Method for constructing social group activity monitoring model based on multiple dimensions

Abstract

The invention discloses a method for constructing a social group activity monitoring model based on multiple dimensions. The method relates to the technical field of electric data processing, and comprises the following steps: collecting and processing social group activity monitoring model construction data, and analyzing, optimizing and adjusting the social group activity monitoring model construction data. The method comprises the following steps: collecting and processing social group activity monitoring model construction data; the construction data of the social group activity monitoring model are analyzed, the social group activity monitoring basic model is constructed, and the construction accuracy method of the activity monitoring model is optimized and adjusted for the social group activity monitoring basic model, so that the reliability of the construction method of the social group activity monitoring model is improved, and the construction efficiency of the social group activity monitoring model is improved. The problem that in the prior art, a construction method of a social group activity monitoring model is insufficient in reliability is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical data processing, and particularly to a method for constructing a social group activity monitoring model based on multiple dimensions. Background Art

[0002] With the rapid development of the Internet and information technology, the speed and scope of information dissemination of social group activities have increased significantly. In order to better manage and supervise these activities, it has become particularly important to construct an effective monitoring model. With the increase in the number of universities and social groups, club management faces more and more challenges, such as member recruitment, activity organization, financial management, etc. However, there are many technical problems in the application of existing social group activity monitoring models, resulting in insufficient reliability evaluation and difficulty in meeting actual needs.

[0003] The construction method of the existing group activity monitoring model uses positioning technology to obtain the precise location information of the members of the tour group in real time by using the Beidou satellite navigation system; and processes and analyzes a large amount of positioning data through big data processing technology to extract useful information.

[0004] For example, a method for constructing a real-time population integration model based on activity trajectory data disclosed in the patent application with the publication number: CN115577052A includes: obtaining population data and corresponding activity trajectory data, and performing preprocessing to form a basic database; using the data in the basic database to construct a knowledge graph with community buildings, portals, and ID cards as entities and portal relationships as edges, marking target personnel to form a knowledge graph of portals and household personnel; based on the activity trajectory data, analyzing the activities and occurrence frequencies of the population, constructing a clustering model that can distinguish the residential types of personnel by using multiple place characteristics of the activity trajectory data, extracting the characteristics of the places entered and exited and the number of times and importing them into the clustering model to obtain the residential types of personnel; constructing a population integration model according to the results obtained from the personnel activity trajectory data and the clustering model.

[0005] For example, a method for realizing a personnel activity graph based on a four-dimensional travel model disclosed in the patent announcement with the announcement number: CN114066431B includes: constructing a four-dimensional travel model; inputting the face image data of travel personnel into the four-dimensional travel model; determining the travel trajectories of target personnel and determining the abnormal personnel among the target personnel; mining the close contacts A who come into contact with the abnormal personnel; mining the high-frequency travel places A of the abnormal personnel; mining the high-frequency visiting personnel of the high-frequency travel places A; mining the personnel who appear in the same frame as the abnormal personnel according to the high-frequency travel places A; mining the high-frequency travel places B of the high-frequency visiting personnel; mining the high-frequency travel places C of the personnel who appear in the same frame; mining the close contacts B of the personnel who appear in the same frame; mining the close contacts C of the close contacts A; mining the high-frequency travel places D of the close contacts A; generating a personnel activity graph.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, it is found that the above technologies have at least the following technical problems:

[0007] When the construction method of the social group activity monitoring model is applied to the application scenario of real-time tracking of the location of members of a tourist group entering a karst cave to ensure that they move along a preset safe route, there is a problem of insufficient reliability evaluation of the construction method of the social group activity monitoring model. When the Beidou signal encounters reflecting surfaces such as buildings, the ground, and water surfaces during propagation, reflection will occur. The reflected signal and the direct signal reach the receiver of the Beidou satellite navigation system at the same time. However, due to the different path lengths, there will be differences in signal phase and amplitude. At the same time, accurately extracting abnormal behavior features such as suddenly deviating from the route and abnormal staying from a large amount of data increases the difficulty and computational amount of feature extraction, and there is a problem of insufficient reliability of the construction method of the social group activity monitoring model. Summary of the Invention

[0008] The embodiments of the present application provide a construction method of a social group activity monitoring model based on multiple dimensions, which solves the problem of insufficient reliability of the construction method of the social group activity monitoring model in the prior art and improves the reliability of the construction method of the social group activity monitoring model.

[0009] The embodiments of the present application provide a construction method of a social group activity monitoring model based on multiple dimensions, including the following steps: collecting and processing the data for constructing the social group activity monitoring model; analyzing the data for constructing the social group activity monitoring model to obtain the multi-path effect influence evaluation value and the abnormal behavior dynamic analysis evaluation value; comprehensively analyzing through the multi-path effect influence evaluation value and the abnormal behavior dynamic analysis evaluation value to obtain the construction accuracy evaluation value of the activity monitoring model; constructing a basic model for monitoring social group activities, and optimizing and adjusting the construction accuracy method of the activity monitoring model for the basic model of social group activity monitoring according to the multi-path effect influence evaluation value, the abnormal behavior dynamic analysis evaluation value, and the construction accuracy evaluation value of the activity monitoring model.

[0010] Further, the specific steps of collecting and processing the data for constructing the social group activity monitoring model are as follows: receiving GPS satellite signals through a handheld GPS locator to collect the original data for constructing the social group activity monitoring model; cleaning and denoising the original data for constructing the social group activity monitoring model to obtain the data for constructing the social group activity monitoring model, and the data for constructing the social group activity monitoring model includes multi-path effect influence data and abnormal behavior dynamic analysis data.

[0011] Further, the specific steps for obtaining the multipath effect influence evaluation value are as follows: The multipath effect influence data includes the number of reflected signals, time delay, reflected signal power, direct signal power, and the phase difference between the reflected signal and the direct signal; Obtain the threshold value of the number of reflected signals, the time difference threshold between the reflected signal and the direct signal arriving at the receiver, the standard phase difference between the reflected signal and the direct signal, the weight factor of the number of reflected signals, the weight factor of the time delay, the weight factor of the reflected signal intensity, and the weight factor of the phase difference between the reflected signal and the direct signal from the group activity monitoring database; The sum average of the ratio of the number of reflected signals to the threshold value of the number of reflected signals in different multipath effect influence detection segments is recorded as the first component of the multipath effect influence; The ratio of the time delay at the multipath effect influence detection point to the time difference threshold between the reflected signal and the direct signal arriving at the receiver is recorded as the second component of the multipath effect influence; The ratio of the reflected signal power at the multipath effect influence detection point to the direct signal power at the multipath effect influence detection point is recorded as the third component of the multipath effect influence; The ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the multipath effect influence detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal is recorded as the fourth component of the multipath effect influence; The multipath effect influence evaluation value represents the quantitative data of the degree of influence of the first component of the multipath effect influence, the second component of the multipath effect influence, the third component of the multipath effect influence, and the fourth component of the multipath effect influence on the reflected signal during the propagation process.

[0012] Further, the specific steps for obtaining the dynamic analysis and evaluation value of abnormal behavior are as follows: The dynamic analysis data of abnormal behavior includes the phase difference, time delay, adjusted threshold, threshold before adjustment, and time difference between the reflected signal and the direct signal; Obtain the standard phase difference between the reflected signal and the direct signal, the time difference threshold for the reflected signal and the direct signal to reach the receiver, the weight factor of the dynamic threshold adaptability coefficient, the weight factor of the phase difference between the reflected signal and the direct signal, and the weight factor of the time delay from the group activity monitoring database; Calculate the ratio of the absolute value of the difference between the adjusted threshold and the threshold before adjustment under the dynamic threshold adaptability time window detection segment to the threshold before adjustment under the dynamic threshold adaptability time window detection segment to obtain the dynamic threshold adaptability coefficient; Sum and average the dynamic threshold adaptability coefficients in different dynamic threshold adaptability time window detection segments, denoted as the first component of the dynamic analysis of abnormal behavior; Denote the ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the abnormal behavior dynamic analysis detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal as the second component of the dynamic analysis of abnormal behavior; Denote the ratio of the time delay at the abnormal behavior dynamic analysis detection point to the time difference threshold for the reflected signal and the direct signal to reach the receiver as the third component of the dynamic analysis of abnormal behavior; The dynamic analysis and evaluation value of abnormal behavior represents the quantitative data of the combined dynamic analysis ability of the first component, the second component, and the third component of the dynamic analysis of abnormal behavior to the characteristics of abnormal behavior.

[0013] Further, the specific steps for comprehensively analyzing and obtaining the construction accuracy evaluation value of the activity monitoring model are as follows: Obtain the data update frequency threshold, the weight factor of the dynamic analysis and evaluation value of abnormal behavior, the weight factor of the data update frequency, and the weight factor of the multi-path effect influence evaluation value from the group activity monitoring database; Sum and average the dynamic analysis and evaluation values of abnormal behavior at different abnormal behavior dynamic analysis detection points, denoted as the first component of the construction accuracy of the activity monitoring model; Sum and average the data update frequency at the construction accuracy detection point of the activity monitoring model and the data update frequency threshold, denoted as the second component of the construction accuracy of the activity monitoring model; Sum and average the multi-path effect influence evaluation values at different multi-path effect influence detection points, denoted as the third component of the construction accuracy of the activity monitoring model; The construction accuracy evaluation value of the activity monitoring model represents the quantitative data of the combined influence degree of the first component, the second component, and the third component of the construction accuracy of the activity monitoring model on the overall construction accuracy of the activity monitoring model in multiple dimensions.

[0014] Further, the specific steps for constructing the basic model for monitoring social group activities are as follows: construct data according to the social group activity monitoring model and initialize the basic model for monitoring social group activities; set the parameters of the basic model for monitoring social group activities, where the parameters of the basic model for monitoring social group activities include the evaluation value of the multi-path effect impact, the evaluation value of the dynamic analysis of abnormal behaviors, and the evaluation value of the construction accuracy of the activity monitoring model; input the data for constructing the social group activity monitoring model as training data into the basic model for monitoring social group activities for training, and continuously adjust the parameters of the basic model for monitoring social group activities to obtain the social group activity monitoring model.

[0015] Further, the specific steps for optimizing and adjusting the method for the construction accuracy of the activity monitoring model for the basic model for monitoring social group activities are as follows: import the first determination value of the evaluation value of the multi-path effect impact, the second determination value of the evaluation value of the dynamic analysis of abnormal behaviors, and the third determination value of the evaluation value of the construction accuracy of the activity monitoring model in the social group activity monitoring database into the social group activity monitoring model, and output the adjustment method through the social group activity monitoring model. The adjustment method output by the social group activity monitoring model includes the method for optimizing the multi-path effect impact, the method for optimizing the dynamic analysis of abnormal behaviors, and the method for optimizing the construction accuracy of the activity monitoring model.

[0016] Further, the specific steps for the method for optimizing the multi-path effect impact are as follows: if the evaluation value of the multi-path effect impact is less than or equal to the first determination value of the evaluation value of the multi-path effect impact, then do not perform the method for optimizing the multi-path effect impact; if the evaluation value of the multi-path effect impact is greater than the first determination value of the evaluation value of the multi-path effect impact, then filter the collected signal by applying a filter and automatically adjust the filter parameters according to the real-time signal characteristics. The filter parameters include the signal frequency and amplitude.

[0017] Further, the specific steps for the method for optimizing the dynamic analysis of abnormal behaviors are as follows: if the evaluation value of the dynamic analysis of abnormal behaviors is greater than or equal to the second determination value of the evaluation value of the dynamic analysis of abnormal behaviors, then there is no need to optimize the method for the dynamic analysis of abnormal behaviors; if the evaluation value of the dynamic analysis of abnormal behaviors is less than the second determination value of the evaluation value of the dynamic analysis of abnormal behaviors, then formulate selective transmission and preferentially transmit key data. The key data includes location and abnormal behaviors; cooperate with a distributed computing cluster to slice and process the massive data and allocate it to different computing nodes for parallel processing.

[0018] Further, the specific steps of the method for optimizing the construction accuracy of the activity monitoring model are as follows: If the evaluation value of the construction accuracy of the activity monitoring model is greater than or equal to the third determination value of the evaluation value of the construction accuracy of the activity monitoring model, then there is no need to optimize the method for the construction accuracy of the activity monitoring model; if the evaluation value of the construction accuracy of the activity monitoring model is less than the third determination value of the evaluation value of the construction accuracy of the activity monitoring model, then the system for real-time monitoring of passenger behavior by cameras is used to identify abnormal behaviors, and the abnormal behavior information is converted into intuitive visual elements, where the intuitive visual elements include color changes and flashing icons. A light signal transmission device composed of LED lights is used to emit visual warning information. Personnel in specific areas are equipped with handheld devices with optoelectronic sensors to receive the devices to decode the light signals and display the visual warning information, and the personnel locate the abnormal passengers according to the warning information; a warning confirmation mechanism is established to allow operators to confirm or reject the warning information to reduce false alarms, and a linkage mechanism between the warning information and the emergency response system is established. Once a warning is issued, corresponding emergency response measures are immediately initiated.

[0019] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0020] 1. By collecting and processing the data for constructing the social group activity monitoring model, analyzing the data for constructing the social group activity monitoring model, constructing the basic model for social group activity monitoring, and optimizing and adjusting the method for the construction accuracy of the activity monitoring model for the basic model of social group activity monitoring, the reliability of the method for constructing the social group activity monitoring model is improved, and the problem of insufficient reliability of the method for constructing the social group activity monitoring model in the prior art is solved.

[0021] 2. By analyzing the data for constructing the social group activity monitoring model, the multi-path effect influence evaluation value and the abnormal behavior dynamic analysis evaluation value are obtained, which helps to correct the positioning errors caused by reasons such as signal reflection and refraction, and helps to distinguish normal activities and potential risk activities, thereby improving the accuracy of the monitoring data.

[0022] 3. By comprehensively analyzing the evaluation value of the construction accuracy of the activity monitoring model, it helps to discover and correct the errors in the model, reduce the phenomena of false alarms and missed alarms, provides a quantitative index for the model performance, and is convenient for comparison and evaluation. Description of the Drawings

[0023] Figure 1 It is a flowchart of the method for constructing a multi-dimensional social group activity monitoring model provided in the embodiments of the present application;

[0024] Figure 2 It is a schematic diagram of the function of the evaluation value of the construction accuracy of the activity monitoring model provided in the embodiments of the present application. Specific Embodiments

[0025] In an embodiment of the present application, by providing a method for constructing a social group activity monitoring model based on multiple dimensions, the problem in the prior art that the reliability of the method for constructing a social group activity monitoring model is insufficient is solved. By collecting and processing the data for constructing the social group activity monitoring model, analyzing the data for constructing the social group activity monitoring model, constructing a basic model for social group activity monitoring, and optimizing and adjusting the method for constructing the activity monitoring model for the basic model of social group activity monitoring, the reliability of the method for constructing the social group activity monitoring model is improved.

[0026] The technical solution in the embodiment of the present application for solving the above problem of insufficient reliability of the method for constructing a social group activity monitoring model has the following general idea:

[0027] By collecting and processing the data for constructing the social group activity monitoring model, analyzing the data for constructing the social group activity monitoring model, constructing a basic model for social group activity monitoring, and optimizing and adjusting the method for constructing the activity monitoring model for the basic model of social group activity monitoring, the reliability of the method for constructing the social group activity monitoring model is improved.

[0028] To better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments.

[0029] As Figure 1 shown, it is a flowchart of a method for constructing a social group activity monitoring model based on multiple dimensions provided by an embodiment of the present application. The method includes the following steps: collecting and processing the data for constructing the social group activity monitoring model; analyzing the data for constructing the social group activity monitoring model to obtain a multi-path effect influence evaluation value and an abnormal behavior dynamic analysis evaluation value; comprehensively analyzing the multi-path effect influence evaluation value and the abnormal behavior dynamic analysis evaluation value to obtain a construction accuracy evaluation value for the activity monitoring model; constructing a basic model for social group activity monitoring, and optimizing and adjusting the method for constructing the activity monitoring model for the basic model of social group activity monitoring according to the multi-path effect influence evaluation value, the abnormal behavior dynamic analysis evaluation value, and the construction accuracy evaluation value of the activity monitoring model.

[0030] Further, the specific steps for collecting and processing the data for constructing the social group activity monitoring model are as follows: receiving GPS satellite signals through a handheld GPS locator to collect the original data for constructing the social group activity monitoring model; cleaning and denoising the original data for constructing the social group activity monitoring model to obtain the data for constructing the social group activity monitoring model, where the data for constructing the social group activity monitoring model includes multi-path effect influence data and abnormal behavior dynamic analysis data.

[0031] Further, the specific steps to obtain the multipath effect influence evaluation value are as follows: The multipath effect influence data includes the number of reflected signals, time delay, reflected signal power, direct signal power, and the phase difference between the reflected signal and the direct signal; Obtain the threshold of the number of reflected signals, the time difference threshold between the reflected signal and the direct signal arriving at the receiver, the standard phase difference between the reflected signal and the direct signal, the weight factor of the number of reflected signals, the weight factor of time delay, the weight factor of the reflected signal intensity, and the weight factor of the phase difference between the reflected signal and the direct signal from the group activity monitoring database; Sum and average the ratio of the number of reflected signals to the threshold of the number of reflected signals in different multipath effect influence detection segments, which is denoted as the first component of the multipath effect influence; Denote the ratio of the time delay at the multipath effect influence detection point to the time difference threshold between the reflected signal and the direct signal arriving at the receiver as the second component of the multipath effect influence; Denote the ratio of the reflected signal power at the multipath effect influence detection point to the direct signal power at the multipath effect influence detection point as the third component of the multipath effect influence; Denote the ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the multipath effect influence detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal as the fourth component of the multipath effect influence; The multipath effect influence evaluation value represents the quantitative data of the degree of influence of the first component of the multipath effect influence, the second component of the multipath effect influence, the third component of the multipath effect influence, and the fourth component of the multipath effect influence on the reflected signal during propagation.

[0032] In this embodiment, the specific method for analyzing and obtaining the multipath effect influence evaluation value is as follows:

[0033] ;

[0034] ;

[0035] Number the preset multipath effect influence detection points in sequence. Denote as the number of the multipath effect influence detection point under the -th multipath effect influence detection segment, where

[0036] Denote the total number of the multipath effect influence detection points as Denote as the number of the multipath effect influence detection segment, where Denote the total number of the multipath effect influence detection segments as

[0037] Denote The multipath effect affects the evaluation value of the multipath effect at the detection point.

[0038] Denote the number of reflected signals under the

[0039] th multipath effect detection segment, and the number of reflected signals is obtained by a high-precision GPS receiver.

[0040] Denote the time delay under the

[0041] th multipath effect detection point, which represents the time difference between the arrival of the reflected signal and the direct signal at the receiver. It refers to the time difference between the signal that directly arrives at the receiver from the transmitter and the signal that arrives at the receiver after one reflection during the same signal transmission process. The signal propagation mainly includes two ways: direct and reflection. The direct signal is the signal that directly arrives at the receiver from the transmitter, while the reflected signal is the signal that arrives at the receiver after one or more reflections. In the actual signal transmission process, due to various objects in the environment such as buildings and the ground that will reflect the signal, the receiver usually receives both the direct signal and the reflected signal at the same time, and the time delay is obtained by a high-precision GPS receiver.

[0042] Denote the power of the reflected signal under the

[0043] th multipath effect detection point, and the power of the reflected signal is obtained by a high-precision GPS receiver. Denote the

[0044] power of the direct signal under the th multipath effect detection point, and the power of the direct signal is obtained by a high-precision GPS receiver.

[0045] represents the standard phase difference between the reflected signal and the direct signal, which is the preset standard phase difference of the reflected signal and the direct signal obtained from the group activity monitoring database, representing the expected phase difference between the reflected signal and the direct signal.

[0046] is the preset weight factor of the number of reflected signals obtained from the group activity monitoring database.

[0047] is the preset time delay weight factor obtained from the group activity monitoring database.

[0048] is the preset reflected signal intensity weight factor obtained from the group activity monitoring database.

[0049] is the preset phase difference weight factor between the reflected signal and the direct signal obtained from the group activity monitoring database.

[0050] The preset weight factor of the number of reflected signals, the preset time delay weight factor, the preset reflected signal intensity weight factor, and the preset phase difference weight factor between the reflected signal and the direct signal are obtained through a mapping relationship. For example, mapping sets of the number of reflected signals, time delay, reflected signal intensity, and the phase difference between the reflected signal and the direct signal and their corresponding weights are established respectively through the relationships between the number of reflected signals, time delay, reflected signal intensity, and the phase difference between the reflected signal and the direct signal and the reflected signal intensity in historical data. The corresponding preset weight factor of the number of reflected signals, preset time delay weight factor, preset reflected signal intensity weight factor, and preset phase difference weight factor between the reflected signal and the direct signal in the mapping set are obtained by inputting the real-time number of reflected signals, time delay, reflected signal intensity, and the phase difference between the reflected signal and the direct signal.

[0051] The larger the number of reflected signals, the more reflectors the signal encounters during propagation, resulting in different reflection paths, and each reflection path has a time delay because the propagation distance and speed of the signal are different. Therefore, an increase in the number of reflected signals is usually accompanied by multiple reflected signals with different time delays. The more the number of reflected signals, the longer the time delay; the direct signal power is the power of the signal directly from the transmitter to the receiver, while the reflected signal power is the power of the signal after reflection. The larger the reflected power, the smaller the direct power; the time delay will cause a change in the phase of the signal. The direct signal power and the phase difference between the reflected signal and the direct signal are the phase differences between the reflected signal and the direct signal. The longer the time delay, the larger the direct signal power and the phase difference between the reflected signal and the direct signal.

[0052] There is a positive correlation between the number of reflected signals and the evaluation value of the influence of multipath effects. The more the number of reflected signals, the more obvious the multipath effects, and the greater the evaluation value of the influence of multipath effects. There is a positive correlation between the time delay and the evaluation value of the influence of multipath effects. The difference in time delay will cause the expansion of the signal in the time domain. The longer the time delay, the greater the evaluation value of the influence of multipath effects. There is a positive correlation between the ratio of the power of the reflected signal to the power of the direct signal and the evaluation value of the influence of multipath effects. The larger the ratio of the power of the reflected signal to the power of the direct signal, the greater the intensity of the multipath interference, and the greater the evaluation value of the influence of multipath effects. There is a positive correlation between the phase difference between the reflected signal and the direct signal and the evaluation value of the influence of multipath effects. The larger the phase difference between the reflected signal and the direct signal, the greater the impact on the overall intensity and stability of the signal, and the greater the evaluation value of the influence of multipath effects.

[0053] Furthermore, the specific steps to obtain the evaluation value of abnormal behavior dynamic analysis are as follows: The abnormal behavior dynamic analysis data includes the phase difference between the reflected signal and the direct signal, the time delay, the adjusted threshold, the threshold before adjustment, and the time difference. Obtain the standard phase difference between the reflected signal and the direct signal, the time difference threshold for the reflected signal and the direct signal to reach the receiver, the weight factor of the dynamic threshold adaptability coefficient, the weight factor of the phase difference between the reflected signal and the direct signal, and the weight factor of the time delay from the group activity monitoring database. Calculate the ratio of the absolute value of the difference between the adjusted threshold and the threshold before adjustment under the dynamic threshold adaptability time window detection section to the threshold before adjustment under the dynamic threshold adaptability time window detection section to obtain the dynamic threshold adaptability coefficient. Sum and average the dynamic threshold adaptability coefficient in different dynamic threshold adaptability time window detection sections, denoted as the first component of abnormal behavior dynamic analysis. Denote the ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the abnormal behavior dynamic analysis detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal as the second component of abnormal behavior dynamic analysis. Denote the ratio of the time delay at the abnormal behavior dynamic analysis detection point to the time difference threshold for the reflected signal and the direct signal to reach the receiver as the third component of abnormal behavior dynamic analysis. The evaluation value of abnormal behavior dynamic analysis represents the quantitative data of the combined dynamic analysis ability of the first component of abnormal behavior dynamic analysis, the second component of abnormal behavior dynamic analysis, and the third component of abnormal behavior dynamic analysis on the characteristics of abnormal behavior.

[0054] In this embodiment, the specific method for analyzing and obtaining the evaluation value of abnormal behavior dynamic analysis is as follows:

[0055] ;

[0056] ; ;

[0057] Number the preset abnormal behavior dynamic analysis detection points in sequence. Indicates the number of the abnormal behavior dynamic analysis detection point. , Indicates the total number of the numbers of the abnormal behavior dynamic analysis detection points.

[0058] Number the preset dynamic threshold adaptability time window detection segments in sequence. Indicates the number of the dynamic threshold adaptability time window detection segment. , Indicates the total number of the numbers of the dynamic threshold adaptability time window detection segments.

[0059] Indicates the th abnormal behavior dynamic analysis evaluation value of the abnormal behavior dynamic analysis detection point.

[0060] Indicates the th dynamic threshold adaptability coefficient under the dynamic threshold adaptability time window detection segment, indicating the change rate of the threshold within the dynamic threshold adaptability time window.

[0061] Indicates the th phase difference between the reflected signal and the direct signal under the abnormal behavior dynamic analysis detection point, and the phase difference between the reflected signal and the direct signal is obtained by a high-precision GPS receiver.

[0062] Indicates the standard phase difference between the reflected signal and the direct signal, which is the preset standard phase difference between the reflected signal and the direct signal obtained from the group activity monitoring database.

[0063] Indicates the th time delay under the abnormal behavior dynamic analysis detection point, indicating the time difference between the arrival of the reflected signal and the direct signal at the receiver, referring to the time difference between the signal that the direct signal directly arrives at the receiver from the transmitter and the signal that the reflected signal arrives at the receiver after one reflection during the same signal transmission process, and the time delay is obtained by a high-precision GPS receiver.

[0064] Indicates the time difference threshold between the arrival of the reflected signal and the direct signal at the receiver, which is the preset time difference threshold between the arrival of the reflected signal and the direct signal at the receiver obtained from the group activity monitoring database, and is used to judge whether the time difference between the reflected signal and the direct signal is within the normal range.

[0065] Indicates the adjusted threshold under the

[0066] Indicates the threshold before adjustment under the

[0067] Indicates the time difference under the

[0068] dynamic threshold adaptation time window detection segment, that is, the time interval between two adjacent threshold adjustments. If no threshold adjustment occurs within a time window, the time difference can be marked as infinity or a specific value. At this time, the dynamic threshold adaptation coefficient should be zero or close to zero, indicating no adjustment, and the time difference is obtained through data processing software.

[0069] Is the preset dynamic threshold adaptation coefficient weight factor obtained from the group activity monitoring database.

[0070] Is the preset phase difference weight factor of the reflected signal and the direct signal obtained from the group activity monitoring database.

[0071] The preset dynamic threshold adaptation coefficient weight factor, the preset phase difference weight factor of the reflected signal and the direct signal, and the preset time delay weight factor are obtained through a mapping relationship. For example, mapping sets of the dynamic threshold adaptation coefficient, the phase difference of the reflected signal and the direct signal, and the time delay and their corresponding weights are established respectively through the relationships between the dynamic threshold adaptation coefficient, the phase difference of the reflected signal and the direct signal, the time delay, and the number of reflected signals in historical data. The corresponding preset dynamic threshold adaptation coefficient weight factor, the preset phase difference weight factor of the reflected signal and the direct signal, and the preset time delay weight factor in the mapping set are obtained by inputting the real-time number of reflected signals, time delay, reflected signal intensity, and the phase difference between the reflected signal and the direct signal.

[0072] Time delay refers to the arrival time difference caused by different path lengths during signal transmission. The phase difference between the reflected signal and the direct signal is due to different path lengths. The longer the time delay, the greater the phase difference between the reflected signal and the direct signal. The threshold before adjustment and the threshold after adjustment are used for signal detection or decision-making. Threshold adjustment is to adapt to the change of signal strength, improve detection performance or reduce misjudgment. The greater the phase difference between the reflected signal and the direct signal, the more necessary it is to adjust the threshold to correctly distinguish these two signals. Time difference usually refers to the arrival time difference between two signals. The greater the time difference, the greater the phase difference between the reflected signal and the direct signal.

[0073] There is a negative correlation between the time delay and the evaluation value of abnormal behavior dynamic analysis. The longer the time delay, the more abnormal there is in the signal transmission process, such as path change. The greater the time delay, the smaller the evaluation value of abnormal behavior dynamic analysis. There is a negative correlation between the phase difference between the reflected signal and the direct signal and the evaluation value of abnormal behavior dynamic analysis. The greater the phase difference between the reflected signal and the direct signal, the more unexpected reflectors or path changes exist. The greater the phase difference between the reflected signal and the direct signal, the smaller the evaluation value of abnormal behavior dynamic analysis. There is a positive correlation between the product of the absolute value of the difference between the threshold after adjustment and the threshold before adjustment and the time difference and the evaluation value of abnormal behavior dynamic analysis. The greater the product of the absolute value of the difference between the threshold after adjustment and the threshold before adjustment and the time difference, the faster it can adapt to sudden changes in the environment. The greater the product of the absolute value of the difference between the threshold after adjustment and the threshold before adjustment and the time difference, the greater the evaluation value of abnormal behavior dynamic analysis.

[0074] Furthermore, the specific steps for comprehensively analyzing and obtaining the evaluation value of the construction accuracy of the activity monitoring model are as follows: Obtain the data update frequency threshold, the weight factor of the evaluation value of abnormal behavior dynamic analysis, the weight factor of data update frequency, and the weight factor of the evaluation value of multipath effect influence from the group activity monitoring database; Sum and average the evaluation value of abnormal behavior dynamic analysis at different abnormal behavior dynamic analysis detection points, which is recorded as the first component of the construction accuracy of the activity monitoring model; Sum and average the data update frequency under the construction accuracy detection point of the activity monitoring model and the data update frequency threshold, which is recorded as the second component of the construction accuracy of the activity monitoring model; Sum and average the evaluation value of multipath effect influence at different multipath effect influence detection points, which is recorded as the third component of the construction accuracy of the activity monitoring model; The evaluation value of the construction accuracy of the activity monitoring model represents the quantitative data of the influence degree of the first component of the construction accuracy of the activity monitoring model, the second component of the construction accuracy of the activity monitoring model, and the third component of the construction accuracy of the activity monitoring model on the overall construction accuracy of the activity monitoring model in multiple dimensions.

[0075] In this embodiment, the specific method for analyzing and obtaining the evaluation value of the construction accuracy of the activity monitoring model is as follows:

[0076] ;

[0077] ;

[0078] Number the construction accuracy detection points of the preset activity monitoring model in sequence. Represents the number of the construction accuracy detection point of the activity monitoring model under the construction accuracy detection point of the activity monitoring model. , Represents the total number of the numbers of the construction accuracy detection points of the activity monitoring model.

[0079] Represents the construction accuracy evaluation value of the activity monitoring model.

[0080] Represents the Abnormal behavior dynamic analysis evaluation value of the abnormal behavior dynamic analysis detection point.

[0081] Represents the Data update frequency under the construction accuracy detection point of the activity monitoring model, and the data update frequency is obtained through fiber optic sensing technology.

[0082] Represents the data update frequency threshold, which is the preset data update frequency threshold obtained from the group activity monitoring database.

[0083] Represents the Multipath effect influence evaluation value of the multipath effect influence detection point.

[0084] Is the preset abnormal behavior dynamic analysis evaluation value weight factor obtained from the group activity monitoring database.

[0085] Is the preset data update frequency weight factor obtained from the group activity monitoring database.

[0086] Is the preset multipath effect influence evaluation value weight factor obtained from the group activity monitoring database.

[0087] The preset abnormal behavior dynamic analysis evaluation value weight factor, the preset data update frequency weight factor, and the preset multipath effect impact evaluation value weight factor are obtained through a mapping relationship. For example, mapping sets of the abnormal behavior dynamic analysis evaluation value, data update frequency, and multipath effect impact evaluation value and their corresponding weights are established respectively based on the relationships between the abnormal behavior dynamic analysis evaluation value, data update frequency, multipath effect impact evaluation value, and data transmission rate in historical data. By inputting the real-time abnormal behavior dynamic analysis evaluation value, data update frequency, and multipath effect impact evaluation value, the corresponding preset abnormal behavior dynamic analysis evaluation value weight factor, preset data update frequency weight factor, and preset multipath effect impact evaluation value weight factor in the mapping set are obtained.

[0088] Table 1 is an example table of the construction accuracy evaluation value of the activity monitoring model. The example parameters in Table 1 are only taken from the parameters under one construction accuracy detection point of an activity monitoring model for illustration. The weight factor is set to 0.4, the weight factor is set to 0.3, the weight factor is set to 0.3, The data update frequency threshold is 2. Table 1 is shown as follows.

[0089] Table 1 Example Table of the Construction Accuracy Evaluation Value of the Activity Monitoring Model Evaluation value of dynamic analysis of abnormal behavior Data update frequency Evaluation value of the influence of multipath effect Evaluation value of the construction accuracy of the activity monitoring model 4 5 3 5.51872090775 4 3 2 4.85493819144 5 4 3 7.08617791316 6 2 7 4.79991345964 3 2 4 2.03713139561

[0090] The abnormal behavior dynamic analysis evaluation value depends on the speed of data update. The faster the data update, the more real-time behavior information can be provided, and the larger the abnormal behavior dynamic analysis evaluation value. The data update frequency affects the real-time nature of the abnormal behavior dynamic analysis evaluation value. The faster the data update frequency, the faster the abnormal behavior of tourists can be captured, and the larger the abnormal behavior dynamic analysis evaluation value. The multipath effect impact evaluation value affects the accuracy of the abnormal behavior dynamic analysis evaluation value. The larger the multipath effect impact evaluation value, the accuracy of the abnormal behavior analysis will be affected, and the smaller the abnormal behavior dynamic analysis evaluation value.

[0091] As can be seen from Table 1, there is a positive correlation between the evaluation value of abnormal behavior dynamic analysis and the evaluation value of the construction accuracy of the activity monitoring model. The higher the evaluation value of abnormal behavior dynamic analysis, the more accurately it can identify and predict the abnormal behavior of tourists. The larger the evaluation value of abnormal behavior dynamic analysis, the larger the evaluation value of the construction accuracy of the activity monitoring model; there is a positive correlation between the data update frequency and the evaluation value of the construction accuracy of the activity monitoring model. The higher the data update frequency, the faster the latest location and behavior information of tourists can be obtained, so that the activity monitoring model can make more real-time responses and judgments. The higher the data update frequency, the larger the evaluation value of the construction accuracy of the activity monitoring model; there is a negative correlation between the evaluation value of the multi-path effect impact and the evaluation value of the construction accuracy of the activity monitoring model. The higher the evaluation value of the multi-path effect impact, the larger the positioning error. The larger the evaluation value of the multi-path effect impact, the smaller the evaluation value of the construction accuracy of the activity monitoring model.

[0092] As shown in Figure 2 Figure 5 is a schematic diagram of the evaluation value function of the construction accuracy of the activity monitoring model provided by the embodiment of the present application; the positive half-axis of the x-axis is the abscissa, and the positive half-axis of the y-axis is the ordinate. The weight factor is set to 0.4, and the weight factor is set to 0.3, and the weight factor is set to 0.3. The data update frequency threshold is 2.

[0093] Curve a indicates that if the data update frequency is set to a fixed value of 1, the evaluation value of the multi-path effect impact is set to a fixed value of 1, the evaluation value of abnormal behavior dynamic analysis is x, and the evaluation value of the construction accuracy of the activity monitoring model is y, the evaluation value of the construction accuracy of the activity monitoring model increases as the evaluation value of abnormal behavior dynamic analysis increases.

[0094] Furthermore, the specific steps for constructing the basic model of social group activity monitoring are as follows: Initialize the basic model of social group activity monitoring according to the data for constructing the social group activity monitoring model; Set the parameters of the basic model of social group activity monitoring, and the parameters of the basic model of social group activity monitoring include the evaluation value of the multi-path effect impact, the evaluation value of abnormal behavior dynamic analysis, and the evaluation value of the construction accuracy of the activity monitoring model; Input the data for constructing the social group activity monitoring model as training data into the basic model of social group activity monitoring for training, and continuously adjust the parameters of the basic model of social group activity monitoring to obtain the social group activity monitoring model.

[0095] Further, the specific steps for optimizing and adjusting the construction accuracy method of the activity monitoring model for social group activities are as follows: Import the first determination value of the multipath effect influence evaluation value, the second determination value of the abnormal behavior dynamic analysis evaluation value, and the third determination value of the construction accuracy evaluation value of the activity monitoring model in the social group activity monitoring database into the social group activity monitoring model. The social group activity monitoring model outputs an adjustment method, and the adjustment method output by the social group activity monitoring model includes an optimized multipath effect influence method, an optimized abnormal behavior dynamic analysis method, and an optimized construction accuracy method of the activity monitoring model.

[0096] Further, the specific steps for optimizing the multipath effect influence method are as follows: If the multipath effect influence evaluation value is less than or equal to the first determination value of the multipath effect influence evaluation value, the multipath effect influence method is not optimized; if the multipath effect influence evaluation value is greater than the first determination value of the multipath effect influence evaluation value, filter the collected signal by applying a filter, and automatically adjust the filter parameters according to the real-time signal characteristics. The filter parameters include signal frequency and amplitude.

[0097] In this embodiment, a first determination value of the multipath effect influence evaluation value is preset to 1, and the multipath effect influence evaluation value is 1.5. Since the multipath effect influence evaluation value is greater than the first determination value of the multipath effect influence evaluation value, optimization and adjustment are performed. First, select a suitable filter, such as a Kalman filter, to filter the collected signal, analyze the characteristics of the real-time signal, including changes in parameters such as signal frequency and amplitude, and automatically adjust the parameters of the filter according to the real-time signal characteristics to adapt to the signal changes and reduce the influence of the multipath effect. For example, in a karst cave environment, a handheld GPS locator is used to monitor social group activities. Due to the complex environment in the karst cave and obvious multipath effect, the GPS signal is interfered. Apply an adaptive filter to filter the GPS signal, analyze the real-time GPS signal, and find that the signal frequency fluctuates greatly and the amplitude is unstable. Automatically adjust the filter parameters: According to the changes in signal frequency and amplitude, automatically adjust the parameters of the adaptive filter. For example, when the signal frequency increases, increase the bandwidth of the filter; when the signal amplitude decreases, adjust the gain of the filter.

[0098] Further, the specific steps for optimizing the abnormal behavior dynamic analysis method are as follows: If the abnormal behavior dynamic analysis evaluation value is greater than or equal to the second determination value of the abnormal behavior dynamic analysis evaluation value, the abnormal behavior dynamic analysis method does not need to be optimized; if the abnormal behavior dynamic analysis evaluation value is less than the second determination value of the abnormal behavior dynamic analysis evaluation value, formulate selective transmission and preferentially transmit key data. The key data includes location and abnormal behavior; cooperate with a distributed computing cluster to slice and process the massive data and allocate it to different computing nodes for parallel processing.

[0099] In this embodiment, a second determination value of the abnormal behavior dynamic analysis evaluation value is preset to be 0.7, and the abnormal behavior dynamic analysis evaluation value is 0.5. Since the abnormal behavior dynamic analysis evaluation value is less than the second determination value of the abnormal behavior dynamic analysis evaluation value, optimization and adjustment are performed. For example, when the members of a tour group enter a karst cave, a monitoring system is used to analyze the behaviors of the tour group members in real time. Due to the huge amount of data, the abnormal behavior dynamic analysis evaluation value is relatively low and fails to reach the preset 0.7. First, a selective transmission strategy is formulated to identify key data, including location information and abnormal behavior identifiers, to ensure that these key data are preferentially processed during transmission to reduce latency. A distributed computing cluster is configured to fragment the massive data for distributed processing. The data fragments are assigned to different computing nodes, and each computing node processes its own data fragment simultaneously to improve the processing efficiency.

[0100] Further, the specific steps of the method for optimizing the construction accuracy of the activity monitoring model are as follows: If the construction accuracy evaluation value of the activity monitoring model is greater than or equal to the third determination value of the construction accuracy evaluation value of the activity monitoring model, there is no need to optimize the method for the construction accuracy of the activity monitoring model; if the construction accuracy evaluation value of the activity monitoring model is less than the third determination value of the construction accuracy evaluation value of the activity monitoring model, the abnormal behavior is identified through the system for real-time monitoring of passenger behaviors by cameras, and the abnormal behavior information is converted into intuitive visual elements. The intuitive visual elements include color changes and flashing icons. A light signal transmission device composed of LED lights is used to emit visual warning information. Personnel in a specific area are equipped with handheld devices with optoelectronic sensors to receive the devices, decode the light signals, and display the visual warning information. Personnel can locate the abnormal passengers based on the warning information. A warning confirmation mechanism is established to allow operators to confirm or reject the warning information to reduce false alarms. A linkage mechanism between the warning information and the emergency response system is established. Once a warning is issued, corresponding emergency response measures are immediately initiated.

[0101] In this embodiment, a third determination value of the construction accuracy evaluation value of an activity monitoring model is preset to be 1.3, and the construction accuracy evaluation value of the activity monitoring model is 1. Since the construction accuracy evaluation value of the activity monitoring model is less than the third determination value of the construction accuracy evaluation value of the activity monitoring model, optimization and adjustment are performed. For example, in a large-scale social group activity, a camera monitors the activity venue in real time. The system identifies abnormal gathering behavior in a certain area, which is displayed in red on the monitoring screen and a flashing warning icon appears. The optical signal transmission device composed of LED lights emits a red flashing optical signal to this area. The handheld devices of the security personnel in the area receive the optical signal and display a red warning and specific location information. The security personnel quickly go to the designated location for verification. The security personnel confirm on-site that the abnormal behavior is true, confirm the early warning through the handheld device, the system records the confirmation information, and further analyzes the cause of the abnormality. After the early warning is confirmed, the emergency response system is immediately activated to evacuate the nearby crowd, block the relevant area, and the security personnel and the emergency team quickly arrive at the scene to handle the abnormal situation.

[0102] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) containing computer-usable program code.

[0103] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for realizing the functions specified in Figure 1 one or more flows Figure 1 or multiple flows and / or blocks

[0104] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device realizes the functions specified in Figure 1 one or more flows Figure 1 or multiple flows and / or blocks

[0105] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions for implementing the functions specified in one process or multiple processes and / or one block or multiple blocks in the process Figure 1 one process or multiple processes and / or Figure 1 the steps of the functions specified in one block or multiple blocks.

[0106] Although the preferred embodiments of the present invention have been described, additional changes and modifications can be made by those skilled in the art once they learn of the basic creative concept. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0107] Obviously, those skilled in the art can make various changes and deformations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and deformations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention is also intended to include these modifications and deformations.

Claims

1. Method for constructing multi-dimensional social group activity monitoring model, characterized in that Including the following steps: Collect and process the data for constructing the social group activity monitoring model; Analyze the data for constructing the social group activity monitoring model to obtain the multi-path effect impact evaluation value and the abnormal behavior dynamic analysis evaluation value; Comprehensively analyze the multi-path effect impact evaluation value and the abnormal behavior dynamic analysis evaluation value to obtain the construction accuracy evaluation value of the activity monitoring model; Construct the basic model for social group activity monitoring, and optimize and adjust the construction accuracy method of the activity monitoring model for the basic model of social group activity monitoring according to the multi-path effect impact evaluation value, the abnormal behavior dynamic analysis evaluation value, and the construction accuracy evaluation value of the activity monitoring model.

2. The construction method of the multi-dimensional social group activity monitoring model according to claim 1, characterized in that The specific steps of collecting and processing the data for constructing the social group activity monitoring model are as follows: Collect the original data for constructing the social group activity monitoring model by receiving GPS satellite signals through a handheld GPS locator; Clean and denoise the original data for constructing the social group activity monitoring model to obtain the data for constructing the social group activity monitoring model, and the data for constructing the social group activity monitoring model includes multi-path effect impact data and abnormal behavior dynamic analysis data.

3. The construction method of the multi-dimensional social group activity monitoring model according to claim 1, wherein The specific steps of obtaining the multi-path effect impact evaluation value are as follows: The multi-path effect impact data includes the number of reflected signals, time delay, reflected signal power, direct signal power, and phase difference between the reflected signal and the direct signal; Obtain the threshold of the number of reflected signals, the time difference threshold between the reflected signal and the direct signal reaching the receiver, the standard phase difference between the reflected signal and the direct signal, the weight factor of the number of reflected signals, the weight factor of time delay, the weight factor of reflected signal strength, and the weight factor of the phase difference between the reflected signal and the direct signal from the group activity monitoring database; Sum and average the ratio of the number of reflected signals to the threshold of the number of reflected signals in different multi-path effect impact detection segments, and record it as the first component of the multi-path effect impact; Record the ratio of the time delay at the multi-path effect impact detection point to the time difference threshold between the reflected signal and the direct signal reaching the receiver as the second component of the multi-path effect impact; Record the ratio of the reflected signal power at the multi-path effect impact detection point to the direct signal power at the multi-path effect impact detection point as the third component of the multi-path effect impact; Record the ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the multi-path effect impact detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal as the fourth component of the multi-path effect impact; The multi-path effect impact evaluation value represents the quantitative data of the degree of influence of the first component of the multi-path effect impact, the second component of the multi-path effect impact, the third component of the multi-path effect impact, and the fourth component of the multi-path effect impact on the reflected signal during the propagation process.

4. The construction method of the multi-dimensional social group activity monitoring model according to claim 1, wherein, The specific steps of obtaining the abnormal behavior dynamic analysis evaluation value are as follows: The abnormal behavior dynamic analysis data includes the phase difference between the reflected signal and the direct signal, time delay, adjusted threshold, threshold before adjustment, and time difference; Obtain the standard phase difference between the reflected signal and the direct signal, the time difference threshold for the reflected signal and the direct signal to reach the receiver, the weight factor of the dynamic threshold adaptability coefficient, the weight factor of the phase difference between the reflected signal and the direct signal, and the weight factor of the time delay from the group activity monitoring database; Calculate the ratio of the absolute value of the difference between the adjusted threshold and the pre-adjusted threshold under the dynamic threshold adaptability time window detection segment to the pre-adjusted threshold under the dynamic threshold adaptability time window detection segment to obtain the dynamic threshold adaptability coefficient; Sum and average the dynamic threshold adaptability coefficients in different dynamic threshold adaptability time window detection segments, denoted as the first component of the abnormal behavior dynamic analysis; Denote the ratio of the absolute value of the difference between the phase difference between the reflected signal and the direct signal at the abnormal behavior dynamic analysis detection point and the standard phase difference between the reflected signal and the direct signal to the standard phase difference between the reflected signal and the direct signal as the second component of the abnormal behavior dynamic analysis; Denote the ratio of the time delay at the abnormal behavior dynamic analysis detection point to the time difference threshold for the reflected signal and the direct signal to reach the receiver as the third component of the abnormal behavior dynamic analysis; The evaluation value of the abnormal behavior dynamic analysis represents the quantitative data of the combined dynamic analysis ability of the first component of the abnormal behavior dynamic analysis, the second component of the abnormal behavior dynamic analysis, and the third component of the abnormal behavior dynamic analysis on the characteristics of abnormal behavior.

5. The method for constructing a multi-dimensional social group activity monitoring model according to claim 1, wherein The specific steps for the comprehensive analysis to obtain the evaluation value of the construction accuracy of the activity monitoring model are as follows: Obtain the data update frequency threshold, the weight factor of the evaluation value of the abnormal behavior dynamic analysis, the weight factor of the data update frequency, and the weight factor of the evaluation value of the multipath effect influence from the group activity monitoring database; Sum and average the evaluation values of the abnormal behavior dynamic analysis at different abnormal behavior dynamic analysis detection points, denoted as the first component of the construction accuracy of the activity monitoring model; Sum and average the data update frequency at the construction accuracy detection point of the activity monitoring model and the data update frequency threshold, denoted as the second component of the construction accuracy of the activity monitoring model; Sum and average the evaluation values of the multipath effect influence at different multipath effect influence detection points, denoted as the third component of the construction accuracy of the activity monitoring model; The evaluation value of the construction accuracy of the activity monitoring model represents the quantitative data of the combined influence degree of the first component of the construction accuracy of the activity monitoring model, the second component of the construction accuracy of the activity monitoring model, and the third component of the construction accuracy of the activity monitoring model on the overall construction accuracy of the activity monitoring model in multiple dimensions.

6. The method for constructing a multi-dimensional social group activity monitoring model according to claim 1, wherein The specific steps for constructing the basic model for monitoring social group activities are as follows: Initialize the basic model for monitoring social group activities according to the data for constructing the social group activity monitoring model; Set the parameters of the basic model for monitoring social group activities, where the parameters of the basic model for monitoring social group activities include the evaluation value of the multipath effect influence, the evaluation value of the abnormal behavior dynamic analysis, and the evaluation value of the construction accuracy of the activity monitoring model; Input the data for constructing the social group activity monitoring model as training data into the basic social group activity monitoring model for training, and continuously adjust the parameters of the basic social group activity monitoring model to obtain the social group activity monitoring model.

7. The construction method of the multi-dimensional social group activity monitoring model according to claim 1, characterized in that, The specific steps for optimizing and adjusting the construction accuracy method of the activity monitoring model for the basic social group activity monitoring model are as follows: Import the first determination value of the multipath effect impact evaluation value, the second determination value of the abnormal behavior dynamic analysis evaluation value, and the third determination value of the construction accuracy evaluation value of the activity monitoring model in the group activity monitoring database into the social group activity monitoring model. The output adjustment method of the social group activity monitoring model is output by the social group activity monitoring model. The output adjustment method of the social group activity monitoring model includes an optimized multipath effect impact method, an optimized abnormal behavior dynamic analysis method, and an optimized construction accuracy method of the activity monitoring model.

8. The construction method of the multi-dimensional social group activity monitoring model according to claim 7, characterized in that, The specific steps of the optimized multipath effect impact method are as follows: If the multipath effect impact evaluation value is less than or equal to the first determination value of the multipath effect impact evaluation value, the optimized multipath effect impact method is not performed; If the multipath effect impact evaluation value is greater than the first determination value of the multipath effect impact evaluation value, filter the collected signal by applying a filter, and automatically adjust the filter parameters according to the real-time signal characteristics. The filter parameters include signal frequency and amplitude.

9. The method for constructing a multi-dimensional social group activity monitoring model according to claim 7, wherein, The specific steps of the optimized abnormal behavior dynamic analysis method are as follows: If the abnormal behavior dynamic analysis evaluation value is greater than or equal to the second determination value of the abnormal behavior dynamic analysis evaluation value, the optimized abnormal behavior dynamic analysis method is not required; If the abnormal behavior dynamic analysis evaluation value is less than the second determination value of the abnormal behavior dynamic analysis evaluation value, formulate selective transmission and preferentially transmit key data. The key data includes location and abnormal behavior; Cooperate with a distributed computing cluster to slice and process massive data and allocate it to different computing nodes for parallel processing.

10. The construction method of the multi-dimensional social group activity monitoring model according to claim 7, characterized in that The specific steps of the optimized construction accuracy method of the activity monitoring model are as follows: If the construction accuracy evaluation value of the activity monitoring model is greater than or equal to the third determination value of the construction accuracy evaluation value of the activity monitoring model, the optimized construction accuracy method of the activity monitoring model is not required; If the construction accuracy evaluation value of the activity monitoring model is less than the third determination value of the construction accuracy evaluation value of the activity monitoring model, identify abnormal behavior through the passenger behavior system monitored by the camera in real time, convert the abnormal behavior information into intuitive visual elements. The intuitive visual elements include color changes and flashing icons. Use the optical signal transmission device composed of LED lights to emit visual warning information. The personnel in the specific area are equipped with handheld devices with optoelectronic sensors to receive the device to decode the optical signal and display the visual warning information. The personnel locate the abnormal passengers according to the warning information; Establish a warning confirmation mechanism to allow operators to confirm or reject the warning information to reduce false alarms. Establish a linkage mechanism between the warning information and the emergency response system. Once a warning is issued, immediately initiate the corresponding emergency response measures.

Images