WO2024005799A1 - A system on a chip comprising a diagnostics module - Google Patents

A system on a chip comprising a diagnostics module Download PDF

Info

Publication number
WO2024005799A1
WO2024005799A1 PCT/US2022/035390 US2022035390W WO2024005799A1 WO 2024005799 A1 WO2024005799 A1 WO 2024005799A1 US 2022035390 W US2022035390 W US 2022035390W WO 2024005799 A1 WO2024005799 A1 WO 2024005799A1
Authority
WO
WIPO (PCT)
Prior art keywords
analytics data
soc
computer
integrated circuit
analytics
Prior art date
Application number
PCT/US2022/035390
Other languages
French (fr)
Inventor
Peter Robertson
Original Assignee
Siemens Industry Software Inc.
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Siemens Industry Software Inc. filed Critical Siemens Industry Software Inc.
Priority to PCT/US2022/035390 priority Critical patent/WO2024005799A1/en
Publication of WO2024005799A1 publication Critical patent/WO2024005799A1/en

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28—Testing of electronic circuits, e.g. by signal tracer
    • G01R31/317—Testing of digital circuits
    • G01R31/3181—Functional testing
    • G01R31/3183—Generation of test inputs, e.g. test vectors, patterns or sequences
    • G01R31/318371—Methodologies therefor, e.g. algorithms, procedures
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
    • G01R31/28—Testing of electronic circuits, e.g. by signal tracer
    • G01R31/2832—Specific tests of electronic circuits not provided for elsewhere
    • G01R31/2836—Fault-finding or characterising
    • G01R31/2846—Fault-finding or characterising using hard- or software simulation or using knowledge-based systems, e.g. expert systems, artificial intelligence or interactive algorithms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G06N3/0455—Auto-encoder networks; Encoder-decoder networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods

Definitions

  • the subject-matter of the present disclosure relates to a system on a chip (SoC) including a diagnostics module, a computer-implemented method of detecting a fault condition of an integrated circuit of a SoC, a computer- implemented method of training a machine learning algorithm to detect an anomaly in analytics data indicative of a fault condition of an integrated circuit of a SoC, and a transitory, or non-transitory, computer readable medium.
  • SoC system on a chip
  • a device including a SoC may be subject to anomalous behaviour either from an external source or tampering with the device itself.
  • Anomalous activities are inherently difficult to identify as anomalous activities are sparse, unpredictable, and include any behaviour that is not typical of the device. Anomalous activities are thus difficult to quantify. Typically, this would require data to be collected and exported off chip to be examined by an individual with significant insight into the normal working behaviour of the chip and prior knowledge that an anomalous event occurred with a rough estimate of the time at which the anomalous event occurred. The collection of this data may not be possible in real world applications as this would require knowledge of ongoing anomalous behaviour (e.g., observation of the ongoing anomalous behaviour and collection of data of the ongoing anomalous behaviour). Even if such data could be obtained, this analysis remains unable to react and prevent further unexpected behaviour on the device, as the analysis is happening off chip after the event in question.
  • the anomalies are to be identified at the time of occurrence, allowing preventative measures to be taken or real-time notifications to be made.
  • a system on a chip includes an integrated circuit, an embedded analytics monitor configured to generate analytics data by monitoring one or more interactions within the integrated circuit, and a diagnostics module including a machine learning algorithm trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit.
  • the machine learning model may be trained to encode the analytics data to a descriptor in a latent space, and to decode the descriptor to generate decoded analytics data.
  • the diagnostics module may be configured to compare the decoded analytics data to the analytics data to generate a difference.
  • the diagnostics module may be configured to classify the analytics data as anomalous when the difference is above a threshold, and to classify the analytics data as non-anomalous when the difference is below the threshold.
  • the difference may be a distance
  • the diagnostics module may be configured to calculate the distance using a Euclidian distance metric or using cosine similarity.
  • the machine learning model may include an autoencoder.
  • a computer- implemented method of detecting a fault condition of an integrated circuit of a system on a chip includes generating analytics data by monitoring one or more interactions within the integrated circuit, and detecting, by a machine learning algorithm, an anomaly in the analytics data indicative of a fault condition of the integrated circuit.
  • the computer-implemented method may further include comparing, by the diagnostics module, the decoded analytics data to the analytics data, generating a difference based on the comparing, classifying the analytics data as anomalous when the difference is above a threshold, and classifying the analytics data as non-anomalous when the difference is below the threshold.
  • the difference may be a distance.
  • the computer-implemented method may further comprise calculating, by the diagnostics module, the distance using a Euclidian distance metric or using cosine similarity.
  • the machine learning model may include an autoencoder.
  • a computer- implemented method of training a machine learning algorithm to detect an anomaly in analytics data, generated by an embedded analytics monitor, indicative of a fault condition of an integrated circuit of a system on a chip includes providing a plurality of analytics data values from the embedded analytics monitor of the SoC.
  • the plurality of data values are obtained by monitoring interactions within the integrated circuit during normal, non- anomalous, operation.
  • the computer-implemented method includes encoding the analytics data values to a descriptor in a latent space, decoding the descriptor to generate decoded analytics data, determining an error between the analytics data and the decoded analytics data, and modifying the machine learning algorithm to reduce the error.
  • the machine learning model may include an autoencoder.
  • a transitory, or non- transitory, computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to perform the method as described above is provided.
  • Figure 1 shows a schematic block diagram of a System on a Chip (SoC) according to an embodiment of the present invention
  • Figure 2 shows a schematic block diagram of a machine learning model from the SoC of Figure 1 , in the form of an autoencoder
  • Figure 3 shows a flow chart of a computer-implemented method of training the machine learning model from Figure 2;
  • Figure 4 shows a flow chart of a computer-implemented method of detecting a fault condition of an integrated circuit of the SoC from Figure 1 .
  • Any methods described herein may be computer-implemented methods.
  • the computer-implemented methods may be provided on a transitory, or non-transitory, computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to perform the method.
  • the processor may be a processor of a computer that also includes storage.
  • the non-transitory computer readable medium may be store in the storage.
  • a SoC 10 includes an integrated circuit 12, an embedded analytics monitor 14, and a diagnostics module 16.
  • the SoC 10 may be any of a variety of different chips for myriad different devices.
  • the SoC 10 may be a traffic bus, for example.
  • a traffic bus is described here for illustrative purposes only and should not be construed as the only type of SoC to which the present subject-matter is applicable.
  • the integrated circuit 12 a conventional integrated circuit where interactions between circuit components occurforthe integrated circuit to perform various operations. In the case of the SoC 10 being a traffic bus, the interactions may include read and/or write interactions.
  • the embedded analytics monitor 14 is independent of the integrated circuit. For example, operating the embedded analytics monitor 14 does not impact on operation of the integrated circuit 12, and vice-versa.
  • the embedded analytics monitor 14 monitors the interactions within the integrated circuit 12.
  • the embedded analytics monitor 14 operates in real time.
  • the embedded analytics monitor 14 monitors the interactions of the integrated circuit to generate analytics data by sampling data values associated with the interactions over a period of time.
  • the diagnostics module 16 includes a machine learning algorithm.
  • the machine learning algorithm is trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit 12.
  • the anomaly may include detecting a read operation when a write operation was expected, and vice-versa.
  • the machine learning algorithm includes an autoencoder 20.
  • the autoencoder 20 includes an encoder
  • the encoder 22 is configured to encode the analytics data
  • the decoder 24 is configured to decode the descriptor to generate decoded analytics data 27.
  • the diagnostics module ( Figure 1 ) includes a data comparison module 26.
  • the data comparison module 26 is configured to compare the decoded analytics data to the analytics data to generate a difference.
  • the data represented in this reduced latent space 25 is used to reconstruct the input data.
  • the decoded analytics data 27 derived from the compressed representation should closely resemble the embedded analytics data 23.
  • the assumption that the data may be reconstructed from the latent space is broken, as the statistical properties of the embedded analytics data 23 without anomalies are different from that which contains anomalies. This is then apparent in the distance metric between the embedded analytics data 23 and the decoded analytics data 27 and may be detected as an anomaly.
  • the data comparison module 26 of the diagnostics module is configured to classify the analytics data as anomalous when a difference is above a threshold.
  • the data comparison module 26 is also configured to classify the analytics data as non-anomalous when the difference is below the threshold.
  • the difference may be a distance.
  • the distance may be calculated by the data comparison module 26 of the diagnostics module using a Euclidian distance metric or using cosine similarity.
  • the machine learning model of the embodiment of Figure 2 may be trained using a computer-implemented method such that the machine learning model detects an anomaly in analytics data, generated by an embedded analytics monitor, indicative of a fault condition of an integrated circuit of a system on a chip (SoC 10).
  • the computer-implemented method includes providing S100 a plurality of analytics data values from the embedded analytics monitor of the SoC 10. The plurality of data values are obtained by monitoring interactions within the integrated circuit 12 during normal, non-anomalous, operation.
  • the computer-implemented method also includes encoding S102 the analytics data values to a descriptor in a latent space.
  • the computer-implemented method also includes decoding S104 the descriptor to generate decoded analytics data.
  • the computer-implemented method also includes determining S106 an error between the analytics data and the decoded analytics data.
  • the computer-implemented method also includes modifying S108 the machine learning algorithm to reduce the error.
  • the invention may be captured, broadly speaking, as a computer-implemented method of detecting a fault condition of an integrated circuit of a system on a chip (SoC).
  • the method may include generating S300 analytics data by monitoring one or more interactions within the integrated circuit 12, and detecting S302, by a machine learning algorithm, an anomaly in the analytics data indicative of a fault condition of the integrated circuit 12.
  • Detecting anomalies on the SoC itself using the machine learning algorithm enables real-time flagging and potentially resolution of the anomaly that would otherwise only be done offline after the SoC has finished operating. Realtime detection may be important for systems such as autonomous vehicles that are to react in real-time.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Theoretical Computer Science (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Medical Informatics (AREA)
  • Debugging And Monitoring (AREA)

Abstract

A system on a chip (SoC) (10) is provided. The SoC (10) includes an integrated circuit (12), an embedded analytics monitor (14) configured to generate analytics data by monitoring one or more interactions within the integrated circuit (12), and a diagnostics module (16) including a machine learning algorithm trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit (12).

Description

A SYSTEM ON A CHIP COMPRISING A DIAGNOSTICS MODULE
FIELD
[0001] The subject-matter of the present disclosure relates to a system on a chip (SoC) including a diagnostics module, a computer-implemented method of detecting a fault condition of an integrated circuit of a SoC, a computer- implemented method of training a machine learning algorithm to detect an anomaly in analytics data indicative of a fault condition of an integrated circuit of a SoC, and a transitory, or non-transitory, computer readable medium.
BACKGROUND
[0002] A device including a SoC may be subject to anomalous behaviour either from an external source or tampering with the device itself.
[0003] Anomalous activities are inherently difficult to identify as anomalous activities are sparse, unpredictable, and include any behaviour that is not typical of the device. Anomalous activities are thus difficult to quantify. Typically, this would require data to be collected and exported off chip to be examined by an individual with significant insight into the normal working behaviour of the chip and prior knowledge that an anomalous event occurred with a rough estimate of the time at which the anomalous event occurred. The collection of this data may not be possible in real world applications as this would require knowledge of ongoing anomalous behaviour (e.g., observation of the ongoing anomalous behaviour and collection of data of the ongoing anomalous behaviour). Even if such data could be obtained, this analysis remains unable to react and prevent further unexpected behaviour on the device, as the analysis is happening off chip after the event in question.
[0004] To prevent these anomalies from causing harm to a system, the anomalies are to be identified at the time of occurrence, allowing preventative measures to be taken or real-time notifications to be made. SUMMARY
[0005] It is an aim of the present disclosure to address the issues discussed above and improve on the prior art.
[0006] According to an aspect of the present invention, a system on a chip (SoC) includes an integrated circuit, an embedded analytics monitor configured to generate analytics data by monitoring one or more interactions within the integrated circuit, and a diagnostics module including a machine learning algorithm trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit.
[0007] In an embodiment, the machine learning model may be trained to encode the analytics data to a descriptor in a latent space, and to decode the descriptor to generate decoded analytics data.
[0008] In an embodiment, the diagnostics module may be configured to compare the decoded analytics data to the analytics data to generate a difference. The diagnostics module may be configured to classify the analytics data as anomalous when the difference is above a threshold, and to classify the analytics data as non-anomalous when the difference is below the threshold.
[0009] In an embodiment, the difference may be a distance.
[0010] In an embodiment, the diagnostics module may be configured to calculate the distance using a Euclidian distance metric or using cosine similarity.
[0011] In an embodiment, the machine learning model may include an autoencoder.
[0012] According to an aspect of the present invention, a computer- implemented method of detecting a fault condition of an integrated circuit of a system on a chip (SoC) includes generating analytics data by monitoring one or more interactions within the integrated circuit, and detecting, by a machine learning algorithm, an anomaly in the analytics data indicative of a fault condition of the integrated circuit.
[0013] In an embodiment, the computer-implemented method may further include comparing, by the diagnostics module, the decoded analytics data to the analytics data, generating a difference based on the comparing, classifying the analytics data as anomalous when the difference is above a threshold, and classifying the analytics data as non-anomalous when the difference is below the threshold.
[0014] In an embodiment, the difference may be a distance.
[0015] In an embodiment, the computer-implemented method may further comprise calculating, by the diagnostics module, the distance using a Euclidian distance metric or using cosine similarity.
[0016] In an embodiment, the machine learning model may include an autoencoder.
[0017] According to an aspect of the present disclosure, a computer- implemented method of training a machine learning algorithm to detect an anomaly in analytics data, generated by an embedded analytics monitor, indicative of a fault condition of an integrated circuit of a system on a chip (SoC) includes providing a plurality of analytics data values from the embedded analytics monitor of the SoC. The plurality of data values are obtained by monitoring interactions within the integrated circuit during normal, non- anomalous, operation. The computer-implemented method includes encoding the analytics data values to a descriptor in a latent space, decoding the descriptor to generate decoded analytics data, determining an error between the analytics data and the decoded analytics data, and modifying the machine learning algorithm to reduce the error.
[0018] In an embodiment, the machine learning model may include an autoencoder.
[0019] According to an aspect of the present invention, a transitory, or non- transitory, computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to perform the method as described above is provided. BRIEF DESCRIPTION OF DRAWINGS
[0020] The embodiments herein may be best understood with reference to the accompanying drawings, in which:
[0021] Figure 1 shows a schematic block diagram of a System on a Chip (SoC) according to an embodiment of the present invention;
[0022] Figure 2 shows a schematic block diagram of a machine learning model from the SoC of Figure 1 , in the form of an autoencoder;
[0023] Figure 3 shows a flow chart of a computer-implemented method of training the machine learning model from Figure 2; and
[0024] Figure 4 shows a flow chart of a computer-implemented method of detecting a fault condition of an integrated circuit of the SoC from Figure 1 .
DESCRIPTION OF EMBODIMENTS
[0025] Any methods described herein may be computer-implemented methods. The computer-implemented methods may be provided on a transitory, or non-transitory, computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to perform the method. The processor may be a processor of a computer that also includes storage. The non-transitory computer readable medium may be store in the storage.
[0026] With reference to Figure 1 , a SoC 10 includes an integrated circuit 12, an embedded analytics monitor 14, and a diagnostics module 16.
[0027] The SoC 10 may be any of a variety of different chips for myriad different devices. The SoC 10 may be a traffic bus, for example. However, a traffic bus is described here for illustrative purposes only and should not be construed as the only type of SoC to which the present subject-matter is applicable.
[0028] The integrated circuit 12 a conventional integrated circuit where interactions between circuit components occurforthe integrated circuit to perform various operations. In the case of the SoC 10 being a traffic bus, the interactions may include read and/or write interactions. [0029] The embedded analytics monitor 14 is independent of the integrated circuit. For example, operating the embedded analytics monitor 14 does not impact on operation of the integrated circuit 12, and vice-versa.
[0030] The embedded analytics monitor 14 monitors the interactions within the integrated circuit 12. The embedded analytics monitor 14 operates in real time. The embedded analytics monitor 14 monitors the interactions of the integrated circuit to generate analytics data by sampling data values associated with the interactions over a period of time.
[0031] The diagnostics module 16 includes a machine learning algorithm. The machine learning algorithm is trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit 12.
[0032] Using the illustrative example of a SoC being a data bus, the anomaly may include detecting a read operation when a write operation was expected, and vice-versa.
[0033] With reference to Figure 2, in an embodiment, the machine learning algorithm includes an autoencoder 20. The autoencoder 20 includes an encoder
22 and a decoder 24. The encoder 22 is configured to encode the analytics data
23 to a descriptor in a latent space 25. The decoder 24 is configured to decode the descriptor to generate decoded analytics data 27.
[0034] The diagnostics module (Figure 1 ) includes a data comparison module 26. The data comparison module 26 is configured to compare the decoded analytics data to the analytics data to generate a difference.
[0035] The data represented in this reduced latent space 25 is used to reconstruct the input data. Provided that the input data contained no anomalous behavior, the decoded analytics data 27 derived from the compressed representation should closely resemble the embedded analytics data 23. By introducing anomalous data, the assumption that the data may be reconstructed from the latent space is broken, as the statistical properties of the embedded analytics data 23 without anomalies are different from that which contains anomalies. This is then apparent in the distance metric between the embedded analytics data 23 and the decoded analytics data 27 and may be detected as an anomaly.
[0036] In this way, the data comparison module 26 of the diagnostics module is configured to classify the analytics data as anomalous when a difference is above a threshold. The data comparison module 26 is also configured to classify the analytics data as non-anomalous when the difference is below the threshold. The difference may be a distance. The distance may be calculated by the data comparison module 26 of the diagnostics module using a Euclidian distance metric or using cosine similarity.
[0037] With reference to Figure 3, the machine learning model of the embodiment of Figure 2 may be trained using a computer-implemented method such that the machine learning model detects an anomaly in analytics data, generated by an embedded analytics monitor, indicative of a fault condition of an integrated circuit of a system on a chip (SoC 10). The computer-implemented method includes providing S100 a plurality of analytics data values from the embedded analytics monitor of the SoC 10. The plurality of data values are obtained by monitoring interactions within the integrated circuit 12 during normal, non-anomalous, operation. The computer-implemented method also includes encoding S102 the analytics data values to a descriptor in a latent space. The computer-implemented method also includes decoding S104 the descriptor to generate decoded analytics data. The computer-implemented method also includes determining S106 an error between the analytics data and the decoded analytics data. In addition, the computer-implemented method also includes modifying S108 the machine learning algorithm to reduce the error.
[0038] With reference to Figure 4, the invention may be captured, broadly speaking, as a computer-implemented method of detecting a fault condition of an integrated circuit of a system on a chip (SoC). The method may include generating S300 analytics data by monitoring one or more interactions within the integrated circuit 12, and detecting S302, by a machine learning algorithm, an anomaly in the analytics data indicative of a fault condition of the integrated circuit 12. [0039] Detecting anomalies on the SoC itself using the machine learning algorithm enables real-time flagging and potentially resolution of the anomaly that would otherwise only be done offline after the SoC has finished operating. Realtime detection may be important for systems such as autonomous vehicles that are to react in real-time.
[0040] Attention is directed to all papers and documents that are filed concurrently with or previous to this specification in connection with this application and that are open to public inspection with this specification, and the contents of all such papers and documents are incorporated herein by reference.
[0041] All of the features disclosed in this specification (including any accompanying claims, abstract and drawings), and/or all of the steps of any method or process so disclosed, may be combined in any combination, except combinations where at least some of such features and/or steps are mutually exclusive.
[0042] Each feature disclosed in this specification (including any accompanying claims, abstract, and drawings) may be replaced by alternative features serving the same, equivalent, or similar purpose, unless expressly stated otherwise. Thus, unless expressly stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.
[0043] The invention is not restricted to the details of the foregoing embodiment(s). The invention extends to any novel one, or any novel combination, of the features disclosed in this specification (including any accompanying claims, abstract, and drawings), or to any novel one, or any novel combination, of the steps of any method or process so disclosed.

Claims

1 . A system on a chip (SoC) (10) comprising: an integrated circuit (12); an embedded analytics monitor (14) configured to generate analytics data by monitoring one or more interactions within the integrated circuit (12); and a diagnostics module (16) comprising a machine learning algorithm trained to detect an anomaly in the analytics data indicative of a fault condition of the integrated circuit (12).
2. The SoC of Claim 1 , wherein the machine learning model is trained to encode the analytics data to a descriptor in a latent space, and to decode the descriptor to generate decoded analytics data.
3. The SoC of Claim 2, wherein the diagnostics module (16) is configured to compare the decoded analytics data (27) to the analytics data (23) to generate a difference, and is configured to classify the analytics data (23) as anomalous when the difference is above a threshold, and to classify the analytics data as non-anomalous when the difference is below the threshold.
4. The SoC of Claim 3, wherein the difference is a distance.
5. The SoC of Claim 4, wherein the diagnostics module (16) is configured to calculate the distance using a Euclidian distance metric or using cosine similarity.
6. The SoC of any of Claims 2 to 5, wherein the machine learning model comprises an autoencoder (20).
7. A computer-implemented method of detecting a fault condition of an integrated circuit of a system on a chip (SoC), the computer-implemented method comprising: generating (S100) analytics data by monitoring one or more interactions within the integrated circuit (12); and detecting (S102), by a machine learning algorithm, an anomaly in the analytics data indicative of a fault condition of the integrated circuit (12).
8. The computer-implemented method of Claim 7, wherein the machine learning model is trained to encode the analytics data (23) to a descriptor in a latent space, and to decode the descriptor to generate decoded analytics data (27).
9. The computer-implemented method of Claim 8, further comprising: comparing, by a diagnostics module of the SoC (10), the decoded analytics data (27) to the analytics data (23) to generate a difference; classifying the analytics data as anomalous when the difference is above a threshold; and classifying the analytics data as non-anomalous when the difference is below the threshold.
10. The computer-implemented method of Claim 9, wherein the difference is a distance.
11. The computer-implemented method of Claim 8, further comprising calculating, by the diagnostics module (16), the distance using a Euclidian distance metric or using cosine similarity.
12. The computer-implemented method of any of Claims 8 to 12, wherein the machine learning model comprises an autoencoder (20).
13. A computer-implemented method of training a machine learning algorithm to detect an anomaly in analytics data, generated by an embedded analytics monitor, indicative of a fault condition of an integrated circuit of a system on a chip (SoC) (10), the computer-implemented method comprising: providing (S200) a plurality of analytics data values from the embedded analytics monitor (14) of the SoC (10), the plurality of data values having been obtained by monitoring interactions within the integrated circuit (12) during normal, non-anomalous, operation; encoding (S202) the analytics data values (23) to a descriptor in a latent space; decoding (S204) the descriptor to generate decoded analytics data (27); determining (S206) an error between the analytics data (23) and the decoded analytics data (27); and modifying (S208) the machine learning algorithm to reduce the error.
14. The computer-implemented method of Claim 8, wherein the machine learning model comprises an autoencoder (20).
15. A transitory, or non-transitory, computer-readable medium having instructions stored thereon that, when executed by a processor, cause the processor to perform the method of any of Claims 7 to 14.
PCT/US2022/035390 2022-06-28 2022-06-28 A system on a chip comprising a diagnostics module WO2024005799A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/US2022/035390 WO2024005799A1 (en) 2022-06-28 2022-06-28 A system on a chip comprising a diagnostics module

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/US2022/035390 WO2024005799A1 (en) 2022-06-28 2022-06-28 A system on a chip comprising a diagnostics module

Publications (1)

Publication Number Publication Date
WO2024005799A1 true WO2024005799A1 (en) 2024-01-04

Family

ID=82702888

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2022/035390 WO2024005799A1 (en) 2022-06-28 2022-06-28 A system on a chip comprising a diagnostics module

Country Status (1)

Country Link
WO (1) WO2024005799A1 (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080040089A1 (en) * 2006-07-18 2008-02-14 Wendemagagnehu Beyene Efficient Characterization of High-Speed Circuits
US20080141072A1 (en) * 2006-09-21 2008-06-12 Impact Technologies, Llc Systems and methods for predicting failure of electronic systems and assessing level of degradation and remaining useful life
US20170031792A1 (en) * 2014-04-16 2017-02-02 Siemens Aktiengesellschaft Transferring failure samples using conditional models for machine condition monitoring
US20190121350A1 (en) * 2016-05-09 2019-04-25 Strong Force Iot Portfolio 2016, Llc Systems and methods for learning data patterns predictive of an outcome
US20200125959A1 (en) * 2019-12-19 2020-04-23 Intel Corporation Autoencoder Neural Network for Signal Integrity Analysis of Interconnect Systems
US20200394506A1 (en) * 2019-06-14 2020-12-17 Robert Bosch Gmbh Machine learning system
US20220171985A1 (en) * 2020-12-01 2022-06-02 International Business Machines Corporation Item recommendation with application to automated artificial intelligence

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080040089A1 (en) * 2006-07-18 2008-02-14 Wendemagagnehu Beyene Efficient Characterization of High-Speed Circuits
US20080141072A1 (en) * 2006-09-21 2008-06-12 Impact Technologies, Llc Systems and methods for predicting failure of electronic systems and assessing level of degradation and remaining useful life
US20170031792A1 (en) * 2014-04-16 2017-02-02 Siemens Aktiengesellschaft Transferring failure samples using conditional models for machine condition monitoring
US20190121350A1 (en) * 2016-05-09 2019-04-25 Strong Force Iot Portfolio 2016, Llc Systems and methods for learning data patterns predictive of an outcome
US20200394506A1 (en) * 2019-06-14 2020-12-17 Robert Bosch Gmbh Machine learning system
US20200125959A1 (en) * 2019-12-19 2020-04-23 Intel Corporation Autoencoder Neural Network for Signal Integrity Analysis of Interconnect Systems
US20220171985A1 (en) * 2020-12-01 2022-06-02 International Business Machines Corporation Item recommendation with application to automated artificial intelligence

Similar Documents

Publication Publication Date Title
CN111967571B (en) Abnormality detection method and device based on MHMA
US20150346066A1 (en) Asset Condition Monitoring
CA2931624A1 (en) Systems and methods for event detection and diagnosis
US20140258187A1 (en) Generating database cluster health alerts using machine learning
CN111080502A (en) Big data identification method for abnormal behavior of regional enterprise data
CN113518011A (en) Abnormality detection method and apparatus, electronic device, and computer-readable storage medium
CN110222765B (en) Method and system for monitoring health state of permanent magnet synchronous motor
CN113066254B (en) Nuclear power equipment working environment abnormity early warning method, device, equipment and storage medium
CN111444072B (en) Abnormality identification method and device for client, computer equipment and storage medium
EP3910437B1 (en) Monitoring apparatus, monitoring method, and computer-readable medium
KR20210017651A (en) Method for Fault Detection and Fault Diagnosis in Semiconductor Manufacturing Process
CN117251812A (en) High-voltage power line operation fault detection method based on big data analysis
CN115454778A (en) Intelligent monitoring system for abnormal time sequence indexes in large-scale cloud network environment
CN108306997B (en) Domain name resolution monitoring method and device
CN113609115A (en) Data cleaning method for slope deformation monitoring data
CN110324207B (en) Detection method and device for data acquisition terminal
CN117336055A (en) Network abnormal behavior detection method and device, electronic equipment and storage medium
US20220156586A1 (en) System for monitoring a circuit breaker
CN116975938B (en) Sensor data processing method in product manufacturing process
WO2024005799A1 (en) A system on a chip comprising a diagnostics module
CN117473436A (en) Preprocessing and alarming method and system for instant data stream
CN111555917A (en) Alarm information processing method and device based on cloud platform
WO2024005797A1 (en) A system on a chip comprising a diagnostics module
KR102486463B1 (en) Method and Apparatus for Real Time Fault Detection Using Time series data According to Degradation
CN111159961B (en) Anomaly detection method and system based on curve data

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: 22747849

Country of ref document: EP

Kind code of ref document: A1