Patent US1885472A successfully refreshed from Google Patents.
Patent US2045171A successfully refreshed from Google Patents.
Patent US5659593A successfully refreshed from Google Patents.
US5659593A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Detection of deviations in monitored patterns
Assignee: Nokia of America Corp
•
Filing: 1994-11-30
•
Granted: 1997-08-19
•
Expires: 2014-11-30
Abstract
Abstract A system and method for detecting significant deviations in a pattern in near real time by making forecasts about the pattern and determining if the actual pattern is within a threshold of the forecast. The preferred embodiment is a method for detecting recording errors in a telephone network that provides services to a plurality of customers, the method comprising: recording information output by the telephone network relating to a service that was provided to the customers in a first time period; forecasting, based on the recorded information, information that will be output by the network in a second time period that has not yet occurred; recording information output by the telephone network relating to the provision of the service to the customers in the second time period; and comparing the result of the forecasting step with the information recorded during the second time period to generate a deviation.
Inventors
- 👤 Dvora Tzvieli
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-11-01 13:57
CPC Classifications
H; H04; H04M; H04M3/00; H04M3/22; H04M3/36; H04M15/00
Claims Analysis
Total Claims: 40
Claim 1
Independent Claim
114 words
A method for detecting recording errors in a telephone network that provides services to a plurality of customers, said method comprising: recording information output by said telephone network relating to the provision of a service to said customers in a first time period; forecasting, based on said recorded information, an expected content of information that will be output by said network for a second time period that has not yet occurred; recording information output by said telephone network relating to the provision of said service to said customers in said second time period; and comparing the result of said forecasting step with the information recorded during said second time period to generate a deviation.
Claim 2
Independent Claim
114 words
A method for detecting recording errors in a telephone network that provides services to a plurality of customers, said method comprising: recording information output by said telephone network relating to the provision of a service to said customers in a first time period; forecasting, based on said recorded information, an expected content of information that will be output by said network for a second time period that has not yet occurred; recording information output by said telephone network relating to the provision of said service to said customers in said second time period; and comparing the result of said forecasting step with the information recorded during said second time period to generate a deviation.
Claim 3
Dependent Claim
93 words
The method of claim 1 further comprising the steps of: forecasting, based on said information recorded during said second time period, the expected content of information that will be output by said network for a third time period that has not yet occurred; recording information output by said telephone network relating to the provision of said service to said customers in said third time period; and comparing the result of said step of forecasting associated with said third time period with the information recorded during said third time period to generate a deviation.
Claim 4
Dependent Claim
22 words
The method of claim 1 further comprising the step of generating an alarm if said deviation is greater than a predetermined threshold.
Claim 5
Dependent Claim
36 words
The method of claim 1 wherein said steps are carried out for a plurality of services provided by said network, and an alarm is automatically generated each time a generated deviation is greater than a threshold.
Claim 6
Dependent Claim
59 words
The method of claim 1 wherein said first time period corresponds to a preceding time period and said second time period corresponds to a future time period, and wherein all of said steps are iteratively repeated to continuously generate deviations corresponding to a comparison of data recently output by said network and a forecast for said recently output data.
Claim 7
Dependent Claim
23 words
The method of claim 5 further comprising the step of generating an alarm if a generated deviation is greater than a predetermined threshold.
Claim 8
Dependent Claim
24 words
The method of claim 6 wherein said step of generating an alarm is suspended for data output by said network during a holiday period.
Claim 9
Dependent Claim
45 words
The method of claim 1 wherein said step of forecasting comprises the steps of: generating a model to fit said information recorded in said first time period; using said model to generate a forecast of data that will be recorded in said second time period.
Claim 10
Dependent Claim
33 words
The method of claim 8 further comprising the steps of periodically making a quality measurement of said model and determining, based on said quality measurement, whether a new forecasting model should be generated.
Claim 11
Dependent Claim
27 words
The method of claim 1 wherein each of said recording steps records information relating to the provision of each of a number of services to said customers.
Claim 12
Dependent Claim
41 words
The method of claim 10, further comprising the step of generating a timeseries for each service provided by said network, each said timeseries comprising data relating to the provision of a said service to said customers as a function of time.
Claim 13
Dependent Claim
48 words
The method of claim 11, wherein said step of forecasting further comprises the steps of: generating a model to fit each of said timeseries; using each said model to generate a forecast for each said timeseries to predict values for each said timeseries during a future time period.
Claim 14
Dependent Claim
60 words
The method of claim 12 wherein, for each said timeseries, said step of generating a model comprises: identifying a number of potential model forms for said timeseries; ordering said potential model forms according to the likelihood that each will fit said timeseries; and iteratively fitting each of said potential model forms to find a model that satisfactorily fits said timeseries.
Claim 15
Dependent Claim
53 words
The method of claim 12 wherein before said step of using each said model to generate a forecast for each said timeseries, said method comprises the steps of: identifying, in each said timeseries, data that differs from forecast data by greater than a threshold; and replacing said identified data in each said timeseries.
Claim 16
Dependent Claim
44 words
The method of claim 13 wherein said step of generating a model further comprises generating a default model for each said timeseries, said default model being chosen as the model for a said timeseries if none of said potential model forms fit said timeseries.
Claim 17
Dependent Claim
28 words
The method of claim 13 wherein, for each said timeseries, said potential model forms are fitted from a stored library of model forms associated with each said timeseries.
Claim 18
Dependent Claim
92 words
The method of claim 11 wherein two timeseries are generated for each said service, including: (1) a volume timeseries identifying the number of times each said service is provided by said network as a function of time; and (2) a ratio timeseries identifying the ratio of the number of times said network connects a call corresponding to each said service over a given period of time to the number of times said network disconnects a call corresponding to each said service over said given period of time, as a function of time.
Claim 19
Independent Claim
142 words
A method for detecting recording errors in a telephone network that provides services to a plurality of customers, said method comprising: (a) recording information output by said telephone network relating to the provision of a service to said customers in a first time period; (b) generating a model to fit said information recorded in said first time period; (c) using said model to generate a forecast of information to be output by said network for a future time period; (d) recording, during the occurrence of said future time period, information output by said telephone network relating to the provision of said service to said customers; (e) comparing the forecast generated in step (c) with the information recorded in step (d) to generate a deviation; and (f) repeating steps (c)-(e) to continuously generate deviations between information output by said network and forecast information.
Claim 20
Independent Claim
142 words
A method for detecting recording errors in a telephone network that provides services to a plurality of customers, said method comprising: (a) recording information output by said telephone network relating to the provision of a service to said customers in a first time period; (b) generating a model to fit said information recorded in said first time period; (c) using said model to generate a forecast of information to be output by said network for a future time period; (d) recording, during the occurrence of said future time period, information output by said telephone network relating to the provision of said service to said customers; (e) comparing the forecast generated in step (c) with the information recorded in step (d) to generate a deviation; and (f) repeating steps (c)-(e) to continuously generate deviations between information output by said network and forecast information.
Claim 21
Dependent Claim
88 words
The method of claim 18 further comprising the steps of: (g) periodically making a quality measurement of said model; (h) determining, based on said quality measurement, whether a new forecasting model should be generated; and (i) if the result of step (h) indicates that a new forecasting model should be generated: (j) generating a new forecasting model to fit information recorded in a predetermined prior time period; and (k) substituting said new forecasting model for the model being used for forecasting in the repeated implementation of steps (c)-(e).
Claim 22
Dependent Claim
23 words
The method of claim 19 wherein step (k) is carried out while maintaining the continuous generation of deviations as specified in step (f).
Claim 23
Independent Claim
169 words
A method for detecting deviations between a monitored pattern and a forecast for said pattern, said method comprising: (a) recording said pattern during a first time period; (b) generating a model to fit said pattern recorded in said first time period; (c) using said model to generate a forecast of said pattern in a future time period; (d) recording said pattern during the occurrence of said future time period; (e) comparing the forecast generated in step (c) with the pattern recorded in step (d) to generate a deviation; (f) repeating steps (c)-(e) to continuously generate deviations between said pattern and a forecast of said pattern; (g) periodically making a quality measurement of said model; (h) determining, based on said quality measurement, whether a new forecasting model should be generated; and, if so: (i) generating a new forecasting model to fit information recorded in a predetermined prior time period; and (j) substituting said new forecasting model for the model being used for forecasting in the repeated implementation of steps (c)-(e).
Claim 24
Independent Claim
169 words
A method for detecting deviations between a monitored pattern and a forecast for said pattern, said method comprising: (a) recording said pattern during a first time period; (b) generating a model to fit said pattern recorded in said first time period; (c) using said model to generate a forecast of said pattern in a future time period; (d) recording said pattern during the occurrence of said future time period; (e) comparing the forecast generated in step (c) with the pattern recorded in step (d) to generate a deviation; (f) repeating steps (c)-(e) to continuously generate deviations between said pattern and a forecast of said pattern; (g) periodically making a quality measurement of said model; (h) determining, based on said quality measurement, whether a new forecasting model should be generated; and, if so: (i) generating a new forecasting model to fit information recorded in a predetermined prior time period; and (j) substituting said new forecasting model for the model being used for forecasting in the repeated implementation of steps (c)-(e).
Claim 25
Dependent Claim
20 words
The method of claim 21 wherein step (j) is carried out while continuously generating deviations as specified in step (f).
Claim 26
Dependent Claim
26 words
The method of claim 21 further comprising the step of generating an alarm if said any of said generated deviations is greater than a predetermined threshold.
Claim 27
Dependent Claim
17 words
The method of claim 21 wherein said steps are carried out simultaneously for a number of patterns.
Claim 28
Dependent Claim
34 words
The method of claim 24 further comprising the step of generating a timeseries for each monitored pattern, each said timeseries comprising data relating to the value of said pattern as a function of time.
Claim 29
Dependent Claim
48 words
The method of claim 25, wherein said step of forecasting further comprises the steps of: generating a model to fit each of said timeseries; using each said model to generate a forecast for each said timeseries to predict values for each said timeseries during a future time period.
Claim 30
Dependent Claim
60 words
The method of claim 26 wherein, for each said timeseries, said step of generating a model comprises: identifying a number of potential model forms for said timeseries; ordering said potential model forms according to the likelihood that each will fit said timeseries; and iteratively fitting each of said potential model forms to find a model that satisfactorily fits said timeseries.
Claim 31
Dependent Claim
53 words
The method of claim 26 wherein before said step of using each said model to generate a forecast for each said timeseries, said method comprises the steps of: identifying, in each said timeseries, data that differs from forecast data by greater than a threshold; and replacing said identified data in each said timeseries.
Claim 32
Dependent Claim
43 words
The method of claim 28 wherein said step of generating a model further comprises generating a default model for each said timeseries, said default model being chosen as the model for a said timeseries if none of said potential models fit said timeseries.
Claim 33
Dependent Claim
28 words
The method of claim 27 wherein, for each said timeseries, said potential model forms are selected from a stored library of model forms associated with each said timeseries.
Claim 34
Independent Claim
160 words
A method for detecting deviations between a plurality of monitored patterns and a forecast for each said pattern, said method comprising: (a) generating a timeseries for each monitored pattern, each said timeseries comprising data relating to the value of said pattern as a function of time; (b) selecting a model to fit each said timeseries, said selection for each timeseries comprising the steps of: (1) identifying a number of potential model forms for said timeseries; (2) iteratively fitting each of said potential model forms to find a model that satisfactorily fits said timeseries; (c) using said model to generate a forecast of each said timeseries in a future time period; (d) recording each said timeseries during the occurrence of said future time period; (e) comparing each forecast generated in step (c) with the timeseries recorded in step (d) to generate a deviation; and (f) repeating steps (c)-(e) to continuously generate deviations between said pattern and a forecast of said pattern.
Claim 35
Independent Claim
160 words
A method for detecting deviations between a plurality of monitored patterns and a forecast for each said pattern, said method comprising: (a) generating a timeseries for each monitored pattern, each said timeseries comprising data relating to the value of said pattern as a function of time; (b) selecting a model to fit each said timeseries, said selection for each timeseries comprising the steps of: (1) identifying a number of potential model forms for said timeseries; (2) iteratively fitting each of said potential model forms to find a model that satisfactorily fits said timeseries; (c) using said model to generate a forecast of each said timeseries in a future time period; (d) recording each said timeseries during the occurrence of said future time period; (e) comparing each forecast generated in step (c) with the timeseries recorded in step (d) to generate a deviation; and (f) repeating steps (c)-(e) to continuously generate deviations between said pattern and a forecast of said pattern.
Claim 36
Dependent Claim
45 words
The method of claim 31 wherein after step (a) and before step (c) said method comprises the additional steps of: identifying, in each said timeseries, data that differs from forecast data by greater than a threshold; and replacing said identified data in each said timeseries.
Claim 37
Dependent Claim
40 words
The method of claim 31 wherein step (b) further comprises generating a default model for each said timeseries, said default model being chosen as the model for a said timeseries if none of said potential model forms fit said timeseries.
Claim 38
Dependent Claim
28 words
The method of claim 31 wherein, for each said timeseries, said potential model forms are selected from a stored library of model forms associated with each said timeseries.
Claim 39
Independent Claim
122 words
A system for detecting recording errors in a telephone network that provides services to a plurality of customers, said system comprising: means for recording information output by said telephone network relating to the provision of a service to said customers in a first time period; means for forecasting, based on said recorded information, an expected content of information that will be output by said network for a second time period that has not yet occurred; means for recording information output by said telephone network relating to the provision of said service to said customers in said second time period; and means for comparing the result of said forecasting step with the information recorded during said second time period to generate a deviation.
Claim 40
Independent Claim
122 words
A system for detecting recording errors in a telephone network that provides services to a plurality of customers, said system comprising: means for recording information output by said telephone network relating to the provision of a service to said customers in a first time period; means for forecasting, based on said recorded information, an expected content of information that will be output by said network for a second time period that has not yet occurred; means for recording information output by said telephone network relating to the provision of said service to said customers in said second time period; and means for comparing the result of said forecasting step with the information recorded during said second time period to generate a deviation.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 2000-08-18 | 2001-02-17 | 2001-08-19 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2004-08-18 | 2005-02-17 | 2005-08-19 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2008-08-18 | 2009-02-17 | 2009-08-19 | $7700.00 | Window Closed / Lapsed |