WO2008038975A1 - Method for analyzing and predicting error generated in broadcasting system and apparatus therefor - Google Patents
Method for analyzing and predicting error generated in broadcasting system and apparatus therefor Download PDFInfo
- Publication number
- WO2008038975A1 WO2008038975A1 PCT/KR2007/004671 KR2007004671W WO2008038975A1 WO 2008038975 A1 WO2008038975 A1 WO 2008038975A1 KR 2007004671 W KR2007004671 W KR 2007004671W WO 2008038975 A1 WO2008038975 A1 WO 2008038975A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- error
- prediction model
- function
- model function
- prediction
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 31
- 230000002068 genetic effect Effects 0.000 claims abstract description 30
- 238000007405 data analysis Methods 0.000 claims abstract description 14
- 238000009795 derivation Methods 0.000 claims abstract description 14
- 238000012795 verification Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 117
- 238000012937 correction Methods 0.000 claims description 49
- 238000004364 calculation method Methods 0.000 claims description 18
- 238000012360 testing method Methods 0.000 claims description 15
- 238000000605 extraction Methods 0.000 claims description 10
- 230000003044 adaptive effect Effects 0.000 description 4
- 230000035772 mutation Effects 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000005457 optimization Methods 0.000 description 3
- 238000000611 regression analysis Methods 0.000 description 3
- 108700003861 Dominant Genes Proteins 0.000 description 2
- 238000013459 approach Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 210000000349 chromosome Anatomy 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000004904 shortening Methods 0.000 description 1
- 238000013179 statistical model Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N17/00—Diagnosis, testing or measuring for television systems or their details
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/24—Testing correct operation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/20—Arrangements for detecting or preventing errors in the information received using signal quality detector
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L2001/0092—Error control systems characterised by the topology of the transmission link
- H04L2001/0093—Point-to-multipoint
Definitions
- the present invention relates to a method and apparatus for analyzing and predicting errors generated in a broadcasting system, and more particularly to a method and apparatus for analyzing and predicting errors occurring in a broadcasting system, which analyzes errors periodically generated in the broadcasting system and predicts the occurrence dates of future errors, thereby preventing the errors from occurring.
- a genetic algorithm that was proposed by John Holland in 1975 is one of search techniques for finding solutions to optimization problems, etc. through models of biological evolutionary phenomena and global search heuristics.
- Such a genetic algorithm consists of an evaluation function (fitness function), a genetic operator, and a chromosome for expressing a solution to a problem.
- an optimal solution is obtained by gradually improving the chromosome.
- This hybrid genetic algorithm employs a scheme in which a solution is positioned near a local optimum through crossover/mutation, and the solution is guided to the local optimum by using the local optimization algorithm, thereby advantageously enhancing the fine adjustment capability of the genetic algorithm and shortening a convergence time.
- the genetic algorithm as mentioned above is applied to image processing, it is used for searching an image to be extracted from the background in the first step of feature extraction. That is, several pointers specified in the entire image are created as an initial population, the values of which are estimated by a fitness function. Subsequently, a given number of dominant genes are selected from among the initial population by means of a selection operation, and the values of the selected dominant genes are subjected to crossover and mutation operations, as a result of which a new population is created. Next, the created new population is reappraised through the fitness function, and a population with an optimal fitness value is selected by repeatedly performing the aforementioned processes. The finally selected population is used for feature extraction.
- a conventional prediction model uses a prediction model function, such as a regression analysis equation, as a statistical model, but such an analysis has a problem in that the reliability of prediction is significantly lowered when the randomness or variation of data is large.
- the conventional prediction model using a regression analysis equation has another problem in that it must use only a fixed prediction model function because of difficulties in applying weights to the prediction model function according to the passage of time and variations of error factors.
- the present invention has been made to solve the above-mentioned problems occurring in the prior art, and the present invention provides a method and apparatus for analyzing and predicting errors, which analyzes errors periodically generated in a broadcasting system and predicts the occurrence dates of future errors through a prediction model function using a genetic algorithm, thereby preventing the errors from occurring.
- the present invention provides a method and apparatus for analyzing and predicting errors, which enhances the reliability of error prediction by finding weights for error factors of a multiple regression equation used as a prediction model function by means of a genetic algorithm.
- the present invention provides an adaptive prediction method and apparatus by deriving error factors and a variable prediction model function adaptively to various system environments.
- a method of analyzing and predicting errors occurring in a broadcasting system including: an error data analysis step of analyzing error data for errors generated in the broadcasting system; a function derivation step of deriving a prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing the prediction model function by applying a genetic algorithm, which uses the error data generated for a given period of time, to the prediction model function derived in the function derivation step; and a prediction step of predicting a specific type of error occurring in the broadcasting system through the prediction model function completed in the function completion step.
- a method of analyzing and predicting errors occurring in a broadcasting system including: an error type definition step of defining error types based on error data for errors generated for a given period of time in the broadcasting system; an error data analysis step of analyzing the error data defined in the error type definition step; a function derivation step of deriving a basic prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing a prediction model function by applying a genetic algorithm, which uses the error data generated for the given period of time, to the basic prediction model function derived in the function derivation step; a prediction verification step of verifying pre-test prediction for a specific type of error generated for the given period of time in the broadcasting system through the prediction model function completed in the function completion step; a correction value calculation step of calculating a correction value by comparing a result value of the pre-test prediction for the error, which is tested for the given period of time
- a system for analyzing and predicting errors occurring in a broadcasting system including: an error type definition module for defining error types based on error data for errors generated for a given period of time in the broadcasting system; an error factor extraction module for extracting error factors by analyzing the error data defined in the error type definition module; a prediction model function module for deriving a basic prediction model function from the error factors extracted in the error factor extraction module, and completing a prediction model function by applying a genetic algorithm to the derived basic prediction model function; a prediction module for verifying a probability of predicting error occurrence through the prediction model function completed in the prediction model function module, and predicting a specific type of error through the prediction model function corrected in a function correction module; a correction value calculation module for calculating a correction value by comparing a result value of prediction tested in the prediction module with actually generated error data; and the function correction module for correcting the prediction model function by applying the correction value calculated in the correction value calculation module to
- the method and apparatus for analyzing and predicting errors occurring in a broadcasting system can prevent errors from occurring by analyzing errors periodically generated in the broadcasting system and predicting the occurrence dates of future errors.
- the present invention can recognize system errors in advance and maintain a stable transmission environment by presenting error prediction methodology beyond a conventional passive transmission monitoring function.
- the present invention can enhance the reliability of error prediction, complete a customized prediction model suitable for various broadcasting system environments, and provide an adaptive prediction model according to the passage of time and the variation of an error occurrence pattern by finding weights for error factors in a multiple regression equation through a genetic algorithm.
- FIG. 1 is a block diagram schematically illustrating an apparatus for analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention.
- FIG. 2 is a flowchart illustrating a method of analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention
- FIG. 1 schematically illustrates an apparatus for analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention.
- the apparatus for analyzing and predicting errors occurring in a broadcasting system includes an error type definition module 100, an error factor extraction module 110, a prediction model function module 120, a prediction verification module 130, a correction value calculation module 140, and a function correction module 150.
- the error type definition module 100 defines various errors that are generated for a given period of time in the broadcasting system. With regard to this, the error type definition module 100 may define a type for each error, or may combine several errors and define a type for the combined errors.
- the error factor extraction module 110 analyzes error data for error types defined in the error type definition module, and extracts error factors. With regard to this, it is preferred that statistically important error sources are extracted as the error factors from the error data by analyzing error sources that are periodically collected.
- the error factors include cycles according to the time windows, dates and frequencies of error occurrence, and may include an error source log generation pattern or system usage, such as CPU, memory and network overloads.
- the error factors are not fixed, and vary according to error type definition. With regard to this, it is preferred to exclude less contributive error factors from among various error factors, thereby simplifying a prediction model function and reducing the amount of calculations.
- the weight values E to E are preferably derived as optimal
- the prediction module 130 gives a pre-test prediction for the occurrence of a specific type of error occurring in the broadcasting system by using the prediction model function completed in the prediction model function module 120, provides the correction value calculation module 140 with a result value of the pre-test prediction, and finally measures and verifies a specific type of error for a given period of time by using the prediction model function that is corrected by the function correction module 150 based on an output value of the correction value calculation module 140.
- the correction value calculation module 140 compares a result value of pre-test prediction for errors occurring for a given period of time, which is provided by the prediction module 130, with the actually generated error data, and calculates a correction value for compensating for a difference between the result value and the actual error data.
- the function correction module 150 corrects the prediction model function by applying a correction value calculated in the correction value calculation module 140 to the prediction model function, and provides the prediction module 130 with the corrected prediction model function.
- the method of analyzing and predicting errors occurring in a broadcasting system includes an error type definition step S200, an error data analysis step S210, a function derivation step S220, a function completion step S230, a pre-test prediction step S240, a correction value calculation step S250, a function correction step S260, and a prediction step S270.
- error types are defined for various errors that have been generated for a given period of time (e.g., six months from January to June) in the broadcasting system. With regard to this, the error type may be defined for each error, or may be defined for several combined errors.
- error data analysis step S210 error data for error types defined in the error type definition step are analyzed. With regard to this, it is preferred that statistically important error sources are extracted as error factors from the error data by analyzing error sources that are periodically collected.
- a prediction model equation can be simplified by varying error factor extraction adaptively to circumstances, extracting factors corresponding to direct causes of errors as the error factors, and excluding less contributive error factors.
- a basic prediction model function is derived by extracting error factors from the error data analyzed in the error data analysis step S210.
- the error factors preferably use cycles according to the time windows, dates and frequencies of error occurrence, and may include an error source log generation pattern or system usage, such as CPU, memory and network overloads.
- the weight values E 1 to E n are preferably derived as optimal weights by means of a genetic algorithm. Also, since an error occurrence pattern may vary according to the passage of time when weight values are derived using a genetic algorithm, optimal weight values must also change. Therefore, it is preferred to use the genetic algorithm in order to complete an adaptive model function suited to the passage of time and the error occurrence pattern.
- a prediction model function is completed by applying a genetic algorithm, which uses error data generated for a given period of time (e.g., next five months from July t November), to the derived basic prediction model function derived in the function derivation step S220.
- the prediction model function completed in the function completion step S230 is applied to pre-test prediction for errors that have been generated for a given past period of time (e.g., next one month of December), and the prediction model function is verified in this way.
- a correction value is calculated by comparing a result value of the pre-test prediction, which is performed for a given period of time in the pre-test prediction step S240, with error data actually generated for that period of time.
- the prediction model function is corrected by applying the correction value calculated in the correction value calculation step to the prediction model function.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Quality & Reliability (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Medical Treatment And Welfare Office Work (AREA)
Abstract
Disclosed is a method and apparatus for analyzing and predicting errors occurring in a broadcasting system, which analyzes errors periodically generated in the broadcasting system and predicts the occurrence dates of future errors, thereby preventing the errors from occurring. The method includes an error data analysis step of analyzing error data for errors occurring in the broadcasting system, a function derivation step of deriving a prediction model function by extracting error factors from the error data analyzed in the error data analysis step, a function completion step of completing the prediction model function through a genetic algorithm using the error data generated for a given period of time, based on the prediction model function derived in the function derivation step, a prediction verification step of correcting a prediction error by verifying the completed prediction model function, and a prediction step of predicting a specific type of error occurring in the broadcasting system through the prediction model function completed in the function completion step and the prediction verification step.
Description
Description
METHOD FOR ANALYZING AND PREDICTING ERROR GENERATED IN BROADCASTING SYSTEM AND APPARATUS
THEREFOR
Technical Field
[1] The present invention relates to a method and apparatus for analyzing and predicting errors generated in a broadcasting system, and more particularly to a method and apparatus for analyzing and predicting errors occurring in a broadcasting system, which analyzes errors periodically generated in the broadcasting system and predicts the occurrence dates of future errors, thereby preventing the errors from occurring.
[2]
Background Art
[3] A genetic algorithm that was proposed by John Holland in 1975 is one of search techniques for finding solutions to optimization problems, etc. through models of biological evolutionary phenomena and global search heuristics.
[4] Such a genetic algorithm consists of an evaluation function (fitness function), a genetic operator, and a chromosome for expressing a solution to a problem. In the genetic algorithm, an optimal solution is obtained by gradually improving the chromosome.
[5] However, although the genetic algorithm allows search in a vast space, and can efficiently use features of yielded solutions, the speed of approach to a local optimum is generally slow because crossover and mutation operations applied to yield a solution are optionally performed.
[6] Thereupon, in order to complement the speed of approach to a local optimum, a hybrid genetic algorithm has been proposed, in which a local optimization algorithm is applied to a solution produced by crossover and mutation.
[7] This hybrid genetic algorithm employs a scheme in which a solution is positioned near a local optimum through crossover/mutation, and the solution is guided to the local optimum by using the local optimization algorithm, thereby advantageously enhancing the fine adjustment capability of the genetic algorithm and shortening a convergence time.
[8] When the genetic algorithm as mentioned above is applied to image processing, it is used for searching an image to be extracted from the background in the first step of feature extraction. That is, several pointers specified in the entire image are created as an initial population, the values of which are estimated by a fitness function. Subsequently, a given number of dominant genes are selected from among the initial
population by means of a selection operation, and the values of the selected dominant genes are subjected to crossover and mutation operations, as a result of which a new population is created. Next, the created new population is reappraised through the fitness function, and a population with an optimal fitness value is selected by repeatedly performing the aforementioned processes. The finally selected population is used for feature extraction.
[9] In addition, a conventional prediction model uses a prediction model function, such as a regression analysis equation, as a statistical model, but such an analysis has a problem in that the reliability of prediction is significantly lowered when the randomness or variation of data is large.
[10] Further, the conventional prediction model using a regression analysis equation has another problem in that it must use only a fixed prediction model function because of difficulties in applying weights to the prediction model function according to the passage of time and variations of error factors.
[11] Further, it is very difficult to predict errors only by a fixed regression analysis function in a broadcasting system.
[12]
Disclosure of Invention Technical Problem
[13] Accordingly, the present invention has been made to solve the above-mentioned problems occurring in the prior art, and the present invention provides a method and apparatus for analyzing and predicting errors, which analyzes errors periodically generated in a broadcasting system and predicts the occurrence dates of future errors through a prediction model function using a genetic algorithm, thereby preventing the errors from occurring.
[14] Further, the present invention provides a method and apparatus for analyzing and predicting errors, which enhances the reliability of error prediction by finding weights for error factors of a multiple regression equation used as a prediction model function by means of a genetic algorithm.
[15] Further, the present invention provides an adaptive prediction method and apparatus by deriving error factors and a variable prediction model function adaptively to various system environments.
[16]
Technical Solution
[17] In accordance with an aspect of the present invention, there is provided a method of analyzing and predicting errors occurring in a broadcasting system, the method including: an error data analysis step of analyzing error data for errors generated in the
broadcasting system; a function derivation step of deriving a prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing the prediction model function by applying a genetic algorithm, which uses the error data generated for a given period of time, to the prediction model function derived in the function derivation step; and a prediction step of predicting a specific type of error occurring in the broadcasting system through the prediction model function completed in the function completion step.
[18] In accordance with another aspect of the present invention, there is provided a method of analyzing and predicting errors occurring in a broadcasting system, the method including: an error type definition step of defining error types based on error data for errors generated for a given period of time in the broadcasting system; an error data analysis step of analyzing the error data defined in the error type definition step; a function derivation step of deriving a basic prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing a prediction model function by applying a genetic algorithm, which uses the error data generated for the given period of time, to the basic prediction model function derived in the function derivation step; a prediction verification step of verifying pre-test prediction for a specific type of error generated for the given period of time in the broadcasting system through the prediction model function completed in the function completion step; a correction value calculation step of calculating a correction value by comparing a result value of the pre-test prediction for the error, which is tested for the given period of time in the prediction verification step, with actually generated error data; a function correction step of correcting the prediction model function by applying the correction value calculated in the correction value calculation step to the prediction model function; and a prediction step of finally predicting the specific type of error occurring for the given period of time in the broadcasting system through the prediction model function corrected in the function correction step.
[19] In accordance with yet another aspect of the present invention, there is provided a system for analyzing and predicting errors occurring in a broadcasting system, the system including: an error type definition module for defining error types based on error data for errors generated for a given period of time in the broadcasting system; an error factor extraction module for extracting error factors by analyzing the error data defined in the error type definition module; a prediction model function module for deriving a basic prediction model function from the error factors extracted in the error factor extraction module, and completing a prediction model function by applying a genetic algorithm to the derived basic prediction model function; a prediction module for verifying a probability of predicting error occurrence through the prediction model
function completed in the prediction model function module, and predicting a specific type of error through the prediction model function corrected in a function correction module; a correction value calculation module for calculating a correction value by comparing a result value of prediction tested in the prediction module with actually generated error data; and the function correction module for correcting the prediction model function by applying the correction value calculated in the correction value calculation module to the prediction model function.
Advantageous Effects
[20] As noted from the forgoing, the method and apparatus for analyzing and predicting errors occurring in a broadcasting system can prevent errors from occurring by analyzing errors periodically generated in the broadcasting system and predicting the occurrence dates of future errors.
[21] Further, the present invention can recognize system errors in advance and maintain a stable transmission environment by presenting error prediction methodology beyond a conventional passive transmission monitoring function.
[22] Furthermore, the present invention can enhance the reliability of error prediction, complete a customized prediction model suitable for various broadcasting system environments, and provide an adaptive prediction model according to the passage of time and the variation of an error occurrence pattern by finding weights for error factors in a multiple regression equation through a genetic algorithm.
[23]
Brief Description of the Drawings
[24] The accompanying drawings are only for the purpose of illustrating preferred embodiments of the present invention, and promote the understanding of the present invention in connection with the following detailed description. Therefore, the present invention should not be construed as being limited to the illustrations in the accompanying drawings.
[25] The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:
[26] FIG. 1 is a block diagram schematically illustrating an apparatus for analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention; and
[27] FIG. 2 is a flowchart illustrating a method of analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention
[28]
[29] <Description of Reference Numerals>
[30] S210: error data analysis step
[31] S 220: function derivation step
[32] S230: function completion step
[33] S240: test prediction step
[34] S250: correction value calculation step
[35] S260: function correction step
[36] S270: prediction step
[37]
Mode for the Invention
[38] Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings.
[39] FIG. 1 schematically illustrates an apparatus for analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention.
[40] The apparatus for analyzing and predicting errors occurring in a broadcasting system includes an error type definition module 100, an error factor extraction module 110, a prediction model function module 120, a prediction verification module 130, a correction value calculation module 140, and a function correction module 150.
[41] The error type definition module 100 defines various errors that are generated for a given period of time in the broadcasting system. With regard to this, the error type definition module 100 may define a type for each error, or may combine several errors and define a type for the combined errors.
[42] The error factor extraction module 110 analyzes error data for error types defined in the error type definition module, and extracts error factors. With regard to this, it is preferred that statistically important error sources are extracted as the error factors from the error data by analyzing error sources that are periodically collected.
[43] Also, the error factors include cycles according to the time windows, dates and frequencies of error occurrence, and may include an error source log generation pattern or system usage, such as CPU, memory and network overloads. The error factors are not fixed, and vary according to error type definition. With regard to this, it is preferred to exclude less contributive error factors from among various error factors, thereby simplifying a prediction model function and reducing the amount of calculations.
[44] The prediction model function module 120 derives a basic prediction model function from the error factors extracted in the error factor extraction module 110, and completes a prediction model function by applying a genetic algorithm to the derived basic prediction model function.
[45] The following multiple regression equation may be used as the basic prediction model function: P(T+ 1) = E +E X +E X (t)+E X (t)+...+E X (t)+e(t) (where, P denotes
0 1 1 2 2 3 3 n n an error rate, t denotes a time, X to X denote error factors, E to E denote weights,
I n I n and e(t) denotes a correction value). [46] With regard to this, the weight values E to E are preferably derived as optimal
1 n weights by means of a genetic algorithm. Also, since an error occurrence pattern may vary according to the passage of time when weight values are derived using a genetic algorithm, optimal weight values must also change. Therefore, it is preferred to use the genetic algorithm in order to complete an adaptive model function suited to the passage of time and the error occurrence pattern.
[47] The prediction module 130 gives a pre-test prediction for the occurrence of a specific type of error occurring in the broadcasting system by using the prediction model function completed in the prediction model function module 120, provides the correction value calculation module 140 with a result value of the pre-test prediction, and finally measures and verifies a specific type of error for a given period of time by using the prediction model function that is corrected by the function correction module 150 based on an output value of the correction value calculation module 140.
[48] The correction value calculation module 140 compares a result value of pre-test prediction for errors occurring for a given period of time, which is provided by the prediction module 130, with the actually generated error data, and calculates a correction value for compensating for a difference between the result value and the actual error data.
[49] The function correction module 150 corrects the prediction model function by applying a correction value calculated in the correction value calculation module 140 to the prediction model function, and provides the prediction module 130 with the corrected prediction model function.
[50] Reference will now be made to a method of analyzing and predicting errors occurring in a broadcasting system according to a preferred embodiment of the present invention, with reference to FIG. 2 illustrating a flowchart of such a method.
[51] The method of analyzing and predicting errors occurring in a broadcasting system according to this embodiment includes an error type definition step S200, an error data analysis step S210, a function derivation step S220, a function completion step S230, a pre-test prediction step S240, a correction value calculation step S250, a function correction step S260, and a prediction step S270.
[52] In the error type definition step S200, error types are defined for various errors that have been generated for a given period of time (e.g., six months from January to June) in the broadcasting system. With regard to this, the error type may be defined for each error, or may be defined for several combined errors.
[53] In the error data analysis step S210, error data for error types defined in the error type definition step are analyzed. With regard to this, it is preferred that statistically important error sources are extracted as error factors from the error data by analyzing error sources that are periodically collected.
[54] Particularly, a prediction model equation can be simplified by varying error factor extraction adaptively to circumstances, extracting factors corresponding to direct causes of errors as the error factors, and excluding less contributive error factors.
[55] In the function derivation step S220, a basic prediction model function is derived by extracting error factors from the error data analyzed in the error data analysis step S210. In particular, the error factors preferably use cycles according to the time windows, dates and frequencies of error occurrence, and may include an error source log generation pattern or system usage, such as CPU, memory and network overloads.
[56] Here, the following multiple regression equation is preferably used as the basic r prediction model function: P(T+ 1) = E 0 +E 1 X1 +E2 X 2 (t)+E 3 X 3 (t)+...+E n X n (t)+e(t)
(where, P denotes an error rate, t denotes a time, X I to Xn denote error factors, E I to En denote weights, and e(t) denotes a correction value). [57] Particularly, the weight values E 1 to E n are preferably derived as optimal weights by means of a genetic algorithm. Also, since an error occurrence pattern may vary according to the passage of time when weight values are derived using a genetic algorithm, optimal weight values must also change. Therefore, it is preferred to use the genetic algorithm in order to complete an adaptive model function suited to the passage of time and the error occurrence pattern.
[58] In the function completion step S230, a prediction model function is completed by applying a genetic algorithm, which uses error data generated for a given period of time (e.g., next five months from July t November), to the derived basic prediction model function derived in the function derivation step S220.
[59] In the pre-test prediction step S240, the prediction model function completed in the function completion step S230 is applied to pre-test prediction for errors that have been generated for a given past period of time (e.g., next one month of December), and the prediction model function is verified in this way.
[60] In the correction value calculation step S250, a correction value is calculated by comparing a result value of the pre-test prediction, which is performed for a given period of time in the pre-test prediction step S240, with error data actually generated for that period of time.
[61] In the function correction step S260, the prediction model function is corrected by applying the correction value calculated in the correction value calculation step to the prediction model function.
[62] In the prediction step S270, a specific type of error occurring in the broadcasting
system is finally predicted through the prediction model function corrected in the function correction step S260.
[63] Although several preferred embodiments of the present invention have been described for illustrative purposes, those skilled in the art will appreciate that various modifications, additions and substitutions are possible, without departing from the scope and spirit of the invention as disclosed in the accompanying claims.
[64]
Industrial Applicability
[65] According to the present invention as described above, system errors are recognized in advance and a stable transmission environment is maintained by analyzing errors periodically generated in a broadcasting system and predicting the occurrence dates of future errors. Accordingly, the present invention has industrial applicability because it has sufficient marketability or business possibility as well as applicability to related services, and is also obviously practicable.
Claims
[1] A method of analyzing and predicting errors occurring in a broadcasting system, the method comprising: an error type definition step of defining error types for errors occurring in the broadcasting system; an error data analysis step of analyzing error data having the error types defined in the error type definition step; a function derivation step of deriving a prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing the prediction model function by applying a genetic algorithm, which uses the error data generated for a given period of time, to the prediction model function derived in the function derivation step; and a prediction step of predicting a specific type of error occurring in the broadcasting system through the prediction model function completed in the function completion step.
[2] The method as claimed in claim 1, wherein statistically important error sources are extracted as the error factors from the error data by analyzing periodically collected error sources.
[3] The method as claimed in claim 2, wherein the error factors are extracted in cycles according to time windows, dates and frequencies of error occurrence, are extracted according to error source log generation patterns, and are extracted according to occurrence frequencies of CPU, memory and network overloads.
[4] The method as claimed in claim 1, wherein the prediction model function comprises a multiple regression equation.
[5] The method as claimed in claim 4, wherein the multiple regression equation is given by a following equation,
P(T+ 1) = E 0 +E1 X1 +E2 X2 (t)+E 3 X3 (t)+...+E n Xn (t)+e(t) where, P denotes an error rate, t denotes a time, X I to Xn denote error factors, E 1 to E denote weights, and e(t) denotes a correction value. n
[6] The method as claimed in claim 5, wherein the weights E to E are derived as
1 n optimal weights by means of a genetic algorithm.
[7] A method of analyzing and predicting errors occurring in a broadcasting system, the method comprising: an error type definition step of defining error types for errors generated for a given period of time in the broadcasting system; an error data analysis step of analyzing error data having the error types defined
in the error type definition step; a function derivation step of deriving a basic prediction model function by extracting error factors from the error data analyzed in the error data analysis step; a function completion step of completing a prediction model function by applying a genetic algorithm, which uses the error data generated for the given period of time, to the basic prediction model function derived in the function deri vation step; a pre-test prediction verification step of verifying pre-test prediction for a specific type of error generated for the given period of time in the broadcasting system through the prediction model function completed in the function completion step; a correction value calculation step of calculating a correction value by comparing a result value of the pre-test prediction for the error, which is tested for the given period of time in the pre-test prediction verification step, with actually generated error data; a function correction step of correcting the prediction model function by applying the correction value calculated in the correction value calculation step to the prediction model function; and a prediction step of finally predicting the specific type of error occurring for the given period of time in the broadcasting system through the prediction model function corrected in the function correction step.
[8] The method as claimed in claim 7, wherein statistically important error sources are extracted as the error factors from the error data by analyzing periodically collected error sources.
[9] The method as claimed in claim 8, wherein the error factors are extracted in cycles according to time windows, dates and frequencies of error occurrence, are extracted according to error source log generation patterns, and are extracted according to occurrence frequencies of CPU, memory and network overloads.
[10] The method as claimed in claim 7, wherein the prediction model function comprises a multiple regression equation P(T+ 1) given by a following equation, P(T+ 1) = E 0 +E1 X1 +E2 X2 (t)+E 3 X 3 (t)+...+E n Xn (t)+e(t) where, P denotes an error rate, t denotes a time, X to X denote error factors, E
1 n 1 to E n denote weights, and e(t) denotes a correction value.
[11] The method as claimed in claim 10, wherein the weights E 1 to E n are derived as optimal weights by means of a genetic algorithm.
[12] A system for analyzing and predicting errors occurring in a broadcasting system, the system comprising:
an error type definition module for defining error types based on error data for errors generated for a given period of time in the broadcasting system; an error factor extraction module for extracting error factors by analyzing the error data defined in the error type definition module; a prediction model function module for deriving a basic prediction model function from the error factors extracted in the error factor extraction module, and completing a prediction model function by applying a genetic algorithm to the derived basic prediction model function; a prediction module for verifying a probability of predicting error occurrence through the prediction model function completed in the prediction model function module, and predicting a specific type of error through the prediction model function corrected in a function correction module; a correction value calculation module for calculating a correction value by comparing a result value of prediction tested in the prediction module with actually generated error data; and the function correction module for correcting the prediction model function by applying the correction value calculated in the correction value calculation module to the prediction model function.
[13] The system as claimed in claim 12, wherein statistically important error sources are extracted as the error factors from the error data by analyzing periodically collected error sources.
[14] The system as claimed in claim 13, wherein the error factors are extracted in cycles according to time windows, dates and frequencies of error occurrence.
[15] The system as claimed in any one of claims 12 to 14, wherein the prediction model function comprises a multiple regression equation P(T+ 1) given by a following equation, P(T+ 1) = E 0 +E1 X1 +E2 X2 (t)+E 3 X3 (t)+...+E n Xn (t)+e(t) where, P denotes an error rate, t denotes a time, X to X denote error factors, E
I n 1 to E denote weights, and e(t) denotes a correction value. n
[16] The system as claimed in claim 15, wherein the weights E to E are derived as
1 n optimal weights by means of a genetic algorithm.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020060094309A KR100818493B1 (en) | 2006-09-27 | 2006-09-27 | Method for analyzing and predicting error generated in broadcasting system and apparatus therefor |
KR10-2006-0094309 | 2006-09-27 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2008038975A1 true WO2008038975A1 (en) | 2008-04-03 |
Family
ID=39230359
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2007/004671 WO2008038975A1 (en) | 2006-09-27 | 2007-09-21 | Method for analyzing and predicting error generated in broadcasting system and apparatus therefor |
Country Status (2)
Country | Link |
---|---|
KR (1) | KR100818493B1 (en) |
WO (1) | WO2008038975A1 (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105444923A (en) * | 2015-11-18 | 2016-03-30 | 浙江工业大学 | Mechanical temperature instrument error prediction method based on genetic-algorithm optimized least square support vector machine |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
KR101538758B1 (en) * | 2014-05-19 | 2015-07-27 | 주식회사 디리아 | Apparatus for forecasting disruption and method thereof in IT system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030046633A1 (en) * | 2001-08-28 | 2003-03-06 | Jutzi Curtis E. | Data error correction based on reported factors and predicted data interference factors |
US20030081671A1 (en) * | 2001-10-26 | 2003-05-01 | Takaharu Ishida | Method for digital broadcast interpolation and digital broadcast receiving system |
US20030198405A1 (en) * | 2002-04-17 | 2003-10-23 | Koninklijke Philips Electronics N. V. | Method and an apparatus to speed the video system optimization using genetic algorithms and memory storage |
Family Cites Families (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2002262260A (en) | 2001-02-26 | 2002-09-13 | Sanyo Electric Co Ltd | Data broadcast transmission/reception system, error management method and data broadcast receiver |
KR100476521B1 (en) * | 2002-08-27 | 2005-03-17 | 하나로텔레콤 주식회사 | Coaxial Cable Network Fault Inference Method in CATV Internet Service |
-
2006
- 2006-09-27 KR KR1020060094309A patent/KR100818493B1/en not_active IP Right Cessation
-
2007
- 2007-09-21 WO PCT/KR2007/004671 patent/WO2008038975A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030046633A1 (en) * | 2001-08-28 | 2003-03-06 | Jutzi Curtis E. | Data error correction based on reported factors and predicted data interference factors |
US20030081671A1 (en) * | 2001-10-26 | 2003-05-01 | Takaharu Ishida | Method for digital broadcast interpolation and digital broadcast receiving system |
US20030198405A1 (en) * | 2002-04-17 | 2003-10-23 | Koninklijke Philips Electronics N. V. | Method and an apparatus to speed the video system optimization using genetic algorithms and memory storage |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105444923A (en) * | 2015-11-18 | 2016-03-30 | 浙江工业大学 | Mechanical temperature instrument error prediction method based on genetic-algorithm optimized least square support vector machine |
Also Published As
Publication number | Publication date |
---|---|
KR100818493B1 (en) | 2008-04-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109242135B (en) | Model operation method, device and business server | |
Guthrie | Real options analysis of climate-change adaptation: investment flexibility and extreme weather events | |
Saghafian et al. | The newsvendor under demand ambiguity: Combining data with moment and tail information | |
CN109889391B (en) | Network short-time flow prediction method based on combined model | |
CN111639798A (en) | Intelligent prediction model selection method and device | |
EP3896622A1 (en) | Sampling device, sampling method, and sampling program | |
CN103729530B (en) | Device and method for processing sequence | |
US7643972B2 (en) | Computer-implemented systems and methods for determining steady-state confidence intervals | |
Ma et al. | The mahalanobis kernel for heritability estimation in genome-wide association studies: fixed-effects and random-effects methods | |
US20200052635A1 (en) | Method for searching excitation signal of motor, and electronic device | |
CN114124260A (en) | Spectrum prediction method, apparatus, medium, and device based on composite 2D-LSTM network | |
WO2008038975A1 (en) | Method for analyzing and predicting error generated in broadcasting system and apparatus therefor | |
CN117574835A (en) | Method, equipment and storage medium for simulating Nand false LLR distribution | |
Kousathanas et al. | A guide to general-purpose ABC software | |
Yang et al. | Adaptive reward computation in reinforcement learning-based continuous integration testing | |
EP2693388A1 (en) | Risk-profile generation device | |
CN112308278B (en) | Optimization method, device, equipment and medium of user off-network prediction model | |
CN103942403A (en) | Method and device for screening mass variables | |
Aliniya et al. | Dynamic constrained multi-objective optimization based on adaptive combinatorial response mechanism | |
Filipovska et al. | Reliable trajectory-adaptive routing strategies in stochastic, time-varying networks with generalized correlations | |
KR20090005506A (en) | System and method for autiomatic distribution fitting | |
Colombelli et al. | Multi-objective prioritization for data center vulnerability remediation | |
Kousathanas et al. | A guide to general-purpose approximate bayesian computation software | |
CN112416789B (en) | Process metric element evaluation method for evolution software | |
Goel et al. | Multi-objective optimization using LS-OPT |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 07808436 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1)EPC, EPO FORM 1205A SENT ON 03/07/09 |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 07808436 Country of ref document: EP Kind code of ref document: A1 |