CN117118431B - Temperature frequency offset compensation method and system - Google Patents
Temperature frequency offset compensation method and system Download PDFInfo
- Publication number
- CN117118431B CN117118431B CN202311143984.4A CN202311143984A CN117118431B CN 117118431 B CN117118431 B CN 117118431B CN 202311143984 A CN202311143984 A CN 202311143984A CN 117118431 B CN117118431 B CN 117118431B
- Authority
- CN
- China
- Prior art keywords
- data
- frequency offset
- temperature
- feature
- compensation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 239000013078 crystal Substances 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims description 3
- 230000000875 corresponding effect Effects 0.000 description 89
- 238000012937 correction Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000002596 correlated effect Effects 0.000 description 2
- 238000005520 cutting process Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006698 induction Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- ZRHANBBTXQZFSP-UHFFFAOYSA-M potassium;4-amino-3,5,6-trichloropyridine-2-carboxylate Chemical compound [K+].NC1=C(Cl)C(Cl)=NC(C([O-])=O)=C1Cl ZRHANBBTXQZFSP-UHFFFAOYSA-M 0.000 description 1
- 239000010453 quartz Substances 0.000 description 1
- 238000009877 rendering Methods 0.000 description 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N silicon dioxide Inorganic materials O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/25—Integrating or interfacing systems involving database management systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03L—AUTOMATIC CONTROL, STARTING, SYNCHRONISATION, OR STABILISATION OF GENERATORS OF ELECTRONIC OSCILLATIONS OR PULSES
- H03L1/00—Stabilisation of generator output against variations of physical values, e.g. power supply
- H03L1/02—Stabilisation of generator output against variations of physical values, e.g. power supply against variations of temperature only
- H03L1/022—Stabilisation of generator output against variations of physical values, e.g. power supply against variations of temperature only by indirect stabilisation, i.e. by generating an electrical correction signal which is a function of the temperature
- H03L1/026—Stabilisation of generator output against variations of physical values, e.g. power supply against variations of temperature only by indirect stabilisation, i.e. by generating an electrical correction signal which is a function of the temperature by using a memory for digitally storing correction values
Abstract
The application relates to the field of clock compensation, and provides a temperature frequency offset compensation method and a system, wherein the method comprises the following steps: generating a temperature compensation database based on sample data, wherein the temperature compensation database comprises a plurality of reference characteristic data and temperature frequency offset compensation data sets which are in one-to-one correspondence with the plurality of reference characteristic data; detecting actual characteristic data of a target clock circuit, and inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data; and carrying out temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit. The method and the device ensure that the enrichment of the temperature compensation database can realize the improvement of the precision of crystal oscillator compensation under the condition of measuring a small amount of temperature point frequency offset data.
Description
Technical Field
The application relates to the field of clock crystal oscillator compensation, in particular to a temperature frequency offset compensation method and a system.
Background
The main frequency characteristics of a crystal oscillator depend on its internal crystal unit. The characteristics of the crystal unit depend on the cutting process, and the temperature frequency characteristics, the resonance frequency range and various parameters of the equivalent model are different in different cutting modes. There are three types of vigilant oscillators currently in the market: tuning Fork (tuning Fork), the frequency is mainly KHz level; AT-Cut type, the frequency is mainly MHz grade; the acoustic surface modes have frequencies from hundred MHz to GHz. In the actual production process of the crystal oscillator, the cut angles of the crystals in each time cannot be identical in percentage due to process deviation, so that the temperature and frequency characteristics of crystal oscillator products on the same production line are obviously different. In order to solve the problem that the crystal frequency changes along with temperature and process, the crystal product for the clock chip needs to quickly and efficiently perform temperature compensation on each quartz crystal, so that the output frequency of the same batch of products can be kept within a certain precision requirement range under a wide temperature condition.
The current frequency offset compensation method of the clock crystal oscillator is mainly based on curve fitting, for example, the real temperature frequency offset compensation method based on orthogonal least square method curve fitting provided by the invention with publication number CN 115904000A, if the accuracy of curve fitting is required to be improved, the frequency offset data at a plurality of temperatures need to be measured for high-order fitting, but the time cost is larger when the frequency offset data at a plurality of temperatures are measured, and the method does not meet the expectations of mass production; it is apparent that in the case where the number of detection of the temperature point frequency offset data is reduced, the detection accuracy is lowered.
Disclosure of Invention
In order to improve the precision of crystal oscillator compensation under the condition of measuring a small amount of temperature point frequency offset data, the application provides a temperature frequency offset compensation method and a system.
The temperature frequency offset compensation method and the system provided by the application adopt the following technical scheme:
a temperature frequency offset compensation method for a crystal oscillator based clock circuit, the method comprising:
generating a temperature compensation database based on sample data, wherein the temperature compensation database comprises a plurality of reference characteristic data and temperature frequency offset compensation data sets which are in one-to-one correspondence with the plurality of reference characteristic data;
Detecting actual characteristic data of a target clock circuit, and inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data; and carrying out temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit.
By adopting the technical scheme, a plurality of datum feature data and temperature frequency offset compensation data sets corresponding to the datum feature data one by one are set in the temperature compensation database, target datum feature data with highest similarity is found according to actual feature data, and clock crystal oscillator compensation is realized according to actual working temperature according to the target frequency offset compensation data set corresponding to the target datum feature data.
According to the method, the target reference characteristic data with the highest similarity degree can be locked through the actual characteristic data, implementation is more convenient, the quantity dependence on the state data in the actual characteristic data is low, and even if only two groups of state data in the actual characteristic data are still target reference characteristic data corresponding to the actual characteristic data, the clock correction can be realized.
Optionally, the sample data includes a plurality of measured temperature-frequency offset data pairs associated with a preset number of sample clock circuits, and the generating the temperature compensation database based on the sample data includes:
classifying the plurality of measured temperature-frequency offset data pairs based on the proximity relation of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data respectively associated with each category and a temperature frequency offset compensation data set in one-to-one correspondence with the plurality of reference feature data.
By the technical scheme, a plurality of sample clock circuits are set in the sample data, each sample clock circuit is provided with a set actually measured temperature-frequency offset data pair, wherein the actually measured temperature-frequency offset data pair represents the frequency offset characteristic of the sample clock circuit generated along with the change of temperature. Classifying the measured temperature-frequency offset data pairs based on the proximity relation of the measured temperature-frequency offset data pairs in a two-dimensional space to obtain a plurality of reference characteristic data respectively associated with each category and a temperature frequency offset compensation data set corresponding to the reference characteristic data one by one.
Optionally, the reference characteristic data includes a plurality of temperature values and frequency offset values corresponding to the plurality of temperature values one to one.
Optionally, classifying the plurality of measured temperature-frequency offset data pairs based on the proximity relation of the plurality of measured temperature-frequency offset data pairs in the two-dimensional space to determine a plurality of reference feature data respectively associated with each category, including: sorting the plurality of actually measured temperature-frequency offset data pairs according to preset characteristic dimensions to obtain a plurality of characteristic subsets corresponding to the characteristic dimensions;
calculating the shortest class spacing corresponding to each feature subset; taking the maximum value of all the shortest class distances under the same characteristic dimension as a reference class distance;
when the reference class spacing is larger than or equal to a preset minimum tolerance class spacing, taking a feature subset corresponding to the reference class spacing as an optimal feature subset, and taking a feature dimension corresponding to the reference class spacing as an optimal feature dimension;
establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
According to the technical scheme, firstly, temperature-frequency offset data pairs in a temperature compensation database are sorted according to feature dimensions to obtain feature subsets corresponding to the feature dimensions, discrimination between actually measured temperature-frequency offset data pairs is measured through shortest class spacing, when a sample feature space is established, the shortest class spacing is selected to serve as the optimal feature dimension and feature subset for establishing the sample feature space, discrimination of other actually measured temperature-frequency offset data pairs in the sample feature space is guaranteed, all actually measured temperature-frequency offset data pairs are projected into the sample feature space to obtain reference feature data of each class, and therefore the actual feature data of a target clock circuit can find the nearest reference feature data in the sample space to serve as target reference feature data.
Optionally, classifying the plurality of measured temperature-frequency offset data pairs based on the proximity relation of the plurality of measured temperature-frequency offset data pairs in the two-dimensional space to determine a plurality of reference feature data respectively associated with each category, including: sorting the plurality of actually measured temperature-frequency offset data pairs according to preset characteristic dimensions to obtain a plurality of characteristic subsets corresponding to the characteristic dimensions;
calculating the shortest class interval corresponding to each feature subset under the current feature dimension, and taking the maximum value of the shortest class interval as a reference class interval;
when the reference class spacing corresponding to the current feature dimension is smaller than the preset minimum tolerance class spacing, increasing the unit feature dimension of the current feature dimension to obtain the next feature dimension, updating the next feature dimension to be a new current feature dimension, calculating the shortest class spacing corresponding to each feature subset under the new current feature dimension, and taking the maximum value of the shortest class spacing as the reference class spacing;
until the corresponding reference class spacing under the current feature dimension is larger than or equal to the preset minimum tolerance class spacing, taking the feature subset corresponding to the reference class spacing as an optimal feature subset, and taking the current feature dimension as an optimal feature dimension;
Establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
By adopting the technical scheme, the temperature-frequency offset data in the temperature compensation database are firstly sorted according to the characteristic dimensions, and the characteristic subsets corresponding to the characteristic dimensions are obtained. When a sample space is set, directly stopping operation when the reference class spacing is larger than or equal to the minimum class spacing under the current feature dimension, directly taking the current feature dimension as the optimal feature dimension, taking the current feature subset as the optimal feature subset, and establishing a sample feature space according to the current feature dimension; and directly stopping operation when the reference class spacing is smaller than the minimum tolerance class spacing under the current feature dimension, directly switching to the next feature dimension, calculating the reference class spacing of the feature subset corresponding to the next feature dimension, and re-comparing the reference class spacing and the minimum tolerance class spacing.
Optionally, the minimum tolerated class spacing is associated with a feature dimension.
By adopting the technical scheme, the setting of the minimum tolerance class interval can be set by the staff. The minimum tolerance class spacing can be a uniform value or a plurality of values can be set, and each value corresponds to one characteristic dimension, so that the directional adjustment of the characteristic dimension is realized.
Optionally, the sorting the plurality of actually measured temperature-frequency offset data pairs according to a preset feature dimension to obtain a plurality of feature subsets corresponding to the feature dimension includes:
obtaining a value range of the characteristic dimension according to the number of actually measured temperature-frequency offset data pairs in the sample data;
traversing the feature dimension according to the value range of the feature dimension, extracting a corresponding number of actually measured temperature-frequency offset data pairs from sample data to form a feature subset corresponding to the feature dimension, wherein a plurality of feature subsets corresponding to the same feature dimension are different from each other.
By adopting the technical scheme, for the induction process of the characteristic dimension and the characteristic subset, the value range of the characteristic dimension is determined according to the number of the actually measured temperature-frequency offset data pairs, so that various characteristic dimensions can be obtained. And traversing the feature dimension according to the value range of the feature dimension, extracting a corresponding number of actually measured temperature-frequency offset data pairs from the temperature compensation database according to the feature dimension, and taking the selected actually measured temperature-frequency offset data pairs as a feature subset, wherein a plurality of feature subsets corresponding to the same feature dimension are mutually different. For example, there are 150 measured temperature-frequency offset data pairs, and the characteristic dimension M has a value of (1, 150 ]When the value of the characteristic dimension M is 2, the arrangement combination of the measured temperature-frequency offset data pairs extracted from the sample data comprisesFor this feature dimension m=2, the number of feature subsets formed is +.>
Optionally, querying the temperature compensation database to determine target baseline characteristic data that matches the actual characteristic data includes:
calculating a distance value between a projection point of the actual feature data in the sample feature space and the reference feature data of each category;
and when the distance value is smaller than a preset value, taking the reference characteristic data corresponding to the distance value as target reference characteristic data.
Optionally, performing temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and an actual working temperature of the clock circuit, including:
and according to the actual working temperature of the target clock circuit, finding out corresponding target frequency offset compensation data from the target temperature frequency offset compensation data set, and writing the corresponding target frequency offset compensation data into a compensation register to perform temperature frequency offset compensation on the clock circuit.
By adopting the technical scheme, because the actually measured temperature-frequency offset data pair comprises the sample temperature value and the sample compensation value, namely the value of the sample compensation value changes along with the temperature change, the corresponding target compensation value can be found according to the actual working temperature, and then the target compensation value is written into the compensation register to be used for adjusting the frequency offset of the real-time clock signal.
The invention also provides a temperature frequency offset compensation system, which comprises:
a database generating unit, configured to generate a temperature compensation database based on sample data, where the temperature compensation database includes a plurality of reference feature data and a temperature frequency offset compensation data set corresponding to the plurality of reference feature data one-to-one;
a feature acquisition unit for detecting actual feature data of the target clock circuit;
the processing unit is used for inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data;
and the temperature frequency offset compensation unit is used for carrying out temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit.
By adopting the technical scheme, a plurality of datum feature data and temperature frequency offset compensation data sets corresponding to the datum feature data one by one are set in the temperature compensation database, target datum feature data with highest similarity is found according to actual feature data, and clock crystal oscillator compensation is realized according to actual working temperature according to the target frequency offset compensation data set corresponding to the target datum feature data.
According to the method, the target reference characteristic data with the highest similarity degree can be locked through the actual characteristic data, implementation is more convenient, the quantity dependence on the state data in the actual characteristic data is low, and even if only two groups of state data in the actual characteristic data are still target reference characteristic data corresponding to the actual characteristic data, the clock correction can be realized.
In summary, the present application includes at least one of the following beneficial technical effects:
1. the accuracy of the accuracy adjustment in the scheme depends on the quantity of the reference characteristic data and the temperature frequency offset compensation data set which are correlated with each other in the temperature compensation database, so that the dependence on the actual characteristic data is not high, and the target frequency offset compensation data set is obtained and is injected into a compensation register to realize the correction of the clock.
2. According to the real characteristic data of the target clock circuit, target reference characteristic data and a target frequency offset compensation data set are found from the compensation database, according to the target reference characteristic data with the highest similarity between the real characteristic data and the target reference characteristic data, and according to the target frequency offset compensation data set corresponding to the target reference characteristic data, clock crystal oscillator compensation is realized through a compensation register according to the real working temperature, and the running speed is faster.
3. The method and the device generate sample data in a temperature compensation database based on the sample data, establish a sample feature space according to a plurality of actually measured temperature-frequency offset data pairs, and land other actually measured temperature-frequency offset data pairs in the sample feature space so as to increase the degree of distinction between the actually measured temperature-frequency offset data pairs, so that the actual feature data can find out target reference feature data which is most similar to the actual feature data through a distance value, and accuracy is improved.
Drawings
Fig. 1 is a schematic flow chart of a temperature frequency offset compensation method according to a first embodiment of the present invention;
FIG. 2 is a schematic flow chart of step 101 in FIG. 1;
FIG. 3 is a schematic diagram of another flow chart of step 101 in FIG. 1;
FIG. 4 is a schematic flow chart of step 102 in FIG. 1;
fig. 5 is a schematic structural diagram of a temperature frequency offset compensation system according to a second embodiment of the present invention.
Detailed Description
The present application is described in further detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application.
In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the inventive concepts. As part of this specification, some of the drawings of the present disclosure represent structures and devices in block diagram form in order to avoid obscuring the principles of the disclosure. In the interest of clarity, not all features of an actual implementation are necessarily described. Reference in the present disclosure to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment, and multiple references to "one embodiment" or "an embodiment" should not be understood as necessarily all referring to the same embodiment.
The terms "a," "an," and "the" are not intended to refer to a singular entity, but rather include the general class of which a particular example may be used for illustration, unless clearly defined. Thus, the use of the terms "a" or "an" may mean any number of at least one, including "one", "one or more", "at least one", and "one or more than one". The term "or" means any of the alternatives and any combination of alternatives, including all alternatives, unless alternatives are explicitly indicated as mutually exclusive. The phrase "at least one of" when combined with a list of items refers to a single item in the list or any combination of items in the list. The phrase does not require all of the listed items unless specifically so defined.
First embodiment:
a first embodiment of the present invention provides a temperature frequency offset compensation method for a clock circuit based on a crystal oscillator, the method including: generating a temperature compensation database based on sample data, wherein the temperature compensation database comprises a plurality of reference characteristic data and temperature frequency offset compensation data sets which are in one-to-one correspondence with the plurality of reference characteristic data; detecting actual characteristic data of a target clock circuit, and inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data; and carrying out temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit.
In the scheme, a temperature compensation database sets a plurality of datum feature data and temperature frequency offset compensation data sets corresponding to the datum feature data one by one.
In the scheme, corresponding matched target reference characteristic data is found in a compensation database according to actual characteristic data of a target clock circuit, and a corresponding target temperature frequency offset compensation data set is found according to target reference characteristic, namely, in the scheme, corresponding reference characteristic data is found according to a matching rule and is used as target temperature reference characteristic data according to actual characteristic data, and the temperature frequency offset compensation data set corresponding to the target reference characteristic data is used as a target temperature frequency offset compensation data set. And then, finding a corresponding frequency offset compensation value from the target temperature frequency offset compensation data set according to the actual working temperature, and further carrying out temperature frequency offset compensation on a target clock circuit according to the frequency offset compensation value so as to realize clock correction.
The implementation details of the temperature frequency offset compensation method of the present embodiment are specifically described below, and the following details are provided only for facilitating understanding, but are not required to be provided in the present embodiment, and a specific flow of the present embodiment is shown in fig. 1, and includes the following steps:
Step 101, generating a temperature compensation database based on sample data, wherein the temperature compensation database comprises a plurality of reference characteristic data and temperature frequency offset compensation data sets corresponding to the reference characteristic data one by one.
Specifically, the temperature compensation database comprises a plurality of reference characteristic databases and a temperature frequency offset compensation database corresponding to the basic characteristic databases. The sample data is a series of data detected and configured by the sample clock circuit in advance, such as frequency deviation detected by the sample clock circuit along with temperature change and corresponding frequency deviation compensation data manually set for the frequency deviation of the sample clock circuit. The temperature frequency offset data set can be a discrete point set of temperature-frequency offset compensation data pairs generated according to sample data, a temperature frequency offset compensation curve generated by fitting the temperature-frequency offset compensation data pairs generated according to the sample data, or a polynomial representing the rule of the temperature-frequency offset data pairs generated according to the sample data.
In some examples, the sample data includes a plurality of measured temperature-frequency offset data pairs associated with a preset number of sample clock circuits, and the generating the temperature compensation database based on the sample data includes:
Classifying the plurality of measured temperature-frequency offset data pairs based on the proximity relation of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data respectively associated with each category and a temperature frequency offset compensation data set in one-to-one correspondence with the plurality of reference feature data.
In the specific implementation of the example, a plurality of sample clock circuits are set in sample data, each sample clock circuit is provided with a set actually measured temperature-frequency offset data-frequency offset compensation data pair, the actually measured temperature-frequency offset data pair comprises a sample actually measured temperature value and a sample actually measured frequency offset value, and the actually measured temperature-frequency offset compensation data pair comprises a sample actually measured temperature value and a frequency offset compensation value. The measured temperature-frequency offset data pair is usually obtained by directly measuring frequency offset values of a sample clock circuit at different temperatures by a worker, and is used for setting frequency offset compensation values of the sample clock circuit at different temperatures by the worker. And, the measured temperature-frequency offset data pair represents the frequency offset characteristic of the sample clock circuit generated by the change of temperature.
And classifying and sorting the measured temperature-frequency offset data pairs based on the proximity relation of the measured temperature-frequency offset data pairs in a two-dimensional space to obtain a plurality of reference characteristic data respectively associated with each category and a temperature frequency offset compensation data set corresponding to the reference characteristic data one by one. The temperature frequency offset compensation data set comprises a plurality of actually measured temperature-frequency offset compensation data pairs.
Further, the reference characteristic data includes a plurality of temperature values and frequency offset values corresponding to the plurality of temperature values one by one.
In a further example, as shown in fig. 2, classifying the plurality of measured temperature-frequency offset data pairs based on proximity relations of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data associated with each category, respectively, includes:
s101, sorting the plurality of measured temperature-frequency offset data pairs according to preset feature dimensions to obtain a plurality of feature subsets corresponding to the feature dimensions;
s102, calculating the shortest class spacing corresponding to each feature subset; taking the maximum value of all the shortest class distances under the same characteristic dimension as a reference class distance;
S103, when the reference class spacing is larger than or equal to a preset minimum tolerance class spacing, taking a feature subset corresponding to the reference class spacing as an optimal feature subset, and taking a feature dimension corresponding to the reference class spacing as an optimal feature dimension;
s104, establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
s105, projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
By adopting the technical scheme, the characteristic dimension M is set, a plurality of actually measured temperature-frequency offset data pairs in the temperature compensation database are induced according to each characteristic dimension M, a plurality of characteristic subsets corresponding to the plurality of actually measured temperature-frequency offset data pairs are obtained, and the characteristic subsets are marked as T0 and T1 … … Tn; then, class spacing among a plurality of actually measured temperature-frequency offset data pairs is calculated, and the minimum value of the class spacing is taken as the shortest class spacing of the feature subset and is respectively marked as d0 and d1 … … dn; taking the maximum dx of the shortest class interval of the elements in the same characteristic dimension as a reference class interval; comparing the reference class spacing dx with the preset minimum tolerance class spacing d under all feature dimensions, taking the feature subset with dx not less than d as an optimal feature subset, and taking the feature dimension corresponding to the reference class spacing dx as an optimal feature dimension.
In this case, there are a plurality of best feature dimensions and best feature subsets corresponding to each other in S103. In executing step S104, a sample feature space is created by directly randomly selecting or selecting a set of best feature dimensions and best feature subsets according to user instructions. In order to facilitate the selection of a set of optimal feature dimensions and an optimal feature subset according to the user instruction in S104, after step S103, the obtained sets of optimal feature dimensions and optimal feature subsets corresponding to each other are displayed, so that the user instruction for selecting the optimal feature dimensions and the optimal feature subsets is generated by means of user input or selection. In the example, multiple dimensions are adopted to obtain feature subsets corresponding to feature dimensions, the degree of distinction between sample features is measured through the shortest class spacing, when a sample feature space is established, the shortest class spacing is selected to be qualified as the optimal feature dimension and feature subset for establishing the sample feature space, and the distinction of other sample feature data in the sample feature space is ensured, so that the actual feature data can find the closest sample feature data in the sample space as target reference feature data.
Firstly, sorting temperature-frequency offset data pairs in a temperature compensation database according to feature dimensions to obtain feature subsets corresponding to the feature dimensions, and measuring the distinguishing degree between actually measured temperature-frequency offset data pairs through shortest class spacing.
In a further example, classifying the plurality of measured temperature-frequency offset data pairs based on their proximity in two-dimensional space to determine a plurality of reference feature data associated with each respective category, as shown in fig. 3, includes:
s211, sorting the measured temperature-frequency offset data pairs according to preset feature dimensions to obtain feature subsets corresponding to the feature dimensions;
s112, calculating the shortest class spacing corresponding to each feature subset under the current feature dimension, and taking the maximum value of the shortest class spacing as a reference class spacing;
s113, judging whether the reference class spacing corresponding to the current feature dimension is smaller than the preset minimum tolerance class spacing or not, if yes, executing S114, and if not, executing S115;
s114, adding a unit feature dimension to the current feature dimension to obtain a next feature dimension, updating the next feature dimension to be a new current feature dimension, and returning to S112;
s115, taking the feature subset corresponding to the reference class distance as an optimal feature subset, taking the current feature dimension as an optimal feature dimension, and executing S116;
s116, establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
S117, projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
Therefore, in S111, sample feature data in the temperature compensation database is generalized according to each feature dimension M, so as to obtain a plurality of feature subsets corresponding to each feature dimension, which are denoted as T0, T1 … … Tn; s212, calculating class intervals among elements in each feature subset, wherein the minimum value of the class intervals is taken as the shortest class interval of the feature subset and is respectively marked as d0 and d1 … … dn; according to the knowledge of S112, S113, S114 and S115, when the reference class distance dx corresponding to the current feature dimension M is smaller than the preset minimum tolerance class distance d, adding a unit feature dimension to the current feature dimension to obtain a next feature dimension, updating the next feature dimension to a new current feature dimension, namely m=m+1, calculating the shortest class distance corresponding to each feature subset under the new current feature dimension M, and taking the maximum value of the shortest class distance as the reference class distance dx, namely returning to step S112 to perform the cycle of "S112-S113-S114-S112"; and taking the feature subset corresponding to the reference class spacing as an optimal feature subset when the reference class spacing corresponding to the current feature dimension is larger than or equal to the preset minimum tolerance class spacing, and taking the current feature dimension as the optimal feature dimension.
When a sample space is set, directly stopping operation when the reference class distance dx is larger than or equal to the minimum class distance d under the current feature dimension, directly taking the current feature dimension M as the optimal feature dimension, taking the current feature subset as the optimal feature subset, and establishing a sample feature space according to the current feature dimension M; and directly stopping operation when the reference class spacing is smaller than the minimum tolerance class spacing under the current feature dimension, directly switching to the next feature dimension, calculating the reference class spacing of the feature subset corresponding to the next feature dimension, and re-comparing the reference class spacing and the minimum tolerance class spacing. Compared with the previous example, the method and the device have the advantages that after the current feature dimensions are sequentially increased, operation is stopped after a group of optimal feature dimensions and feature subsets are obtained, data calculation pressure of an instantaneous computer is reduced, running time is shortened, and overall efficiency is improved.
Therefore, in the example, firstly, the temperature-frequency offset data in the temperature compensation database are sorted according to the feature dimensions, so that feature subsets corresponding to the feature dimensions are obtained. When a sample space is set, directly stopping operation when the reference class spacing is larger than or equal to the minimum class spacing under the current feature dimension, directly taking the current feature dimension as the optimal feature dimension, taking the current feature subset as the optimal feature subset, and establishing a sample feature space according to the current feature dimension; and directly stopping operation when the reference class spacing is smaller than the minimum tolerance class spacing under the current feature dimension, directly switching to the next feature dimension, calculating the reference class spacing of the feature subset corresponding to the next feature dimension, and re-comparing the reference class spacing and the minimum tolerance class spacing.
And a plurality of actually measured temperature-frequency offset data pairs are projected into the sample feature space, so that the difference between the projections of the plurality of actually measured temperature-frequency offset data pairs in the sample feature space is increased, the effect of improving the distinction degree among the plurality of actually measured temperature-frequency offset data pairs is realized, so that the actual feature data can conveniently find the reference feature data which is most similar to the actual feature data as target reference feature data, and the current crystal oscillator to be measured is subjected to clock correction according to the target frequency offset compensation data set corresponding to the target reference feature data.
Further example, the minimum tolerated class spacing is associated with a feature dimension.
The setting of the minimum tolerated class spacing d in this example may be set by the staff. The minimum tolerance class distance d can be a uniform value or a plurality of values can be set, and each value corresponds to one characteristic dimension M, so that the directional adjustment of the characteristic dimension is realized.
In a further example, in S111 or S101, the sorting the plurality of measured temperature-frequency offset data pairs according to a preset feature dimension to obtain a plurality of feature subsets corresponding to the feature dimension includes:
obtaining a value range of the characteristic dimension according to the number of actually measured temperature-frequency offset data pairs in the sample data;
Traversing the feature dimension according to the value range of the feature dimension, extracting a corresponding number of actually measured temperature-frequency offset data pairs from sample data to form a feature subset corresponding to the feature dimension, wherein a plurality of feature subsets corresponding to the same feature dimension are different from each other.
In this example, for the induction process of the feature dimension and the feature subset, the value range of the feature dimension is determined according to the number of measured temperature-frequency offset data pairs, so that multiple feature dimensions can be obtained. And traversing the feature dimension according to the value range of the feature dimension, extracting a corresponding number of actually measured temperature-frequency offset data pairs from the temperature compensation database according to the feature dimension, and taking the selected actually measured temperature-frequency offset data pairs as a feature subset, wherein a plurality of feature subsets corresponding to the same feature dimension are mutually different. For example, there are 150 measured temperature-frequency offset data pairs, and the characteristic dimension M has a value of (1, 150]When the value of the characteristic dimension M is 2, the arrangement combination of the measured temperature-frequency offset data pairs extracted from the sample data comprisesFor this feature dimension m=2, the number of feature subsets formed is +.>
Step 102, detecting actual characteristic data of a target clock circuit, and querying the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data.
Specifically, the frequency deviation values of the clock circuit (namely, the target clock circuit) of the crystal oscillator to be measured at different temperatures are measured, so that one or more sets of state data are obtained, and the one or more sets of state data are used as actual characteristic data. The actual characteristic data represents the frequency deviation condition of the clock circuit of the crystal oscillator to be measured at different temperatures, and the state data corresponding to the actual characteristic data usually measure at least two groups of state data in order to ensure the measurement accuracy. And inquiring a temperature compensation database according to the actual characteristic data so as to determine target reference characteristic data matched with the temperature compensation database, finding a corresponding target temperature frequency offset compensation data set according to the target reference characteristic data, and then carrying out temperature frequency offset compensation on a clock circuit of the crystal oscillator to be detected according to the target temperature frequency offset data set.
In some examples, querying the temperature compensation database to determine target baseline characteristic data that matches the actual characteristic data, as shown in fig. 4, includes:
s2-1, calculating distance values between projection points of the actual characteristic data in the sample characteristic space and the reference characteristic data of each category;
S2-2, when the distance value is smaller than a preset value, taking the reference characteristic data corresponding to the distance value as target reference characteristic data.
Specifically, in S117, the plurality of measured temperature-frequency offset data pairs are projected into the sample feature space to obtain reference feature data of each category; the preset value in S2-2 is preset by a worker. It is intended to find the closest reference feature data after the actual feature data falls in the sample space.
And 103, performing temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit.
In some examples, performing temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation dataset and an actual operating temperature of the clock circuit includes:
and according to the actual working temperature of the target clock circuit, finding out corresponding target frequency offset compensation data from the target temperature frequency offset compensation data set, and writing the corresponding target frequency offset compensation data into a compensation register to perform temperature frequency offset compensation on the target clock circuit.
Specifically, according to the actual working temperature of the target clock circuit, the target frequency offset compensation data (target compensation value) corresponding to the actual working temperature is found by combining the temperature-compensation value correspondence rule in the target frequency offset compensation data set, and the target frequency offset compensation data (target compensation value) is written into a compensation register to be used for adjusting the frequency offset of the real-time clock signal.
The method realizes real-time adjustment of target frequency offset compensation data (target compensation value) according to the environment where the target clock circuit is located, so as to ensure that the clock circuit of each crystal oscillator is subjected to temperature compensation quickly and efficiently, and the output frequency of the same batch of crystal oscillators can be kept within a certain precision requirement range under the wide temperature condition.
Typically the actual operating temperature is sensed by a temperature sensor in the crystal oscillator. Of course, the actual working temperature can also be obtained by other channels, for example, the actual working temperature can be obtained by transmission after detection by other devices connected with the device implemented by the scheme, or can be obtained by A/D conversion of a temperature detection signal simulated by an maintainer, and the like.
The above steps of the various methods are divided, for clarity of description, and may be combined into one step or split into multiple steps when implemented, so long as they include the same logic relationship, and all the steps are within the scope of protection of this patent, and adding insignificant modifications or introducing insignificant designs to the algorithm or the process, but not changing the core designs of the algorithm and the process are within the scope of protection of this patent.
Second embodiment:
a second embodiment of the present invention provides a clock crystal compensation system 50, as shown in fig. 5, comprising:
a database generating unit 501, configured to generate a temperature compensation database based on sample data, where the temperature compensation database includes a plurality of reference feature data and a temperature frequency offset compensation data set corresponding to the plurality of reference feature data in a one-to-one manner;
a storage unit 502 for storing the temperature compensation database
A feature acquisition unit 503 for detecting actual feature data of the target clock circuit;
a processing unit 504, configured to query the temperature compensation database to determine target reference feature data that matches the actual feature data, and a target temperature frequency offset compensation dataset corresponding to the target reference feature data;
and the temperature frequency offset compensation unit 505 is configured to perform temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit.
In the specific implementation, a plurality of reference characteristic data and temperature frequency offset compensation data sets corresponding to the reference characteristic data are set in a temperature compensation database, target reference characteristic data with highest similarity is found according to actual characteristic data, and clock crystal oscillator compensation is realized according to actual working temperature according to the target frequency offset compensation data set corresponding to the target reference characteristic data.
According to the method, the target reference characteristic data with the highest similarity degree can be locked through the actual characteristic data, implementation is more convenient, the quantity dependence on the state data in the actual characteristic data is low, and even if only two groups of state data in the actual characteristic data are still target reference characteristic data corresponding to the actual characteristic data, the clock correction can be realized. Wherein the actual operating temperature is typically sensed by a temperature sensor. Of course, the current temperature value may be obtained by other channels, for example, obtained by detecting and transmitting other devices connected with the device executed by the scheme, or obtained by converting a temperature detection signal simulated by an maintainer through A/D, etc.
Therefore, the target reference characteristic data with the highest similarity degree can be locked through the actual characteristic data, the quantity dependence of the state data in the actual characteristic data is not high, even if only two groups of state data still can judge the target reference characteristic data corresponding to the actual characteristic data, so that a target frequency offset compensation data set is obtained and is injected into a compensation register to realize the correction of a clock.
Therefore, the accuracy of the accuracy adjustment is more dependent on the number of the sample characteristic data and the sample compensation data which are correlated with each other in the sample database, but not the state data group quantity of the actual characteristic data, so that the dependence on the state data in the actual characteristic data is not high, even if only two groups of state data are still target reference characteristic data which can be judged to correspond to the actual characteristic data, so that a target frequency offset compensation data set is obtained and is injected into a compensation register to realize the correction of the clock. The method realizes the quick and efficient temperature compensation of each crystal oscillator, and ensures that the output frequency of the crystal oscillator products in the same batch can be kept within a certain precision requirement range under a wide temperature condition.
Other implementation details and working manners of the clock crystal oscillator compensation device disclosed in the present application are the same as or similar to the above-described webpage generation method capable of configuring rendering, and are not described herein in detail.
The foregoing are all preferred embodiments of the present application, and are not intended to limit the scope of the present application in any way, therefore: all equivalent changes in structure, shape and principle of this application should be covered in the protection scope of this application.
Claims (8)
1. A method of temperature frequency offset compensation for a crystal oscillator based clock circuit, the method comprising:
generating a temperature compensation database based on sample data, wherein the temperature compensation database comprises a plurality of reference characteristic data and temperature frequency offset compensation data sets which are in one-to-one correspondence with the plurality of reference characteristic data;
detecting actual characteristic data of a target clock circuit, and inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data;
performing temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit;
The sample data includes a plurality of measured temperature-frequency offset data pairs associated with a preset number of sample clock circuits, and the generating a temperature compensation database based on the sample data includes:
classifying the plurality of measured temperature-frequency offset data pairs based on proximity relations of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data respectively associated with each category and a temperature frequency offset compensation data set in one-to-one correspondence with the plurality of reference feature data;
classifying the plurality of measured temperature-frequency offset data pairs based on their proximity in two-dimensional space to determine a plurality of baseline characteristic data associated with each respective category, comprising:
sorting the plurality of actually measured temperature-frequency offset data pairs according to preset characteristic dimensions to obtain a plurality of characteristic subsets corresponding to the characteristic dimensions;
calculating the shortest class spacing corresponding to each feature subset; taking the maximum value of all the shortest class distances under the same characteristic dimension as a reference class distance;
when the reference class spacing is larger than or equal to a preset minimum tolerance class spacing, taking a feature subset corresponding to the reference class spacing as an optimal feature subset, and taking a feature dimension corresponding to the reference class spacing as an optimal feature dimension;
Establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
2. The method of temperature frequency offset compensation according to claim 1, wherein the reference characteristic data includes a plurality of temperature values and frequency offset values corresponding to the plurality of temperature values one to one.
3. The temperature frequency offset compensation method of claim 1, wherein the plurality of measured temperature-frequency offset data pairs are classified based on their proximity in two-dimensional space to determine a plurality of reference characteristic data associated with each respective class, further comprising:
sorting the plurality of actually measured temperature-frequency offset data pairs according to preset characteristic dimensions to obtain a plurality of characteristic subsets corresponding to the characteristic dimensions;
calculating the shortest class interval corresponding to each feature subset under the current feature dimension, and taking the maximum value of the shortest class interval as a reference class interval;
when the reference class spacing corresponding to the current feature dimension is smaller than the preset minimum tolerance class spacing, increasing the unit feature dimension of the current feature dimension to obtain the next feature dimension, updating the next feature dimension to be a new current feature dimension, calculating the shortest class spacing corresponding to each feature subset under the new current feature dimension, and taking the maximum value of the shortest class spacing as the reference class spacing;
Until the corresponding reference class spacing under the current feature dimension is larger than or equal to the preset minimum tolerance class spacing, taking the feature subset corresponding to the reference class spacing as an optimal feature subset, and taking the current feature dimension as an optimal feature dimension;
establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
4. A temperature frequency offset compensation method according to any one of claims 1 or 3, wherein the minimum tolerated class spacing is associated with a characteristic dimension.
5. A temperature frequency offset compensation method according to any one of claims 1 or 3, wherein the sorting the plurality of measured temperature-frequency offset data pairs according to a preset feature dimension to obtain a plurality of feature subsets corresponding to the feature dimension includes:
obtaining a value range of the characteristic dimension according to the number of actually measured temperature-frequency offset data pairs in the sample data;
traversing the feature dimension according to the value range of the feature dimension, extracting a corresponding number of actually measured temperature-frequency offset data pairs from sample data to form a feature subset corresponding to the feature dimension, wherein a plurality of feature subsets corresponding to the same feature dimension are different from each other.
6. The method of temperature frequency offset compensation according to claim 1, wherein said querying said temperature compensation database to determine target baseline characteristic data that matches said actual characteristic data comprises:
calculating a distance value between a projection point of the actual feature data in the sample feature space and the reference feature data of each category;
and when the distance value is smaller than a preset value, taking the reference characteristic data corresponding to the distance value as target reference characteristic data.
7. The method of temperature frequency offset compensation as set forth in claim 1 wherein performing temperature frequency offset compensation on the clock circuit based on the target temperature frequency offset compensation dataset and an actual operating temperature of the clock circuit comprises:
and according to the actual working temperature of the target clock circuit, finding out corresponding target frequency offset compensation data from the target temperature frequency offset compensation data set, and writing the corresponding target frequency offset compensation data into a compensation register to perform temperature frequency offset compensation on the clock circuit.
8. A clock crystal compensation system, comprising:
a database generating unit, configured to generate a temperature compensation database based on sample data, where the temperature compensation database includes a plurality of reference feature data and a temperature frequency offset compensation data set corresponding to the plurality of reference feature data one-to-one;
A feature acquisition unit for detecting actual feature data of the target clock circuit;
the processing unit is used for inquiring the temperature compensation database to determine target reference characteristic data matched with the actual characteristic data and a target temperature frequency offset compensation data set corresponding to the target reference characteristic data;
the temperature frequency offset compensation unit is used for performing temperature frequency offset compensation on the clock circuit according to the target temperature frequency offset compensation data set and the actual working temperature of the clock circuit;
the sample data includes a plurality of measured temperature-frequency offset data pairs associated with a preset number of sample clock circuits, and the generating a temperature compensation database based on the sample data includes: classifying the plurality of measured temperature-frequency offset data pairs based on proximity relations of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data respectively associated with each category and a temperature frequency offset compensation data set in one-to-one correspondence with the plurality of reference feature data;
wherein classifying the plurality of measured temperature-frequency offset data pairs based on proximity relations of the plurality of measured temperature-frequency offset data pairs in a two-dimensional space to determine a plurality of reference feature data associated with each category, respectively, comprises:
Sorting the plurality of actually measured temperature-frequency offset data pairs according to preset characteristic dimensions to obtain a plurality of characteristic subsets corresponding to the characteristic dimensions;
calculating the shortest class spacing corresponding to each feature subset; taking the maximum value of all the shortest class distances under the same characteristic dimension as a reference class distance;
when the reference class spacing is larger than or equal to a preset minimum tolerance class spacing, taking a feature subset corresponding to the reference class spacing as an optimal feature subset, and taking a feature dimension corresponding to the reference class spacing as an optimal feature dimension;
establishing a sample feature space according to the optimal feature subset and the optimal feature dimension;
projecting the measured temperature-frequency offset data pairs into the sample feature space to obtain reference feature data of each category.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311143984.4A CN117118431B (en) | 2023-09-06 | 2023-09-06 | Temperature frequency offset compensation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311143984.4A CN117118431B (en) | 2023-09-06 | 2023-09-06 | Temperature frequency offset compensation method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117118431A CN117118431A (en) | 2023-11-24 |
| CN117118431B true CN117118431B (en) | 2024-03-29 |
Family
ID=88805445
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311143984.4A Active CN117118431B (en) | 2023-09-06 | 2023-09-06 | Temperature frequency offset compensation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117118431B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102065040A (en) * | 2011-01-06 | 2011-05-18 | 意法·爱立信半导体（北京）有限公司 | Adjustment method of terminal frequency offset, terminal and TDD (Time Division Duplexing) system |
| CN114928355A (en) * | 2022-07-20 | 2022-08-19 | 广东大普通信技术股份有限公司 | Voltage compensation method, device and equipment based on crystal oscillator and storage medium |
| CN115133924A (en) * | 2022-08-10 | 2022-09-30 | 广东大普通信技术股份有限公司 | Frequency compensation method, device, equipment and storage medium based on crystal oscillator |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020005765A1 (en) * | 2000-03-17 | 2002-01-17 | William Ashley | Digital indirectly compensated crystal oscillators |
-
2023
- 2023-09-06 CN CN202311143984.4A patent/CN117118431B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102065040A (en) * | 2011-01-06 | 2011-05-18 | 意法·爱立信半导体（北京）有限公司 | Adjustment method of terminal frequency offset, terminal and TDD (Time Division Duplexing) system |
| CN114928355A (en) * | 2022-07-20 | 2022-08-19 | 广东大普通信技术股份有限公司 | Voltage compensation method, device and equipment based on crystal oscillator and storage medium |
| CN115133924A (en) * | 2022-08-10 | 2022-09-30 | 广东大普通信技术股份有限公司 | Frequency compensation method, device, equipment and storage medium based on crystal oscillator |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117118431A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Park et al. | G-CARE: A framework for performance benchmarking of cardinality estimation techniques for subgraph matching | |
| US8874548B2 (en) | Predicting query execution time | |
| Gao et al. | Efficient reverse top-k boolean spatial keyword queries on road networks | |
| Soliman et al. | Ranking with uncertain scoring functions: semantics and sensitivity measures | |
| Qiao et al. | Top-k nearest keyword search on large graphs | |
| US9218391B2 (en) | Scalable computation of data | |
| WO2017223125A1 (en) | Aggregate-query database system and processing | |
| Vlachou et al. | Skyline query processing over joins | |
| CN108920786B (en) | Section uncertainty analysis method based on Chebyshev polynomial fitting | |
| US8880511B2 (en) | Database query optimization and cost estimation | |
| CN102722554B (en) | Randomness weakening method of location-sensitive hash | |
| Khemmarat et al. | Fast top-k path-based relevance query on massive graphs | |
| CN117118431B (en) | Temperature frequency offset compensation method and system | |
| Christensen et al. | STORM: Spatio-temporal online reasoning and management of large spatio-temporal data | |
| Tran et al. | Conditioning and aggregating uncertain data streams: Going beyond expectations | |
| Ma et al. | An efficient approach to finding dense temporal subgraphs | |
| Wong et al. | Fast estimation of fractal dimension and correlation integral on stream data | |
| CN111079788A (en) | K-means clustering method based on density Canopy | |
| Mouratidis et al. | Tree-based partition querying: a methodology for computing medoids in large spatial datasets | |
| Tong et al. | Efficient probabilistic supergraph search over large uncertain graphs | |
| US8478762B2 (en) | Ranking system | |
| CN115406377A (en) | Method and device for establishing theoretical spectrum library and method and device for measuring morphological parameters | |
| US10176221B2 (en) | Data processing apparatus and prediction method of pattern frequency thereof | |
| Müller et al. | Memory-efficient key/foreign-key join size estimation via multiplicity and intersection size | |
| Guo et al. | Summarizing RDF graphs using node importance and query history |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |