Search

EP-4305822-B1 - METHODS AND APPARATUSES FOR REPORTING ANOMALIES AND FORECASTING KEY PERFORMANCE INDICATOR (KPI) MEASUREMENTS IN A TELECOMMUNICATIONS NODE

EP4305822B1EP 4305822 B1EP4305822 B1EP 4305822B1EP-4305822-B1

Inventors

  • BASHBAGHI, Saman
  • SELLIER, Jean Michel
  • GOLEA, MOSTEFA
  • MARATH, Sathi

Dates

Publication Date
20260506
Application Date
20210312

Claims (15)

  1. A method (700) for reporting an anomaly in a telecommunications node, comprising: - obtaining (702) a measurement of a key performance indicator, KPI, of the telecommunication node; - upon receiving the measurement of the KPI, updating (704) coefficients of a polynomial function, comprising, for each of the coefficients of the polynomial function: ∘ computing a loss function for the measurement of the KPI; ∘ computing a gradient of the loss function; and ∘ updating the coefficient as a function of the gradient of the loss function; - based on the updated coefficients of the polynomial function, computing (706) an expected measurement of the KPI ; - computing (708) a confidence band for the expected measurement of the KPI; and - reporting (710) the anomaly when the measurement of the KPI is outside of the confidence band.
  2. A method (800) for forecasting a plurality of expected measurements for a key performance indicator, KPI, in a telecommunications node, comprising: - obtaining (802) a measurement of the KPI of the telecommunication node; - upon receiving the measurement of the KPI, updating (804) coefficients of a polynomial function, comprising, for each of the coefficients of the polynomial function: ∘ computing a loss function for the measurement of the KPI; ∘ computing a gradient of the loss function; and ∘ updating the coefficient as a function of the gradient of the loss function; - based on the updated coefficients of the polynomial function, computing (806) the plurality of expected measurements for the KPI, over a time-horizon; - computing (808) a confidence band for each of the plurality of expected measurements for the KPI, using accuracy measurements obtained from past predictions; and - reporting (810) the plurality of expected measurements and corresponding confidence bands for the KPI to a management system.
  3. The method of claim 1 or 2, wherein initial coefficients of the polynomial function are computed based on M initial measurements of the KPI and the coefficients a are computed using: a = polyfit(x[0:M], y[0:M], degree) where polyfit is a function that computes the coefficients using a least square method, x is the time at which is taken the measurement of the KPI, y is the measurement of the KPI and degree is the degree of the polynomial function.
  4. The method of claim 1 or 2, wherein computing the confidence band comprises: - computing an average and a standard deviation for the expected measurement of the KPI based on previous measurements; and - setting the confidence band to minus three times the standard deviation from the average to plus three times the standard deviation from the average.
  5. The method of claim 4, wherein: - the confidence band is computed for each measurement of the KPI in a time series and includes computing the average and the standard deviation; or - computing the average and the standard deviation is based on a predetermined number of most recent previous measurements which does not include all the previous measurements.
  6. The method of claim 1 or 2, wherein the polynomial function is a 7 th order polynomial function.
  7. The method of claim 1 or 2, wherein - computing the loss function for the measurement of the KPI is done using: L(a, x i , y i ) = / P(x i ) - y i / 2 , where L is the loss function, a is the coefficient of the polynomial function, y i is the data point, P(a, x i ) is the expected measurement of the KPI and i is an index of the measurement of the KPI; or - computing the gradient of the loss function is done using: ∇ aj L(x i ) = 2 (x i ) / P(x i ) - y i / , where L is the loss function, x i is the time at which is taken the measurement of the KPI, y i is the measurement of the KPI, P(x i ) is the expected measurement of the KPI and i is an index of the measurement of the KPI.
  8. An apparatus operative to report an anomaly in a telecommunications node comprising processing circuits and a memory, the memory containing instructions executable by the processing circuits whereby the apparatus is operative to: - obtain a measurement of a key performance indicator, KPI, of the telecommunication node; - upon receiving the measurement of the KPI, update coefficients of a polynomial function, comprising, for each of the coefficients of the polynomial function: ∘ computing a loss function for the measurement of the KPI; ∘ computing a gradient of the loss function; and ∘ updating the coefficient as a function of the gradient of the loss function; - based on the updated coefficients of the polynomial function, compute an expected measurement of the KPI ; - compute a confidence band for the expected measurement of the KPI; and - report the anomaly when the measurement of the KPI is outside of the confidence band.
  9. An apparatus operative to forecast a plurality of expected measurements for a key performance indicator, KPI, in a telecommunications node comprising processing circuits and a memory, the memory containing instructions executable by the processing circuits whereby the apparatus is operative to: - obtain a measurement of the KPI of the telecommunication node; - upon receiving the measurement of the KPI, update coefficients of a polynomial function, comprising, for each of the coefficients of the polynomial function: ∘ computing a loss function for the measurement of the KPI; ∘ computing a gradient of the loss function; and ∘ updating the coefficient as a function of the gradient of the loss function; - based on the updated coefficients of the polynomial function, compute the plurality of expected measurements for the KPI, over a time-horizon; - compute a confidence band for each of the plurality of expected measurements for the KPI, using accuracy measurements obtained from past predictions; and - report the plurality of expected measurements and corresponding confidence bands for the KPI to a management system.
  10. The apparatus of claim 8 or 9, wherein initial coefficients of the polynomial function are computed based on M initial measurements of the KPI and the coefficients a are computed using: a = polyfit(x[0:M], y[0:M], degree) where polyfit is a function that computes the coefficients using a least square method, x is the time at which is taken the measurement of the KPI, y is the measurement of the KPI and degree is the degree of the polynomial function.
  11. The apparatus of claim 8 or 9, wherein computing the confidence band comprises: - computing an average and a standard deviation for the expected measurement of the KPI based on previous measurements; and - setting the confidence band to minus three times the standard deviation from the average to plus three times the standard deviation from the average.
  12. The apparatus of claim 11, wherein: - the confidence band is computed for each measurement of the KPI in a time series and includes computing the average and the standard deviation; or - computing the average and the standard deviation is based on a predetermined number of most recent previous measurements which does not include all the previous measurements.
  13. The apparatus of claim 8 or 9, wherein the polynomial function is a 7 th order polynomial function.
  14. The apparatus of claim 8 or 9, wherein: - computing the loss function for the measurement of the KPI is done using: L(a, x i , y i ) := / P(x i ) - y i / 2 , where L is the loss function, a is the coefficient of the polynomial function, y i is the data point, P(a, x i ) is the expected measurement of the KPI and i is an index of the measurement of the KPI; or - computing the gradient of the loss function is done using: ∇ aj L(x i ) = 2 (x i ) / P(x i ) - y i / , where L is the loss function, x i is the time at which is taken the measurement of the KPI, y i is the measurement of the KPI, P(x i ) is the expected measurement of the KPI and i is an index of the measurement of the KPI.
  15. A non-transitory computer readable medium having stored thereon instructions in a telecommunications node, the instructions comprising any of the steps of claims 1 to 7.

Description

TECHNICAL FIELD The present disclosure relates to surveillance of key performance indicators in telecommunications nodes. BACKGROUND We are now living in the big data and real-time processing era. In telecommunications networks management domains, large number of metrics, key performance indicators (KPIs), are continuously monitored, on almost every network device. The resulting data streams are then pipelined and analyzed, in near real-time, for anomalies, trends, correlations, etc. Network operators combine those real-time analytics to react and correct issues, to keep the networks running smoothly. Machine Learning (ML) and Artificial Intelligence (AI) are becoming key components of network management solutions. In the time series analysis community, there has been a long-standing consensus that sophisticated methods do not necessarily produce better forecast and/or anomaly detection (AD), when compared to simpler methods. This was one of the conclusions of the influential M3 forecasting competition held in 1999. Simpler and noise-insensitive models, with reasonable assumptions about the data, will typically perform very well, e.g. Exponential Smoothing techniques and the well-known AutoRegressive Integrated Moving Average (ARIMA) method. Anomaly-detection in time-series is a well-established field of research. A large number of models and techniques are documented in the literature. Anomaly-detection has applications in many domains, including telecommunications networks management, fraud detection, health, etc. Document US 2021 / 0 051 503 A1 may be construed to disclose an anomaly detection and analysis system that generates analysis or summary of the anomalies detected from key performance indicators (KPIs). The system receives anomaly data reporting anomalies detected in key performance indicator (KPI) data. The system classifies the reported anomalies into a plurality of anomaly items, wherein anomalies from KPI data that share a set of features are assigned to one anomaly item. The system computes a ranking score for each anomaly item by assigning predefined weights for different anomaly types that are present in the anomaly item. The system sorts a list of anomaly items from the plurality of anomaly items into a sorted list of anomaly items according to the ranking scores computed for the plurality of anomaly items. The system sends the sorted list of anomaly items to a user device for presentation. SUMMARY Existing models do not fully address the complexities of deploying and maintaining a productized anomaly-detection solution, including complexities such as memory footprint, processing speed, and handling of data-drift. According to the present disclosure, there are provided methods, apparatuses and a computer-readable medium according to the independent claims. Further developments are set forth in the dependent claims. The method, apparatus and non-transitory computer readable medium provided herein present improvements to the way reporting anomalies and forecasting of expected measurements for a key performance indicator (KPI) operate. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a graph illustrating results obtained with different methods.Figure 2 is a schematic illustration of an anomaly detection process.Figure 3 is a graph illustrating anomaly detection results of an offline polynomial approach without re-training.Figures 4 and 5 are graphs illustrating anomaly detection results of an offline polynomial approach with retraining on a scheduled weekly basis (parts 1 and 2).Figure 6 is a graph illustrating anomaly detection results according to the methods described herein.Figure 7 is a flowchart of a method for reporting an anomaly in a telecommunications node.Figure 8 is a flowchart of a method for forecasting a plurality of expected measurements for a key performance indicator (KPI) in a telecommunications node.Figure 9 is a schematic illustration of a virtualization environment in which the different methods, apparatuses and non-transitory computer readable media described herein can be deployed. DETAILED DESCRIPTION Various features will now be described with reference to the drawings to fully convey the scope of the disclosure to those skilled in the art. Sequences of actions or functions may be used within this disclosure. It should be recognized that some functions or actions, in some contexts, could be performed by specialized circuits, by program instructions being executed by one or more processors, or by a combination of both. Further, computer readable carrier or carrier wave may contain an appropriate set of computer instructions that would cause a processor to carry out the techniques described herein. The functions/actions described herein may occur out of the order noted in the sequence of actions or simultaneously. Furthermore, in some illustrations, some blocks, functions or actions may be optional and may or may not be executed; these are generally illustrated with das