Time sequence signal similarity measurement and classification method based on multi-feature fusion

Abstract

The invention discloses a time sequence signal similarity measurement and classification method based on multi-feature fusion. The classification method comprises the following steps: firstly, taking input time sequence signals as a reference, screening out signals with great difference through hierarchical filtering, and selecting a measurement index for measuring the similarity of the time sequence signals; the multi-feature fusion-based time sequence signal similarity measurement method is used for counting the importance of a data set on each similarity measurement index, improving a scaling method in an analytic hierarchy process, and constructing a more accurate judgment matrix to obtain a weight coefficient of the data set about each similarity measurement index. And finally, similarity measurement is carried out on an actually measured data set by using the method, a classification test is realized, comparison tests of different similarity measurement methods are carried out, and a test result shows that the proposed method can accurately reflect the similarity between time sequence signals in different data sets, and is excellent in classification accuracy and calculation speed.

Description

Technical Field

[0001] The present invention belongs to the technical field of data mining, and relates to data similarity analysis, and specifically to a time series signal similarity measurement and classification method based on multi-feature fusion. Background Art

[0002] Time series signals refer to a set of data points arranged in chronological order, which are commonly found in various fields such as medicine, economics, meteorology and engineering. With the development of big data and big models and the popularization of sensor technology, using data mining to mine useful information from a large number of time series signals has become increasingly important and has broad application prospects.

[0003] So far, data mining of time series signals has mainly focused on tasks such as prediction, classification, clustering and anomaly detection. In these tasks, one of the key links is the similarity measurement of time series signals. The method of time series similarity measurement can help quickly locate and analyze new time series signals, provide a basis for decision-making, and improve the efficiency and quality of data mining.

[0004] Although existing technologies have optimized the efficiency and effectiveness of metrics at different levels through feature extraction, dimensionality reduction, and algorithm optimization, there are still some shortcomings: First, the adaptability to data sets in different application scenarios is poor, and the stability of the results and the robustness of the algorithm need to be improved. Second, when processing long time series, especially when real-time analysis is required, it still takes a lot of computing time. Summary of the invention

[0005] In view of the shortcomings of the prior art, the present invention proposes a similarity measurement and classification method for time series signals based on multi-feature fusion, introduces the concepts of multi-feature fusion and hierarchical filtering, uses multiple similarity feature indicators to measure the similarity between time series signals, and provides a method for calculating feature weight coefficients to accurately characterize the importance of the selected similarity feature indicators on different data sets. This method not only enhances the versatility and generalization of the method, but also greatly improves the calculation rate, accuracy and stability of the measurement results.

[0006] The time series signal similarity measurement method based on multi-feature fusion specifically includes the following steps: Step 1: Select N similarity feature indicators to measure the similarity between different time series signals.

[0007] Step 2: Set the similarity interval threshold for each signal in the data set Construct a similarity interval. Calculate the similarity of each signal in the data set under different similarity feature indicators, sort them from large to small according to similarity, and take the top signal, construct signal In the high similarity interval under different similarity feature indicators, compare the elements in the similar interval and the high similarity interval to determine the signal Contribution value to the Dth similarity feature index:

[0008] in Indicates signal The similarity interval of Indicates signal The high similarity interval of Represents similarity interval The first A signal.

[0009] Normalize the contribution value of each signal to obtain the percentage of the Dth similarity feature index in the data set : Step 3: Based on the hierarchical analysis method, the difference between the maximum and minimum percentages of the similarity feature index in a data set is calculated. It is divided into 9 distribution intervals of importance, forming a scaling method based on interval constraints. According to the importance between each similarity feature index, a judgment matrix M is constructed to allocate the weight distribution results of each similarity feature index in the data set.

[0010] Step 4: When measuring the similarity between time series signals X and Y, first calculate the similarity between X and Y under multiple similarity feature indicators in step 1. Then, according to the methods of steps 2 and 3, calculate the weight values ​​of the data set where the time series signal Y is located for different similarity feature indicators Finally, the results of multiple similarity feature indicators are weighted and summed according to the weight value as the similarity measurement result of time series signals X and Y : The classification method of time series signal similarity measurement based on multi-feature fusion first determines the weight value of the similarity feature index under different known data sets. Then obtain the length of the time series signal T to be classified , remove the known data sets whose length range is in the interval Signals other than is the set length threshold parameter. Further obtain the total amplitude of the time series signal T to be classified , remove the remaining signals with a total amplitude of Signals other than is the total amplitude threshold parameter, is the total signal amplitude of the time series signal T.

[0011] Finally, the similarity measurement result between the time series signal T and the remaining signals is calculated, and the similarity measurement result is used as the distance between the time series signal T and the remaining signals. Based on the idea of ​​KNN, Z signals closest to the time series signal T are selected from the remaining signals, and the categories of these signals are counted. The category with the highest number of occurrences is taken as the category of the time series signal T to complete the classification.

[0012] The present invention has the following beneficial effects: (1) It eliminates the limitation of traditional methods on single features and provides multiple similarity measurement indicators and their corresponding similarity measurement methods. In particular, a hierarchical analysis method based on similarity interval constraints is proposed to solve the problem of weight coefficient allocation for each similarity measurement indicator. This method optimizes the scaling method in the traditional hierarchical analysis method by quantifying the importance of multiple similarity measurement indicators, and then obtains a judgment matrix suitable for time series signal data sets, thereby calculating accurate weight coefficients.

[0013] (2) The concept of hierarchical filtering is introduced to reduce unnecessary calculations. Experimental results on actual data sets show that this method achieves a classification accuracy of over 90% in classification tasks, and the calculation time is reduced by one order of magnitude compared to other methods. The overall solution takes into account excellent usability and strong scalability while also ensuring high generalization. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the time series signal similarity measurement method based on multi-feature fusion; Figure 2 is an example of the similarity feature index percentage result used in Example 1; Figure 3 is a schematic diagram of the target placement of the test site obtained by the data set in Example 2; Figure 4 is the original acquisition signal of the data set DS1 in Example 2 of the method of the present invention; Figure 5 is the original acquisition signal of the data set DS2 in Example 2 of the method of the present invention; Figure 6 is the original acquisition signal of the data set DS3 in Example 2 of the method of the present invention; Figure 7is the original acquisition signal of the data set DS4 in Example 2 of the method of the present invention; Figure 8 The data set DS1 in Example 2 of the method of the present invention is a signal after data preprocessing; Fig. 9 It is the signal after the data set DS2 in Example 2 of the method of the present invention is preprocessed; Fig.10 It is the signal after the data set DS3 in Example 2 of the method of the present invention is preprocessed; Fig.11 It is the signal after the data set DS4 in Example 2 of the method of the present invention is preprocessed; Fig.12 It is a flow chart of the classification method based on the similarity measurement of time series signals based on multi-feature fusion. DETAILED DESCRIPTION

[0015] The present invention will be further explained below with reference to the accompanying drawings; Example 1 This embodiment introduces a time series signal similarity measurement method based on multi-feature fusion, such as Figure 1 As shown, it includes the following steps: Step 1: First, preprocess the data, including standardization, data form conversion and noise removal, to improve the quality, accuracy and availability of the data, and lay the foundation for subsequent processing steps. Then select signal length, total signal amplitude, signal discreteness, maximum signal correlation coefficient, signal shape and signal distance as similarity feature indicators to measure the similarity between time series signals X and Y:

[0016] s1.1, Similarity of signal length Pay attention to the length difference between the two signals. The smaller the length difference, the higher the similarity of the two signals: (1) Where n and m represent the lengths of the time series signals X and Y respectively.

[0017] s1.2, signal amplitude covers a lot of information such as signal energy, trend information and noise level. The similarity of the total signal amplitude Focus on the amplitude difference between the two signals. The smaller the total amplitude of the signal, the higher the similarity of the two signals: (2) in, represents the i-th element of the time series signal X, Represents the jth element of the time series signal Y; It means to find the absolute value. It means to find the maximum value; s1.3, signal dispersion represents the degree of change or volatility of the signal, and the similarity of signal dispersion The smaller it is, the more similar the two signals are: (3) in, and Represent the standard deviation of time series signals X and Y respectively.

[0018] s1.4, the maximum cross-correlation coefficient measures the similarity of two signals under a certain time delay. The similarity of the maximum cross-correlation coefficient The larger it is, the more similar the shapes and changing trends of the two signals are at the same time: (4) in, , K is the number of time series signals from the same dataset as the time series signal Y, represents the maximum correlation coefficient between the time series signal X and the kth time series signal in the data set, k=1,2,…,K, Indicates the maximum cross-correlation coefficient of time series signals X and Y.

[0019] s1.5. Similarity of signal shapes Focus on the difference between the shapes or waveforms of two signals: First, calculate the upper and lower envelopes of the time series signal X and , and the overlap length of time series signals X and Y Then intercept , and the first L elements of Y, we get a subsequence of length L , , Using subsequences Falling within the envelope and The number of elements between To reflect the shape similarity between time series signals X and Y, and use the time series signal of the data set to which time series signal Y belongs to normalize, and obtain the signal shape similarity of time series signals X and Y :

[0020] (5) in, , Indicates the shape similarity between the time series signal X and the kth time series signal in the dataset.

[0021] s1.6. Use dynamic time warping method to measure the signal distance between time series , starting from the first data point of the two time series, each alignment operation between data points is regarded as a decision stage, so as to The distance value is converted to the subsequence distance value: (6) in, Represents the last element of the time series signal X, Y. Respectively represent the time series signals X and Y removed After that, the remaining elements are a sequence.

[0022] After normalization, the similarity of the signal distance of time series signals X and Y is obtained : (7) in, , Represents the signal distance between the time series signal X and the kth time series signal in the data set.

[0023] Step 2: For different data sets, the weights of the above N similarity feature indicators in measuring the similarity of time series signals need to be determined based on the data characteristics in the data set: s2.1, for a data set , set the similarity interval threshold to , find the location away from the signal Recent signals, together forming the similarity interval of the signal : (8) s2.2, Calculation signal The similarity between each similarity feature indicator and other signals in the dataset S ~ : (9) s2.3. Since the signals in the similar interval are close in position, the similarity of the signals in the similar interval will be higher in theory. In order to avoid the interference of abnormal signals, the concept of high similarity interval is proposed. Similarity Sort from largest to smallest, take the first signals, forming signals High similarity interval .

[0024] Comparison signal Similar interval High similarity interval The signal within, confirm the signal Contribution value to the Dth similarity feature index: (10) (11) s2.4. Determine the contribution value of data set S to the Dth similarity feature index: (12) After normalization, the percentage of the Dth similarity feature index is obtained : (13) Step 3: Based on the analytic hierarchy process, determine the weight coefficients of different similarity feature indicators on the data set S: s3.1. According to the calculation results of step 2, determine the percentage of similarity feature indicators with the greatest importance in data set S , the percentage of similarity feature indicators with the least importance , calculate the difference between the two Then the difference Divided into 9 distribution intervals of importance, the width of each distribution interval is , forming a scaling method based on interval constraints. The importance scales corresponding to different distribution intervals are shown in Table 1:

[0025] Table 1 s3.2. Based on the scaling method given in Table 1, compare the importance of each similarity feature index and construct the judgment matrix M. Figure 2 The similarity feature index percentage results shown, the corresponding judgment matrix is:

[0026] s3.3. Calculate the maximum eigenvalue of the judgment matrix M And the order E, then calculate its consistency index CI and consistency ratio CR, and perform consistency test on the judgment matrix M: (14) (15) Among them, RI is the average random consistency index, which is determined according to the order of the judgment matrix M. If the consistency ratio , then the judgment matrix M passes the consistency test, otherwise it is necessary to adjust the size of the similarity interval and the high similarity interval in step 2 and reconstruct the judgment matrix M.

[0027] s3.4, according to the judgment matrix M, determine Figure 2 The weight distribution results of each similarity feature index under the similarity feature index percentage shown are shown in Table 2: Table 2 Step 4: When measuring the similarity between time series signals X and Y, first calculate the similarity between X and Y under multiple similarity feature indicators in step 1. Then, according to the methods of steps 2 and 3, calculate the weight values ​​of the data set where the time series signal Y is located for different similarity feature indicators Finally, the results of multiple similarity feature indicators are weighted and summed according to the weight value as the similarity measurement result of time series signals X and Y : (16) Example 2 This embodiment introduces a classification method based on the similarity measurement of time series signals based on multi-feature fusion. The known data set used in this embodiment is the active sonar target echo signal in the underwater acoustic field. Figure 3 As shown in the figure, the test scenario for obtaining measured data is as follows: the transmission signal is a linear frequency modulation (LFM) signal, the transducer, hydrophone and target are all located 7m underwater, the distance between the transducer and the target is 8m, the transmission signal pulse width is 2ms, the signal frequency range is 40-80kHz, and the sampling frequency is 1Mhz. The experimental equipment collects the active sonar target echo signal in the form of a time series signal.

[0028] The target is a diver that rotates slowly with the platform. The data is classified according to the target posture to form four data sets: DS1, DS2, DS3, and DS4. The original collected signals in the four data sets are as follows: Figures 4 to 7 As shown in the figure, after the original collected signal is subjected to data preprocessing operations such as filtering, noise reduction and segmentation, the signal processing results are as follows Figures 8 to 11 The detailed information in each dataset is shown in Table 3:

[0029] Table 3 like Figure 3 As shown, firstly, the weight values ​​of the similarity feature indexes in the four data sets DS1, DS2, DS3, and DS4 are determined according to the methods of steps 2 and 3 in Example 1, and the calculation results are shown in Table 4: Table 4 The threshold values ​​involved in the weight calculation process are shown in Table 5: Table 5 Then get the length of the time series signal T to be classified , remove the known data sets whose length range is in the interval Signals other than =0.5, which is the length threshold parameter. Further obtain the total amplitude of the time series signal T to be classified , remove the remaining signals with a total amplitude of Signals other than =0.5, is the total amplitude threshold parameter, is the total signal amplitude of the time series signal T.

[0030] Finally, the similarity measurement result between the time series signal T and the remaining signals is calculated according to the method of step 4 in Example 1, and the similarity measurement result is used as the distance between the time series signal T and the remaining signals. Based on the idea of ​​KNN, Z signals closest to the time series signal T are selected from the remaining signals, and the categories of these signals are counted. The category with the highest number of occurrences is used as the category of the time series signal T to complete the classification.

[0031] Multiple samples are randomly selected from the data sets DS1, DS2, DS3, and DS4 to test the accuracy of this classification method. The results are shown in Table 6: Table 6 It can be seen that this method achieves good results when classifying the four types of measured data set sample signals, which reflects that this method has high stability and robustness, and can maintain good separation between different categories and maintain high similarity consistency within the same category.

[0032] In addition, this embodiment also selects several traditional similarity measurement methods for comparative analysis, and the selected classifier is the KNN classifier, and the Z value is set to 1. The final classification accuracy and calculation time results are shown in Table 7:

[0033] Table 7 From the results in Table 7, it can be seen that compared with other time series signal similarity measurement methods, this method has more advantages in terms of classification accuracy and calculation time.

[0034] 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, and 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 to be protected. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A time series signal similarity measurement method based on multi-feature fusion, characterized by: The specific steps include: Step 1: Select N similarity feature indicators to measure the similarity between different time series signals; Step 2: Set the similarity interval threshold for each signal in the data set Construct a similarity interval ; Calculate the similarity of each signal in the data set under different similarity feature indicators, sort them from large to small according to the similarity, and take the top signal, construct signal High similarity interval under the Dth similarity feature index ; Compare the elements in the similar interval with the high similarity interval to determine the signal Contribution value to the D-th similarity feature index, D=1,2,…,N: （1） （2） Normalize the contribution value of each signal to obtain the percentage of the Dth similarity feature index in the data set : （3） （4） Step 3: Based on the hierarchical analysis method, the difference between the maximum and minimum percentages of the similarity feature index in a data set is calculated. Divide it into multiple distribution intervals of importance, form a scaling method based on interval constraints, construct a judgment matrix M according to the importance between each similarity feature index, and thus allocate the weight distribution results of each similarity feature index under the data set; Step 4: When measuring the similarity between time series signals X and Y, first calculate the similarity between X and Y under multiple similarity feature indicators in step 1. Then, according to the methods of steps 2 and 3, calculate the weight values ​​of the data set where the time series signal Y is located for different similarity feature indicators Finally, the results of multiple similarity feature indicators are weighted and summed according to the weight value as the similarity measurement result of time series signals X and Y : （5） The time series signal similarity measurement method based on multi-feature fusion as described in claim 1 is characterized by selecting signal length, signal total amplitude, signal discreteness, signal maximum mutual correlation coefficient, signal shape and signal distance as similarity feature indicators.

2. The method for measuring similarity of time series signals based on multi-feature fusion according to claim 2, characterized in that: For a time series signal X with a length of n and a time series signal Y with a length of m, the calculation method of each similarity feature index is: s1.1, Similarity of signal length : （6） s1.2, similarity of total signal amplitude : （7） in, represents the i-th element of the time series signal X, Represents the jth element of the time series signal Y; It means to find the absolute value. It means to find the maximum value; s1.3, Similarity of signal dispersion : （8） in, and Respectively represent the standard deviation of time series signals X and Y; s1.4, Similarity of Maximum Correlation Coefficient : （9） in, , K is the number of time series signals from the same dataset as the time series signal Y, represents the maximum correlation coefficient between the time series signal X and the kth time series signal in the data set, k=1,2,…,K, Indicates the maximum mutual correlation coefficient of time series signals X and Y; s1.

5. Similarity of signal shapes : （10） in, , Indicates the shape similarity between the time series signal X and the kth time series signal in the data set; s1.6, Similarity of signal distance : （11） in, , Represents the signal distance between the time series signal X and the kth time series signal in the data set.

3. The method for measuring similarity of time series signals based on multi-feature fusion as claimed in claim 3, characterized in that: The shape similarity calculation method is: first calculate the upper and lower envelopes of the time series signal X and , and the overlap length of time series signals X and Y ; then intercept , and the first L elements of Y, we get a subsequence of length L , , ; Using subsequence Falling within the envelope and The number of elements between To reflect the shape similarity between time series signals X and Y.

4. The method for measuring similarity of time series signals based on multi-feature fusion as claimed in claim 3, characterized in that: Use dynamic time warping to measure the signal distance between time series : （12） in, Represents the last element of the time series signal X, Y; Respectively represent the time series signals X and Y removed The sequence consisting of the remaining elements after .

5. The method for measuring similarity of time series signals based on multi-feature fusion according to claim 1, characterized in that: Based on the interval-constrained scaling method, the importance scales corresponding to different distribution intervals are shown in Table 1: Table 1 in, .

6. The method for measuring similarity of time series signals based on multi-feature fusion according to claim 1, characterized in that: If the constructed judgment matrix M cannot pass the consistency test, the size of the similarity interval and the high similarity interval in step 2 is adjusted, and the judgment matrix M is reconstructed.

7. A classification method for time series signal similarity measurement based on multi-feature fusion, characterized by: First, according to any of claims 1 to 7, the time series signal similarity measurement method based on multi-feature fusion is used to determine the weight values ​​of similarity feature indicators under different known data sets; Then get the length of the time series signal T to be classified , remove the known data sets whose length range is in the interval Signals other than is the set length threshold parameter; further obtain the total amplitude of the time series signal T to be classified , remove the remaining signals with a total amplitude of Signals other than is the total amplitude threshold parameter, is the total signal amplitude of the time series signal T; Finally, the similarity measurement result between the time series signal T and the remaining signals is calculated, and the similarity measurement result is used as the distance between the time series signal T and the remaining signals. Based on the idea of ​​KNN, Z signals closest to the time series signal T are selected from the remaining signals, and the categories of these signals are counted. The category with the highest number of occurrences is taken as the category of the time series signal T to complete the classification.

Images