An adjustable indoor detection device and a method of using the same
By using time window adjustment, environmental change identification and autoregressive model processing techniques in indoor detection devices, dynamically adjusting sensor parameters, the problems of insufficient environmental change identification and poor signal processing in the existing technology are solved, and the accuracy of monitoring is improved and reliability.
Patent Information
- Application Number
- CN202411273313.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-12
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2044-09-12
AI Technical Summary
Existing indoor detection devices are difficult to accurately identify the continuity and trends of environmental changes, and the sensor sensitivity and sampling frequency are fixed, resulting in false alarms or missed alarms in complex environments, insufficient signal processing, and affecting monitoring effects.
The sensor data filtering module is used to adjust the time window and analyze the data correlation. The control signal correlation detection module recognizes environmental changes, the environmental parameter optimization module dynamically adjusts the sensor sensitivity and sampling frequency, and the output signal processing module uses an autoregressive model to process signals to separate noise and enhance the effective signal.
It improves the accuracy and continuity of environmental monitoring, enhances signal quality, improves the reliability and user experience of monitoring devices, and can better cope with complex and changeable indoor environmental conditions.
Smart Images

Figure CN118862015B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of measurement technology, and in particular to an adjustable indoor detection device and a use method thereof. Background Art
[0002] Indoor detection devices are devices used to monitor different parameters in indoor environments. Their main purpose is to ensure the safety and comfort of the environment by monitoring these parameters. For example, the device can detect indoor temperature, humidity, air quality, etc. When a parameter exceeds a predetermined safety threshold, the device will sound an alarm or take corresponding control measures to remind the user or automatically adjust environmental conditions. This device is widely used in indoor environments such as homes, offices, and public places.
[0003] However, in the prior art, due to the lack of effective detection and correlation analysis of environmental change trends, it is difficult for the prior art to accurately identify the continuity of environmental changes, which may lead to delayed response measures. In addition, the prior art usually adopts fixed sensor sensitivity and sampling frequency, and fails to dynamically adjust according to actual environmental changes, resulting in poor performance of monitoring equipment in complex environments, and false alarms or missed alarms. Insufficient noise suppression and extraction of effective signals during signal processing can easily lead to a decrease in the quality of the output signal, affecting the overall monitoring effect. These shortcomings limit the application effect of the prior art in complex environments and reduce the reliability of the equipment and user experience. Summary of the invention
[0004] The purpose of the present invention is to solve the shortcomings of the prior art and to propose an adjustable indoor detection device and a method of using the same.
[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: an adjustable indoor detection device comprises:
[0006] The sensor data filtering module, based on the sensor signal, adjusts the length of the current time window, determines the time range, obtains the time window setting parameters, analyzes and calculates the event probability of each data point within the set time window, determines the relevance of the data points, filters out low-relevance data, retains high-relevance data, and marks high-probability data points to generate a marked data set;
[0007] The control signal association detection module performs time series analysis on the marked events in adjacent time windows based on the marked data set, extracts time series information, generates event time series information, and determines the continuity and trend change of events in adjacent time windows by performing correlation analysis on the time series information, obtains event correlation information, analyzes events with high correlation, and establishes environmental change recognition signals;
[0008] The environmental parameter optimization module evaluates the entropy value of the measurement parameter based on the environmental change recognition signal, obtains the parameter entropy value information, and then optimizes and adjusts the setting of the measurement parameter, modifies the sensitivity and sampling frequency of the sensor, generates an optimized parameter configuration, and then performs a joint correction operation according to the synergistic relationship between the measurement parameters in the optimized parameter configuration to establish a corrected parameter configuration;
[0009] The output signal processing module, based on the corrected parameter configuration, establishes the autoregressive model parameters under the current environment by analyzing the autoregressive model of the measurement signal, generates the autoregressive model parameters, identifies and separates the noise components in the signal, obtains the noise suppression signal, enhances the effective signal part, and generates an enhanced signal output.
[0010] As a further solution of the present invention, the step of obtaining the time window setting parameters is:
[0011] Based on the sensor signal, determine the length of the time window using the formula: Calculate the time window length in represents the number of data points, Indicates the sampling frequency; using the obtained time window length, combined with the total observation time , calculate the probability of an event occurring within the entire time window, using the formula: Get the setting parameters ; Based on the setting parameters , using parameter adjustment coefficient and To further optimize the time window length, use the formula: Get time window setting parameters ,in and Represent the adjustment coefficients, Values are set based on prior knowledge or goals and represent ideal or desired states.
[0012] As a further solution of the present invention, the step of acquiring the event timing information is:
[0013] Based on the labeled data set, the time series analysis of the labeled events in adjacent time windows is performed using the formula: Get timing information ,in Indicates at a point in time The timing information of Representation time window The weight parameter, Indicates time point data points; perform correlation analysis on the time series information, using the formula: Calculate the correlation information of events ,in and is the adjustment coefficient, express The average value of
[0014] Based on the correlation information, the correlation analysis is further refined using the formula:
[0015] Optimize the correlation analysis process to obtain event timing information ,in and Represents the adjustment coefficient.
[0016] As a further solution of the present invention, the step of acquiring the environmental change identification signal is:
[0017] Based on the event timing information, the preliminary signal of environmental change is calculated by a composite operation method using the formula:
[0018] Calculate preliminary environmental change identification signals ,in represents the weight of event data, represents the square of each event;
[0019] Based on the preliminary environmental change identification signal, the comprehensive environmental change identification signal is calculated using the formula:
[0020] Obtain comprehensive environmental change identification signals ,in and represent the standard deviation adjustment coefficient and the nonlinear adjustment factor respectively;
[0021] The comprehensive environmental change recognition signal is used to adjust the sensitivity and sampling frequency of the sensor, using the formula:
[0022] Get environmental change recognition signals ,in is the performance adjustment factor.
[0023] As a further solution of the present invention, the step of obtaining the optimization parameter configuration is:
[0024] First, the efficiency and adaptability of multiple measurement parameters are evaluated by analyzing the sensor response data, using the formula:
[0025] Generate entropy information , where the logarithm base is 2, which improves the sensitivity and accuracy of entropy calculation. represents the probability distribution of each parameter;
[0026] Combined with the comprehensive environmental change identification signal , using the formula:
[0027] Generate parameter configuration ,in and It is a regulating factor used to balance the influence of entropy value and environmental signals;
[0028] Based on the parameter configuration, parameter correction is performed using the formula:
[0029] Generate optimized parameter configuration ,in is the synergistic adjustment coefficient.
[0030] As a further solution of the present invention, the step of obtaining the enhanced signal output is:
[0031] Based on the optimization parameter configuration , perform autoregressive analysis on the measured signal, using the formula: Get the initial enhanced signal ,in Indicates at a point in time The enhanced signal To measure the effective component of the signal, is the weight coefficient of the autoregressive model, is the signal enhancement adjustment coefficient;
[0032] Based on the preliminary enhanced signal , for noise separation and suppression, use the formula:
[0033] Get the denoised signal ,in is the signal mean, is the signal standard deviation, is the noise suppression correction factor;
[0034] Then based on the denoised signal , to ensure signal stability and enhancement through nonlinear amplification, using the formula:
[0035] Generates enhanced signal output ,in is the signal amplification factor.
[0036] As a further solution of the present invention, the step of acquiring the event association information is:
[0037] Based on the labeled data set, the event sequences in adjacent time windows are analyzed, and the correlation strength of events in the time windows is calculated by weighted accumulation and nonlinear adjustment. The formula is as follows:
[0038] Generate preliminary event correlation strength ,in Indicates time point Event data, is the weight coefficient of the time window, To adjust the power term of the absolute value of the event data;
[0039] Based on the preliminary event association strength, the association degree of events in adjacent time windows is calculated using the formula:
[0040] Generate event correlation ,in Indicates at a point in time The event correlation degree, is the adjustment coefficient of the rate of change, is the adjustment coefficient of the square root term.
[0041] As a further solution of the present invention, it also includes:
[0042] Based on the event correlation, the continuity and trend change of the event are judged, and the double weighting and nonlinear adjustment formula is used to enhance the event trend judgment. The formula is as follows:
[0043] Get event related information ,in is the adjustment coefficient of the correlation, It is the adjustment coefficient for trend judgment.
[0044] A method for using an adjustable indoor detection device, the method for using the adjustable indoor detection device is performed based on the above-mentioned adjustable indoor detection device, and comprises the following steps:
[0045] S1, based on the sensor data filtering module, adopts a dynamic time window adjustment algorithm, adjusts the length of the current time window, determines the time range, obtains the time window setting parameters, and uses a probability analysis algorithm to analyze and calculate the event probability of each data point within the set time window, determines the relevance of the data points, and filters out low-relevance data through a relevance screening method, retains high-relevance data, marks high-probability data points, and generates a marked data set;
[0046] S2, based on the marked data set, adopts the time series analysis algorithm, and controls the signal association detection module to perform time series analysis on the marked events in adjacent time windows, extract time series information, and generate event time series information. Subsequently, the time series information is analyzed using the correlation analysis method to determine the continuity and trend changes of events in adjacent time windows, obtain event association information, and analyze events with higher correlation to establish environmental change recognition signals;
[0047] S3, based on the environmental change recognition signal, adopts the parameter optimization algorithm, through the environmental parameter optimization module, evaluates the entropy value of the measurement parameter, obtains the parameter entropy value information, uses the entropy optimization method to optimize and adjust the measurement parameter settings, modifies the sensitivity and sampling frequency of the sensor, generates the optimized parameter configuration, and then uses the joint correction method to perform the joint correction operation according to the synergistic relationship between the measurement parameters of the optimized parameter configuration, and establishes the corrected parameter configuration. Subsequently, based on the corrected parameter configuration, the autoregressive analysis method is adopted, through the output signal processing module, the autoregressive model of the measurement signal is analyzed, the autoregressive model parameters under the current environment are established, the autoregressive model parameters are generated, the noise components in the signal are identified and separated, and the noise suppression signal is obtained. Finally, the effective signal part is enhanced to generate an enhanced signal output.
[0048] Compared with the prior art, the advantages and positive effects of the present invention are:
[0049] In the present invention, by adjusting the time window of sensor data, highly correlated data points are screened out, thereby improving the accuracy of data analysis. By associating the event sequence in adjacent time windows, the trend of environmental changes is identified, and the continuity and accuracy of environmental change detection are enhanced. The entropy value evaluation and optimization adjustment of the measurement parameters help to dynamically adjust the sensitivity and sampling frequency of the sensor, realize the coordinated optimization of the measurement parameters, and further improve the effectiveness of environmental monitoring. The measurement signal is processed and noise suppressed by the autoregressive model, which ensures the clarity and reliability of the output signal and improves the overall performance of the monitoring device. In summary, the present invention makes environmental monitoring more accurate, flexible, and reliable, can better cope with complex and changeable indoor environmental conditions, and ensures the safety and comfort of environmental monitoring. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 It is a system flow chart of the present invention. DETAILED DESCRIPTION
[0051] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.
[0052] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.
[0053] Example
[0054] See also Figure 1 The present invention provides a technical solution: an adjustable indoor detection device comprising:
[0055] The sensor data filtering module, based on the sensor signal, adjusts the length of the current time window, determines the time range, obtains the time window setting parameters, analyzes and calculates the event probability of each data point within the set time window, determines the relevance of the data points, filters out low-relevance data, retains high-relevance data, and marks high-probability data points to generate a marked data set;
[0056] The control signal association detection module performs time series analysis on the marked events in adjacent time windows based on the marked data set, extracts time series information, generates event time series information, and determines the continuity and trend change of events in adjacent time windows by performing correlation analysis on the time series information, obtains event correlation information, analyzes events with high correlation, and establishes environmental change recognition signals;
[0057] The environmental parameter optimization module evaluates the entropy value of the measurement parameter based on the environmental change recognition signal, obtains the parameter entropy value information, and then optimizes and adjusts the setting of the measurement parameter, modifies the sensitivity and sampling frequency of the sensor, generates an optimized parameter configuration, and then performs a joint correction operation according to the synergistic relationship between the measurement parameters in the optimized parameter configuration to establish a corrected parameter configuration;
[0058] The output signal processing module, based on the corrected parameter configuration, establishes the autoregressive model parameters under the current environment by analyzing the autoregressive model of the measurement signal, generates the autoregressive model parameters, identifies and separates the noise components in the signal, obtains the noise suppression signal, enhances the effective signal part, and generates an enhanced signal output.
[0059] In this embodiment, the steps for obtaining the time window setting parameters are:
[0060] Based on the sensor signal, determine the length of the time window using the formula: Calculate the time window length ,in represents the number of data points, Indicates the sampling frequency;
[0061] Using the obtained time window length, combined with the total observation time , calculate the probability of an event occurring within the entire time window, using the formula:
[0062] Get the setting parameters ;
[0063] Based on the setting parameters , using parameter adjustment coefficient and To further optimize the time window length, use the formula:
[0064] Get time window setting parameters ,in and Represent the adjustment coefficients, Values are set based on prior knowledge or goals and represent ideal or desired states.
[0065] 1000 data points are collected, and the sampling frequency is set to 1000 Hz. The time window length is calculated as follows:
[0066] This result represents the total duration of 1000 data points collected by the sensor within 1 second;
[0067] The current observation time is 10 seconds. Substitute the previously calculated parameter values for calculation:
[0068] In the calculation, the inner part is calculated first:
[0069] Then calculate:
[0070] Substituting this value into the exponentiation operation:
[0071] Therefore, we get:
[0072] This result indicates that within the 10-second observation time, the probability of an event occurring is very close to 1, that is, the event can be observed almost throughout the entire observation time.
[0073] Set parameters for time windows Calculation:
[0074] in is the probability of the target event occurring, which is obtained through prior knowledge or the set ideal state value. It is currently set to 0.95, indicating that the event is expected to occur most of the time. and The adjustment coefficient is obtained by adjusting the relative influence of event probability and sampling parameters. The current setting is , , and then bring in the parameter values calculated in the previous steps to calculate:
[0075] First calculate:
[0076] Then calculate:
[0077]
[0078] This result indicates that after optimization and adjustment, the time window setting parameter is approximately 1.0105 seconds, which is slightly higher than the initial time window length, indicating that the adjusted time window is more in line with the set target.
[0079] In this embodiment, the steps for acquiring the event timing information are:
[0080] Based on the labeled data set, the time series analysis of the labeled events in adjacent time windows is performed using the formula:
[0081] Get timing information ,in Indicates at a point in time The timing information of Representation time window The weight parameter, Indicates time point data points;
[0082] The correlation analysis of the timing information is performed using the formula:
[0083] Calculate the correlation information of events ,in and is the adjustment coefficient, express The average value of
[0084] Based on the correlation information, the correlation analysis is further refined using the formula: Optimize the correlation analysis process to obtain event timing information ,in and Represents the adjustment coefficient.
[0085] Representation time window The weight parameter is obtained through statistical analysis of historical data. The weight parameter is usually determined by analyzing the contribution of different time windows to the event. The specific value can be calculated based on the correlation of past data. The weight parameter obtained through analysis is: .
[0086] Indicates at a point in time The data points are obtained through the historical data points collected by the sensor. The data points obtained in the past three time windows are: .
[0087] Substitute actual values for calculation:
[0088]
[0089] The calculation results are expressed at the time point The time series information for is 10.2, which reflects the weighted sum of the data points in the past three time windows.
[0090] The adjustment coefficient is obtained by calibrating the model accuracy. . The adjustment coefficient is used to adjust the impact of the variance part and adjust the model sensitivity through historical data. It is set to . express The average value is obtained by calculating the average value of the collected time series information. The average value is 9.
[0091] Calculation process:
[0092] First calculate the cumulative sum:
[0093]
[0094] Then calculate the variance part:
[0095]
[0096] Enter the formula to calculate:
[0097]
[0098] The calculation results show that the correlation information of the event is 20.649, which reflects the cumulative effect of the time series information at multiple time points and the degree of its deviation from the average value.
[0099] The adjustment coefficient is used to adjust the influence of correlation information on time series information. Currently, it is obtained through model adjustment. , The adjustment coefficient is used to adjust the impact of the reverse accumulation part. It is currently obtained through model adjustment. .
[0100] Calculation process:
[0101] Calculate the reverse cumulative part:
[0102]
[0103] Enter the formula to calculate:
[0104]
[0105] The calculation results show that the final event timing information is 12.3963, which comprehensively considers the impact of event correlation information and its reverse accumulation part on the entire system.
[0106] In this embodiment, the steps of acquiring the environmental change identification signal are:
[0107] Based on the event timing information, the preliminary signal of environmental change is calculated by a composite operation method using the formula:
[0108] Calculate preliminary environmental change identification signals ,in represents the weight of event data, represents the square of each event;
[0109] Based on the preliminary environmental change identification signal, the comprehensive environmental change identification signal is calculated using the formula:
[0110]
[0111] Obtain comprehensive environmental change identification signals ,in and Respectively represent the standard deviation adjustment coefficient and the nonlinear adjustment factor; the sensitivity and sampling frequency of the sensor are adjusted by using the comprehensive environmental change identification signal, and the formula is adopted:
[0112] Get environmental change recognition signals ,in is the performance adjustment factor.
[0113] in, Represents the weight of event data, which is obtained by analyzing the impact of different events on the environment. The weight values currently obtained through statistical analysis are: .
[0114] Represents the value of each event, which is directly obtained through sensor monitoring data. The event values currently collected by the sensor are
[0115] Substitute actual values for calculation:
[0116] First, the square of each event is calculated:
[0117] Then calculate the weighted value:
[0118]
[0119] The calculation results show that the preliminary environmental change identification signal is 167.5, which reflects the weighted square sum of each event in the environment.
[0120] is the standard deviation adjustment factor, obtained by analyzing the volatility of the initial signal and currently obtained through model adjustment . is a nonlinear adjustment factor used to nonlinearly amplify or reduce the preliminary signal. .
[0121] Substitute the parameter values calculated in the previous steps for calculation:
[0122]
[0123] Calculate the parts:
[0124] The calculation results show that the comprehensive environmental change identification signal is 167.5, which indicates that after standard deviation adjustment and nonlinear adjustment, the signal maintains an amplitude consistent with the preliminary signal.
[0125] in is the performance adjustment factor, which is obtained by adjusting the sensitivity and sampling frequency of the sensor. The current setting is .
[0126] Substitute the parameter values calculated in the previous steps for calculation:
[0127] The calculation results show that the final environmental change recognition signal is 201, which means that the environmental change signal detected by the system is strong. Therefore, the system needs to increase the sensitivity of the sensor accordingly in order to more accurately capture subtle environmental changes.
[0128] In this embodiment, the steps of obtaining the optimization parameter configuration are:
[0129] First, the efficiency and adaptability of multiple measurement parameters are evaluated by analyzing the sensor response data, using the formula:
[0130] Generate entropy information , where the logarithm base is 2, which improves the sensitivity and accuracy of entropy calculation. represents the probability distribution of each parameter;
[0131] Combined with the comprehensive environmental change identification signal , using the formula:
[0132] Generate parameter configuration ,in and It is a regulating factor used to balance the influence of entropy value and environmental signals;
[0133] Based on the parameter configuration, parameter correction is performed using the formula:
[0134] Generate optimized parameter configuration ,in is the synergistic adjustment coefficient.
[0135] in Represents the probability distribution of each measurement parameter. This value is obtained by analyzing the response data of the sensor. Currently, by monitoring the data of multiple measurement parameters, the probability distribution of each parameter is obtained as follows: . Substitute the probability distribution of each parameter into the formula for calculation:
[0136] Calculate the parts:
[0137]
[0138] Calculate the entropy value:
[0139]
[0140] The calculation result represents the entropy value information , which is used to measure the efficiency and adaptability of multiple measurement parameters and indicates the distribution status of system parameters.
[0141] The adjustment factor is used to adjust the effect of entropy on parameter configuration and is set to . The adjustment factor is used to adjust the impact of environmental signals on parameter configuration and is set to . The value calculated in the previous step is .
[0142] Substitute the parameter values calculated in the previous steps for calculation:
[0143]
[0144] Calculation of environmental signal part:
[0145] Then calculate the whole formula:
[0146]
[0147] The calculation results represent the optimized parameter configuration , which comprehensively considers the impact of entropy information and environmental change signals on the system.
[0148] is the synergistic adjustment coefficient, which is used to further optimize the parameter configuration and is set as .
[0149] Substitute the parameter values calculated in the previous steps for calculation:
[0150]
[0151] First calculate the denominator:
[0152] Then calculate the whole formula:
[0153]
[0154] The calculation results represent the final optimized parameter configuration. ,This value is relatively moderate, indicating that the system, in the optimized state, has balanced the influence of various parameters, does not favor a specific factor, but takes all relevant parameters into consideration.
[0155] In this embodiment, the steps of obtaining the enhanced signal output are:
[0156] Based on the optimization parameter configuration , perform autoregressive analysis on the measured signal, using the formula:
[0157] Get the initial enhanced signal ,in Indicates at a point in time The enhanced signal To measure the effective component of the signal, is the weight coefficient of the autoregressive model, is the signal enhancement adjustment coefficient;
[0158] Based on the preliminary enhanced signal , for noise separation and suppression, use the formula:
[0159] Get the denoised signal ,in is the signal mean, is the signal standard deviation, is the noise suppression correction factor;
[0160] Then based on the denoised signal , to ensure signal stability and enhancement through nonlinear amplification, using the formula:
[0161] Generates enhanced signal output ,in is the signal amplification factor.
[0162] is the weight coefficient of the autoregressive model, which is obtained through historical data regression analysis. The current weight values obtained through analysis are: . The effective component of the measurement signal is obtained through real-time acquisition by the sensor. The effective signal value at the current time point is The valid signal values at the first few time points are: ,and . is the signal enhancement adjustment coefficient, which is calibrated through experiments and is currently set to .
[0163] Substitute the above parameters into the formula for calculation:
[0164]
[0165] Step by step calculation:
[0166] Calculation results Indicates at a point in time Next, the initial signal value is obtained by comprehensive weighting of past historical signals and current signals. This value reflects the initial processing result of the signal based on the autoregressive model and is the basis for subsequent signal processing steps.
[0167] is the noise suppression correction factor, which is obtained by analyzing the noise signal characteristics and suppression requirements and is set to . is the signal mean, which is calculated by averaging the signal over a period of time and set to . is the signal standard deviation, which is obtained by calculating the volatility of the signal over a period of time and is set to .
[0168] Substitute the above calculation results into And other parameters are calculated:
[0169]
[0170] Calculate the denoising part:
[0171] Calculate the overall formula:
[0172]
[0173] Calculation results Represents the signal value after denoising and adjustment.
[0174] is the signal amplification factor, which is determined by experiment and is used to control the amplification intensity of the final signal and is set to .
[0175] Substitute the above calculation results into And other parameters are calculated:
[0176] Calculate the magnified part:
[0177]
[0178] Final calculation results Represents the final enhanced signal output after denoising and amplification. This value reflects the optimal signal strength after multiple processing, suitable for further analysis, control or other application scenarios. Signal values close to 1 indicate that the system has a good processing effect, retaining the effective information of the signal after denoising and amplification, and is suitable for accurate subsequent analysis and control decisions.
[0179] In this embodiment, the steps for obtaining the event association information are:
[0180] Based on the labeled data set, the event sequences in adjacent time windows are analyzed, and the correlation strength of events in the time windows is calculated by weighted accumulation and nonlinear adjustment. The formula is as follows:
[0181] Generate preliminary event correlation strength ,in Indicates time point Event data, is the weight coefficient of the time window, To adjust the power term of the absolute value of the event data;
[0182] Based on the preliminary event association strength, the association degree of events in adjacent time windows is calculated using the formula:
[0183] Generate event correlation ,in Indicates at a point in time The event correlation degree, is the adjustment coefficient of the rate of change, is the adjustment coefficient of the square root term.
[0184] Based on the event correlation, the continuity and trend change of the event are judged, and the double weighting and nonlinear adjustment formula is used to enhance the event trend judgment. The formula is as follows:
[0185] Get event related information ,in is the adjustment coefficient of the correlation, It is the adjustment coefficient for trend judgment.
[0186] is the weight coefficient of the time window, which is determined by analyzing the impact of events at different time points. The weight values obtained through historical data analysis are: . Indicates time point Event data is directly obtained through sensor or log data collection, currently at the time point The event data are . To adjust the power term of the absolute value of the event data, it is obtained by adjusting the influence of different event data on the correlation strength. The current setting is .
[0187] Substitute the specific values of each parameter into the formula for calculation:
[0188]
[0189] First, calculate the power value of each event data:
[0190]
[0191] Then calculate the weight product:
[0192]
[0193] Calculation results Indicates at a point in time The initial event association strength under .
[0194] is the adjustment coefficient of the change rate, which is obtained by adjusting the impact of the change in the initial association strength on the association degree. The current setting is . is the adjustment coefficient of the square root term, which is used to adjust the response of event correlation to intensity changes. The current setting is . and are the initial event association strengths at the current time point and the previous time point, and the association strength at the previous time point .
[0195] Bring it into the previous step And other parameters are calculated:
[0196] First calculate the rate of change part:
[0197]
[0198] Then calculate the square root term:
[0199] Combine all the parts:
[0200]
[0201] Calculation results Indicates at a point in time This value quantifies the correlation between events in adjacent time windows. A higher value indicates a stronger temporal correlation between events, which is an important basis for judging event trends.
[0202] is the adjustment coefficient of correlation, which is obtained by adjusting the impact of event correlation on the final correlation information. The current setting is . is the adjustment coefficient for trend judgment, which is used to adjust the impact of the difference between event correlations on the final correlation information. The current setting is . and are the event correlations at the current time point and the previous time point, respectively. .
[0203] Substitute the above calculation results into And other parameters are calculated:
[0204] First calculate the correlation part:
[0205] Then calculate the trend judgment part:
[0206] Combine all the parts:
[0207]
[0208] Calculation results Indicates at a point in time This value integrates the event correlation and its changing trend, providing a comprehensive indicator for judging the correlation and trend changes between events.
[0209] The present invention provides a method for using an adjustable indoor detection device, comprising the following steps:
[0210] Based on the sensor data filtering module, a dynamic time window adjustment algorithm is used to adjust the length of the current time window, determine the time range, obtain the time window setting parameters, and use the probability analysis algorithm to analyze and calculate the event probability of each data point within the set time window, determine the relevance of the data points, and filter out low-relevance data through the relevance screening method, retain high-relevance data, mark high-probability data points, and generate a marked data set;
[0211] Based on the marked data set, the time series analysis algorithm is used to perform time series analysis on the marked events in adjacent time windows through the control signal association detection module, extract time series information, and generate event time series information. Subsequently, the time series information is analyzed using the correlation analysis method to determine the continuity and trend changes of events in adjacent time windows, obtain event correlation information, and analyze events with high correlation to establish environmental change recognition signals.
[0212] Based on the environmental change recognition signal, a parameter optimization algorithm is adopted. Through the environmental parameter optimization module, the entropy value of the measurement parameter is evaluated, the parameter entropy value information is obtained, and the entropy value optimization method is used to optimize and adjust the setting of the measurement parameter, modify the sensitivity and sampling frequency of the sensor, and generate the optimized parameter configuration. Then, through the joint correction method, according to the synergistic relationship between the measurement parameters of the optimized parameter configuration, a joint correction operation is performed to establish the corrected parameter configuration. Subsequently, based on the corrected parameter configuration, an autoregressive analysis method is adopted. Through the output signal processing module, the autoregressive model of the measurement signal is analyzed, the autoregressive model parameters under the current environment are established, the autoregressive model parameters are generated, the noise components in the signal are identified and separated, and the noise suppression signal is obtained. Finally, the effective signal part is enhanced to generate an enhanced signal output.
[0213] It should be understood that, although each step in the flow chart of each embodiment of the present invention is shown in sequence according to the indication of the arrow, these steps are not necessarily performed in sequence according to the order indicated by the arrow. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.
[0214] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0215] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.
Claims
1. An adjustable indoor detection device, It is characterized in that include: The sensor data filtering module, based on the sensor signal, adjusts the length of the current time window, determines the time range, obtains the time window setting parameters, analyzes and calculates the event probability of each data point within the set time window, determines the relevance of the data points, filters out low-relevance data, retains high-relevance data, and marks high-probability data points to generate a marked data set; The control signal association detection module performs time series analysis on the marked events in adjacent time windows based on the marked data set, extracts time series information, generates event time series information, and determines the continuity and trend change of events in adjacent time windows by performing correlation analysis on the time series information, obtains event correlation information, analyzes events with high correlation, and establishes environmental change recognition signals; The environmental parameter optimization module evaluates the entropy value of the measurement parameter based on the environmental change recognition signal, obtains the parameter entropy value information, and then optimizes and adjusts the setting of the measurement parameter, modifies the sensitivity and sampling frequency of the sensor, generates an optimized parameter configuration, and then performs a joint correction operation according to the synergistic relationship between the measurement parameters in the optimized parameter configuration to establish a corrected parameter configuration; The output signal processing module, based on the corrected parameter configuration, establishes the autoregressive model parameters under the current environment by analyzing the autoregressive model of the measurement signal, generates the autoregressive model parameters, identifies and separates the noise components in the signal, obtains the noise suppression signal, enhances the effective signal part, and generates an enhanced signal output.
2. The adjustable indoor detection device according to claim 1, It is characterized in that The steps for obtaining the time window setting parameters are as follows: Based on the sensor signal, determine the length of the time window using the formula: Calculate the time window length ,in represents the number of data points, Indicates the sampling frequency; using the obtained time window length, combined with the total observation time , calculate the probability of an event occurring within the entire time window, using the formula: Get the setting parameters ; Based on the setting parameters , using parameter adjustment coefficient and To further optimize the time window length, use the formula: Get time window setting parameters ,in and Represent the adjustment coefficients, Values are set based on prior knowledge or goals and represent ideal or desired states.
3. The adjustable indoor detection device according to claim 1, It is characterized in that The steps for obtaining the event timing information are as follows: Based on the labeled data set, the time series analysis of the labeled events in adjacent time windows is performed using the formula: Get timing information ,in Indicates at a point in time The timing information of Representation time window The weight parameter, Indicates time point data points; The correlation analysis of the timing information is performed using the formula: Calculate the correlation information of events ,in and is the adjustment coefficient, express The average value of Based on the correlation information, the correlation analysis is further refined using the formula: Optimize the correlation analysis process to obtain event timing information ,in and Represents the adjustment coefficient.
4. The adjustable indoor detection device according to claim 3, It is characterized in that The steps of acquiring the environmental change identification signal are: Based on the event timing information, the preliminary signal of environmental change is calculated by a composite operation method using the formula: Calculate preliminary environmental change identification signals ,in represents the weight of event data, represents the square of each event; Based on the preliminary environmental change identification signal, the comprehensive environmental change identification signal is calculated using the formula: Obtain comprehensive environmental change identification signals ,in and represent the standard deviation adjustment coefficient and the nonlinear adjustment factor respectively; The comprehensive environmental change recognition signal is used to adjust the sensitivity and sampling frequency of the sensor, using the formula: Get environmental change recognition signals ,in is the performance adjustment factor.
5. The adjustable indoor detection device according to claim 4, It is characterized in that The steps for obtaining the optimization parameter configuration are: First, the efficiency and adaptability of multiple measurement parameters are evaluated by analyzing the sensor response data, using the formula: Generate entropy information , where the logarithm base is 2, which improves the sensitivity and accuracy of entropy calculation. represents the probability distribution of each parameter; Combined with the comprehensive environmental change identification signal , using the formula: Generate parameter configuration ,in and It is a regulating factor used to balance the influence of entropy value and environmental signals; Based on the parameter configuration, parameter correction is performed using the formula: Generate optimized parameter configuration ,in is the synergistic adjustment coefficient.
6. The adjustable indoor detection device according to claim 5, It is characterized in that The steps of obtaining the enhanced signal output are: Based on the optimization parameter configuration , perform autoregressive analysis on the measured signal, using the formula: Get the initial enhanced signal ,in Indicates at a point in time The enhanced signal To measure the effective component of the signal, is the weight coefficient of the autoregressive model, is the signal enhancement adjustment coefficient; Based on the preliminary enhanced signal , for noise separation and suppression, use the formula: Get the denoised signal ,in is the signal mean, is the signal standard deviation, is the noise suppression correction factor; Then based on the denoised signal , to ensure signal stability and enhancement through nonlinear amplification, using the formula: Generates enhanced signal output ,in is the signal amplification factor.
7. The adjustable indoor detection device according to claim 1, It is characterized in that The steps for obtaining the event association information are as follows: Based on the labeled data set, the event sequences in adjacent time windows are analyzed, and the correlation strength of events in the time windows is calculated by weighted accumulation and nonlinear adjustment. The formula is as follows: Generate preliminary event correlation strength ,in Indicates time point Event data, is the weight coefficient of the time window, To adjust the power term of the absolute value of the event data; Based on the preliminary event association strength, the association degree of events in adjacent time windows is calculated using the formula: Generate event correlation ,in Indicates at a point in time The event correlation degree, is the adjustment coefficient of the rate of change, is the adjustment coefficient of the square root term.
8. The adjustable indoor detection device according to claim 7, It is characterized in that Also includes: Based on the event correlation, the continuity and trend change of the event are judged, and the double weighting and nonlinear adjustment formula is used to enhance the event trend judgment. The formula is as follows: Get event related information ,in is the adjustment coefficient of the correlation, It is the adjustment coefficient for trend judgment.
9. A method for using an adjustable indoor detection device, It is characterized in that The adjustable indoor detection device according to any one of claims 1 to 8 comprises the following steps: Based on the sensor data filtering module, a dynamic time window adjustment algorithm is used to adjust the length of the current time window, determine the time range, obtain the time window setting parameters, and use the probability analysis algorithm to analyze and calculate the event probability of each data point within the set time window, determine the relevance of the data points, and filter out low-relevance data through the relevance screening method, retain high-relevance data, mark high-probability data points, and generate a marked data set; Based on the marked data set, the time series analysis algorithm is used to perform time series analysis on the marked events in adjacent time windows through the control signal association detection module, extract time series information, and generate event time series information. Subsequently, the time series information is analyzed using the correlation analysis method to determine the continuity and trend changes of events in adjacent time windows, obtain event correlation information, and analyze events with high correlation to establish environmental change recognition signals. Based on the environmental change recognition signal, a parameter optimization algorithm is adopted. Through the environmental parameter optimization module, the entropy value of the measurement parameter is evaluated, the parameter entropy value information is obtained, and the entropy value optimization method is used to optimize and adjust the setting of the measurement parameter, modify the sensitivity and sampling frequency of the sensor, and generate the optimized parameter configuration. Then, through the joint correction method, according to the synergistic relationship between the measurement parameters of the optimized parameter configuration, a joint correction operation is performed to establish the corrected parameter configuration. Subsequently, based on the corrected parameter configuration, an autoregressive analysis method is adopted. Through the output signal processing module, the autoregressive model of the measurement signal is analyzed, the autoregressive model parameters under the current environment are established, the autoregressive model parameters are generated, the noise components in the signal are identified and separated, and the noise suppression signal is obtained. Finally, the effective signal part is enhanced to generate an enhanced signal output.
