KR20240074932A - System for monitoring employing working of edge computing and working of security - Google Patents

Abstract

An integrated monitoring system using security technology and edge computing technology comprises: an IoT sensor module for sensing a state around each place; a Wi-Fi sensor installed for each place; an electronic display board device installed for each place to display warning broadcasting information; a speaker device installed for each place to output the warning broadcasting information by voice; a facility device installed for each place to control the state around each place; an IoT platform for transmitting a control command of a management information processing device to the facility device; and a management information processing device. The present invention secures a rapid risk situation recognition and response system through normal situation recognition.

Description

Integrated monitoring system using security technology and edge computing technology {System for monitoring employing working of edge computing and working of security}

The content disclosed in this specification is to enable the management device to perform appropriate situational recognition and respond when various dangerous situations occur in various places, for example, around residential areas or on trails.

Unless otherwise indicated herein, the material described in this section is not prior art to the claims of this application, and is not admitted to be prior art by inclusion in this section.

In general, as life has become more complex and changes rapidly in recent years, various difficulties have arisen in public safety, and interest in this has been shown in many ways. And, due to this situation, we are seeing a lot of movement in the fields of broadcasting, media, public welfare, and disaster safety.

So, when looking at this point, it would be useful to secure related technologies that can properly detect and recognize various external situations in various environments, such as manufacturing and complex places (living spaces), and respond smoothly. do.

Meanwhile, these technologies have traditionally centered on CCTV and video analysis technology, and have been mainly used in the case of disaster and social safety services. With the development of IoT-based safety services, it is expected that system introduction costs will be reduced and product substitution effects will be achieved. You will be able to.

Looking at these technologies, platform technologies for collecting sensor data for safety services, etc. are broadly classified into management devices and manufacturing data collection devices. In addition, management devices are developed as solutions that provide safety, security, and services in the production area, or are used in the manufacturing process to handle production, quality, and operations. In addition, it is not easy to develop manufacturing data collection devices from the device side and major parts of sensors and control systems linked to detailed equipment development.

In addition, unlike before, IoT technology also uses technologies such as lightweight IoT security gateways that provide lightweight device functions, device-customized security methods that resolve differences in HW limitations and performance, and lightweight/low-power security HW modules in the security area. can be developed

Therefore, it is also desirable to prevent risks in advance and provide secure services through the use of IoT sensors.

For example, it can be used as a monitoring and system that can recognize dangerous situations for tourists and inform them of dangerous situations in real time in tourist attractions with residential areas nearby, beach promenades where night traffic is prohibited, and places where fishing is prohibited.

In addition, it can be used not only in tourist destinations but also in the field of recognizing and notifying abnormal situations for safety.

The prior art literature against this background is the following.

(Patent Document 0001) KR1020150074060 A

(Patent Document 0002) KR1020180061841 A

For reference, Document 1 above relates to an apparatus and method for providing control services for IoT security, and roughly relates to technology for detecting and controlling abnormal packet behavior between IoT sensors and servers in an IoT-based network.

In addition, Document 2 is also about an edge-cloud system for automatic control of IoT devices, and shows the characteristics of the degree to which it is controlled when a specific state occurs through the pattern and situation of IoT sensor signals.

The disclosed content is intended to provide an integrated monitoring system using security technology and edge computing technology that can recognize dangerous situations and find quick response measures through normal situation recognition in tourist attractions with residential areas nearby and trails where night traffic is prohibited. do.

And in this case, when edge technology is used in various places, information is distributed and exchanged using an appropriate encryption method, allowing information to be exchanged quickly and smoothly in various places.

An integrated monitoring system using security technology and edge computing technology according to the embodiment,

By using an edge-type IoT sensor module to analyze visitor patterns, detect abnormal stay patterns, and use noise detection sensors to detect noise patterns of visitors, etc. in the late-night hours of the demonstration area, various disasters and safety are prevented. Judge the situation, etc. In addition, when an abnormal stay is detected, a situation notification is provided that displays a dangerous situation through a speaker and display connected to the verification equipment.

In addition, a format for determining the safety situation is created based on the information collected and analyzed from these sensors, different alarms are determined depending on the safety situation, and notifications are provided for various situations.

In particular, when information is exchanged with the administrator in such various places, a partial secret key is created and shared for each corresponding number of modules in multiple different places, and this is scrambled and calculated using the corresponding information and exclusive OR characteristics. In addition, whenever information is exchanged, the scramble is canceled according to the characteristics of exclusive OR to transmit commercial encryption information, for example, AES information.

According to embodiments, a rapid recognition and response system for dangerous situations is secured through normal situation recognition, and a rapid response system is established through situation recognition, for example, long-term movement in a specific place and approach (entry) to a dangerous place. By securing, rapid relief and rescue is possible.

In particular, in this case, when edge technology is used in various places, information is exchanged quickly and smoothly in various places by dispersing information and exchanging information through an appropriate encryption method.

1 and 2 are diagrams to conceptually explain an integrated monitoring system using security technology and edge computing technology according to an embodiment.
Figure 3 is a diagram illustrating the overall integrated monitoring system using security technology and edge computing technology according to an embodiment.
Figure 4 is a block diagram showing the configuration of management information processing applied to an integrated monitoring system using security technology and edge computing technology according to an embodiment.
Figure 5 is a flow chart sequentially showing the operation of an integrated monitoring system using security technology and edge computing technology according to an embodiment.

Figures 1 and 2 are diagrams to conceptually explain an integrated monitoring system using security technology and edge computing technology according to an embodiment.

Specifically, Figure 1 is an overall conceptual diagram of this configuration, and Figure 2 is an exemplary diagram showing an edge board of this configuration.

As shown in Figures 1 and 2, in one embodiment, temperature/humidity sensors and noise/vibration sensors are installed in a number of different places, such as tourist attractions with residential areas nearby and trails where night traffic is prohibited. Includes IoT sensor module, electronic signboard device, and speaker device. And, it includes an IoT platform and a management information processing device in a remote location.

Specifically, in this case, the system of this embodiment performs visitor pattern analysis using an edge-type IoT sensor module within each location, detects abnormal stay patterns, and utilizes noise detection sensors, etc. to detect late-night traffic in the demonstration area. By detecting noise patterns from visitors, etc. at different times of the day, various disasters and safety situations are determined. In addition, when an abnormal stay is detected, a situation notification is sent to indicate a dangerous situation at the location through a speaker and display connected to the verification equipment.

Looking at the characteristics of this system, in this case, it is based on edge computing that can solve the shortcomings of existing products (such as large capacity and security), processing large amounts of unstructured information and improving security vulnerabilities (see Figure 2).

Also, in this case, a separate interface is provided between the disaster alert and the edge sensor to notify the site of the disaster situation (for reference, more details are explained below).

Meanwhile, in addition to this, one embodiment secures recognition of dangerous situations and a quick response plan through recognition of abnormal situations, for example, a quick response plan through no movement in a specific place for a long time and approach (entry) to a dangerous place, etc. By securing, rapid relief and rescue is possible.

For this purpose, resident information, such as resident/residence time information, is collected through a Wi-Fi sensor. However, in this case, the conventional Wi-Fi detection sensor collects information while the user turns on Wi-Fi or connects to the AP. Therefore, it provides relatively low counting accuracy.

However, the proposed sensor utilizes a method of detecting finely transmitted signals at a high level even when Wi-Fi is turned off, and in response to the use of random Macs in recent smartphones, Mac de-landization (De- By applying randomization, it is recognized as the same device without duplicate detection and is effectively applied to external intrusion detection. A more detailed description may be made with reference to FIG. 6 below.

On the other hand, the configuration according to this embodiment additionally configures a format for determining a disaster situation based on information collected and analyzed from the above-described sensor, determines an alert according to this disaster situation, and determines a warning in various situations. Provides notification functions according to the following.

For example, in these cases, Wi-Fi signal detection can detect unusual activity, such as a person located in a place that should be empty, but this unusual activity varies depending on the location and scenario. For example, if you stay near a lake for more than an hour at 3 a.m., a warning will be issued, but if the same situation occurs in a crowded place, it is considered normal behavior.

And, in these cases, scenarios for such abnormal behavior are defined separately.

Therefore, through this, one embodiment provides safety care tailored to the situation through information learning and diagnosis about places and people's behavior, etc., and minimizes damage due to disasters and safety accidents due to dissemination.

In this state, when multiple modules are used in multiple locations, one embodiment performs distributed information and encryption processing using the method below to smoothly process information according to the situation in which multiple modules are used in multiple locations. Make it possible (see Figure 2).

For this purpose, first, the IoT sensor module operates as follows.

In other words, when providing surrounding status information, a partial secret key is created for each number (n) of sensor modules corresponding to each of a number of different locations from the setting format and shared by the target device. Next, when exchanging this information, this information is scrambled and exclusive OR is performed on the distributed information to encrypt it. Then, automatic destruction of this scramble is performed using the exclusive OR feature and transmitted, allowing information to be distributed and provided to various locations.

Here, the setting format is as follows.

a) First, when collecting surrounding status information, a different secret key is selected according to the shape or length of the plaintext based on the AES (Advanced Encryption Standard) encryption method.

Then, according to the size of the secret key, one of a number of different rounds is selected and executed repeatedly to obtain the ciphertext.

In this case, the function for each round applies the ARIA (Academy Research Institute Agency) encryption format, which includes round key addition, a permutation layer, and a spreading layer.

b) Then, before performing the round key addition operation, the plaintext and the secret key are modulated by first performing exclusive OR with n-1 scramblers (n is each number).

c) Next, the altered information is set as n partial secret keys, and a different partial secret key is provided to each of the n targets.

d) When modulation is performed, the ciphertext is obtained by repeatedly executing it as a function for each round including one round of round key addition operations according to the secret key size.

So, whenever the ciphertext is obtained, the scrambler is systematically destroyed in a round key addition operation.

To briefly explain this background, as terminals become more high-performance and high-capacity, the importance of protecting and managing important data inside the terminal is emerging.

To protect and manage such important data, a method of backing up the important data of a terminal by encrypting it and storing it in an external storage device has been used.

However, if important data is stored in an external storage device, a problem may arise where important data is easily leaked if the storage device is hacked.

Therefore, when storing important data in an external storage device, a more secure method is required.

Meanwhile, in addition to this, the above-described distribution and encryption processing method performs this processing from the configuration below using conventional technology as another example in this case, thereby enabling effective distribution, etc. in other aspects.

For example, this configuration first constructs a secret key with n-1 scrambles, that is, the number of targets minus one, and uses this secret key to create n pieces of distributed information.

To explain more specifically, a t-1st order random polynomial (f(x)) with a constant term as secret information is selected, and values from 1 to n are assigned to x of the t-1st order random polynomial (f(x)). By substituting, n distributed information can be created. At this time, t represents the number of distributed data required to recover secret data.

Then, configure a specific secret key.

Next, n-1 random keys, that is, n-1 scrambles, are generated, and both the secret key and the n-1 random keys are bitwise XORed.

Then, this information can be set into n partial secret keys, and a different partial secret key can be sent to each of the n destinations.

Then, each n piece of distributed information is encrypted with a secret key to create n pieces of distributed information.

Then, n encrypted distributed information is transmitted to n destinations corresponding to each and stored. For reference, this method is also used in the same way as the above method.

Therefore, before transmitting the encrypted distributed information, a process of authenticating whether n targets are legitimate users can be performed.

Figure 3 is a diagram illustrating the overall integrated monitoring system using security technology and edge computing technology according to an embodiment.

As shown in FIG. 3, the system of one embodiment includes IoT sensor modules (100-1 to 100-n) and Wi-Fi sensors (110-1 to 110-1 to 100-n) for temperature/humidity and noise/vibration, respectively installed in a number of different locations. 110-n), electronic display devices (120-1 to 120-n) and speaker devices (130-1 to 130-n), IoT platform (200), and management information processing device (300).

The IoT sensor modules (100-1 to 100-n) are used in a number of different places, such as tourist attractions with residential areas nearby and trails where night traffic is prohibited, for example, place a and place b, ..., place n. It is installed in and measures temperature, humidity, noise, etc. within each location and provides surrounding status information to managers, etc. through the management information processing device 300.

The Wi-Fi sensors (110-1 to 110-n) are installed at each location and provide Wi-Fi signals according to the resident information and residence time information of the person staying in each location through the management information processing device 300. Therefore, through this, the management information processing device 300 determines information about the resident, for example, whether there is a resident in a sparsely populated place late at night.

The electronic sign devices (120-1 to 120-n) specifically receive a situation notification message from the disaster warning information according to an embodiment through the management information processing device 300, and output it through a display medium to warn the user. provides.

The speaker device 130 receives a situation notification message from a disaster warning (management information processing device 300) according to an embodiment, and outputs the message to notify the alarm.

The IoT platform 200 enables information transfer and management throughout the entire range from the sensor module (!00) to the management information processing device 300. In this case, it is used to store sensing information in the management information processing device 300, and provides functions related to device management such as registration, search, and modification of the corresponding resource. Additionally, these devices may be connected through their own networks, such as TCP/IP networks.

The management information processing device 300 collects information about the surrounding conditions and residents of each place from the IoT sensor modules (100-1 to 100-n) and Wi-Fi sensors (110-1 to 110-n), When the status is confirmed, an alarm broadcast is provided through the electronic signboard devices (120-1 to 120-n) and speaker devices (130-1 to 130-n) in the corresponding location. Additionally, in this case, the management information processing device 300 transmits control commands to equipment in the relevant location to appropriately adjust the conditions of ventilation, lighting, heating and cooling, etc. Additionally, the management information processing device 300 may arbitrarily control the operation of each facility device as desired by the manager from a remote location, for example, depending on the schedule or whether an event occurs. In addition, the management information processing device 300 determines a disaster situation based on the collected and analyzed information, determines an alarm according to the disaster situation, and provides notification functions according to various situations. In this case, the management information processing device 300 provides better services by pre-registering and linking with external organizations in relation to such information, and is also a hospital information processing device and a police station information processing device. Additionally, the management information processing device 300 performs map-based real-time monitoring and notification, easily informing the surrounding conditions and resident status of each location through figures and maps, and also displays operation status and history.

Figure 4 is a block diagram showing the configuration of a management information processing device applied to an integrated monitoring system using security technology and edge computing technology according to an embodiment.

As shown in FIG. 4, the management information processing device 300 according to one embodiment largely includes an interface unit 301, a control unit 302, a database 303, a key signal input unit 304, and a display unit 305. Includes.

The interface unit 301 is connected to the IoT sensor module 100, Wi-Fi sensor 110, electronic sign device 120, and speaker device 130 located at each location, and is also connected to various equipment 140 to provide various Collects status information and resident information (including residence time information) around various locations and performs warning broadcasts and facility control.

The control unit 302 compares the status information, resident information, and stay time information around each place with each set normal status information, and operates the electronic sign device 120 and the speaker device through the IoT platform 200 according to whether there is an abnormality. Warning broadcast information is provided at (130). And, in this case, the control unit 302 provides control commands to the equipment 140 from a remote location. In addition, the control unit 302 determines a disaster situation based on the collected and analyzed information, determines an alarm according to the disaster situation, and provides notification functions according to various situations. Specifically, first, when controlling from a) a remote location, the control unit 302 internally records the surrounding status and visitor behavior, including status information around each location, resident information, and residence time information for a number of different locations and situations. Register in advance the disaster/safety management model that is accumulated and calculated according to. The surrounding state behavior and the visitor behavior are calculated separately according to different locations and scenarios. b) Then, whenever state information around each place, resident information, and residence time information are input for each location, the current surrounding state behavior and visitor behavior are set to normal behavior and visitor behavior from the above disaster/safety management model, respectively. Compare. c) As a result of the above comparison, if the current surrounding state behavior and visitor behavior correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided, and if they do not correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided. Provides control commands.

Meanwhile, the disaster/safety management model is as follows.

a) First, when controlling from the remote location, a model that classifies and learns the surrounding state behavior and visitor behavior in each situation for each location according to the state information around each place, the resident information, and the residence time information. Define.

When classifying the learning model, the learning parameters are classified from a vector made of N variables including multiple different locations, start time, end time, device type, and device registration status.

b) Next, extract a dataset showing the characteristics of surrounding state behavior and visitor behavior according to state information around each of a number of different places, resident information, and residence time information.

c) Then, the dataset is characterized by reflecting a number of different places, time zones, surrounding conditions, and situation information.

d) Then, based on this attribution information, determine the state information around each place for each learning model, and the attributes of the surrounding state behavior and visitor behavior according to the resident information and residence time information.

e) Next, normalize the decision information.

f) So, based on the above normalization information, the surrounding state behavior and visitor behavior according to the state information around each place, the resident information, and the residence time information are set for each learning model. So, in order to calculate the surrounding state behavior and visitor behavior, the independent variables of the surrounding state behavior and visitor behavior and the dependent variables of state information around each place, resident information, and residence time information are set.

g) Then, make this configuration information into learning and training data.

h) So, create the corresponding format from the above information.

The database 303 registers device information such as various sensor modules, electronic signboard devices, and speaker devices, and registers status information around each location, resident information, and residence time information in a database. Additionally, the database 303 stores operation status and history, and also stores the above-described learning model information.

When performing disaster situation judgment and situation notification operations, the key signal input unit 304 provides various setting information according to the user's key operation, for example, monitoring information such as location settings and normal state information for abnormality confirmation. It receives input and performs various functions.

When the display unit 305 performs disaster situation judgment and situation notification operations, it displays various notification information on the screen, such as abnormality messages, operation status, and history, etc., so that the manager can easily see them.

Figure 5 is a procedural flowchart sequentially showing the operation of an integrated monitoring system using security technology and edge computing technology according to an embodiment.

As shown in Figure 5, one embodiment is first connected to the IoT sensor module 100 that detects the condition around each place, including sensors for detecting various noises and vibrations, and provides status information around each place. Collect.

It is installed at each location and connected to the Wi-Fi sensor 110, which provides a Wi-Fi signal according to the resident information and residence time information of the resident within each location, and collects the resident information and residence time information around each location.

Therefore, the status information, resident information, and stay time information around each place received by the IoT sensor module 100 and the Wi-Fi sensor 110 are compared with each set normal state information.

Then, based on the compared information, any abnormality is determined and the corresponding set warning broadcasting information or control information is provided from a remote location.

In this case, when providing the information collected from each module to the administrator according to one embodiment, the information is quickly and smoothly secured at various locations through the method below and distributed appropriately at the edge. By sending it, it is provided.

Specifically, when providing collected information, as described above, a partial secret key is created for each number of sensor modules (n) in multiple different locations and shared by the target device.

Next, when exchanging this information, this information is scrambled and exclusive OR is performed on the distributed information to encrypt it. Then, automatic destruction of this scramble is performed using the exclusive OR feature and transmitted, allowing information to be distributed and provided to various locations.

Meanwhile, in this state, the management information processing device 300 performs the following operations according to one embodiment.

a) In the case of control from a remote location, the above-mentioned method is calculated by accumulating internal surrounding state behavior and visitor behavior according to the state information around each place and the resident information and residence time information for a number of different places and situations. Register the disaster/safety management model in advance.

In addition, the surrounding state behavior and the visitor behavior are calculated separately according to different locations and scenarios.

b) Next, whenever status information around each location, resident information, and stay time information are input for each location, the current surrounding status behavior and visitor behavior are compared with each set normal behavior from the disaster/safety management model. do.

c) As a result of the above comparison, if the current surrounding state behavior and visitor behavior correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided, and if they do not correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided. Provides control commands.

For example:

First, collect status information around each location by connecting to IoT sensor modules that detect the conditions around each location, including multiple temperature sensors, humidity sensors, noise sensors, and vibration sensors installed in multiple different locations. do.

It is installed at each location and connected to a Wi-Fi sensor that provides a Wi-Fi signal according to the resident information and residence time information of the resident within each location, and collects resident information and residence time information around each location.

Therefore, the status information, resident information, and stay time information around each place received by these IoT sensor modules and the Wi-Fi sensor are compared with each set normal state information.

Then, based on the compared information, any abnormality is determined and the corresponding set warning broadcasting information or control information is provided from a remote location.

In this state, the following operations are performed according to one embodiment.

a) First, in the case of control from a remote location, the surrounding state behavior and visitor behavior are accumulated and calculated internally according to the state information around each place, the resident information, and the residence time information for a number of different places and situations. Register the bar’s disaster/safety management model in advance.

In addition, the surrounding state behavior and the visitor behavior are calculated separately according to different locations and scenarios.

b) Next, whenever status information around each location, resident information, and stay time information are input for each location, the current surrounding status behavior and visitor behavior are set for each location from the above-mentioned disaster/safety management model. Compare with behavior.

c) As a result of the above comparison, if the current surrounding state behavior and visitor behavior correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided, and if they do not correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided. Provides control commands.

As described above, in one embodiment, visitor pattern analysis is performed using an edge-type IoT sensor module, abnormal stay patterns are detected, and noise detection sensors are used to detect noise patterns of visitors, etc. in the late-night hours of the demonstration area. By detecting, various disasters and safety situations are judged. Also, at this time, when an abnormal stay is detected, a situation notification is sent to indicate a dangerous situation through a speaker and display connected to the verification equipment.

So, through this, a format for determining the safety situation is created based on the collected and analyzed information, different alarms are determined depending on the safety situation, and notifications are provided for each of the various situations.

In particular, when determining the safety situation, sensors are set in response to various locations, that is, the surrounding state, and the safety situation is quickly determined according to the surrounding state (or type of surrounding state behavior). For example, these sensors include temperature, humidity, noise, and vibration sensors, and the type of sensor, that is, the type of surrounding state and behavior, is set to suit each location and judged accordingly.

Therefore, through this, one embodiment secures a rapid recognition and response system for dangerous situations through normal situation recognition, and quickly recognizes situations such as no movement in a specific place for a long time and approach (entry) to a dangerous place. By securing a response system, we provide rapid relief and rescue.

In particular, in this case, as described above, the surrounding conditions are specified for various places and situation recognition is performed more smoothly and quickly in a customized manner, thereby enabling a better rescue.

As described above, in one embodiment, visitor pattern analysis is performed using an edge-type IoT sensor module, abnormal stay patterns are detected, and noise detection sensors are used to detect noise patterns of visitors, etc. in the late-night hours of the demonstration area. By detecting, various disasters and safety situations are judged. Also, at this time, when an abnormal stay is detected, a situation notification is sent to indicate a dangerous situation through a speaker and display connected to the verification equipment.

And, when such collected information is provided, a format for determining the safety situation is constructed based on this, different alarms are determined depending on the safety situation, and notifications are provided for each of the various situations.

In particular, when information is exchanged with the administrator in such various places, a partial secret key is created and shared for each corresponding number of modules in multiple different places, and this is scrambled and calculated using the corresponding information and exclusive OR characteristics. And, whenever information is exchanged, the scramble is canceled according to the characteristics of exclusive OR to transmit commercial encryption information, for example, AES information.

Therefore, through this, one embodiment allows recognition of dangerous situations and finding quick response measures through normal situation recognition in tourist attractions with residential areas nearby and trails where night traffic is prohibited.

Additionally, when multiple modules are used in multiple locations, distributed information and encryption processing are performed using the method below, allowing the information to be processed smoothly according to the situation in which multiple modules are used in multiple locations.

Meanwhile, in addition, the MAC address de-identification configuration for collecting resident information applied to this configuration is explained as follows.

First, in the configuration of de-identifying the MAC address for collecting resident information according to one embodiment, the MAC address of the user terminal is pseudonymized in the Wi-Fi sensor and transmitted/stored to the server. Moreover, the processed information is no longer personally identifiable information and therefore does not violate relevant laws. Additionally, the Mac information collected in this way is de-identified using a 512-bit single-path hash password before being transmitted.

Additionally, the MAC address de-identification configuration for collecting resident information is as follows.

First, the WiFi sensor performs the following operations.

a) The detection conditions for sensing the above Wi-Fi signal are divided into ■ a state in which Wi-Fi is connected to a specific AP, ■ a state in which Wi-Fi is turned on but not connected to the AP, and ■ a state in which Wi-Fi is turned off.

b) Next, in this state, if connected to the Wi-Fi network, the real MAC address of the resident terminal is used, and if not connected to the Wi-Fi network, a random MAC address is used.

c) Then, when the resident terminal is recognized by the random MAC address, the random MAC addresses are grouped from the set derandomization format and recognized as one resident terminal.

d) So, after recognizing the resident terminal, the MAC address of the resident terminal is de-identified from the set One-Way Hash encryption format and transmitted to the management information processing device.

In addition, in this case, a temporary MAC address is created to correspond to a number of different user terminals, operating formats, and router types, and is transmitted with information that creates a manufacturing name based on a hash table of the MAC address.

So, the management information processing device performs the following operations accordingly.

That is, when receiving resident information and residence time information of the resident, visitor behavior is obtained by collecting resident information and residence time information based on the MAC address of the de-identified resident terminal.

301: interface unit 302: control unit
303: database 304: key signal input unit
305: display unit

Claims (2)

IoT sensor modules that detect the conditions around each location, including multiple temperature sensors, humidity sensors, noise sensors, and vibration sensors installed in multiple different locations;
A Wi-Fi sensor installed at each location to provide a Wi-Fi signal according to the resident information and residence time information of the person staying in each location;
Electronic signboard devices installed at each location to display warning broadcasting information;
Speaker devices installed at each location to output audio warning broadcasting information;
Equipment installed at each location to control conditions around each location;
Status information, resident information, and stay time information around each location are collected by the IoT sensor module and the Wi-Fi sensor, and transmitted to a pre-registered management information processing device, and the warning broadcast information of the management information processing device is sent to the electronic signboard device. an IoT platform that transmits to the speaker device and also transmits control commands of the management information processing device to the equipment; and
The status information, resident information, and stay time information around each place received by the IoT platform and the Wi-Fi sensor are compared with each set normal status information, and depending on whether there is an abnormality, warning broadcast information and control commands are sent remotely to the IoT platform. Management information processing device provided by; It includes,

The IoT sensor module is,
When providing surrounding status information, create a partial secret key for each number of sensor modules (n) in a number of different places from the setting format, perform an exclusive OR with the distributed information, and encrypt it to provide it.
Here, the setting format is:
a) When collecting surrounding status information, select a different secret key according to the shape or length of the plaintext based on the AES (Advanced Encryption Standard) encryption method.
Depending on the size of the secret key, one of several different rounds is selected and executed repeatedly to obtain the ciphertext.
The function for each round applies the ARIA (Academy Research Institute Agency) encryption format, which includes round key addition, substitution layer, and diffusion layer.
b) Before performing the round key addition operation, modulation is performed by first performing n-1 scramblers and exclusive OR on the plaintext and secret key,
c) Set the altered information as n partial secret keys, and provide one different partial secret key to each of the n targets.
d) When modulation is performed, obtain the ciphertext by repeatedly executing it as a round-by-round function including one round of round key addition operations according to the secret key size,
Each time the ciphertext is obtained, the scrambler is systematically destroyed in the round key addition operation, so that the operation result is obtained in the same format as commercial AES,

The management information processing device,
a) In the case of controlling from the above remote location, disaster/safety is calculated by accumulating internal surrounding state behavior and visitor behavior according to the state information around each place, resident information, and residence time information for a number of different places and situations. Register the management model in advance,
The surrounding state behavior and the visitor behavior are calculated separately according to different locations and scenarios for each location and situation,
b) Whenever status information around each location, resident information, and residence time information are input for each location, the current surrounding status behavior and visitor behavior are compared with each set normal behavior from the disaster/safety management model,
c) As a result of the above comparison, if the current surrounding state behavior and visitor behavior correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided, and if they do not correspond to each setting's normal behavior, warning broadcasting information and control commands are not provided. Provide control commands,

Here, the disaster/safety management model is,
a) When controlling from the remote location, a model that classifies and learns the surrounding state behavior and visitor behavior in each situation for each location according to the state information around each place, the resident information, and the residence time information. define ,
When classifying the learning model, the learning parameters are classified from a vector made of N variables including multiple different locations, start time, end time, device type, and device registration status,
b) Extract data sets showing the characteristics of surrounding state behavior and visitor behavior according to state information around a number of different places, resident information and residence time information, respectively,
c) Attribute the dataset by reflecting multiple different locations, times, surrounding conditions, and situational information,
d) Based on the attribute-processed information, determine the attributes of surrounding state behavior and visitor behavior according to state information around each place, resident information, and residence time information for each learning model,
e) By regular processing of the information determined above,
f) Based on the above regularly processed information, the surrounding state behavior and visitor behavior are set for each learning model according to the state information around each place, the resident information, and the residence time information, and the surrounding state behavior and visitor behavior are determined. To calculate it, set independent variables of surrounding state behavior and visitor behavior, and dependent variables of state information around each place, resident information, and stay time information,
g) learning the above setting information and creating training data; An integrated monitoring system using security technology and edge computing technology. In claim 1,
The WiFi sensor is,
a) The detection conditions for sensing the Wi-Fi signal are divided into ■ a state in which Wi-Fi is connected to a specific AP, ■ a state in which Wi-Fi is turned on but not connected to the AP, ■ a state in which Wi-Fi is turned off,
b) In the above state, if connected to the Wi-Fi network, the real MAC address of the resident terminal is used, and if not connected to the Wi-Fi network, a random MAC address is used,
c) When the resident terminal is recognized by the random MAC address, the random MAC addresses are grouped from the setting derandomization format and recognized as one resident terminal,
d) After recognizing the resident terminal, de-identify the MAC address of the resident terminal from the set One-Way Hash encryption format and transmit it to the management information processing device,
e) When transmitting the MAC address, a temporary MAC address is created to transmit in response to a number of different user terminals, operating formats, and router types in combination, and a manufacturing name is created based on a hash table of the MAC address. transmitting,
Whether or not it is a router device is judged based on the type of Wi-Fi message.

In addition, the management information processing device,
When receiving resident information and residence time information of the resident, collecting the resident information and residence time information based on the MAC address of the de-identified resident terminal to obtain visitor behavior; An integrated monitoring system using security technology and edge computing technology.