Internet of Things monitoring method based on data fusion mechanism

Abstract

The invention belongs to the technical field of data fusion, and provides an Internet of Things monitoring method and system based on a data fusion mechanism, and the method comprises the steps: obtaining all target data in a target data group and all existing data in an existing data group, carrying out the processing and analysis of the target data and the existing data, and obtaining a data deviation value PC, carrying out PC processing analysis based on the data deviation value to generate a replacement analysis signal, carrying out processing analysis on missing data and existing processing data based on the replacement analysis signal to obtain a data combination, carrying out processing analysis based on the data combination to obtain a similar value XSZ, marking the data combination as a similar data set and a non-similar data set based on the similar value XSZ, and carrying out data combination identification based on the similar data set. According to the method, the missing data meeting the condition is obtained through further judgment of the replacement analysis signal, the replacement signal is generated, the missing data in the target data is replaced with the existing corresponding data, then data fusion is carried out, and the data fusion rate is increased.

Description

Technical Field

[0001] The invention belongs to the technical field of data fusion, and in particular to an Internet of Things monitoring method based on a data fusion mechanism. Background Art

[0002] The data in the Internet of Things comes from a wide range of sources and is characterized by diversity and complexity. Different types of sensors may collect data in different formats, with different precisions and frequencies, such as temperature data collected by temperature sensors, humidity data collected by humidity sensors, and image data collected by image sensors. Various types of sensors and smart devices are deployed in large numbers in different scenarios, such as industrial production, smart homes, environmental monitoring, and smart transportation. This has led to a sharp increase in the amount of data in the Internet of Things, resulting in a data explosion. If these massive amounts of data are not effectively processed and integrated, they will bring tremendous pressure to data transmission, storage, and analysis.

[0003] A Chinese patent application with publication number CN111385854B discloses an Internet of Things air monitoring system based on a data fusion mechanism, including: the Internet of Things air monitoring system includes an access node and more than two sensor nodes, the sensor nodes include isolated nodes, cluster head nodes and intra-cluster nodes; isolated nodes are sensor nodes that are neither cluster head nodes nor intra-cluster nodes; users can quickly obtain air parameters through the implementation method of the Internet of Things air monitoring system provided by the present invention, thereby effectively shortening the delay and cost of obtaining air data, and can monitor air quality in real time. The present invention can be applied to fields such as air monitoring and has broad application prospects.

[0004] In the above-mentioned prior art, users can quickly obtain air parameters through the implementation method of the Internet of Things air monitoring system, thereby effectively shortening the delay and cost of obtaining air data and being able to monitor air quality in real time. However, in the above-mentioned prior art, there is a lack of replacement of similar data. If the data does not conform to the fusion mechanism, it needs to be re-collected. Data collection consumes manpower and material resources, which greatly reduces the fusion rate.

[0005] To this end, the present invention provides an Internet of Things monitoring method based on a data fusion mechanism. Summary of the invention

[0006] In order to make up for the deficiencies of the prior art, at least one technical problem raised in the background technology is solved.

[0007] The technical solution adopted by the present invention to solve the technical problem is: an Internet of Things monitoring method based on a data fusion mechanism: comprising:

[0008] Step 1: Acquire all target data in the target data group and all existing data in the existing data group, process and analyze the target data and the existing data to obtain a data deviation value PC, and generate a replacement analysis signal based on the data deviation value PC;

[0009] Step 2: Based on the replacement analysis signal, the missing data and the existing processed data are processed and analyzed to obtain a data combination, and a similarity value XSZ is obtained based on the data combination processing and analysis. The data combination is marked based on the similarity value XSZ, and is marked as a similar data group and a non-similar data group;

[0010] Step 3: Based on the similar data group, mark the missing data to obtain replaceable data. Based on the replaceable data, process and analyze to obtain the missing data replacement value QDZ. According to the missing data replacement value QDZ, determine whether to perform a replacement operation on all missing data groups. If so, generate a replacement signal.

[0011] Step 4: Based on the replacement signal, perform replacement operations on all missing data groups.

[0012] As a further solution of the present invention: the data deviation value PC is obtained in the following manner:

[0013] Obtain the missing data quantity ratio LC and the missing data degree ratio SC;

[0014] The data deviation value PC is calculated by the formula: PC=a1×LC+a2×SC, where a1 and a2 are both preset proportional coefficients, and a1 and a2 are both greater than 0.

[0015] As a further solution of the present invention: the data missing quantity ratio LC and the data missing degree ratio SC are obtained as follows:

[0016] In the target data group, all target data are traversed, and each target data is compared with each existing data in the existing data group. If there is no existing data of the same type as the target data in the existing data group, the corresponding target data is marked as missing data;

[0017] Count the number of missing data in the target data group to obtain the number of missing data;

[0018] Counting the number of target data in the target data group to obtain the number of target data;

[0019] The missing data number is processed by ratio processing with the target data number to obtain the missing data number ratio LC;

[0020] The number of bytes corresponding to all target data in the target data group is summed up to obtain the total target bytes, the bytes corresponding to all missing data are summed up to obtain the missing bytes, and the missing bytes are ratioed to the total target bytes to obtain the data missing degree ratio SC.

[0021] As a further solution of the present invention: the generation method of the substitution analysis signal is:

[0022] Compare the data deviation value to the data deviation threshold:

[0023] If the data deviation value is less than the data deviation threshold, a replacement analysis signal is generated.

[0024] As a further solution of the present invention: the similarity value XSZ is obtained as follows:

[0025] Get the relevant period value SJ and the relevant change variance XGC;

[0026] By formula: The similarity value XSZ of the current data combination is calculated, where b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0.

[0027] As a further solution of the present invention: the relevant time period value SJ and the relevant change variance XGC are obtained in the following manner:

[0028] Based on any missing data, the missing data is combined with each existing processed data to obtain multiple data combinations;

[0029] Among them, the method of obtaining the existing processing data is:

[0030] In the existing data group, the existing data of the same type as the target data is deleted to obtain the existing processed data group, and the data contained in the existing processed data group is the existing processed data;

[0031] Based on any data combination, the missing data in the data combination and the existing processed data are processed and analyzed to obtain the similarity value corresponding to the data combination, specifically:

[0032] Obtain the same historical collection period for the missing data and existing data in the data combination, and divide the historical collection period into several sub-periods of equal length;

[0033] Obtain the data change value corresponding to the missing data of the data combination in the sub-period, and mark it as the missing change value;

[0034] The missing change value is obtained by performing difference processing on the missing data corresponding to the end time point of the sub-period and the missing data corresponding to the start time point of the sub-period;

[0035] Obtaining data change values ​​corresponding to existing data in the data combination within the sub-period, and marking them as existing change values;

[0036] The existing change value is obtained by performing difference processing on the existing data corresponding to the end time point of the sub-period and the existing data corresponding to the start time point of the sub-period;

[0037] Compare the missing change value and the existing change value corresponding to the sub-period for positive or negative information;

[0038] If the missing change value corresponding to the sub-period has the same positive or negative correlation as the existing change value, the sub-period with the same positive or negative correlation is marked as the relevant time period;

[0039] Count the total number of relevant time periods and compare it with the number of sub-periods to obtain the relevant time period value SJ;

[0040] Obtain the missing change value and existing change value of each relevant time period, perform ratio processing on the missing change value of the relevant time period and the existing change value to obtain the relevant change ratio of the relevant time period, integrate the relevant change ratios of all relevant time periods into a relevant change ratio data group, obtain the variance value of the relevant change ratio data group, and mark it as the relevant change variance XGC.

[0041] As a further solution of the present invention: the process of marking the data combination based on the similarity value is:

[0042] Compare the similarity value to the similarity threshold:

[0043] If the similarity value is greater than the similarity threshold, the data combination is marked as a similar data group;

[0044] If the similarity value is less than or equal to the similarity threshold, the data combination is marked as a non-similar data group.

[0045] As a further solution of the present invention: the missing data replacement value QDZ is obtained as follows:

[0046] Obtain the replaceable data number ratio SJB and the replaceable degree ratio CDB;

[0047] The missing data replacement value QDZ is calculated by the formula: QDZ = c1 × SJB + c2 × CDB, where c1 and c2 are both preset proportional coefficients, and c1 and c2 are both greater than 0;

[0048] Compare the replacement value to the replacement threshold:

[0049] If the replacement value is less than or equal to the replacement threshold, it means that many missing data in the missing data group cannot be replaced, and no replacement operation is performed;

[0050] If the replacement value is greater than the replacement threshold, a replacement signal is generated.

[0051] As a further solution of the present invention: the method of obtaining the replaceable data number ratio SJB and the replaceable degree ratio CDB is:

[0052] If the number of similar data groups in the multiple data combinations corresponding to the missing data is not 0, the missing data is marked as replaceable data;

[0053] Count the number of replaceable data in all missing data and mark them as replaceable data number. Ratio the replaceable data number to the total number of missing data to obtain the replaceable data number ratio SJB.

[0054] The number of bytes corresponding to all replaceable data is summed up to obtain replaceable bytes, the number of bytes corresponding to all missing data is summed up to obtain missing bytes, and the replaceable bytes are compared with the missing bytes to obtain a replaceability ratio CDB.

[0055] As a further solution of the present invention: the specific method of performing the replacement operation on all missing data sets is:

[0056] If the number of similar data groups corresponding to the missing data is 1, the similar data group corresponding to the missing data is marked as a replaceable data group, and the missing data in the replaceable data group is replaced with the existing data in the replaceable data group;

[0057] If the number of similar data groups corresponding to the missing data is greater than 1, obtain the similarity values ​​of each similar data group corresponding to the missing data, compare the similarity values ​​of the similar data groups corresponding to the missing data, select the similar data group with the largest similarity value as the replaceable data group, and replace the missing data in the replaceable data group with the existing data in the replaceable data group.

[0058] The beneficial effects of the present invention are as follows:

[0059] 1. Obtain all target data in the target data group and all existing data in the existing data group, obtain data deviation values ​​PC based on processing and analysis of the target data and the existing data, and generate replacement analysis signals based on processing and analysis of the data deviation values ​​PC. The present invention facilitates the generation of replacement analysis signals for data with low importance in the target data by processing and analyzing the target data and the existing data, which is beneficial for further determining whether the missing data in the target data can be replaced by the existing data;

[0060] 2. Based on the replacement analysis signal, the missing data and the existing processed data are processed and analyzed to obtain a data combination, based on the data combination processing and analysis, a similarity value XSZ is obtained, based on the similarity value XSZ, the data combination is marked as a similar data group and a non-similar data group, based on the similar data group, the missing data is marked to obtain replaceable data, based on the replaceable data, a missing data replacement value QDZ is obtained through processing and analysis, and according to the missing data replacement value QDZ, it is determined whether to perform a replacement operation on all missing data groups, if so, a replacement signal is generated, and based on the replacement signal, a replacement operation is performed on all missing data groups. The present invention obtains replacement data and irreplaceable data through further determination of the replacement analysis signal, generates a replacement signal for missing data that meets the conditions, replaces the missing data in the target data with the corresponding existing data, and then performs data fusion, thereby improving the data fusion rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] The present invention will be further described below in conjunction with the accompanying drawings.

[0062] Figure 1 It is a flowchart of the steps of an Internet of Things monitoring method based on a data fusion mechanism according to an embodiment of the present invention;

[0063] Figure 2 It is a flow chart of the steps of obtaining the missing data replacement value QDZ of the Internet of Things monitoring method based on the data fusion mechanism described in an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.

[0065] Example 1

[0066] like Figure 1 The Internet of Things monitoring method based on a data fusion mechanism described in an embodiment of the present invention includes:

[0067] Step 1: Acquire all target data in the target data group and all existing data in the existing data group, process and analyze the target data and the existing data to obtain a data deviation value PC, and generate a replacement analysis signal based on the data deviation value PC;

[0068] In the target data group, all target data are traversed, and each target data is compared with each existing data in the existing data group. If there is no existing data of the same type as the target data in the existing data group, the corresponding target data is marked as missing data;

[0069] Count the number of missing data in the target data group to obtain the number of missing data;

[0070] Counting the number of target data in the target data group to obtain the number of target data;

[0071] The missing data number is processed by ratio processing with the target data number to obtain the missing data number ratio LC;

[0072] The number of bytes corresponding to all target data in the target data group is summed to obtain the total target bytes, the number of bytes corresponding to all missing data is summed to obtain the missing bytes, and the missing bytes are ratioed to the total target bytes to obtain the data missing degree ratio SC;

[0073] The data deviation value PC is calculated by the formula: PC = a1 × LC + a2 × SC, where a1 and a2 are both preset proportional coefficients, and a1 and a2 are both greater than 0;

[0074] It should be noted that the target data is the data required to monitor the Internet of Things through the data fusion mechanism, and the existing data is the data obtained by preliminary data collection of the Internet of Things. The higher the proportion of missing bytes, the higher the importance of the missing data in the target data;

[0075] Compare the data deviation value to the data deviation threshold:

[0076] If the data deviation value is greater than or equal to the data deviation threshold, the missing data needs to be recollected;

[0077] If the data deviation value is less than the data deviation threshold, a replacement analysis signal is generated;

[0078] The technical solution of an embodiment of the present invention is: acquiring all target data in a target data group and all existing data in an existing data group, obtaining a data deviation value PC based on processing and analysis of the target data and the existing data, and generating a replacement analysis signal based on processing and analysis of the data deviation value PC. The present invention facilitates generating a replacement analysis signal for data with low importance in the target data by processing and analyzing the target data and the existing data, which is beneficial to further determine whether the missing data in the target data can be replaced by the existing data.

[0079] Example 2

[0080] like Figure 2 As shown, based on Example 1, an Internet of Things monitoring method based on a data fusion mechanism according to an embodiment of the present invention includes:

[0081] Step 2: Based on the replacement analysis signal, the missing data and the existing processed data are processed and analyzed to obtain a data combination, and a similarity value XSZ is obtained based on the data combination processing and analysis. The data combination is marked based on the similarity value XSZ, and is marked as a similar data group and a non-similar data group;

[0082] Based on any missing data, the missing data is combined with each existing processed data to obtain multiple data combinations;

[0083] Among them, the method of obtaining the existing processing data is:

[0084] In the existing data group, the existing data of the same type as the target data is deleted to obtain the existing processed data group, and the data contained in the existing processed data group is the existing processed data;

[0085] Based on any data combination, the missing data in the data combination and the existing processed data are processed and analyzed to obtain the similarity value corresponding to the data combination, specifically:

[0086] Obtain the same historical collection period for the missing data and existing data in the data combination, and divide the historical collection period into several sub-periods of equal length;

[0087] Obtain the data change value corresponding to the missing data of the data combination in the sub-period, and mark it as the missing change value;

[0088] The missing change value is obtained by performing difference processing on the missing data corresponding to the end time point of the sub-period and the missing data corresponding to the start time point of the sub-period;

[0089] Obtaining data change values ​​corresponding to existing data in the data combination within the sub-period, and marking them as existing change values;

[0090] The existing change value is obtained by performing difference processing on the existing data corresponding to the end time point of the sub-period and the existing data corresponding to the start time point of the sub-period;

[0091] Compare the missing change value and the existing change value corresponding to the sub-period for positive or negative information;

[0092] If the missing change value corresponding to the sub-period has the same positive or negative correlation as the existing change value, the sub-period with the same positive or negative correlation is marked as the relevant time period;

[0093] If the missing change value corresponding to the sub-period is different in positive or negative correlation from the existing change value, the sub-period with different positive or negative correlation is marked as a non-correlated time period;

[0094] Count the total number of relevant time periods and compare it with the number of sub-periods to obtain the relevant time period value SJ;

[0095] Obtain the missing change value and the existing change value of each relevant time period, perform ratio processing on the missing change value of the relevant time period and the existing change value to obtain the relevant change ratio of the relevant time period, integrate the relevant change ratios of all relevant time periods into a relevant change ratio data group, obtain the variance value of the relevant change ratio data group, and mark it as the relevant change variance XGC;

[0096] By formula: Calculate and obtain the similarity value XSZ of the current data combination, where b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0;

[0097] Compare the similarity value to the similarity threshold:

[0098] If the similarity value is greater than the similarity threshold, the data combination is marked as a similar data group;

[0099] If the similarity value is less than or equal to the similarity threshold, the data combination is marked as a non-similar data group;

[0100] Step 3: Based on the similar data group, mark the missing data to obtain replaceable data. Based on the replaceable data, process and analyze to obtain the missing data replacement value QDZ. According to the missing data replacement value QDZ, determine whether to perform a replacement operation on all missing data groups. If so, generate a replacement signal.

[0101] If the number of similar data groups in the multiple data combinations corresponding to the missing data is 0, the missing data will be marked as irreplaceable data;

[0102] If the number of similar data groups in the multiple data combinations corresponding to the missing data is not 0, the missing data is marked as replaceable data;

[0103] Count the number of replaceable data in all missing data and mark them as replaceable data number. Ratio the replaceable data number to the total number of missing data to obtain the replaceable data number ratio SJB.

[0104] The number of bytes corresponding to all replaceable data is summed up to obtain replaceable bytes, the number of bytes corresponding to all missing data is summed up to obtain missing bytes, and the replaceable bytes are compared with the missing bytes to obtain a replaceability ratio CDB;

[0105] The missing data replacement value QDZ is calculated by the formula: QDZ = c1 × SJB + c2 × CDB, where c1 and c2 are both preset proportional coefficients, and c1 and c2 are both greater than 0;

[0106] It should be noted that the missing data replacement value QDZ means that: the higher the proportion of replaceable data in the missing data and the higher the importance of replaceable data in the missing data, the larger the missing replacement value QDZ is, indicating that the possibility of missing data generating replacement signals is greater;

[0107] Compare the replacement value to the replacement threshold:

[0108] If the replacement value is less than or equal to the replacement threshold, it means that many missing data in the missing data group cannot be replaced, and no replacement operation is performed;

[0109] If the replacement value is greater than the replacement threshold, a replacement signal is generated;

[0110] Step 4: Based on the replacement signal, perform replacement operation on all missing data groups;

[0111] Specifically, if the number of similar data groups corresponding to the missing data is 1, the similar data group corresponding to the missing data is marked as a replaceable data group, and the missing data in the replaceable data group is replaced with the existing data in the replaceable data group;

[0112] If the number of similar data groups corresponding to the missing data is greater than 1, then obtain the similarity values ​​of the similar data groups corresponding to the missing data, compare the similarity values ​​of the similar data groups corresponding to the missing data, select the similar data group with the largest similarity value as the replaceable data group, and replace the missing data in the replaceable data group with the existing data in the replaceable data group;

[0113] The technical solution of the embodiment of the present invention is: based on the replacement analysis signal, the missing data and the existing processed data are processed and analyzed to obtain a data combination, based on the data combination processing and analysis, a similarity value XSZ is obtained, based on the similarity value XSZ, the data combination is marked as a similar data group and a non-similar data group, based on the similar data group, the missing data is marked to obtain replaceable data, based on the replaceable data, a missing data replacement value QDZ is obtained by processing and analyzing, and according to the missing data replacement value QDZ, it is determined whether to perform a replacement operation on all missing data groups, and if so, a replacement signal is generated, and based on the replacement signal, a replacement operation is performed on all missing data groups. The present invention obtains replacement data and irreplaceable data through further determination of the replacement analysis signal, generates a replacement signal for the missing data that meets the conditions, replaces the missing data in the target data with the corresponding existing data, and then performs data fusion, thereby improving the data fusion rate.

[0114] The above shows and describes the basic principles, main features and advantages of the present invention. It should be understood by those skilled in the art that the present invention is not limited to the above embodiments. The above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. An Internet of Things monitoring method based on a data fusion mechanism, characterized in that: include: Step 1: Acquire all target data in the target data group and all existing data in the existing data group, process and analyze the target data and the existing data to obtain a data deviation value PC, and generate a replacement analysis signal based on the data deviation value PC; Step 2: Based on the replacement analysis signal, the missing data and the existing processed data are processed and analyzed to obtain a data combination, and a similarity value XSZ is obtained based on the data combination processing and analysis. The data combination is marked based on the similarity value XSZ, and is marked as a similar data group and a non-similar data group; Step 3: Based on the similar data group, mark the missing data to obtain replaceable data. Based on the replaceable data, process and analyze to obtain the missing data replacement value QDZ. According to the missing data replacement value QDZ, determine whether to perform a replacement operation on all missing data groups. If so, generate a replacement signal. Step 4: Based on the replacement signal, perform replacement operations on all missing data groups.

2. The Internet of Things monitoring method based on data fusion mechanism according to claim 1 is characterized in that: The data deviation value PC is obtained in the following manner: Obtain the missing data quantity ratio LC and the missing data degree ratio SC; The data deviation value PC is calculated by the formula: PC=a1×LC+a2×SC, where a1 and a2 are both preset proportional coefficients, and a1 and a2 are both greater than 0.

3. The Internet of Things monitoring method based on data fusion mechanism according to claim 2 is characterized in that: The method for obtaining the data missing quantity ratio LC and the data missing degree ratio SC is as follows: In the target data group, all target data are traversed, and each target data is compared with each existing data in the existing data group. If there is no existing data of the same type as the target data in the existing data group, the corresponding target data is marked as missing data; Count the number of missing data in the target data group to obtain the number of missing data; Counting the number of target data in the target data group to obtain the number of target data; The missing data number is processed by ratio processing with the target data number to obtain the missing data number ratio LC; The number of bytes corresponding to all target data in the target data group is summed up to obtain the total target bytes, the bytes corresponding to all missing data are summed up to obtain the missing bytes, and the missing bytes are ratioed to the total target bytes to obtain the data missing degree ratio SC.

4. The Internet of Things monitoring method based on data fusion mechanism according to claim 2 is characterized in that: The substitution analysis signal is generated by: Compare the data deviation value to the data deviation threshold: If the data deviation value is less than the data deviation threshold, a replacement analysis signal is generated.

5. The Internet of Things monitoring method based on data fusion mechanism according to claim 1 is characterized in that: The similarity value XSZ is obtained as follows: Get the relevant period value SJ and the relevant change variance XGC; By formula: The similarity value XSZ of the current data combination is calculated, where b1 and b2 are both preset proportional coefficients, and b1 and b2 are both greater than 0.

6. The Internet of Things monitoring method based on data fusion mechanism according to claim 5 is characterized in that: The relevant time period value SJ and the relevant change variance XGC are obtained as follows: Based on any missing data, the missing data is combined with each existing processed data to obtain multiple data combinations; Among them, the method of obtaining the existing processing data is: In the existing data group, the existing data of the same type as the target data is deleted to obtain the existing processed data group, and the data contained in the existing processed data group is the existing processed data; Based on any data combination, the missing data in the data combination and the existing processed data are processed and analyzed to obtain the similarity value corresponding to the data combination, specifically: Obtain the same historical collection period for the missing data and existing data in the data combination, and divide the historical collection period into several sub-periods of equal length; Obtain the data change value corresponding to the missing data of the data combination in the sub-period, and mark it as the missing change value; The missing change value is obtained by performing difference processing on the missing data corresponding to the end time point of the sub-period and the missing data corresponding to the start time point of the sub-period; Obtaining data change values ​​corresponding to existing data in the data combination within the sub-period, and marking them as existing change values; The existing change value is obtained by performing difference processing on the existing data corresponding to the end time point of the sub-period and the existing data corresponding to the start time point of the sub-period; Compare the missing change value and the existing change value corresponding to the sub-period for positive or negative information; If the missing change value corresponding to the sub-period has the same positive or negative correlation as the existing change value, the sub-period with the same positive or negative correlation is marked as the relevant time period; Count the total number of relevant time periods and compare it with the number of sub-periods to obtain the relevant time period value SJ; Obtain the missing change value and existing change value of each relevant time period, perform ratio processing on the missing change value of the relevant time period and the existing change value to obtain the relevant change ratio of the relevant time period, integrate the relevant change ratios of all relevant time periods into a relevant change ratio data group, obtain the variance value of the relevant change ratio data group, and mark it as the relevant change variance XGC.

7. The Internet of Things monitoring method based on data fusion mechanism according to claim 5 is characterized in that: The process of marking data combinations based on similarity values ​​is as follows: Compare the similarity value to the similarity threshold: If the similarity value is greater than the similarity threshold, the data combination is marked as a similar data group; If the similarity value is less than or equal to the similarity threshold, the data combination is marked as a non-similar data group.

8. The Internet of Things monitoring method based on data fusion mechanism according to claim 1 is characterized in that: The missing data replacement value QDZ is obtained as follows: Obtain the replaceable data number ratio SJB and the replaceable degree ratio CDB; The missing data replacement value QDZ is calculated by the formula: QDZ = c1 × SJB + c2 × CDB, where c1 and c2 are both preset proportional coefficients, and c1 and c2 are both greater than 0; Compare the replacement value to the replacement threshold: If the replacement value is less than or equal to the replacement threshold, it means that many missing data in the missing data group cannot be replaced, and no replacement operation is performed; If the replacement value is greater than the replacement threshold, a replacement signal is generated.

9. The Internet of Things monitoring method based on data fusion mechanism according to claim 8 is characterized in that: The method of obtaining the replaceable data number ratio SJB and the replaceable degree ratio CDB is as follows: If the number of similar data groups in the multiple data combinations corresponding to the missing data is not 0, the missing data is marked as replaceable data; Count the number of replaceable data in all missing data and mark them as replaceable data number. Ratio the replaceable data number to the total number of missing data to obtain the replaceable data number ratio SJB. The number of bytes corresponding to all replaceable data is summed up to obtain replaceable bytes, the number of bytes corresponding to all missing data is summed up to obtain missing bytes, and the replaceable bytes are compared with the missing bytes to obtain a replaceability ratio CDB.

10. The Internet of Things monitoring method based on data fusion mechanism according to claim 7 is characterized in that: The specific method of replacing all missing data groups is: If the number of similar data groups corresponding to the missing data is 1, the similar data group corresponding to the missing data is marked as a replaceable data group, and the missing data in the replaceable data group is replaced with the existing data in the replaceable data group; If the number of similar data groups corresponding to the missing data is greater than 1, obtain the similarity values ​​of each similar data group corresponding to the missing data, compare the similarity values ​​of the similar data groups corresponding to the missing data, select the similar data group with the largest similarity value as the replaceable data group, and replace the missing data in the replaceable data group with the existing data in the replaceable data group.

Images