TW201642677A - System and method for dynamic management of context awareness service to multiple users - Google Patents

Abstract

The present invention provides a system and method for dynamic management of context awareness service to multiple users. The system includes a data retrieval sub-system for executing a context-aware data retrieval mode, a data analysis and prediction sub-system for executing a context-aware data prediction mode, a push notification sub-system for executing a push notification mode based on the data collected through mobile crowd sensing and context aware devices, a navigation application sub-system for executing an IoT navigation application mode, and a database for storing the data collected in different modes. The context awareness service most suitable to the needs of the users is provided based on the context data such as identity, events, time, location and things.

Description

Dynamic multi-person situational awareness system and method thereof

The present invention relates to an Internet of Things (IoT) application system architecture based on Action Mass Perception (MCS) and Context Awareness, so as to solve the problem that traditional operation methods need to construct various sensors in the environment beforehand. A situational awareness system and method for cost and maintenance issues.

In recent years, due to the rapid development of scientific and technological information, so-called "situational awareness" technology has become popular. Contextual awareness is the provision of services that best suit the user's situation through the sensing of various sensors. Through context-aware technology, it not only improves service quality, but also provides a better user experience. For example, to simulate a situation, when a visitor stands in front of a monument, what visitors can see is just a cold ice-cold monument, but through context-aware service technology, using GPS positioning or devices such as smart glasses, it is possible to instantly make the monuments relevant. The information is displayed on the mobile phone or glasses. For the first time visitors to the monument, even if there is no introduction of the tour guide, they can understand the historical background.

Therefore, as disclosed in the "Multi-Mode Proximity Detection" patent of the invention No. 201112052 published on April 1, 100, the disclosure of the invention discloses that one or more electronic devices transmit signals to adjacent servers, which determine the first electronic Whether the device is likely to be adjacent to the second electronic device. The proximity server transmits a signal to the first electronic device and the second electronic device, and in response to the signal, the first and second electronic devices activate an environmental sensor, collect samples of at least one environmental data, and retrieve the At least one characteristic group of the environmental data, generating a first hidden characteristic group from the characteristic group, and transmitting the first and second hidden characteristic groups to the neighboring server. The proximity server uses the first hidden feature set and the second hidden feature set to determine whether the first electronic device and the second electronic device are adjacent.

There is also a new type of "Situation Awareness System" patent case No. M405025 announced on June 1, 100, which includes a server, a information collection unit and a handheld electronic unit. The information collection unit is used for Collecting an external situation information and storing it in a database of the server, the handheld electronic unit includes a satellite positioning unit for receiving a location information and an input unit for inputting an internal situation information, and the location information is And the internal situation information is stored in the database; the server extracts the external situation information, the location information and the internal situation information from the database, and the operation result is returned to the handheld electronic unit, and then the An output unit of the handheld electronic unit outputs the result of the operation; the user can appropriately respond to the situation of the surrounding environment according to the operation result.

However, the above patents have the following disadvantages.

1. The scope of use and the achievable effect are relatively narrow and can only be used by static single people.

2. It is also necessary to build multiple sensors in the physical environment, resulting in increased usage costs.

3. The analysis technology is only provided by the server on the back end, lacking the application to the front end of the client.

4. Moreover, the parameters returned by the sensors produced by various manufacturers are different and incompatible with each other.

5. Maintenance, overhaul and adjustment of many sensors.

6. The sensing result is limited by the type and quantity of sensors set in advance, which makes the accuracy of the sensor poor.

Accordingly, the present situational awareness system and method have the above disadvantages. Therefore, the present invention provides a dynamic multi-person context sensing system, which is implemented in conjunction with a client and a server in a sensing environment, and is provided with: a data retrieval subsystem built in the user terminal, the data is The subsystem includes a trigger module, a dynamic sensor participation module, and a full-automatic background service module for performing a context-aware data capture mode; a data analysis and prediction subsystem is built in In the server, the data analysis and prediction subsystem includes a person-time feature information analysis module and a behavior data prediction module to execute a context-aware data and prediction mode; an information-pushing subsystem is separately Built on the client and the server, the information push subsystem includes a person-time feature message screening module, a promotion information push module, an emergency information push module, and a lottery result push mode. a group, in order to implement an information promotion mode based on action mass perception and context awareness; a navigation application subsystem is built in the user terminal, and the navigation application subsystem includes An intent list module, a list guide menu module, a link activity module and a situation learning module for executing an Internet of Things navigation application mode; a database connected to the server, the data The library is used to store the information obtained when each of the above subsystems executes different modes.

The present invention may also be a dynamic multi-person context sensing method, which is implemented in conjunction with the dynamic multi-person context sensing system, and at least one trigger is set in the sensing environment, and the trigger is transmitted by a blue signal, the user end After entering the sensing environment, the trigger is sensed, thereby performing one or a combination of the following modes: A. starting the data retrieval subsystem to execute the context aware data capture mode; B. starting the The data analysis and prediction subsystem executes the situational awareness data and prediction mode; C. starts the information promotion subsystem to perform the information promotion mode based on the action mass perception and context awareness; D. starts the navigation application subsystem to execute the IoT navigation application mode.

The context-aware data capture mode determines whether the Bluetooth signal conforms to a preset value by using the trigger module, and the dynamic sensor participant module opens a sensor built in the user terminal after the judgment is met. In order to generate a perceived function, and to activate the automatic background service module, the environmental information in the perceived environment and the personal data that can be disclosed by the client are captured.

The above-mentioned situational awareness data and prediction mode are first classified according to the situation by the personnel time feature information analysis module, and are weighted according to the situation, and finally the prediction result is outputted by the behavior data prediction module.

The information dissemination mode of the above-mentioned action mass perception and context perception is first classified according to the situation through the personnel time information filtering module, and it is confirmed whether the server sends a difficult message, and after the emergency message is issued, the emergency information is launched. The broadcast module is used to integrate the user terminals located at the same time point, in the same area or in the same situation, and analyze the location of the user in the neighboring area or the same situation, and generate a list, and display a push-and-disaster message on the user end.

The promotion information promotion module screens the environmental information in the sensing environment, and according to the preference of the user, pushes the promotion information of the appropriate product that meets the screening result.

The above-mentioned lottery result pushing module is based on the situation classification, and based on the result of screening the environmental data in the sensing environment, the lottery information is found from the server and displayed through the client, the lottery information includes a lottery event and a The result of the draw.

The intent list module is configured to create an intent list and parse it into a corresponding list of physical locations, and the list guide navigation module displays a navigation screen according to the intent list to guide the user to arrive at the target location.

The above-mentioned activation of the activity module is performed in a related activity, and the location of the user terminal in the sensing environment is perceived as a criterion for determining the activity of the activity, and the screen of the activity is displayed on the user terminal.

The scenario learning module is activated to perform a context learning, and a scenario learning screen is displayed on the user end, and the related learning materials of the product are displayed according to the result of the environmental information in the sensing environment. .

The invention has the following advantages:

1. Set the Bluetooth signal transmitter at some fixed point to act as a trigger, so that the smart mobile device of the user can recognize that it has entered a specific area, and automatically wake up the front end application of the user for subsequent sensing service and Analysis, users do not need to deliberately open the front-end application, can use the perception service in a natural situation environment, which can achieve the operation of automatic sensing and analysis.

2. The smart mobile device of the user terminal (which can be regarded as the sensor) does not need to be configured in a specific place in advance, and it is dynamic participation throughout the process, and all smart mobile devices that support the system and have a dedicated front-end application installed. With its built-in sensor for a series of sensing and analysis, it is highly scalable, and each smart mobile device is regarded as a composite sensor, so each composite sensor enters a certain area, that is, Add a new sensor to the area.

3. Since the sensor is dynamically involved, the system administrator does not need to set the sensor in advance, and there is no disadvantage that the sensor returns the parameter format incompatibility produced by each manufacturer.

4. With the trigger and the design of the master-slave architecture, the trigger will continue to transmit the Bluetooth signal. The smart mobile device will detect and filter the Bluetooth signal, and automatically launch the corresponding front-end application and process. After the initial analysis and processing, the data is sent to the server at the back end for collection and storage.

5. Capture and analyze data through the proprietary front-end application, including environment (temperature, luminosity, sound, indoor relative position, GPS coordinates) and user data (public age, gender, occupation, number of communication lists, intent list) And according to the region and interval time, determine whether it is a meaningful behavior, if it is not meaningful, it is directly excluded, and is not sent to the server storage at the back end, thereby reducing the load on the server and saving the power consumed by the user. Resources.

6. Consider the behavior analysis and prediction of users, people, time, place, and material users. By extracting users' public and environmental data, they can classify, analyze, and analyze people, events, time, place, and material. Screening, defining weighting patterns by occasion, finding out their behavior patterns and predicting them.

7. At the same time, consider the urgent information of people, events, time, place and material. By extracting the publicly available data and environmental data, the server can find out the urgent information that has occurred and use the client-side application. Remind the user in an eye-catching manner, and at the same time produce a user location correspondence table that can be referenced by the rescue team.

8. At the same time, consider the promotion information of people, events, time, place and property. By extracting the user's publicly available information and environmental information, the server can find out the appropriate information and display it through the client's front-end application. User, this information will contain the results of the promotion related to the product in this situation or the lottery that has participated in the sweepstakes.

9. Support the lottery, customs and various activities. By entering the specific area as the identification of the event participation, the server will identify the identity of the participating users from the available information, and identify it as a lottery. It can also be directly pushed. The winning message is given to the winning user.

10. Support contextual learning, which includes language and related knowledge, and considers people, events, time, places, and objects to carry out the most suitable learning information.

11. Through the system to guide the user to inventory their behavior intentions, simplify the follow-up process, the client front-end application will analyze the list, and connect to the list guide wizard module to guide users conveniently and quickly Arrive at the location of the item.

12. The architecture of the system of the present invention is streamlined and clear, and the construction cost is low and the maintenance is easy.

13. The invention is mainly guided by software, and the hardware only participates at an appropriate time, greatly improving the application change capability, and is more suitable for various occasions.

Referring to the first figure and the second figure, the embodiment of the present invention is a dynamic multi-person context sensing system, which is jointly executed by a client (A) and a server (B), the client (A) It is a smart mobile device of a smart phone, a tablet computer or a smart watch, and the server (B) is a server computer. The client (A) downloads and stores a front-end application (A1), and the server (B) downloads and stores a back-end application (B1) via the front-end application (A1) and the back-end application. (B1), whereby the context aware system of the present invention is jointly implemented. Embodiments of the present invention include a data capture subsystem (1), a data analysis and prediction subsystem (2), an information push subsystem (3), a navigation application subsystem (4), and a database (5), wherein :

The data retrieval subsystem (1) is built in the user terminal (A). The data capture subsystem (1) includes a trigger module (11), a dynamic sensor participation module (12), and a fully automatic background service module (13) for performing a context aware data. Capture mode.

The data analysis and prediction subsystem (2) is built in the server (B). The data analysis and prediction subsystem (2) includes a person-time feature information analysis module (21) and a behavior data prediction module (22) to perform a context-aware data and prediction mode.

The information push subsystem (3) is built in the user terminal (A) and the server terminal (B) respectively. The information promotion subsystem (3) includes a person-time feature message screening module (31), a promotion information promotion module (32), an emergency information promotion module (33), and a lottery result push. The broadcast module (34) can perform an information push mode based on action mass perception and context awareness.

The navigation application subsystem (4) is built into the client (A). The navigation application subsystem (4) includes an intent list module (41), a list guide menu module (42), a gate activity module (43), and a context learning module (44). An IoT navigation application mode can be executed.

The database (5) is connected to the server (B). The database (5) is used to store the data obtained when the various subsystems perform different modes.

The invention is a dynamic multi-person context sensing method and is jointly executed in conjunction with the above dynamic multi-person context sensing system. As shown in the first figure and the second figure, a trigger (C1) can be set in advance in each fixed point of a sensing environment (C), and the trigger (C1) is transmitted by a blue signal, mainly by providing Identification of the perceived environment (C). When the user terminal (A) enters the sensing environment (C), the data capturing subsystem (1) is started to execute the context aware data capturing mode. The operation process of the context-aware data capture mode is as shown in the third figure. First, when the client (A) enters the trigger range of the sensing environment (C), the client (A) is intelligent. The mobile device will execute the front-end application (A1) to activate the signal detection service, which mainly detects the Bluetooth signal sent by the trigger (C1) when the data capture subsystem (1) When the detection signal is detected, the trigger module (11) of the data capture subsystem (1) is activated, and it is determined whether the Bluetooth signal is the same as the signal source pre-configured by the context aware system. When the values are different, the program for detecting the Bluetooth signal will be returned. If it is judged to be the same, it will judge whether the Bluetooth signal is equal to the source of the last found signal. If the judgment is equal, it will judge whether the interval time is less than a preset value. If the judgment is no, then Go back to the process of detecting the Bluetooth signal. If it is determined that the Bluetooth signal is not equal to the source of the last detected signal, or the interval is less than the preset value, the dynamic sensor participation module (12) is activated to enable the user terminal (A). The smart mobile device is originally built and available for each sensor to produce a perceptual function. And establishing a connection between the Wi-Fi wireless access point (D) in the sensing environment (C) and the back end application (B1) executed by the server (B), that is, equal to the context aware system of the present invention. Dynamically added a composite sensor. Then, the automatic background service module (13) is activated, so that the environmental data of the sensing environment (C) and the personal data that can be disclosed by the user terminal (A) can be retrieved as the subsequent sensing environment ( In C), the identification of the lottery, the related activities, and the like, and the environmental data includes one or a combination of people, events, time, place, and thing. At the same time, the data obtained by executing the context-aware data acquisition mode is transmitted to the server (B) and stored in the database (5). The server (B) can simultaneously respond to the client (A), for example, applying the modules of each subsystem for front-end reaction, which can include basic knowledge of various commodities, multi-language learning of products, and completed tasks. Information such as checkpoints can be used to achieve a process that is fully automated. The trigger module (11) continuously detects the Bluetooth message to determine whether the service still exists. If the Bluetooth signal is not detected, the service is no longer present, and the server (B) may be powered off. The Bluetooth signal is turned off or the client (A) has left the triggering range of the sensing environment (C), and thus the context aware data capturing mode is ended.

When the data acquisition subsystem (1) has the environmental information of the person, the thing, the time, the place, the object and the personal data of the user (A), the data analysis is performed. The cum prediction subsystem (2) will implement the situational awareness data and prediction mode. The operational process of the situational awareness data and prediction mode is shown in the fourth figure. First, the personnel information analysis module (21) is classified according to the situation to determine the classification of people, events, time, place, and objects, and is weighted according to the situation, and the weighted index is according to different users or environmental conditions. The weighting is performed, and finally the prediction result is outputted by the behavior data prediction module (22).

At the same time, when the data retrieval subsystem (1) has the environmental information of the person, the thing, the time, the place, the object, or the personal data of the user (A), the user terminal (A) and the information dissemination subsystem (3) of the client (B) jointly perform the information promotion mode of the action mass perception and context awareness. The operational flow of the information dissemination mode of the action mass perception and context perception is as shown in the fifth figure. Firstly, according to the situation classification, the personnel information screening module (31) sorts the results of "person, event, time, place, and object" according to the result, and whether the manager of the server (B) has the situation in the situation. When a difficult message is sent, if there is a difficult message, the emergency information push module (33) will be activated to integrate the user (A) in the same region or situation at that point in time, and analyze the neighboring area or the same The location of the user in the situation, the list of outputs, for use by the rescue team, and the display of the emergency message on the user side (A) (as shown in Figure 6). The following information is displayed on the screen of the user's (A) emergency information push.

1. Eye-catching area: Allows users to quickly identify events.

2. Description of the emergency incident.

3. Confirm button: The emergency message will continue to be displayed before the button is pressed.

4. Map: Indicate the current location of the client (A) and plan the most appropriate evacuation route (the distance between the distance and the crowd at the same time can be considered).

If the server (B) does not send any emergency messages, the promotion information pushing module (32) is activated to find out the user's results based on the results of screening "people, events, time, place, and objects". Preference, to promote promotional information on appropriate products in the context (as shown in Figure 7). In the screen of the promotional information of the user terminal (A), the user terminal (A) is located in the cleaning product area of the hypermarket, and a cleaning product has been set in the shopping list, and the situation time is in the rainy season. This speculates on other related promotional items. The promotion information is presented in a square format list, with two columns and multiple columns, and a single product item is presented in a single grid. The user can also directly select the square to obtain more detailed product information.

When the lottery result pushing module (34) is activated, the user (A) is searched according to the situation, according to the results of screening "people, things, time, place, and things". The publicly available information and environmental information, from the server (B) to find the appropriate lottery information, and presented to the user through the client (A) front-end application (A1), the lottery information is related to the situation The next lottery and the results of the draw.

The client (A) can also create a list according to its own needs, and can activate the navigation application subsystem (4) to execute the IoT navigation application mode. The operation flow of the Internet of Things navigation application mode is as shown in the eighth figure, by which the intent list module (41) is activated for the user to establish an intent list and parse it into a physical location correspondence list. The screen display of the intent list of the user terminal (A) is as shown in the ninth figure, taking the "hypermarket" application as an example. The user's intention is "purchasing goods"; if "public transportation" is taken as an example, the user's intention is It may be "the number of times you want to ride a car", and it will be converted into a list of intentions no matter where it is applied. And the following information is displayed.

1. List item: Each column is an intent, and the estimated price (the average price of the item) will be displayed on the right after the input is completed.

2. The input will be guessed by characters to reduce the user's typing time, and the content of the intention can be limited to a reasonable range (because the system administrator must first establish possible intentions, such as commodity items, train schedules).

3. Replace with the lowest price button: When pressed, the product with the lowest price in the store will be automatically compared to all products in the intent list and replaced with the intent list.

4. Estimated Total Price: Shows the estimated amount of all items in the intent list plus.

After the intent list is established, the sub-system (1) can be used to sense the current location of the user terminal (A), and the list guide navigation module (42) is based on the location node. A navigation screen is displayed for the intent list. As shown in the tenth figure, the navigation screen of the user terminal (A) compares the map resources established in advance by the system administrator according to the contents of the "intention list" and displays them on the screen of the user terminal (A). For example, in the case of the above-mentioned "hypermarket", the products "printer", "pull disk" and "toilet paper" which the user wants to purchase are marked with the shelf position of the display, and the route is guided by the arrow. And the following information is displayed.

1. The user's current location.

2. Route guidance, and display the distance and estimated consumption time on the route (unit m=meter, s=second).

3. Intent project location (this is the product you want to buy).

4. The total distance of the route planning and the total estimated elapsed time, the elapsed time will additionally calculate the purchase time, preset 3 minutes for each product (180s), and 3 minutes for checkout. The total consumption time shown in the figure = distance 40s + toilet paper 180s + distance 20s + portable disk 180s + printer 180s + distance 50s + checkout 180s = 830s, estimated to be about 14 minutes.

The list guide module (42) detects whether the user terminal (A) has reached the set target position at any time, and when it arrives at the target position, it will respond to the target position. If the target location has not been reached, the current location will continue to be updated until the target location is reached.

The related activity module (43) is also activated, and if the hypermarket can hold a customs clearance activity according to different regions, the customs activity module (43) is to perceive the position of the user terminal (A) as a preliminary criterion for the identification of the activity, and actually holds the level. At the same time, QR Code (or NFC) can be used as a verification mechanism. And at the user terminal (A), a screen showing the activities, as shown in the eleventh figure, is provided in the "Bathroom Supplies Area", "3C Area" and "Home Appliances Area" and the operation of the customs, the client ( A) There is currently no entry into any relevant area, so the number of passes has been 0. And the following information is displayed.

1. Level icon: Remind users that there is ongoing activity in the area.

2. Activity message: Calculate the current total number of levels, and the number of completed levels.

When the situational learning module (44) is activated, a context learning can be performed. And displaying a situational learning screen on the user terminal (A). As shown in the twelfth figure, the product is displayed according to the result of the above-mentioned screening by "person, event, time, place, object". Related learning materials. For example, in the "hypermarket" environment, the "3D printer" product in the "3C area" will be displayed as an example.

1. Product Name: Display the product name in bilingual mode.

2. Schematic: The representative picture of the item in the “environment” is mainly used. Press the center play button to play the related display or teaching video.

3. Product-related knowledge: Users can learn about relevant product information, so that they can choose products more rationally.

4. Multi-language selection button: Users can freely choose the content language. When pressed, all the characters in the screen will be replaced with the target (selected) language.

5. Pronunciation button: Press to pronounce the content in the currently set language.

Through the situational learning method, users can learn foreign languages and enhance their understanding of related products during the shopping process, and can also apply multi-language functions to foreigners for shopping translation.

In addition, in order to facilitate the management of the server (B) for the management and implementation of the context-aware system of the present invention, the distribution information of the user may be displayed on the server (B), as shown in the thirteenth figure, the content data and The project also takes the hypermarket as an example and will show the following information.

1. System Status: Shows the number of sensors (ie, user terminals (A)) entering all of the sensing environment (C) (ie, hypermarkets), which refers to the smart handheld device or wearable in the present invention. Device.

2. Regional statistics: The current number of users in each region is presented in terms of data matching the bar graph.

Thus, the present invention can provide the most appropriate information service mode for users according to different environmental factors such as people, events, time, place, and materials, so as to meet the needs of all users.

However, the above description is only one of the embodiments of the present invention. When the scope of the patent application of the present invention is not limited thereto, the simple equivalent changes and substitutions made by the scope of the patent application and the contents of the specification of the present invention are It is still within the scope of the protection covered by the scope of the invention.

(1) ‧ ‧ data acquisition subsystem
(11)‧‧‧Trigger module
(12)‧‧‧Dynamic Sensor Participation Module
(13) ‧ ‧ tends to fully automatic background service module
(2) ‧‧‧Data Analysis and Prediction Subsystem
(21) ‧‧‧ Personnel Information Analysis Module
(22) ‧‧‧ Behavioral Data Prediction Module
(3) ‧‧‧Information Push Subsystem
(31) ‧‧‧ Personnel Information Screening Module
(32)‧‧‧Promotional Information Push Module
(33) ‧‧‧ urgent information push module
(34) ‧‧‧Drawing Results Push Module
(4) ‧ ‧ navigation application subsystem
(41)‧‧‧Intent list module
(42) ‧ ‧ List guide guide module
(43) ‧ ‧ 闯 活动 activity module
(44) ‧‧‧Situational Learning Module
(5) ‧ ‧ database
(A) ‧ ‧ client
(A1)‧‧‧ front-end applications
(B) ‧‧‧Server
(B1)‧‧‧ Backend application
(C) ‧ ‧ perceived environment
(C1)‧‧‧ Trigger
(D)‧‧‧Wi-Fi wireless access point

[First figure] is a system architecture diagram of the present invention.

[Second figure] is a configuration diagram of the system used in the present invention.

[Third image] is a flow chart of the operation of the context-aware data capture mode of the present invention.

[Fourth figure] is a flow chart of the operation of the situational awareness data and prediction mode of the present invention.

[Fifth] is a flow chart of the operation of the information dissemination mode of the action mass perception and context perception of the present invention.

[Sixth figure] is a schematic diagram showing a screen for pushing a difficult message according to the present invention.

[Seventh figure] is a schematic diagram showing a promotional information screen of the present invention.

[Eighth image] is a flowchart of the operation of the Internet of Things navigation application mode of the present invention.

[9th] is a schematic diagram showing a screen of an intent list of the present invention.

[Tenth Graph] is a schematic diagram showing a navigation screen of the present invention.

[11th] is a schematic diagram showing the screen of the activity of the present invention.

[Twelfth Figure] is a schematic diagram showing a situational learning screen of the present invention.

[Thirteenth Diagram] is a schematic diagram showing the distribution information of users according to the present invention.

(1) ‧ ‧ data acquisition subsystem

(11)‧‧‧ Trigger Module

(12)‧‧‧Dynamic Sensor Participation Module

(13) ‧ ‧ tends to fully automatic background service module

(2) ‧‧‧Data Analysis and Prediction Subsystem

(21) ‧‧‧ Personnel Information Analysis Module

(22) ‧‧‧ Behavioral Data Prediction Module

(3) ‧‧‧Information Push Subsystem

(31) ‧‧‧ Personnel Information Screening Module

(32)‧‧‧Promotional Information Push Module

(33) ‧‧‧ urgent information push module

(34) ‧‧‧Drawing Results Push Module

(4) ‧ ‧ navigation application subsystem

(41)‧‧‧Intent list module

(42) ‧ ‧ List guide guide module

(43) ‧ ‧ 闯 活动 activity module

(44) ‧‧‧Situational Learning Module

(5) ‧ ‧ database

Claims (10)

A dynamic multi-person context awareness system, which cooperates with a client and a server in a sensing environment, and includes: a data retrieval subsystem built in the user terminal, and the data retrieval subsystem includes There is a trigger module, a dynamic sensor participation module and a fully automatic background service module for performing a context aware data capture mode; a data analysis and prediction subsystem is built in the server, The data analysis and prediction subsystem includes a person-time feature information analysis module and a behavior data prediction module to implement a context-aware data and prediction mode; an information-pushing subsystem is built in the user terminal respectively. And the server, the information promotion subsystem comprises a person-time feature message screening module, a promotion information push module, a emergency information push module and a lottery result push module, thereby executing a An information dissemination mode based on action mass perception and context perception; a navigation application subsystem built in the user terminal, the navigation application subsystem includes an intention to clear a single module, a list guide module, a link activity module and a situation learning module for executing an Internet of Things navigation application mode; a database connected to the server, the database In order to store the information obtained when each of the above subsystems executes different modes. A dynamic multi-person context sensing method is implemented in conjunction with the dynamic multi-person context sensing system according to claim 1 of the patent application scope, wherein at least one trigger is set in the sensing environment, and the trigger is transmitted by using a Bluetooth signal. When the user enters the sensing environment, the trigger is sensed, thereby performing one or a combination of the following modes: A. starting the data retrieval subsystem to execute the context aware data capture mode; Initiating the data analysis and prediction subsystem to execute the situational awareness data and prediction mode; C. initiating the information promotion subsystem to perform the information promotion mode based on the action mass perception and context awareness; D. launching the navigation application The system executes the IoT navigation application mode. The dynamic multi-person context sensing method according to claim 2, wherein the context-aware data acquisition mode determines whether the Bluetooth signal conforms to a preset value by using the trigger module, and the dynamic sensor participation module After the judgment is met, one of the sensors built in the client is opened to generate a sensing function, and the automatic background service module is activated to capture the environmental data in the sensing environment and the client. Publicly available personal data. The dynamic multi-person context sensing method according to claim 2, wherein the context-aware data and the prediction mode are first classified according to the situation by the personnel time feature information analysis module, and are weighted by the context, and finally The behavioral data prediction module outputs prediction results. For example, the dynamic multi-person context sensing method described in claim 2, wherein the action mass perception and context-aware information promotion mode are first classified according to the context by the personnel time feature information screening module, and the servo is confirmed. Whether the terminal sends a difficult message and activates the emergency information push module after the emergency message is sent, so as to integrate the user terminal at the same time point, in the same area or the same situation, and analyze the user location of the neighboring area or the same situation. , output a list, and display a push message on the user side. The dynamic multi-person context sensing method according to claim 2, wherein the promotional information pushing module filters the environmental data in the sensing environment, and pushes the screening result according to the preference of the user terminal. Promotional information on appropriate products. The dynamic multi-person context sensing method according to claim 2, wherein the lottery result pushing module is configured according to the situation, and according to the result of screening the environmental data in the sensing environment, the lottery information is found from the server. And through the user terminal, the lottery information includes a lottery event and a lottery result. The dynamic multi-person context sensing method according to claim 2, wherein the intent list module is configured to establish an intent list and parse it into a corresponding list of physical locations, and the list guide navigation module is based on A navigation screen is displayed on the intent list to guide the client to a target location. The dynamic multi-person context sensing method according to claim 2, wherein the activation of the activity module is performed in a related activity, and the location of the user terminal in the sensing environment is perceived as a criterion for determining the activity of the activity. And the user screen displays the screen of the activity. The dynamic multi-person context sensing method according to claim 2, wherein the situation learning module is activated to perform a situational learning, and a scene learning picture is displayed on the user end, according to the sensing environment The results of the screening of environmental data show the relevant learning materials of a commodity.