A home safety monitoring system based on artificial intelligence

Abstract

The present invention relates to the field of home safety technology, and discloses a home safety supervision system based on artificial intelligence, including a device connection module, which is used to establish data connection between multiple smart home appliances and a preset terminal to obtain multiple monitoring units; a monitoring module, which is used to set multiple identical data transmission windows within a monitoring time period, and within the data transmission window, alternately send preset codes to the smart home appliances through the preset terminal, and obtain transmission feedback of the preset codes; a testing module, which is used to add P wave interference to the monitoring unit within a preset test time period to obtain characteristic transmission feedback of the preset codes; a correction module, which is used to perform spatiotemporal analysis on the transmission feedback of the preset codes to obtain spatiotemporal analysis results, and correct the transmission feedback of the preset codes according to the spatiotemporal analysis results to obtain standard feedback; and a judgment module, which is used to fit the standard feedback and the characteristic transmission feedback to obtain a fitting result.

Description

Technical Field

[0001] The present invention relates to the field of home safety, and more specifically, to a home safety monitoring system based on artificial intelligence. Background Art

[0002] Before an earthquake occurs, there are often smaller foreshocks, which are usually precursors to larger earthquakes. However, due to their small magnitude, these foreshocks are almost imperceptible when people are sleeping.

[0003] The current earthquake warning system mainly relies on the Internet to send alarm information to users through mobile terminals. However, the alarm system of mobile terminals may have delays at critical moments and cannot achieve timely pre-earthquake warnings. In addition, at night, the alarm volume of mobile phones is limited, and the alarm sound is difficult to wake up residents. This means that in an emergency, users may not be able to receive earthquake alarms in time, thus affecting the time of evacuation and emergency response. Summary of the invention

[0004] The present invention provides a home safety monitoring system based on artificial intelligence to solve the technical problems raised in the background technology.

[0005] In a first aspect, the present invention provides a home safety monitoring system based on artificial intelligence, comprising:

[0006] A device connection module is used to establish data connections between multiple smart home appliances and a preset terminal to obtain multiple monitoring units;

[0007] A monitoring module, used to set a plurality of identical data transmission windows within a monitoring period, and in the data transmission windows, send preset codes to the smart home appliance device through a preset terminal alternately, and obtain transmission feedback of the preset codes;

[0008] A test module, used to add P-wave interference to the monitoring unit within a preset test time period to obtain characteristic transmission feedback of a preset code;

[0009] A correction module, used for performing spatiotemporal analysis on the transmission feedback of the preset code to obtain a spatiotemporal analysis result, and correcting the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain a standard feedback;

[0010] A judgment module is used to fit the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judge the earthquake risk according to the fitting result;

[0011] The alarm module is used to control smart home appliances to alarm people in the house according to the earthquake risk.

[0012] Furthermore, the preset terminal includes a data transfer unit:

[0013] The data transfer unit is deployed at a fixed terminal in the house;

[0014] The data transfer unit is configured with a variety of handshake protocols, which are used to establish data connections with smart home appliances through handshake protocols;

[0015] The data transfer unit is configured with a startup layer, which is used to activate the data connection of the monitoring unit and enter the data monitoring state after receiving the user's startup command;

[0016] The data transfer unit is configured with an interference layer for adding uniform artificial noise to the preset code during data transmission.

[0017] Furthermore, the preset encoding includes:

[0018] Step 1, obtaining all instruction sets of the brand of the corresponding smart home appliance and the supported instruction sets of the corresponding smart home appliance;

[0019] Step 2, taking the difference between the entire instruction set and the supported instruction set as the non-supported instructions in the entire instruction set;

[0020] Step 3: configure the non-supported instruction into a data packet, and use the data packet as a preset code for the corresponding smart home appliance.

[0021] Furthermore, sending a preset code to the smart home appliance through a preset terminal and obtaining transmission feedback of the preset code includes:

[0022] Step 4, sending a preset code to the smart home appliance through the fixed terminal, and obtaining an orderly response from the smart home appliance;

[0023] Step 5.1: If the smart home appliance obtains a preset code, the transmission feedback is a first character string, and the dimension of the first character string is the same as the preset code;

[0024] Step 5.2: If the smart home appliance obtains part of the preset code, the transmission feedback is a second character string, and the dimension of the second character string is smaller than the preset code.

[0025] Furthermore, during a preset test period, P wave interference is added to the monitoring unit to obtain a characteristic transmission feedback of a preset code, including:

[0026] Step 6, setting a P wave generating device in the house within a preset test time period;

[0027] Step 7, executing a start command on the monitoring unit to obtain a characteristic character string fed back by the smart home appliance within a preset test time period;

[0028] Step 8, performing delay analysis on the characteristic character string, determining the delay between adjacent characters in the characteristic character string, and sorting the characters from long to short according to the length of the delay to obtain a characteristic sequence;

[0029] Step 9, determining the dimension difference between the feature string and the preset code, and intercepting the delay of the dimension difference from the front to the back of the feature sequence;

[0030] Step 10, fill the empty set in the character interval corresponding to the delay to obtain feature transmission feedback.

[0031] Further, the transmission feedback of the preset code is corrected according to the spatiotemporal analysis result to obtain standard feedback, including:

[0032] Step 11, configuring a noise monitoring device for the monitoring unit, the noise monitoring device responding to the start layer of the data transfer unit;

[0033] Step 12, setting the sampling interval of the noise monitoring device to be the same as the character sending time interval of the preset code;

[0034] Step 13, matching the noise in the monitoring time period with the transmission feedback to obtain a noise segment;

[0035] Step 14: Extract the intensity of the noise fragment at multiple sampling moments ,frequency , rate of change and phase shift PD, we get the noise vector ;

[0036] Step 15, based on the noise vector , to correct the transmission feedback, the formula is as follows:

[0037] ;in, Indicates the update timestamp of each character in the transmission feedback, Indicates the initial timestamp of each character in the transmission feedback, represents the first weight matrix, represents the linear correction term of the noise, represents the second weight matrix, The nonlinear sinusoidal correction term representing the noise characteristics, represents the third weight matrix, The nonlinear cosine correction term representing the noise characteristics, Represents the natural base;

[0038] Step 16: Update the transmission feedback according to the update timestamp of each character in the transmission feedback to obtain standard feedback.

[0039] Furthermore, the standard feedback and the characteristic transmission feedback are fitted to obtain fitting results, including:

[0040] Step 17, performing the operations of step 8 to step 10 on the second character string corresponding to the standard feedback to obtain updated standard feedback;

[0041] Step 18, calculating the cosine similarity of the update standard feedback and the feature transmission feedback, and taking the cosine similarity as the fitting result;

[0042] Step 19, setting an alarm warning value and an alarm maximum value, and using the alarm maximum value as the maximum earthquake risk value;

[0043] Step 20, normalize the fitting results by the maximum and minimum values ​​to obtain the earthquake risk.

[0044] Furthermore, according to the earthquake risk, the smart home appliances are controlled to alarm the people in the house, including:

[0045] Step 21, comparing the earthquake risk with the alarm warning value;

[0046] Step 22: If the earthquake risk is greater than the alarm threshold, the smart home appliance is controlled to alarm the people in the house, as follows:

[0047] Wake up the audio equipment in smart home appliances and output alarm sound at maximum power;

[0048] Wake up the lighting devices in smart home appliances and illuminate them at maximum power.

[0049] In a second aspect, a home safety supervision method based on artificial intelligence is applied to the system, comprising:

[0050] Establish data connection between multiple smart home appliances and preset terminals to obtain multiple monitoring units;

[0051] A plurality of identical data transmission windows are set within the monitoring time period, and within the data transmission windows, a preset code is sent to the smart home appliance device through a preset terminal alternately, and transmission feedback of the preset code is obtained;

[0052] In a preset test period, P wave interference is added to the monitoring unit to obtain characteristic transmission feedback of preset coding;

[0053] Performing a spatiotemporal analysis on the transmission feedback of the preset code to obtain a spatiotemporal analysis result, and correcting the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain a standard feedback;

[0054] Fitting the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judging the earthquake risk according to the fitting result;

[0055] According to the earthquake risk, smart home appliances are controlled to alert people in the house.

[0056] In a second aspect, a storage medium stores a program, and the program implements the system when executed.

[0057] The beneficial effects of the present invention are: through real-time monitoring and multi-level noise correction, the system can accurately distinguish between earthquake P waves and environmental interference, trigger an alarm in time when an earthquake occurs, and link audio and lighting equipment to emit sound and light alarms at maximum power, quickly awakening residents and reminding them to take protective measures, thereby significantly improving home safety and reducing personal injury and property losses that may be caused by earthquakes. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 It is a module diagram of a home safety monitoring system based on artificial intelligence of the present invention;

[0059] Figure 2 It is a flow chart of a home safety supervision method based on artificial intelligence of the present invention. DETAILED DESCRIPTION

[0060] The subject matter described herein will now be discussed with reference to example implementations. It should be understood that the discussion of these implementations is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Various examples may omit, replace, or add various processes or components as needed. In addition, the features described in some examples may also be combined in other examples.

[0061] like Figure 1 As shown, a home safety monitoring system based on artificial intelligence includes:

[0062] A device connection module is used to establish data connections between multiple smart home appliances and a preset terminal to obtain multiple monitoring units;

[0063] A monitoring module, used to set a plurality of identical data transmission windows within a monitoring period, and in the data transmission windows, send preset codes to the smart home appliance device through a preset terminal alternately, and obtain transmission feedback of the preset codes;

[0064] A test module, used to add P-wave interference to the monitoring unit within a preset test time period to obtain characteristic transmission feedback of a preset code;

[0065] A correction module, used for performing spatiotemporal analysis on the transmission feedback of the preset code to obtain a spatiotemporal analysis result, and correcting the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain a standard feedback;

[0066] A judgment module is used to fit the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judge the earthquake risk according to the fitting result;

[0067] The alarm module is used to control smart home appliances to alarm people in the house according to the earthquake risk.

[0068] Specifically, the house includes multiple smart home appliances, which are connected through preset terminals to build a local area network. A preset terminal is connected to multiple smart home appliances in parallel to obtain multiple monitoring units. Therefore, in multiple data transmission windows, real-time monitoring effects are achieved through the alternation of monitoring units.

[0069] In one embodiment of the present invention, the preset terminal includes a data transfer unit:

[0070] The data transfer unit is deployed at a fixed terminal in the house;

[0071] The data transfer unit is configured with a variety of handshake protocols, which are used to establish data connections with smart home appliances through handshake protocols;

[0072] The data transfer unit is configured with a startup layer, which is used to activate the data connection of the monitoring unit and enter the data monitoring state after receiving the user's startup command;

[0073] The data transfer unit is configured with an interference layer for adding uniform artificial noise to the preset code during data transmission.

[0074] Specifically, different brands of smart home appliances have different handshake protocols for establishing data connections, so the data transfer unit needs to be compatible with smart home appliances and be configured with multiple handshake protocols.

[0075] The user's start command can be a preset time period, such as the working time period between 11 pm and 8 am for the monitoring unit, or it can be a user-defined start command, such as sending a start command via a mobile phone, or the above two schemes can be used in parallel.

[0076] Since the handshake protocol of smart home appliances includes a specific data transmission algorithm to ensure the stability of data transmission, the interference caused by P waves is limited and it is difficult to capture the corresponding features. Therefore, uniform artificial noise is added to the preset code, which makes the data transmission susceptible to interference from P waves and thus captures the corresponding features.

[0077] In one embodiment of the present invention, the preset encoding includes:

[0078] Step 1, obtaining all instruction sets of the brand of the corresponding smart home appliance and the supported instruction sets of the corresponding smart home appliance;

[0079] Step 2, taking the difference between the entire instruction set and the supported instruction set as the non-supported instructions in the entire instruction set;

[0080] Step 3: configure the non-supported instruction into a data packet, and use the data packet as a preset code for the corresponding smart home appliance.

[0081] Specifically, the entire instruction set of a brand of home appliance smart device is assigned instructions to a home appliance smart device under the brand according to its functions. In most cases, a home appliance smart device will not have all the functions corresponding to the entire instruction set. Therefore, for a home appliance smart device, it can receive the entire instruction set, but cannot make a specific response to the instructions that are not supported by the function. Therefore, using the preset code configured with unsupported instructions will not have a real impact on the home appliance smart device.

[0082] In one embodiment of the present invention, sending a preset code to a smart home appliance through a preset terminal and obtaining transmission feedback of the preset code includes:

[0083] Step 4, sending a preset code to the smart home appliance through the fixed terminal, and obtaining an orderly response from the smart home appliance;

[0084] Step 5.1: If the smart home appliance obtains a preset code, the transmission feedback is a first character string, and the dimension of the first character string is the same as the preset code;

[0085] Step 5.2: If the smart home appliance obtains part of the preset code, the transmission feedback is a second character string, and the dimension of the second character string is smaller than the preset code.

[0086] Specifically, after receiving the command, the smart home appliance device will feedback a signal. Assuming that the signal is 1 and the preset encoding dimension is 10, the smart home appliance device feedback is .

[0087] P waves are the waves that travel farther than other seismic waves and are the fastest (arrive first) among all seismic waves. P waves are short-duration and high-frequency, so they are easy to monitor.

[0088] For example, P wave interference makes some instructions in the preset code invalid, and the string is obtained .

[0089] In one embodiment of the present invention, adding P wave interference to the monitoring unit within a preset test time period to obtain a preset coded characteristic transmission feedback includes:

[0090] Step 6, setting a P wave generating device in the house within a preset test time period;

[0091] Step 7, executing a start command on the monitoring unit to obtain a characteristic character string fed back by the smart home appliance within a preset test time period;

[0092] Step 8, performing delay analysis on the characteristic character string, determining the delay between adjacent characters in the characteristic character string, and sorting the characters from long to short according to the length of the delay to obtain a characteristic sequence;

[0093] Step 9, determining the dimension difference between the feature string and the preset code, and intercepting the delay of the dimension difference from the front to the back of the feature sequence;

[0094] Step 10, fill the empty set in the character interval corresponding to the delay to obtain feature transmission feedback.

[0095] For example, the string Based on the timestamp of each character, construct a vector , then the characteristic sequence is .exist The corresponding character intervals are filled, then , and Fill empty sets between ,in, represents the empty set. Represents feature transmission feedback.

[0096] In one embodiment of the present invention, the transmission feedback of the preset code is corrected according to the spatiotemporal analysis result to obtain the standard feedback, including:

[0097] Step 11, configuring a noise monitoring device for the monitoring unit, the noise monitoring device responding to the start layer of the data transfer unit;

[0098] Step 12, setting the sampling interval of the noise monitoring device to be the same as the character sending time interval of the preset code;

[0099] Step 13, matching the noise in the monitoring time period with the transmission feedback to obtain a noise segment;

[0100] Step 14: Extract the intensity of the noise fragment at multiple sampling moments ,frequency , rate of change and phase shift , and get the noise vector ;

[0101] Step 15, based on the noise vector , to correct the transmission feedback, the formula is as follows:

[0102] ;in, Indicates the update timestamp of each character in the transmission feedback, Indicates the initial timestamp of each character in the transmission feedback, represents the first weight matrix, represents the linear correction term of the noise, represents the second weight matrix, The nonlinear sinusoidal correction term representing the noise characteristics, represents the third weight matrix, The nonlinear cosine correction term representing the noise characteristics, Represents the natural base;

[0103] Step 16: Update the transmission feedback according to the update timestamp of each character in the transmission feedback to obtain standard feedback.

[0104] Assume that the following data are obtained at four sampling times t=1, 2, 3, 4:

[0105] The initial timestamp of each character in the transmission feedback , .

[0106] Noise signature matrix , , ,

[0107] in, , , and Indication strength , , , and Indicates frequency , , , and Indicates the rate of change , , , and Indicates phase shift .

[0108] is a 1×4 matrix used for linear correction, assuming .

[0109] is a 1×4 matrix used for nonlinear correction. Assume .

[0110] is a 1×4 matrix used for nonlinear correction. Assume .

[0111] Calculated according to the correction formula , , and The updated timestamp will give standard feedback.

[0112] In one embodiment of the present invention, the standard feedback and the characteristic transmission feedback are fitted to obtain a fitting result, including:

[0113] Step 17, performing the operations of step 8 to step 10 on the second character string corresponding to the standard feedback to obtain updated standard feedback;

[0114] Step 18, calculating the cosine similarity of the update standard feedback and the feature transmission feedback, and taking the cosine similarity as the fitting result;

[0115] Step 19, setting an alarm warning value and an alarm maximum value, and using the alarm maximum value as the maximum earthquake risk value;

[0116] Step 20, normalize the fitting results by the maximum and minimum values ​​to obtain the earthquake risk.

[0117] Specifically, assuming the standard feedback is , the characteristic transmission feedback is ; Then calculate the cosine similarity, assuming it is 0.65;

[0118] Based on the experience of experts, the maximum alarm value of cosine similarity is set to 0.8, and the alarm threshold value is set to 70%;

[0119] By normalizing the maximum and minimum values, 0.65 is converted into a probability of 81.25%, and the earthquake risk is 81.25%.

[0120] In one embodiment of the present invention, according to the earthquake risk, controlling the smart home appliance to alarm people in the house includes:

[0121] Step 21, comparing the earthquake risk with the alarm warning value;

[0122] Step 22: If the earthquake risk is greater than the alarm threshold, the smart home appliance is controlled to alarm the people in the house, as follows:

[0123] Wake up the audio equipment in smart home appliances and output alarm sound at maximum power;

[0124] Wake up the lighting devices in smart home appliances and illuminate them at maximum power.

[0125] Specifically, if it is judged that the earthquake risk is 81.25%>the alarm warning value is 70%, the smart home appliance will be controlled to alarm the people in the house.

[0126] The specific operation of the alarm includes the following two aspects: waking up the audio device and outputting the alarm sound at maximum power; when the audio device wakes up, the system sends a command to the smart audio device in the house to switch it from standby to working state. Maximum power output alarm sound: After the audio device is awakened, the alarm sound is played at maximum volume. The alarm sound is usually designed to be high-volume, rapid and continuous, with the aim of attracting the attention of all people in the house in a short period of time to ensure that they can be aware of potential earthquake risks in time. In terms of sound design, the alarm sound can use continuous beeps, sirens, etc., which are usually high-decibel and highly stimulating sound effects to improve the effectiveness of the alarm.

[0127] Wake up the lighting equipment and illuminate at maximum power. Wake up the lighting equipment: The system sends instructions to the intelligent lighting equipment in the house to switch it from sleep or low-power mode to working state. Maximum power lighting: After the lighting equipment is awakened, it illuminates at the highest brightness. The purpose of this is to quickly light up the space in the house in an emergency, even at night or in a low-light environment, to ensure that people can clearly see the surrounding environment, so as to facilitate rapid evacuation and action. Flashing lighting: In order to enhance the alarm effect, the system can also make the lighting equipment flash to remind people with bright lights, so as to further attract the attention of people in the house. The purpose and effect of the alarm, by controlling the maximum power output of the sound and lighting equipment, the system can quickly and effectively remind people in the house when the earthquake risk reaches a certain level. The alarm system designed in this way has the following purposes: Rapid reminder: The sound and lighting are output at maximum power to maximize the alarm effect, ensuring that they can immediately alert people even in a noisy environment or when people are not concentrating. Dual visual and auditory reminders: Through the dual stimulation of sound and light, everyone in the house can be aware of the risk, even people with limited hearing or vision can be reminded. Facilitates escape at night or in low-light environments: At night or in low-light environments, maximum-power lighting helps illuminate the path, helping residents evacuate the house quickly and safely, and reducing secondary damage caused by earthquakes.

[0128] In a preferred embodiment of the present invention, a mobile terminal device (such as a mobile phone) is connected to a preset terminal. When the mobile terminal device receives an earthquake alarm, the transmission feedback of the monitoring unit is not required, and the smart home appliance device is directly controlled to alarm the people in the house.

[0129] Specifically, when an area enters a geologically active period and earthquakes begin to occur frequently, this system can be used to monitor earthquakes of lower magnitude at night. At the same time, it avoids the problem of mobile phones being unable to respond to earthquake alarms in a timely manner due to signal problems in the house, and maximizes the use of existing functions of smart home appliances in the house, thereby reducing alarm costs and improving the success rate of alarms.

[0130] like Figure 2 As shown, a home safety supervision method based on artificial intelligence is applied to the system, including:

[0131] Establish data connection between multiple smart home appliances and preset terminals to obtain multiple monitoring units;

[0132] A plurality of identical data transmission windows are set within the monitoring time period, and within the data transmission windows, a preset code is sent to the smart home appliance device through a preset terminal alternately, and transmission feedback of the preset code is obtained;

[0133] In a preset test period, P wave interference is added to the monitoring unit to obtain characteristic transmission feedback of preset coding;

[0134] Performing a spatiotemporal analysis on the transmission feedback of the preset code to obtain a spatiotemporal analysis result, and correcting the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain a standard feedback;

[0135] Fitting the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judging the earthquake risk according to the fitting result;

[0136] According to the earthquake risk, smart home appliances are controlled to alert people in the house.

[0137] A storage medium stores a program, and the system is implemented when the program is executed.

[0138] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. A home safety monitoring system based on artificial intelligence, characterized in that: include: The device connection module is used to establish data connection between multiple smart home appliances and a preset terminal to obtain multiple monitoring units; wherein the preset terminal includes a data transfer unit: The data transfer unit is deployed at a fixed terminal in the house; The data transfer unit is configured with a variety of handshake protocols, which are used to establish data connections with smart home appliances through handshake protocols; The data transfer unit is configured with a startup layer, which is used to activate the data connection of the monitoring unit and enter the data monitoring state after receiving the user's startup command; The data transfer unit is configured with an interference layer for adding uniform artificial noise to the preset code during data transmission; The monitoring module is used to set a plurality of identical data transmission windows within the monitoring time period, and in the data transmission windows, send preset codes to the smart home appliance device through the preset terminal alternately, and obtain transmission feedback of the preset codes; wherein the preset codes include: Step 1, obtaining all instruction sets of the brand of the corresponding smart home appliance and the supported instruction sets of the corresponding smart home appliance; Step 2, taking the difference between the entire instruction set and the supported instruction set as the non-supported instructions in the entire instruction set; Step 3, configuring the non-supported command into a data packet, and using the data packet as a preset code of the corresponding smart home appliance; Get transmission feedback of preset encoding, including: Step 4, sending a preset code to the smart home appliance through the fixed terminal, and obtaining an orderly response from the smart home appliance; Step 5.1: If the smart home appliance obtains a preset code, the transmission feedback is a first character string, and the dimension of the first character string is the same as the preset code; Step 5.2: If the smart home appliance obtains part of the preset code, the transmission feedback is a second string, and the dimension of the second string is smaller than the preset code; The test module is used to add P-wave interference to the monitoring unit within a preset test time period to obtain characteristic transmission feedback of a preset code, including: Step 6, setting a P wave generating device in the house within a preset test time period; Step 7, executing a start command on the monitoring unit to obtain a characteristic character string fed back by the smart home appliance within a preset test time period; Step 8, performing delay analysis on the characteristic character string, determining the delay between adjacent characters in the characteristic character string, and sorting the characters from long to short according to the length of the delay to obtain a characteristic sequence; Step 9, determining the dimension difference between the feature string and the preset code, and intercepting the delay of the dimension difference from the front to the back of the feature sequence; Step 10, filling an empty set in the character interval corresponding to the delay to obtain feature transmission feedback; The correction module is used to perform spatiotemporal analysis on the transmission feedback of the preset code to obtain the spatiotemporal analysis result, and correct the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain the standard feedback, including: Step 11, configuring a noise monitoring device for the monitoring unit, the noise monitoring device responding to the start layer of the data transfer unit; Step 12, setting the sampling interval of the noise monitoring device to be the same as the character sending time interval of the preset code; Step 13, matching the noise in the monitoring time period with the transmission feedback to obtain a noise segment; Step 14: Extract the intensity of the noise fragment at multiple sampling moments ,frequency , rate of change and phase shift , and get the noise vector ; Step 15, based on the noise vector , correct the transmission feedback; A judgment module is used to fit the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judge the earthquake risk according to the fitting result; The alarm module is used to control smart home appliances to alarm people in the house according to the earthquake risk.

2. The artificial intelligence-based home safety monitoring system according to claim 1, characterized in that: Based on the noise vector , to correct the transmission feedback, the formula is as follows: ;in, Indicates the update timestamp of each character in the transmission feedback, Indicates the initial timestamp of each character in the transmission feedback, represents the first weight matrix, represents the linear correction term of the noise, represents the second weight matrix, The nonlinear sinusoidal correction term representing the noise characteristics, represents the third weight matrix, The nonlinear cosine correction term representing the noise characteristics, Represents the natural base; Step 16: Update the transmission feedback according to the update timestamp of each character in the transmission feedback to obtain standard feedback.

3. The artificial intelligence-based home safety monitoring system according to claim 2, characterized in that: The standard feedback and characteristic transmission feedback are fitted to obtain fitting results, including: Step 17, performing the operations of step 8 to step 10 on the second character string corresponding to the standard feedback to obtain updated standard feedback; Step 18, calculating the cosine similarity of the update standard feedback and the feature transmission feedback, and taking the cosine similarity as the fitting result; Step 19, setting an alarm warning value and an alarm maximum value, and using the alarm maximum value as the maximum earthquake risk value; Step 20, normalize the fitting results by the maximum and minimum values ​​to obtain the earthquake risk.

4. The artificial intelligence-based home safety monitoring system according to claim 3, characterized in that: According to the earthquake risk, smart home appliances are controlled to alert people in the house, including: Step 21, comparing the earthquake risk with the alarm warning value; Step 22: If the earthquake risk is greater than the alarm threshold, the smart home appliance is controlled to alarm the people in the house, as follows: Wake up the audio equipment in smart home appliances and output alarm sound at maximum power; Wake up the lighting devices in smart home appliances and illuminate them at maximum power.

5. A home safety monitoring method based on artificial intelligence, characterized in that: The system applied to any one of claims 1 to 4, comprising: Establish data connection between multiple smart home appliances and preset terminals to obtain multiple monitoring units; A plurality of identical data transmission windows are set within the monitoring time period, and within the data transmission windows, a preset code is sent to the smart home appliance device through a preset terminal alternately, and transmission feedback of the preset code is obtained; In a preset test period, P wave interference is added to the monitoring unit to obtain characteristic transmission feedback of preset coding; Performing a spatiotemporal analysis on the transmission feedback of the preset code to obtain a spatiotemporal analysis result, and correcting the transmission feedback of the preset code according to the spatiotemporal analysis result to obtain a standard feedback; Fitting the standard feedback and the characteristic transmission feedback to obtain a fitting result, and judging the earthquake risk according to the fitting result; According to the earthquake risk, smart home appliances are controlled to alert people in the house.

6. A storage medium, characterized in that: The storage medium stores a program, and when the program is executed, the system according to any one of claims 1 to 4 is implemented.

Images