An abnormal situation alarm method based on target activity frequency regularity

Abstract

The application provides an abnormal situation warning method based on target activity frequency regularity, belongs to the field of intelligence reconnaissance, and is particularly in the situation analysis stage. The method extracts the target activity frequency regularity from the target historical activity database according to the time and date, corrects the warning parameters and the activity frequency threshold value in combination with the experience data of commanders, calculates the target monitored in real time, judges whether the activity frequency of the target at the current time and date exceeds the activity frequency threshold value, and generates threat warnings of different levels. The application can automatically monitor the activity frequency of the target of interest based on the historical activity data of the target, generate threat warnings, improve the efficiency of threat research and judgment, can be used in the field of situation analysis, and supports threat early warning analysis on enemy behaviors.

Description

An abnormal situation alarm method based on the frequency pattern of target activities Technical Field

[0001] This invention relates to an abnormal situation alarm method based on the frequency pattern of target activity, and studies its application in threat assessment. It belongs to the field of intelligence reconnaissance, specifically in the situation analysis stage. Background Technology

[0002] Common methods for issuing alerts for target anomalies include: 1. issuing alerts when a target enters, stays in, or leaves a region of interest; 2. issuing alerts when there are significant changes in the target's speed, altitude, and direction; 3. generating alert information by assessing the target's threat level using expert systems or mathematical models. These alerting methods primarily reflect the current threat situation based on changes in the target's state and parameters, lacking data-driven decision support from historical target activity knowledge and failing to reflect threats arising from patterned changes in the target. Therefore, it is necessary to extract activity patterns from the historical activities of key targets, and issue intelligent warnings when target activity deviates from these patterns, alerting commanders to pay close attention and implement accurate predictions and responses. Summary of the Invention

[0003] The purpose of this invention is to provide an abnormal situation alarm method based on the frequency pattern of target activity. By setting the area of ​​interest, the target of interest and alarm conditions, the activity frequency pattern is extracted from the historical activity of the target. The real-time activity of the target and the activity frequency pattern are compared and calculated. When the alarm conditions are met, different levels of threat warning are issued, thereby improving the efficiency and intelligence of threat warning.

[0004] This invention can be achieved through the following technical means:

[0005] An abnormal situation alarm method based on the frequency pattern of target activities includes the following steps:

[0006] Step 1: Set up alarm rules for abnormal situations. The rules include the key areas of focus and the set of targets of focus. The areas of focus are irregular polygons, and the attributes of the targets of focus include the type and model of the targets.

[0007] Step 2: Set the parameters corresponding to the target of interest when different levels of alarms occur. The alarm levels include high, medium and low, and the alarm parameter is the proportion that exceeds the normal activity frequency threshold.

[0008] Step 3: Based on the preset time period, statistically analyze the frequency of the target activity from the target historical activity database according to time and date to obtain the activity frequency threshold matrix for time and the activity frequency threshold matrix for date, thereby generating the target activity frequency pattern.

[0009] Step 4: Monitor the target's activity location in real time, determine whether the target's location is within the area of ​​interest and whether the target type and model meet the alarm rules, count the number of target activities that meet the conditions at the current time and date, compare them with the corresponding activity frequency threshold, and generate a threat alarm of the corresponding level.

[0010] Furthermore, step 4 specifically includes the following steps:

[0011] Step 4-1: Monitor the target activity in real time and extract the target's location information, target type, and target model;

[0012] Step 4-2: Determine whether the target is in the target set based on its type and model. If it is in the target set... i If they are the same, continue with the alarm judgment; otherwise, return to step 4-1 to monitor other targets; where 1≤i≤n, and n is the number of targets in the target set.

[0013] Step 4-3: Determine whether the target's location is within the area of ​​interest. If the target is within the area of ​​interest, continue with the alarm determination; otherwise, return to step 4-1 to monitor other targets.

[0014] Step 4-4: Obtain the current time t and date d, and calculate the number E of locations where the target appears within the area of ​​interest at the current time t and date d. i and F i Where 1≤i≤n, 1≤t≤24, 1≤d≤31;

[0015] Steps 4-5: Calculate the quantity E of the target at the current moment. i With target s i The alarm conditions are as follows: The activity frequency thresholds at corresponding times are compared.

[0016]

[0017] The quantity F of the target within the current date i With target s i The alarm conditions are as follows: The corresponding date activity frequency thresholds are compared.

[0018]

[0019] In the formula, c i，t For target s i At the activity frequency threshold at time t, g i，d For target s i The activity frequency threshold on date d, ΔD hi ΔDmi ΔD li Targets s i The alarm parameters corresponding to the high, medium, and low threat levels generated.

[0020] This invention provides anomaly alerts based on the frequency patterns of target activity. By setting areas of interest and targets of interest, it extracts frequency patterns from historical target activity data. Combined with commander experience data, it adjusts alert parameters and activity frequency thresholds, calculates alert levels for targets monitored in real time, and generates threat alerts of varying severity. This solves the problems of excessive reliance on manual analysis, low efficiency, and heavy workload associated with previous target activity alerts. It enables efficient and intelligent target activity warnings and is applicable to fields such as situation analysis and assessment. Attached Figure Description

[0021] Figure 1 is a flowchart of the abnormal situation alarm based on the frequency pattern of target activities according to the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail with reference to Figure 1.

[0023] This invention provides an anomaly alerting method based on the frequency pattern of target activities, comprising the following steps:

[0024] Step 1: Set up alarm rules for abnormal situations. The rules include a key area of ​​focus and a set of targets of interest. The area of ​​focus is an irregular polygon, and the targets of interest include the type and model of the target. Specifically, this includes the following steps:

[0025] Step 1-1: Add an irregular polygon to the alarm rules on the situation map;

[0026] Steps 1-2: Extract dictionary data such as target types and target models from the target knowledge base. Users select targets to be monitored based on their level of interest, and add the corresponding target types and models to the alarm rules, forming a set of targets of interest S = {s1, s2, ..., s...}. n}, where n is the number of targets of interest.

[0027] Step 2: Set the parameters corresponding to different alarm levels for the target of interest. Alarm levels include high, medium, and low. The alarm parameter is the percentage exceeding the normal activity frequency threshold. An alarm is triggered when the activity frequency of the target of interest exceeds the alarm parameter range. This includes the following steps:

[0028] Based on the set of targets of interest, each target of interest s is first read from the prior knowledge base. iThe parameters corresponding to alarms of different threat levels are as follows: threat levels are divided into high, medium, and low, and the corresponding alarm parameters are ΔD. hi ΔD mi ΔD li The meaning is the proportion exceeding the normal threshold. Users can modify the alarm parameters corresponding to different threat situations, where i∈[1,n].

[0029] Step 3: Based on the preset time period, statistically analyze the activity frequency of the target from the target historical activity database by time and date to obtain the activity frequency threshold matrix by time and the activity frequency threshold matrix by date, thus generating the target activity frequency pattern. This specifically includes the following steps:

[0030] Step 3-1: Based on the preset time period, calculate the activity frequency of each target from the historical activity data of the target, categorized by time and date. The activity frequency threshold matrix for each time period is as follows:

[0031]

[0032] The activity frequency threshold matrix for each date is as follows:

[0033]

[0034] Step 3-2: Display the activity frequency of each target in a line graph format. Users can adjust the activity frequency threshold based on their experience.

[0035] Step 4: Monitor the target's activity location in real time, determine if the target's location is within the area of ​​interest, and whether the target type and model meet the alarm rules. Count the number of target activities that meet the conditions at the current time and date, compare this count with the corresponding activity frequency threshold, and generate a threat alarm of the appropriate level. This specifically includes the following steps:

[0036] Step 4-1: Monitor battlefield targets in real time and extract attribute information such as target location, target type, and target model;

[0037] Step 4-2: Determine whether the target is in the target set based on its type and model. If it is in the target set... i If they are the same, continue the judgment; otherwise, return to step 4-1 to monitor other targets, where 1≤i≤n;

[0038] Step 4-3: Determine whether the target's location is within the area of ​​interest. If the target is within the area of ​​interest, continue with alarm analysis; otherwise, return to step 4-1 to monitor other targets.

[0039] Step 4-4: Obtain the current time t and date d, and calculate the number E of locations where the target appears within the area of ​​interest at the current time t and date d. i and F i Where 1≤i≤n, 1≤t≤24, 1≤d≤31;

[0040] Steps 4-5: Calculate the quantity E of the target at the current moment. i With target s i The alarm conditions are as follows: The activity frequency thresholds at corresponding times are compared.

[0041]

[0042] The quantity F of the target within the current date i With target s i The alarm conditions are as follows: The corresponding date activity frequency thresholds are compared.

[0043]

[0044] No alarm is generated under other circumstances; return to step 4-1 to monitor other targets. An alarm is issued for targets that meet the alarm conditions. Based on the alarm level and corresponding alarm notification method, the alert is displayed on the situation map using sound, flashing, or animation.

[0045] This invention proposes an anomaly alerting method based on the frequency patterns of target activity. It primarily focuses on the intelligent monitoring and calculation of anomalies in target activity frequency. The calculation is mainly based on historical target activity data, enabling the statistical analysis of frequency patterns from historical activity data. This allows commanders to modify activity frequency thresholds based on experience and knowledge, and generates different levels of threat alerts based on alarm conditions, thus improving the intelligence and efficiency of threat assessment. It can be used in situation analysis to calculate the threat situation of enemy targets, supporting the analysis of hostile maps and situation assessment.

Claims

1. An abnormal situation alarm method based on the frequency pattern of target activities, characterized in that, Includes the following steps: Step 1: Set alarm rules for abnormal situations, including a key area of ​​focus and a set of targets of focus; the area of ​​focus is an irregular polygon, and the attributes of the targets of focus include the type and model of the target; Step 2: Set parameters corresponding to different levels of alarms for the targets of focus, with alarm levels including high, medium, and low, and alarm parameters being the proportion exceeding the normal activity frequency threshold; Step 3: Based on a preset time period, statistically analyze the activity frequency of the targets of focus from the target historical activity database according to time and date, obtaining the activity frequency threshold matrix for time and the activity frequency threshold matrix for date, generating target activity frequency patterns; Step 4: Monitor the activity location of the targets in real time, determine whether the target location is within the area of ​​focus and whether the target type and model meet the alarm rules, statistically analyze the number of target activities that meet the conditions at the current time and date, and compare them with the corresponding activity frequency threshold to generate a threat alarm of the corresponding level; Specifically, Step 4 includes the following steps: Step 4-1: Monitor the target activity in real time and extract the target's location information, target type, and target model; Step 4-2: Determine whether the target is in the set of targets of focus based on the target type and model, and if it is in the target set... If they are the same, continue with the alarm judgment; otherwise, return to step 4-1 to monitor other targets. ， To determine the number of targets in the target set; Step 4-3: Determine if the target's location is within the target area. If the target is within the target area, continue with alarm determination; otherwise, return to Step 4-1 to monitor other targets; Step 4-4: Obtain the current time. and date Calculate the target at the current moment and date Number of locations appearing within the area of ​​interest and ， , , Steps 4-5: Calculate the quantity of the target at the current moment. With the goal The alarm conditions are as follows: The activity frequency thresholds at corresponding times are compared. The quantity of the target within the current date. With the goal The alarm conditions are as follows: The corresponding date activity frequency thresholds are compared. In the formula, For the goal At any moment The activity frequency threshold, For the goal On the date The activity frequency threshold, △ 、△ and △ The target The alarm parameters corresponding to the high, medium, and low threat levels generated.

Images