real time. In dynamic environments, network load, signal quality, and user activity are constantly changing, and static cluster number selection methods cannot adjust the number of clusters in a timely manner according to these changes. Therefore, when network traffic fluctuates drastically, cluster number selection often fails to achieve optimal results, leading to low system efficiency and poor performance in detecting active users.

[0008] 4. High computational complexity: With the expansion of network scale and the increase in the number of devices in decellularized massive MIMO networks and large-scale machine-type communication scenarios, the computational complexity of cluster selection and active user detection increases exponentially. Traditional cluster selection methods, when dealing with large-scale networks, generate significant computational overhead due to the frequent calculation of the distance between users and cluster centers, resulting in slow system response and poor real-time performance. In the active user detection process, the complex cluster partitioning procedure introduces unnecessary delays, reducing detection accuracy and efficiency.

[0009] To address the aforementioned problems, this invention proposes a clustered active user detection system and method based on particle swarm optimization algorithm. Summary of the Invention

[0010] The purpose of this invention is to provide a clustered active user detection system and method based on particle swarm optimization algorithm to solve the problems mentioned in the background art.

[0011] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0012] A clustered active user detection system based on particle swarm optimization algorithm includes the following modules:

[0013] The K-means network function module uses the K-means algorithm to cluster the geographical locations of access points and user devices, calculates the distance between users and access points, and assigns access points to different clusters based on the distance. This module not only reduces computational complexity but also provides input data for subsequent large-scale fading coefficient calculation and active user detection.

[0014] The covariance matrix optimization and active mode decoding function module is used to decode active user modes. It calculates the user's activity parameters based on the received signals. By minimizing the negative log-likelihood function and optimizing the covariance matrix, it gradually adjusts the activity state parameters of each user to accurately determine whether the user is active.

[0015] Large-scale fading coefficient calculation module: Calculates the large-scale fading coefficient within each cluster based on the distance between the user and the access point, and obtains the overall large-scale fading coefficient of each cluster through a weighted average method; this process helps the model more accurately evaluate the signal strength of different users and optimize subsequent active user detection;

[0016] Signature sequence generation module: Generates a unique signature sequence to identify each user. By generating a signature sequence with a complex Gaussian distribution, it provides a unique signal identifier for the user equipment to ensure the accuracy of signal decoding and active state detection.

[0017] Complex Gaussian noise generation module: used to simulate noise in the wireless channel, generate complex Gaussian noise and add it to the received signal; in this way, the system can simulate the actual wireless communication environment and help evaluate the active user detection performance under different signal-to-noise ratio conditions;

[0018] The missed detection and false alarm probability calculation module is used to calculate the two key performance indicators of the active user detection system: the missed detection probability and the false alarm probability, and to adjust the detection results according to different thresholds. Through this module, the system can evaluate the detection performance under different threshold settings, thereby optimizing the accuracy of active user detection.

[0019] The constant parameter calculation module is responsible for calculating polynomial coefficients used to decode the user's activity status. By calculating polynomial coefficients related to the large-scale fading coefficient, it optimizes the system's decoding process and provides mathematical support for accurate estimation of the user's activity status.

[0020] Particle Swarm Optimization Function Module: Implements dynamic optimization of cluster number selection based on the particle swarm optimization algorithm;

[0021] Fitness function calculation module: used to calculate the fitness value for a given number of clusters. The fitness value comprehensively considers multiple factors such as the probability of missed detection, the probability of false alarm, computational complexity, and the variance of the large-scale fading coefficient.

[0022] Large-scale fading coefficient variance calculation module: used to calculate the large-scale fading coefficient variance within each cluster, measuring the uniformity of signal quality within the cluster;

[0023] Main program module: Responsible for initializing parameters and performing Monte Carlo simulation, setting the cluster number selection range according to different access point numbers, and running particle swarm optimization algorithm to optimize the cluster number.

[0024] Preferably, the specific execution flow of the particle swarm optimization function module is as follows:

[0025] In particle swarm optimization, the position of a particle represents the cluster number, and the velocity represents the change in the cluster number; each particle adjusts its position and velocity based on the fitness value of the current cluster number.

[0026] During the search process, the particle updates its individual optimal and global optimal solutions and updates its velocity according to the inertia weight;

[0027] The particle swarm optimization algorithm obtains the optimal number of clusters through multiple iterations.

[0028] Preferably, the specific execution flow of the fitness function calculation module is as follows:

[0029] For a given number of clusters, calculate the large-scale fading coefficient between the access point and the user within the cluster;

[0030] Calculate the variance of the large-scale fading coefficient to reflect the uniformity of signal quality within the cluster;

[0031] Calculate the probabilities of missed detections and false alarms;

[0032] The fitness value of the cluster number is obtained by weighting the indicators calculated above.

[0033] Preferably, the specific execution flow of the large-scale fading coefficient variance calculation module is as follows:

[0034] The large-scale fading coefficient between the access point and the user is calculated based on the cluster allocation, and a weighted average is calculated based on the large-scale fading coefficient.

[0035] The variance of the large-scale fading coefficient is calculated to reflect the signal quality differences of access points within each cluster: if the variance of the large-scale fading coefficient of a cluster is small, it indicates that the signal quality of access points within the cluster is uniform, which is beneficial for collaborative detection; if the variance is large, it means that the signal quality differences of access points within the cluster are large, resulting in poor detection performance.

[0036] Preferably, the main program execution module executes as follows:

[0037] For each number of access points, the main program optimizes the number of clusters by calling the particle swarm optimization function and calculates the large-scale fading coefficient and complexity performance indicators.

[0038] Through multiple experiments and accumulated calculation results, the average optimal cluster number, computational complexity, and large-scale fading coefficient variance were obtained for different APs and access points.

[0039] The main program generates charts to show the relationship between the number of clusters, complexity, and variance of the large-scale fading coefficient as a function of the number of access points, in order to analyze and optimize network design.

[0040] A clustered active user detection method based on particle swarm optimization algorithm implemented using the system described above includes the following steps:

[0041] S1. Design an objective function that comprehensively considers the probability of missed detection, the probability of false alarm, computational complexity, and the variance of the large-scale fading coefficient, and use it to evaluate the quality of cluster partitioning; during the evaluation process, different weight coefficients are introduced to adjust the objective function to adapt to different scenario requirements;

[0042] S2. Introduce large-scale fading coefficient variance constraints into the system optimization process to ensure uniform channel quality between access points and users within the cluster, thereby improving the collaborative detection capability of the detection system.

[0043] S3. Based on the particle swarm optimization function module, the selection of the number of clusters is dynamically optimized through the particle swarm optimization algorithm. The selection of the number of clusters is further optimized through the fitness function calculation module and the large-scale fading coefficient variance calculation module.

[0044] S4. After optimization, the optimal number of clusters and the corresponding fitness value are output through the particle swarm optimization algorithm to optimize the cluster division in the active user detection work and realize adaptive cluster number selection.

[0045] Preferably, the false alarm probability and the missed detection probability affect the system's active user detection performance; the computational complexity affects the system's resource consumption and latency.

[0046] Preferably, the large-scale fading coefficient variance is used to measure the uniformity of channel quality between access points and users within a cluster: low variance indicates that the signal quality of access points within the cluster is uniform, which helps to improve the accuracy of cooperative detection; high variance indicates that the signal quality is poor, which affects the reliability of the system.

[0047] Preferably, S3 specifically includes the following:

[0048] The particle swarm optimization algorithm is used to simulate the foraging process of bird flocks. Each particle represents a cluster number, and the optimal cluster number is found by iteratively adjusting the position and speed of the particles.

[0049] During each iteration, the particle updates its position and velocity based on its current fitness and the global optimal position in order to approximate the global optimal solution. During the position update process, the particle considers the constraints of the objective function and the variance of the large-scale fading coefficient, and introduces a boundary handling mechanism to ensure that the particle always searches within the physically feasible range.

[0050] The quality of cluster selection is evaluated using a fitness function, and the position of each particle is optimized based on the objective function and constraints.

[0051] By calculating the variance of the particle swarm fitness, we can determine whether the particle swarm has converged. When the fitness variance is close to zero, it means that the particle swarm has found the global optimum and the optimization process ends.

[0052] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0053] (1) This invention dynamically selects the number of clusters based on the particle swarm optimization algorithm, which can adaptively adjust the number of clusters according to the needs of active user detection, so that the number of clusters can flexibly respond to changes in network conditions. This dynamic adaptability significantly improves network performance and scalability in decellularized massive MIMO networks and massive machine-type communications.

[0054] (2) This invention introduces a multi-objective optimization function, which integrates factors such as missed detection probability, false alarm probability, computational complexity, and large-scale fading coefficient variance into the optimization process for comprehensive consideration. Through this method, the cluster number selection not only optimizes the geometric distribution of clusters, but also considers communication quality such as the balance of large-scale fading coefficients, thereby achieving more efficient user detection and resource allocation.

[0055] (3) This invention introduces a large-scale fading coefficient variance constraint, using the uniformity of the large-scale fading coefficient within a cluster as an optimization constraint to ensure uniform signal quality at access points within the cluster, thereby improving network stability and the accuracy of cooperative detection. This innovation ensures that cluster number selection not only meets mathematical optimality but also satisfies the reliability requirements of the physical layer, especially in ultra-densely deployed networks.

[0056] (4) By adding a computational complexity term to the optimization objective function and flexibly adjusting the complexity weight according to actual needs, this invention can achieve a good balance between high-precision detection and low computational complexity. This means that in low-power or high-precision scenarios, the system can intelligently select an appropriate number of clusters while reducing computational complexity, thereby effectively coping with the high computational requirements in large-scale network environments.

[0057] In summary, this invention provides a clustered active user detection system and method based on particle swarm optimization algorithm, which is suitable for active user detection in decellularized large-scale multiple-input multiple-output networks and can effectively improve detection accuracy and computational efficiency. Attached Figure Description

[0058] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings involved in the embodiments are now briefly described. Obviously, the drawings in the following description are merely illustrative of some embodiments of the present invention. For those skilled in the art, other forms of drawings can be constructed based on these drawings without creative effort.

[0059] Figure 1 This is a flowchart of the clustered active user detection method based on particle swarm optimization algorithm proposed in Embodiment 1 of the present invention. Detailed Implementation

[0060] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0061] This invention proposes a clustered active user detection system and method based on particle swarm optimization algorithm to solve the following problems:

[0062] (1) Limitations of static cluster number selection: Existing K-means algorithms rely on manually preset fixed cluster numbers, which may lead to overload of access points within a cluster and idle resources in other clusters when the network load changes dynamically. Static cluster numbers cannot adapt to network changes, and clustering based solely on geographical location during communication is insufficient to guarantee communication quality, reducing the flexibility and efficiency of the system.

[0063] (2) Local Optimality Problem: The traditional K-means algorithm lacks a systematic consideration of communication performance in the selection of the number of clusters, and is prone to getting trapped in local optima. It cannot fully optimize the actual communication conditions of the network and affects the overall performance of the system.

[0064] (3) Problems caused by ultra-dense deployment: In ultra-dense deployment scenarios, the spatial relationship between access points and users is highly complex. Static clustering may cause cross-cluster interference accumulation and energy efficiency decline, thereby affecting the scalability and intelligence level of decellular networks in large-scale machine communication connection scenarios.

[0065] (4) Large-scale fading coefficient imbalance problem: Existing methods fail to fully consider the large-scale fading coefficient differences between access points and users during cluster partitioning, which leads to uneven signal distribution in some clusters, thus affecting the accuracy of cooperative detection and system performance. The following describes the clustered active user detection system and method based on particle swarm optimization algorithm proposed in this invention with reference to relevant figures and specific examples.

[0066] Example 1:

[0067] Please see Figure 1 This invention proposes a clustered active user detection method based on particle swarm optimization algorithm, specifically including:

[0068] First, an objective function was designed that comprehensively considers the false alarm probability, computational complexity, and large-scale fading coefficient variance to evaluate the quality of cluster partitioning. The false alarm probability and false alarm probability are key indicators affecting the system's active user detection performance; computational complexity affects system resource consumption and latency; and the large-scale fading coefficient variance measures the volatility of the large-scale fading coefficient within a cluster. By introducing different weighting coefficients, the objective function can be flexibly adjusted to adapt to different scenario requirements, such as high-precision detection or low-power scenarios. This objective function balances detection performance and computational complexity through a multi-objective optimization framework.

[0069] Secondly, to ensure uniform channel quality between access points and users within a cluster, this invention introduces a large-scale fading coefficient variance constraint during the optimization process. The large-scale fading coefficient variance measures the balance of channel quality between access points and users within a cluster. Low variance indicates relatively uniform signal quality among access points, which helps improve the accuracy of cooperative detection. If the large-scale fading coefficient variance is large, it may lead to excessive differences in signal quality, affecting system reliability. By introducing the constraint of the large-scale fading coefficient variance, the large-scale fading coefficient among access points within a cluster is ensured to be relatively uniform, thereby improving the network's cooperative detection capability.

[0070] Next, this invention employs a particle swarm optimization (PSO) algorithm to dynamically optimize the selection of the number of clusters. The PSO algorithm simulates the foraging process of a flock of birds, where each particle represents a possible cluster number. By iteratively adjusting the particle's position and velocity, it finds the optimal number of clusters. In each iteration, the particle updates its position and velocity based on its current fitness and the global optimum, thus approximating the global optimum. The particle position update not only considers the constraints of the objective function and the variance of the large-scale fading coefficient, but also introduces a boundary handling mechanism to ensure that the particle always searches within a physically feasible range. In the PSO algorithm, the fitness function is used to evaluate the quality of the cluster selection and optimize the position of each particle according to the objective function and constraints. By calculating the variance of the particle swarm fitness, it can be determined whether the particle swarm has converged. When the fitness variance approaches zero, it indicates that the particle swarm has found the global optimum, and the optimization process ends.

[0071] Ultimately, the particle swarm optimization algorithm outputs the optimal number of clusters and the corresponding fitness value. By adaptively selecting the number of clusters, the cluster partitioning in the active user detection algorithm is optimized, satisfying both the requirements of efficient computation and ensuring the reliability and performance of the system. Large-scale fading coefficient variance constraints ensure balanced signal quality within clusters, thereby reducing false detections caused by excessive signal differences and improving the accuracy of active user detection.

[0072] Example 2:

[0073] Based on Example 1 but with a difference, this invention proposes a clustered active user detection method based on particle swarm optimization algorithm, including the following:

[0074] Step 1: Initialize the particle swarm. In this stage, the position of each particle in the swarm represents a candidate cluster number C, and the particle velocity represents the change in the cluster number, respectively:

[0075]

[0076]

[0077] in, For a locally optimal cluster number solution, C best The solution represents the global optimal cluster count, where i represents the current cluster index and t represents the current iteration number. When the particle swarm optimization algorithm provides the current optimal cluster count, it needs to use the fitness function to determine whether to update the particle's local or global optimal cluster count based on the current fitness function value. The particle adjusts its search direction in this way, tending to find the solution that minimizes the fitness function value.

[0078] The initial positions of particles are randomly generated within a preset cluster size range, ensuring that all particle positions cover the entire possible cluster size space. For each particle, its velocity is also initialized to ensure that the particle can adjust the cluster size according to the objective function requirements in subsequent iterations. The fitness value of each particle is calculated based on a multi-objective optimization function, which comprehensively considers factors such as missed detection probability, false alarm probability, computational complexity, and large-scale fading coefficient variance.

[0079]

[0080] Among them, μ1, μ2, μ3, and μ4 are weighting coefficients used to balance the priorities of detection performance, computational complexity, and communication quality uniformity. md (C), P fa (C) represents the probability of a missed detection and the probability of a false alarm, and Comp(C) is the computational complexity term. It is the variance of the large-scale fading coefficient.

[0081] Step 2: After particle swarm initialization, the fitness value of each particle is calculated. The fitness value is obtained through an objective function that considers factors such as the probability of missed detection, the probability of false alarms, computational complexity, and the variance of the large-scale fading coefficient. Each particle's fitness value is calculated based on its current cluster position, as shown in the following expression:

[0082]

[0083] Where, Δ max η represents the maximum permissible variance of the large-scale fading coefficient uniformity. η is a penalty factor used to impose additional penalties on particles that do not meet the large-scale fading coefficient uniformity constraint, explicitly guiding particles to be more inclined to meet the constraint. Introducing the penalty term also allows the fitness function to explicitly penalize cluster partitioning schemes that exceed the uniformity threshold, suppressing the generation of inconsistencies.

[0084] Step 3: In each iteration, each particle compares its current fitness value with its historical best value. If the current fitness value is better, the particle's personal best position is updated. The personal best position represents the best cluster number the particle has obtained during the historical iterations.

[0085] Step 4: The global optimal position is the cluster number represented by the particle with the best fitness in the particle swarm. After each iteration, all particles update their global optimal positions. If a particle's fitness value is better than the current global optimal value, then the global optimal position is updated.

[0086] Step 5: In each iteration, update the particle's velocity and position.

[0087] Step 6: At the end of each iteration, the fitness variance of the particle swarm needs to be checked. Fitness variance measures the difference in fitness within the particle swarm. If the fitness variance is small enough, it indicates that the particle swarm has converged, the algorithm can stop, and the globally optimal number of clusters is output. If the fitness variance is large, it indicates that the particle swarm has not yet converged, and the particle swarm continues to iterate and optimize until the convergence condition is met, as follows:

[0088]

[0089] Among them, I PSO Let represent the total number of particles in the particle swarm, and in this problem, let represent the total number of cluster solutions. Fitness is the fitness value of the i-th particle, and Mean(Fitness) is the average fitness of the particles. When the variance approaches zero, it indicates that all particles in the particle swarm are close to the global optimum or a near-optimal solution, and the search process is complete. Conversely, if the variance is large, it indicates that there are still some fitness differences in the particle swarm, and the search has not yet reached final convergence.

[0090] Step 7: Once the particle swarm converges, the algorithm will output the number of clusters corresponding to the globally optimal particle, which is the optimal number of clusters C. * .

[0091] In practical applications, suppose a decellularized network operates under different loads, such as a high-density urban environment or a low-density rural environment. Traditional methods might require manually setting different cluster numbers for different network environments. However, with the particle swarm optimization algorithm of this invention, the system can adaptively select the number of clusters in real time. For example, under high network load, the system will select more clusters to distribute the load, while under low network load, the number of clusters will be reduced, thereby reducing computational complexity.

[0092] It should be noted that in this invention patent, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0093] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A clustered active user detection system based on particle swarm optimization algorithm, characterized in that, Includes the following modules: K-means network function module: Uses the K-means algorithm to cluster the geographical locations of access points and user equipment, calculates the distance between users and access points, and assigns access points to different clusters based on the distance; Covariance matrix optimization and active mode decoding function module: used to decode active user modes and calculate the user's active state parameters based on the received signals; By minimizing the negative log-likelihood function and optimizing the covariance matrix, the activity status parameters of each user are gradually adjusted to accurately determine whether a user is active. Large-scale fading coefficient calculation module: Calculates the large-scale fading coefficient within each cluster based on the distance between the user and the access point, and obtains the overall large-scale fading coefficient of each cluster through a weighted average method; Signature sequence generation module: Generates a unique signature sequence to identify each user. By generating a signature sequence with a complex Gaussian distribution, it provides a unique signal identifier for the user equipment to ensure the accuracy of signal decoding and active state detection. Complex Gaussian noise generation module: used to simulate noise in the wireless channel, generate complex Gaussian noise and add it to the received signal; The module for calculating the probability of missed detection and false alarm is used to calculate the two key performance indicators of the active user detection system: the probability of missed detection and the probability of false alarm, and to adjust the detection results according to different thresholds. The constant parameter calculation module is responsible for calculating polynomial coefficients, which are used to decode the user's activity status. Particle Swarm Optimization Function Module: Implements dynamic optimization of cluster number selection based on the particle swarm optimization algorithm; Fitness function calculation module: used to calculate the fitness value for a given number of clusters. The fitness value comprehensively considers multiple factors such as the probability of missed detection, the probability of false alarm, computational complexity, and the variance of the large-scale fading coefficient. Large-scale fading coefficient variance calculation module: used to calculate the large-scale fading coefficient variance within each cluster, measuring the uniformity of signal quality within the cluster; Main program module: Responsible for initializing parameters and performing Monte Carlo simulation, setting the cluster number selection range according to different access point numbers, and running particle swarm optimization algorithm to optimize the cluster number.

2. The clustered active user detection system based on particle swarm optimization algorithm according to claim 1, characterized in that, The specific execution flow of the particle swarm optimization function module is as follows: In particle swarm optimization, the position of a particle represents the cluster number, and the velocity represents the change in the cluster number; each particle adjusts its position and velocity based on the fitness value of the current cluster number. During the search process, the particle updates its individual optimal and global optimal solutions and updates its velocity according to the inertia weight; The particle swarm optimization algorithm obtains the optimal number of clusters through multiple iterations.

3. The clustered active user detection system based on particle swarm optimization algorithm according to claim 1, characterized in that, The specific execution flow of the fitness function calculation module is as follows: For a given number of clusters, calculate the large-scale fading coefficient between the access point and the user within the cluster; Calculate the variance of the large-scale fading coefficient to reflect the uniformity of signal quality within the cluster; Calculate the probabilities of missed detections and false alarms; The fitness value of the cluster number is obtained by weighting the variance of the large-scale fading coefficient, the probability of missed detection, the probability of false alarm, and the computational complexity.

4. The clustered active user detection system based on particle swarm optimization algorithm according to claim 1, characterized in that, The specific execution flow of the large-scale fading coefficient variance calculation module is as follows: The large-scale fading coefficient between the access point and the user is calculated based on the cluster allocation, and a weighted average is calculated based on the large-scale fading coefficient. The variance of the large-scale fading coefficient is calculated to reflect the signal quality differences of access points within each cluster: if the variance of the large-scale fading coefficient of a cluster is small, it indicates that the signal quality of access points within the cluster is uniform, which is beneficial for collaborative detection; if the variance is large, it means that the signal quality differences of access points within the cluster are large, resulting in poor detection performance.

5. The clustered active user detection system based on particle swarm optimization algorithm according to claim 1, characterized in that, The specific execution flow of the main program running module is as follows: For each number of access points, the main program optimizes the number of clusters by calling the particle swarm optimization function and calculates the large-scale fading coefficient and complexity performance indicators. Through multiple experiments and accumulated calculation results, the average optimal cluster number, computational complexity, and large-scale fading coefficient variance were obtained for different APs and access points. The main program generates charts to show the relationship between the number of clusters, complexity, and variance of the large-scale fading coefficient as a function of the number of access points, in order to analyze and optimize network design.

6. A method for detecting clustered active users based on particle swarm optimization algorithm implemented using the system described in claim 1, characterized in that, Includes the following steps: S1. Design an objective function that comprehensively considers the probability of missed detection, the probability of false alarm, computational complexity, and the variance of the large-scale fading coefficient, and use it to evaluate the quality of cluster partitioning; during the evaluation process, different weight coefficients are introduced to adjust the objective function to adapt to different scenario requirements; S2. Introduce large-scale fading coefficient variance constraints into the system optimization process to ensure uniform channel quality between access points and users within the cluster, thereby improving the collaborative detection capability of the detection system. S3. Based on the particle swarm optimization function module, the selection of the number of clusters is dynamically optimized through the particle swarm optimization algorithm. The selection of the number of clusters is further optimized through the fitness function calculation module and the large-scale fading coefficient variance calculation module. S4. After optimization, the optimal number of clusters and the corresponding fitness value are output through the particle swarm optimization algorithm to optimize the cluster division in the active user detection work and realize adaptive cluster number selection.

7. The method for detecting clustered active users based on particle swarm optimization algorithm according to claim 6, characterized in that, The probability of missed detection and the probability of false alarm affect the system's performance in detecting active users; the computational complexity affects the system's resource consumption and latency.

8. The method for detecting clustered active users based on particle swarm optimization algorithm according to claim 7, characterized in that, The large-scale fading coefficient variance is used to measure the uniformity of channel quality between access points and users within a cluster: low variance indicates that the signal quality of access points within the cluster is uniform, which helps to improve the accuracy of cooperative detection; high variance indicates that the signal quality is poor, which affects the reliability of the system.

9. The method for detecting clustered active users based on particle swarm optimization algorithm according to claim 6, characterized in that, S3 specifically includes the following: The particle swarm optimization algorithm is used to simulate the foraging process of bird flocks. Each particle represents a cluster number, and the optimal cluster number is found by iteratively adjusting the position and speed of the particles. During each iteration, the particle updates its position and velocity based on its current fitness and the global optimal position in order to approximate the global optimal solution. During the position update process, the particle considers the constraints of the objective function and the variance of the large-scale fading coefficient, and introduces a boundary handling mechanism to ensure that the particle always searches within the physically feasible range. The quality of cluster selection is evaluated using a fitness function, and the position of each particle is optimized based on the objective function and constraints. By calculating the variance of the particle swarm fitness, we can determine whether the particle swarm has converged. When the fitness variance is close to zero, it means that the particle swarm has found the global optimum and the optimization process ends.

Images