Multi-source telemetry data fusion method, device, electronic device and storage medium

Abstract

The present invention discloses a multi-source telemetry data fusion method, device, electronic device and storage medium. The method comprises: obtaining telemetry parameters of satellite equipment sensors; for each telemetry parameter, generating a semantic representation including a time series pattern, and then combining to generate a semantic representation sequence; calculating a first state transfer ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and a second state transfer ratio relationship matrix of the previous semantic representation to the next semantic representation between two adjacent semantic representations; multiplying each second state transfer ratio relationship matrix in sequence to obtain a first matrix; iteratively fusing matrices according to the first state transfer ratio relationship matrix and the first matrix until a preset number of iterations is reached, and using the final first state transfer ratio relationship matrix as a fusion environment comprehensive matrix; more accurate multi-source data fusion can be achieved by implementing the present invention.

Description

Technical Field

[0001] The invention relates to the technical field of software engineering, and in particular to a multi-source telemetry data fusion method. Background Art

[0002] With the development of the global information society, satellite technology has become an important force in promoting the development of informatization and intelligence. It is widely used in many fields such as communication, navigation, weather forecasting, and environmental monitoring, greatly promoting global interconnection and resource sharing. In this context, multi-source satellite telemetry data fusion technology, as a research frontier, aims to integrate telemetry data resources from different satellites, broaden the boundaries and depth of data applications, and serve human society more comprehensively.

[0003] However, the lack of interpretability is a huge challenge facing current multi-source data fusion methods. Under high-dimensional data, the distribution of latent space is difficult to achieve an understandable and interpretable description, which makes the model prediction and decision-making process lack a solid theoretical foundation, reduces the user's trust and understanding of the model results, and limits the scope of application of multi-source data fusion technology. Summary of the invention

[0004] The embodiments of the present invention provide a multi-source telemetry data fusion method, device, electronic device and computer-readable storage medium. By implementing the present invention, the conversion relationship between different telemetry parameters can be obtained more comprehensively, thereby achieving more accurate multi-source data fusion.

[0005] An embodiment of the present invention provides a multi-source telemetry data fusion method, including: acquiring telemetry parameters of satellite device sensors that need to be fused.

[0006] For each of the telemetry parameters, a semantic representation comprising a plurality of temporal patterns is generated.

[0007] All the semantic representations are combined in the order of generation to generate a semantic representation sequence.

[0008] The first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence and the second state transition ratio relationship matrix of the previous semantic representation to the next semantic representation between two adjacent semantic representations are calculated.

[0009] The second state transition proportion relationship matrices are multiplied in sequence to obtain a first matrix.

[0010] According to the first state transfer ratio relationship matrix and the first matrix, the matrix is ​​iteratively fused until a preset number of iterations is reached, and the first state transfer ratio relationship matrix finally generated is used as the fusion environment comprehensive matrix.

[0011] Among them, at each iterative fusion:

[0012] Calculate the difference between the first matrix and the current first state transition ratio relationship matrix to obtain a difference matrix;

[0013] The absolute value of each element in the difference matrix is ​​calculated, and the corresponding exponent value is calculated according to the absolute value, and the obtained exponent value is used as the updated element value.

[0014] Generate an updated difference matrix based on all updated element values ​​in the difference matrix.

[0015] An updated first state transition proportion relationship matrix is ​​calculated according to the updated difference matrix, the first matrix and the current first state transition proportion relationship matrix.

[0016] Furthermore, for each of the telemetry parameters, a semantic representation including several time series patterns is generated, including: for each telemetry parameter, segmented aggregation approximation processing is performed on it to generate a mean sequence.

[0017] For each mean sequence, a symbol aggregation approximation process is performed on it to generate a symbol sequence.

[0018] For each symbol sequence, symbols are merged according to the symbol pairs with the highest frequency to generate a semantic representation.

[0019] Furthermore, a transfer ratio relationship matrix between two semantic representations is generated in the following way.

[0020] The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula:

[0021] ,

[0022] in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the overlapping time interval of the i-th timing pattern in the first semantic representation and the j-th timing pattern in the second semantic representation on the time axis.

[0023] According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

[0024] Furthermore, the updated first state transition ratio relationship matrix is ​​calculated according to the following formula:

[0025]

[0026] in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

[0027] Based on the above method embodiment, the present invention provides a corresponding device embodiment.

[0028] An embodiment of the present invention provides a multi-source telemetry data fusion device, including: a semantic representation generation module, a matrix generation module, and a matrix iteration fusion module.

[0029] The semantic representation generation module is used to obtain the telemetry parameters of the satellite equipment sensors that need to be fused;

[0030] For each of the telemetry parameters, a semantic representation including a plurality of time series patterns is generated; all the semantic representations are combined in the order of generation to generate a semantic representation sequence.

[0031] The matrix generation module is used to calculate the first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and the second state transition ratio relationship matrix of the previous semantic representation to the next semantic representation between two adjacent semantic representations.

[0032] The matrix iterative fusion module is used to multiply each of the second state transfer proportion relationship matrices in sequence to obtain a first matrix; iteratively fuse the matrices according to the first state transfer proportion relationship matrix and the first matrix until a preset number of iterations is reached, and the first state transfer proportion relationship matrix finally generated is used as a fusion environment comprehensive matrix; wherein, in each iterative fusion: the difference between the first matrix and the current first state transfer proportion relationship matrix is ​​calculated to obtain a difference matrix; for each element in the difference matrix, its absolute value is calculated, and the corresponding exponential value is calculated according to the absolute value, and the obtained exponential value is used as the updated element value; according to all updated element values ​​in the difference matrix, an updated difference matrix is ​​generated.

[0033] Furthermore, the semantic representation generation module includes: a mean sequence generation unit, a symbol sequence generation unit, and a symbol merging unit.

[0034] The mean sequence generating unit is used to perform segmented aggregation approximation processing on each telemetry parameter to generate a mean sequence.

[0035] The symbol sequence generating unit is used for performing symbol aggregation approximation processing on each mean value sequence to generate a symbol sequence.

[0036] The coincidence merging unit is used to merge symbols according to the symbol pair with the highest frequency for each symbol sequence to generate a semantic representation.

[0037] Furthermore, the state transition ratio relationship matrix generation module generates a transition ratio relationship matrix between two semantic representations in the following manner:

[0038] The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula:

[0039] ,

[0040] in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the overlapping time interval of the i-th timing pattern in the first semantic representation and the j-th timing pattern in the second semantic representation on the time axis.

[0041] According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

[0042] Furthermore, the matrix iteration fusion module calculates the updated first state transition ratio relationship matrix according to the following formula:

[0043]

[0044] in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

[0045] Based on the above method embodiment, the present invention provides a corresponding electronic device embodiment.

[0046] An embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the multi-source telemetry data fusion method described in any one of the above method embodiments can be implemented.

[0047] Based on the above method embodiment, the present invention provides a corresponding computer-readable storage medium embodiment.

[0048] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the multi-source telemetry data fusion method described in any one of the above method embodiments can be implemented.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] The present invention extracts the semantic representation corresponding to each satellite telemetry parameter based on a variety of satellite telemetry parameters. According to the semantic representation of each parameter, the present invention further constructs a state transition ratio relationship matrix between the parameters, and then according to the state transition ratio relationship matrix, fuses the transfer conditions of various states through matrix multiplication, and then continuously iterates and updates to generate a fusion environment comprehensive matrix under various state transfer conditions for multi-source telemetry data fusion. Compared with the prior art, the present invention fuses different semantic features in the matrix space based on the state transition relationship of multi-source data, and can more comprehensively obtain the conversion relationship between different telemetry parameters, thereby realizing more accurate multi-source telemetry data fusion. In particular, since the fusion environment comprehensive matrix realizes the statistical correlation between multi-source state transitions in different telemetry data, this correlation is based on basic statistics and therefore has a certain degree of interpretability, making the model prediction and decision results more transparent and understandable. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a flowchart of a multi-source telemetry data fusion method provided by an embodiment of the present invention.

[0052] Figure 2 It is a schematic diagram of semantic information corresponding to a telemetry parameter provided by an embodiment of the present invention.

[0053] Figure 3 It is a flowchart of iterative fusion provided by an embodiment of the present invention.

[0054] Figure 4 It is a schematic diagram of iterative summation of difference matrix elements provided by an embodiment of the present invention.

[0055] Figure 5 It is a structural schematic diagram of a multi-source telemetry data fusion device provided in one embodiment of the present invention. DETAILED DESCRIPTION

[0056] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0057] like Figure 1 As shown, an embodiment of the present invention provides a multi-source telemetry data fusion method, which at least includes the following steps:

[0058] Step S1, obtaining telemetry parameters of satellite equipment sensors that need to be fused; for each of the telemetry parameters, generating a semantic representation including several time series patterns; combining all the semantic representations in the order of generation to generate a semantic representation sequence.

[0059] It should be noted that the telemetry parameters obtained here can be time series data, including but not limited to the following parameters: battery voltage and current, attitude angle, orbit altitude, temperature and humidity, signal strength, and light intensity. The telemetry data of multiple different satellite equipment sensors are obtained as the telemetry parameters of the above-mentioned fused satellite equipment sensors.

[0060] In an optional embodiment, for each of the telemetry parameters, a semantic representation including a plurality of time series patterns is generated, including:

[0061] For each telemetry parameter, segmented aggregation and approximation processing is performed to generate a mean sequence.

[0062] For each mean sequence, a symbol aggregation approximation process is performed on it to generate a symbol sequence.

[0063] For each symbol sequence, symbols are merged according to the symbol pairs with the highest frequency to generate a semantic representation.

[0064] Optionally, the piecewise aggregation approximation processing may be processed by a PAA (Piecewise Aggregate Approximation) algorithm, including:

[0065] Divide the time series of telemetry parameters into a preset number of segments.

[0066] For each segment, calculate the mean of the data in the segment and use the mean result as the representative value of the segment.

[0067] According to the representative values, a mean value sequence is formed in chronological order.

[0068] Optionally, the symbolic aggregation approximation processing may be processed by a SAX (Symbolic Aggregate ApproXimation) algorithm, including:

[0069] The mean sequence is normalized to generate a standardized sequence.

[0070] Each element of the standardized sequence is mapped into a corresponding symbol according to a preset corresponding relationship.

[0071] According to the corresponding symbols, a symbol sequence is combined in time order.

[0072] Optionally, the symbol merging according to the symbol pair with the highest frequency may be processed by a BPE (Byte Pair Encoding) algorithm, including:

[0073] Step S1.1: Count the frequencies of all character pairs in the current symbol sequence, merge the most frequently appearing character pairs into a new symbol, and generate an updated symbol sequence.

[0074] Step S1.2, determine whether the frequencies of all character pairs in the updated symbol sequence are 1 or whether the number of iterations reaches a preset value (here the preset number of iterations is 10). If so, go to step S1.3; if not, go to step S1.1.

[0075] Step S1.3: Use the updated symbol sequence as a semantic representation.

[0076] In another preferred embodiment, the step of combining symbols according to the symbol pair with the highest frequency comprises:

[0077] Step S1.4, calculate the similarity of each symbol in the current symbol sequence; wherein the similarity is the proportion of symbols with the same sequence.

[0078] Step S1.5: Merge the symbol combinations whose similarity is greater than a preset threshold and whose distance is only one separating character to generate a first symbol sequence, wherein the distance between the symbols is a preset value, and the preset threshold is 0.6.

[0079] Step S1.6: Count the symbol pairs with the highest frequency in the first symbol sequence, merge them into a new symbol, and generate an updated symbol sequence.

[0080] Step S1.7, determine whether the frequencies of all character pairs in the updated symbol sequence are 1 or whether the number of iterations reaches a preset value (the preset number of iterations is 10 here), if so, go to step S1.8, if not, go to step S1.4.

[0081] Step S1.8: Use the updated symbol sequence as a semantic representation.

[0082] By fusing the neighborhood symbols as described above, the expression of the vocabulary can be further compressed, which has strong noise resistance and anti-interference ability. At the same time, it can also reduce the problems of mode splitting, state capture abnormality, mode mixing, etc. caused by the fixed sliding window in the satellite telemetry data.

[0083] According to step 1, an embodiment of the present invention generates a corresponding semantic representation for a telemetry parameter, such as Figure 2 As shown in Figure 1, the telemetry parameters are discretized into semantic information containing four temporal patterns of ABCD. It should be noted that the semantic information still retains the temporal information.

[0084] Step S2, calculating the first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and the second state transition ratio relationship matrix of the first semantic representation to the second semantic representation between two adjacent semantic representations.

[0085] In order to solve the above problem, a transfer ratio relationship matrix between two semantic representations is generated in the following way: For ease of understanding, the first semantic representation and the second semantic representation are used here to represent two semantic representations, which are used to generate a transfer ratio relationship matrix from the first semantic representation to the second semantic representation.

[0086] The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula:

[0087] ,

[0088] in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the overlapping time interval of the i-th timing pattern in the first semantic representation and the j-th timing pattern in the second semantic representation on the time axis.

[0089] According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

[0090] Step S3: multiply the second state transition ratio relationship matrices in order to obtain the first matrix; perform matrix iterative fusion according to the first state transition ratio relationship matrix and the first matrix until a preset number of iterations is reached, and use the first state transition ratio relationship matrix finally generated as the fusion environment comprehensive matrix. The preset number of iterations here is 100.

[0091] like Figure 3 As shown, in one embodiment of the present invention, the iterative fusion includes at least the following steps:

[0092] Step S3.1, calculate the difference between the first matrix and the current first state transition proportion relationship matrix to obtain a difference matrix.

[0093] Step S3.2: Calculate the absolute value of each element in the difference matrix, and calculate the corresponding index value according to the absolute value, and use the obtained index value as the updated element value.

[0094] Step S3.3: Generate an updated difference matrix according to all updated element values ​​in the difference matrix.

[0095] Step S3.4: Calculate an updated first state transition proportion relationship matrix according to the updated difference matrix, the first matrix and the current first state transition proportion relationship matrix.

[0096] Specifically, the updated first state transition ratio relationship matrix is ​​calculated according to the following formula:

[0097]

[0098] in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

[0099] Step S3 essentially transforms the state space for different semantic features by multiplying each of the second state transfer ratio relationship matrices in sequence, so that the transformation becomes stable in the continuous iterative fusion, thereby obtaining information from multi-source data in a more comprehensive and accurate manner.

[0100] like Figure 4 As shown, all elements of the current difference matrix are summed to obtain the sum result, and the sum result of each iteration is recorded. Figure 4 It can be seen that after iterative summation, the result will tend to be stable.

[0101] Specifically, after obtaining the fusion environment comprehensive matrix, the fusion environment comprehensive matrix can be used in comprehensive time series prediction. The matrix is ​​input into a preset time series prediction model. The time series prediction model predicts the remote sensing data in the future period according to the fusion environment comprehensive matrix to obtain a comprehensive time series prediction result.

[0102] It should be noted that the time series prediction model is an existing model, which can be trained by linear regression model, support vector machine (SVM), decision tree model, random forest model, gradient boosting tree (GBM), neural network model and other models.

[0103] In addition to being used for comprehensive time series prediction, the present invention can also be used in other scenarios, including: missing value completion, state analysis, and target classification.

[0104] The above method can perform multi-source telemetry data fusion based on multi-source satellite telemetry parameters, and can statistically calculate the correlation between multi-source state transitions in different telemetry data. This correlation is based on basic statistics and therefore has a certain degree of interpretability.

[0105] In another embodiment of the present invention, a telemetry data fusion method includes:

[0106] Acquire telemetry parameters of a single satellite device sensor within a time period.

[0107] For the telemetry parameters, a semantic representation including a plurality of temporal patterns is generated.

[0108] Determine the number of types of temporal patterns in the semantic representation and map the temporal patterns to row and column indices of the matrix.

[0109] The adjacent time series patterns in the semantic representation are traversed in sequence, and the values ​​of the corresponding row and column positions in the matrix are updated according to their order to generate the state transition matrix of the telemetry parameter.

[0110] The values ​​of the corresponding row and column positions in the update matrix include:

[0111] Each time a pair of adjacent timing patterns is encountered, the value of the corresponding matrix position is increased by 1 until the traversal is completed.

[0112] In this embodiment, telemetry data fusion is performed based on a single satellite telemetry parameter. By generating a state transition matrix of a single satellite telemetry parameter, state transition analysis can be performed on the time axis of the satellite telemetry data.

[0113] Based on the above method embodiment, the present invention provides a corresponding device embodiment.

[0114] like Figure 5 As shown, an embodiment of the present invention provides a multi-source telemetry data fusion device, including: a semantic representation generation module 101, a matrix generation module 102, and a matrix iterative fusion module 103.

[0115] The semantic representation generation module 101 is used to obtain telemetry parameters of satellite equipment sensors that need to be fused; for each of the telemetry parameters, generate a semantic representation containing several time series patterns; and combine all the semantic representations in the order of generation to generate a semantic representation sequence.

[0116] The matrix generation module 102 is used to calculate the first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and the second state transition ratio relationship matrix of the first semantic representation to the second semantic representation between two adjacent semantic representations.

[0117] The matrix iterative fusion module 103 is used to multiply each of the second state transfer proportion relationship matrices in sequence to obtain a first matrix; iteratively fuse the matrices according to the first state transfer proportion relationship matrix and the first matrix until a preset number of iterations is reached, and the first state transfer proportion relationship matrix finally generated is used as a fusion environment comprehensive matrix; wherein, in each iterative fusion: the difference between the first matrix and the current first state transfer proportion relationship matrix is ​​calculated to obtain a difference matrix; for each element in the difference matrix, its absolute value is calculated, and the corresponding exponential value is calculated according to the absolute value, and the obtained exponential value is used as the updated element value; according to all updated element values ​​in the difference matrix, an updated difference matrix is ​​generated.

[0118] In actual operation, the semantic representation generation module includes: a mean sequence generation unit, a symbol sequence generation unit, and a symbol merging unit.

[0119] The mean sequence generating unit is used to perform segmented aggregation approximation processing on each telemetry parameter to generate a mean sequence.

[0120] The symbol sequence generating unit is used for performing symbol aggregation approximation processing on each mean value sequence to generate a symbol sequence.

[0121] The symbol merging unit is used to perform symbol merging according to the symbol pair with the highest frequency for each symbol sequence to generate a semantic representation.

[0122] In actual operation, the state transition ratio relationship matrix generation module generates a transition ratio relationship matrix between two semantic representations in the following manner:

[0123] The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula:

[0124] ,

[0125] in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the overlapping time interval of the i-th timing pattern in the first semantic representation and the j-th timing pattern in the second semantic representation on the time axis.

[0126] According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

[0127] In actual operation, the matrix iteration fusion module calculates the updated first state transition ratio relationship matrix according to the following formula:

[0128]

[0129] in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

[0130] It should be noted that the embodiments of the device described above correspond to the above embodiments of the present invention, and can implement any of the methods described above in the present invention. In addition, the embodiments of the above-mentioned device are merely schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0131] Based on the above method embodiment of the present invention, a corresponding electronic device embodiment is provided.

[0132] An embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the multi-source telemetry data fusion method described in any one of the present invention is implemented, or when the processor executes the computer program, the functions of the modules in the above-mentioned device embodiments are implemented.

[0133] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0134] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0135] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0136] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0137] Based on the above method embodiment, the present invention provides a storage medium embodiment;

[0138] Another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is executed, the device where the storage medium is located is controlled to execute any one of the multi-source telemetry data fusion methods of the present invention.

[0139] The storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0140] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0141] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A multi-source telemetry data fusion method, characterized in that: include: Obtain telemetry parameters of satellite equipment sensors that need to be fused; For each of the telemetry parameters, a semantic representation including a plurality of time series patterns is generated; all the semantic representations are combined in the order of generation to generate a semantic representation sequence; Calculate the first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and the second state transition ratio relationship matrix of the first semantic representation to the second semantic representation between two adjacent semantic representations; Multiply the second state transition proportion relationship matrices in sequence to obtain a first matrix; According to the first state transfer ratio relationship matrix and the first matrix, iterative fusion of the matrices is performed until a preset number of iterations is reached, and the first state transfer ratio relationship matrix finally generated is used as the fusion environment comprehensive matrix; Among them, at each iterative fusion: Calculate the difference between the first matrix and the current first state transition ratio relationship matrix to obtain a difference matrix; For each element in the difference matrix, the absolute value is calculated, and the corresponding index value is calculated according to the absolute value, and the obtained index value is used as the updated element value; Generate an updated difference matrix according to all updated element values ​​in the difference matrix; An updated first state transition proportion relationship matrix is ​​calculated according to the updated difference matrix, the first matrix and the current first state transition proportion relationship matrix.

2. The multi-source telemetry data fusion method according to claim 1, characterized in that: For each of the telemetry parameters, a semantic representation including several time series patterns is generated, including: For each telemetry parameter, perform segmented aggregation approximation processing on it to generate a mean sequence; For each mean value sequence, perform symbol aggregation approximation processing on it to generate a symbol sequence; For each symbol sequence, symbols are merged according to the symbol pairs with the highest frequency to generate a semantic representation.

3. The multi-source telemetry data fusion method according to claim 1, characterized in that: The transfer ratio relationship matrix between two semantic representations is generated in the following way: The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula: ， in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the time interval of the overlap of the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation on the time axis; According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

4. The multi-source telemetry data fusion method according to claim 1, characterized in that: The updated first state transition ratio relationship matrix is ​​calculated according to the following formula: in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

5. A multi-source telemetry data fusion device, characterized in that: include: Semantic representation generation module, matrix generation module, matrix iteration fusion module; The semantic representation generation module is used to obtain telemetry parameters of satellite equipment sensors that need to be fused; For each of the telemetry parameters, a semantic representation including a plurality of time series patterns is generated; all the semantic representations are combined in the order of generation to generate a semantic representation sequence; The matrix generation module is used to calculate the first state transition ratio relationship matrix of the first semantic representation to the last semantic representation in the semantic representation sequence, and the second state transition ratio relationship matrix of the first semantic representation to the second semantic representation between two adjacent semantic representations; The matrix iteration fusion module is used to multiply the second state transition proportion relationship matrices in sequence to obtain a first matrix; According to the first state transfer ratio relationship matrix and the first matrix, iterative fusion of the matrix is ​​performed until a preset number of iterations is reached, and the first state transfer ratio relationship matrix finally generated is used as the fusion environment comprehensive matrix; wherein, in each iterative fusion: the difference between the first matrix and the current first state transfer ratio relationship matrix is ​​calculated to obtain a difference matrix; for each element in the difference matrix, its absolute value is calculated, and the corresponding exponential value is calculated according to the absolute value, and the obtained exponential value is used as the updated element value; Generate an updated difference matrix based on all updated element values ​​in the difference matrix.

6. The multi-source telemetry data fusion device according to claim 5, characterized in that: The semantic representation generation module includes: a mean sequence generation unit, a symbol sequence generation unit, and a symbol merging unit; The mean sequence generating unit is used to perform segmented aggregation approximation processing on each telemetry parameter to generate a mean sequence; The symbol sequence generating unit is used to perform symbol aggregation approximation processing on each mean value sequence to generate a symbol sequence; The symbol merging unit is used to merge symbols according to the symbol pair with the highest frequency for each symbol sequence to generate a semantic representation.

7. The multi-source telemetry data fusion device according to claim 5, characterized in that: The state transition ratio relationship matrix generation module generates a transition ratio relationship matrix between two semantic representations in the following manner: The value of each element in the transfer ratio relationship matrix is ​​calculated using the following formula: ， in, is the element value of the i-th row and j-th column in the transfer ratio relationship matrix of the first semantic representation to the second semantic representation; is the i-th temporal pattern in the first semantic representation; is the jth temporal pattern in the second semantic representation; is the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the time interval of the jth temporal pattern on the time axis in the second semantic representation; is the time interval in which the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation overlap on the time axis; is the length of the time interval of the i-th temporal pattern on the time axis in the first semantic representation; is the length of the time interval of the overlap of the i-th temporal pattern in the first semantic representation and the j-th temporal pattern in the second semantic representation on the time axis; According to the value of each element, a transfer ratio relationship matrix of the first semantic representation to the second semantic representation is generated.

8. The multi-source telemetry data fusion device according to claim 5, characterized in that: The matrix iteration fusion module calculates the updated first state transition ratio relationship matrix according to the following formula: in, is the updated first state transition ratio relationship matrix, is the updated difference matrix, is the first matrix, is the current first state transfer ratio relationship matrix, is element-wise addition.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the multi-source telemetry data fusion method described in any one of claims 1 to 4 can be implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the multi-source telemetry data fusion method described in any one of claims 1 to 4 can be implemented.

Images