Enhancement of safety communication model: Preserving the black channel concept

Frank Schiller; Dan Judd; Peerasan Supavatanakul; Tina Hardt; Felix Wieczorek

doi:10.1515/auto-2021-0098

Article

Enhancement of safety communication model

Preserving the black channel concept

Frank Schiller

Prof. Dr.-Ing. Frank Schiller is Scientific Director for Functional Safety at Beckhoff Automation. He studied Electrical Engineering at Technical University Dresden and earned his doctoral degree at Technical University Hamburg-Harburg (supervisor: Prof. Jan Lunze). Besides various positions at Siemens, he was professor for automation from 2004 until 2011 at Technical University Munich. His areas include safety communication, software-based safety controllers, and interdependencies between safety and security solutions. He is member of IEC TC65/WG12 “Profiles for functional safe communications in industrial networks” and leads its German mirror committee DKE 914.1. He teaches as visiting professor at East China University of Science and Technology, Shanghai, China.
, Dan Judd

Dan Judd is Chief Technical Officer for Arlington Laboratory Corporation, USA, a leading technology consulting company focused on industrial networks which he founded in 1993. He did his academic research at Massachusetts Institute of Technology (MIT) where he earned his degree in electrical engineering and was inducted by the Eta Kappa Nu honor society for outstanding scholastic achievement in electrical engineering. In 2005, he became active in the IEC fieldbus standards and today still serves as appointed expert from the US National Committee in working groups for protocol, installation, and functional safety. As an active contributing author, he was awarded the IEC 1906 Award in 2018 as “an important technical contributor to developments on extended models and equations”.
, Peerasan Supavatanakul

Dr.-Ing. Peerasan Supavatanakul is Senior Product Specialist and Technical Certifier for safety automation components at TÜV SÜD. He studied Mechatronics at Technical University Hamburg-Harburg and earned his doctoral degree at Ruhr University Bochum (supervisor: Prof. Jan Lunze). His areas of interest include safety communication, quantification of safety and reliability for safety products as well as evaluation of diagnostic approaches. He is member of IEC TC65/WG12 “Profiles for functional safe communications in industrial networks” and of its German mirror committee DKE 914.1.
, Tina Hardt

Dr.-Ing. Tina Hardt, née Mattes, studied Applied Mathematics at University of Trier and earned her doctoral degree at Technical University Munich (supervisor: Prof. Frank Schiller) in 2010. After parental leave, she worked from 2014 until 2016 at Trier University of Applied Sciences as a scientific assistant. In 2017, she joined the research and development department of Arend Process Automation in Wittlich where she worked on a cybersecure industrial edge-gateway. She was a member of the team generating the specification “Requirements and reference architecture of a security gateway for the exchange of industry data and services”, DIN SPEC 27070:2020-03. Since March 2021, she has been research coordinator in Arend’s subsidiary company Arendar IT-Security GmbH.
and Felix Wieczorek

Felix Wieczorek, M. Sc., is Senior Security Analyst at Siemens. He studied Informatics at Technical University Munich and earned his master’s degree in 2011. His research in combinations and interactions of safety and security communication started at Beckhoff Automation. Afterwards he worked as auditor in the field of industrial security at TÜV SÜD. Currently he is responsible for Vulnerability Intelligence at Siemens. He leads the DKE platform TBINK Safety & Security.

Published/Copyright: January 13, 2022

Published by

Become an author with De Gruyter Brill

Submit Manuscript Author Information Explore this Subject

From the journal at - Automatisierungstechnik Volume 70 Issue 1

Abstract

A fundamental measure of safety communication is the residual error probability, i. e., the probability of undetected errors. For the detection of data errors, typically a Cyclic Redundancy Check (CRC) is applied, and the resulting residual error probability is determined based on the Binary Symmetric Channel (BSC) model. The use of this model had been questioned since several error types cannot be sufficiently described. Especially the increasing introduction of security algorithms into underlying communication layers requires a more adequate channel model. This paper introduces an enhanced model that extends the list of considered data error types by combining the BSC model with a Uniformly Distributed Segments (UDS) model. Although models beyond BSC are applied, the hitherto method of the calculation of the residual error probability can be maintained.

Zusammenfassung

Ein grundlegendes Maß für die Safety-Kommunikation ist die Restfehlerwahrscheinlichkeit, d. h. die Wahrscheinlichkeit unentdeckter Fehler. Zur Erkennung von Datenfehlern wird typischerweise ein Cyclic Redundancy Check (CRC) verwendet und die daraus resultierende Restfehlerwahrscheinlichkeit auf der Grundlage des Modells des Binären Symmetrischen Kanals (BSC) bestimmt. Die Verwendung dieses Modells wurde gelegentlich in Frage gestellt, da manche Fehlertypen nicht ausreichend beschrieben werden können. Insbesondere der zunehmende Einsatz von Security-Algorithmen in den unteren Kommunikationsschichten erfordert ein adäquateres Kanalmodell. In diesem Beitrag wird ein erweitertes Modell vorgestellt, das die bisher berücksichtigten Datenfehlertypen durch die Kombination des BSC-Modells mit einem UDS-Modell (Uniformly Distributed Segments) erweitert. Obwohl Modelle angewendet werden, die über den BSC hinausgehen, kann die bisherige Methode zur Berechnung der Restfehlerwahrscheinlichkeit beibehalten werden.

Keywords: functional safety; safety communication; communication error; residual error probability; binomial distribution; uniform distribution

Schlagwörter: Funktionale Sicherheit; Safety-Kommunikation; Kommunikationsfehler; Restfehlerwahrscheinlichkeit; Binomialverteilung; Gleichverteilung

About the authors

Frank Schiller

Prof. Dr.-Ing. Frank Schiller is Scientific Director for Functional Safety at Beckhoff Automation. He studied Electrical Engineering at Technical University Dresden and earned his doctoral degree at Technical University Hamburg-Harburg (supervisor: Prof. Jan Lunze). Besides various positions at Siemens, he was professor for automation from 2004 until 2011 at Technical University Munich. His areas include safety communication, software-based safety controllers, and interdependencies between safety and security solutions. He is member of IEC TC65/WG12 “Profiles for functional safe communications in industrial networks” and leads its German mirror committee DKE 914.1. He teaches as visiting professor at East China University of Science and Technology, Shanghai, China.

Dan Judd

Dan Judd is Chief Technical Officer for Arlington Laboratory Corporation, USA, a leading technology consulting company focused on industrial networks which he founded in 1993. He did his academic research at Massachusetts Institute of Technology (MIT) where he earned his degree in electrical engineering and was inducted by the Eta Kappa Nu honor society for outstanding scholastic achievement in electrical engineering. In 2005, he became active in the IEC fieldbus standards and today still serves as appointed expert from the US National Committee in working groups for protocol, installation, and functional safety. As an active contributing author, he was awarded the IEC 1906 Award in 2018 as “an important technical contributor to developments on extended models and equations”.

Peerasan Supavatanakul

Dr.-Ing. Peerasan Supavatanakul is Senior Product Specialist and Technical Certifier for safety automation components at TÜV SÜD. He studied Mechatronics at Technical University Hamburg-Harburg and earned his doctoral degree at Ruhr University Bochum (supervisor: Prof. Jan Lunze). His areas of interest include safety communication, quantification of safety and reliability for safety products as well as evaluation of diagnostic approaches. He is member of IEC TC65/WG12 “Profiles for functional safe communications in industrial networks” and of its German mirror committee DKE 914.1.

Tina Hardt

Dr.-Ing. Tina Hardt, née Mattes, studied Applied Mathematics at University of Trier and earned her doctoral degree at Technical University Munich (supervisor: Prof. Frank Schiller) in 2010. After parental leave, she worked from 2014 until 2016 at Trier University of Applied Sciences as a scientific assistant. In 2017, she joined the research and development department of Arend Process Automation in Wittlich where she worked on a cybersecure industrial edge-gateway. She was a member of the team generating the specification “Requirements and reference architecture of a security gateway for the exchange of industry data and services”, DIN SPEC 27070:2020-03. Since March 2021, she has been research coordinator in Arend’s subsidiary company Arendar IT-Security GmbH.

Felix Wieczorek

Felix Wieczorek, M. Sc., is Senior Security Analyst at Siemens. He studied Informatics at Technical University Munich and earned his master’s degree in 2011. His research in combinations and interactions of safety and security communication started at Beckhoff Automation. Afterwards he worked as auditor in the field of industrial security at TÜV SÜD. Currently he is responsible for Vulnerability Intelligence at Siemens. He leads the DKE platform TBINK Safety & Security.

Acknowledgment

The authors thank Mr. Xiaobo Peng from Rockwell Automation, Shanghai, member of IEC TC65/WG12 “Profiles for functional safe communications in industrial networks”, for his constructive criticism.

Appendix A Residual error probability with probability of occurrence of a combined UDS fault

For introduction of a combined UDS fault, f UDS ↔ f UDS 1 ∨ f UDS 2 , eqn. (7), Section 5.4, is transformed:

P r e BSC / UDS 1 / UDS 2 ≤ P r e Binom ( p = p BSC ) · P ( ¬ f UDS 1 ∧ ¬ f UDS 2 ) + P r e Binom ( p = 0.25 + 0.5 · p BSC ) · P ( f UDS 1 ∧ ¬ f UDS 2 ) + 2 − r · P ( f UDS 2 ) ≤ P r e Binom ( p = p BSC ) · P ( ¬ f UDS 1 ∧ ¬ f UDS 2 ) + max P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r · P ( f UDS 1 ∧ ¬ f UDS 2 ) + P ( f UDS 2 ) ≤ P r e Binom ( p = p BSC ) · P ( ¬ f UDS 1 ∧ ¬ f UDS 2 ) + max P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r · P ( f UDS 1 ∨ f UDS 2 ) ≤ P r e Binom ( p = p BSC ) · P ( ¬ f UDS ) + max P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r · P ( f UDS ) ≤ P r e Binom ( p = p BSC ) · ( 1 − P ( f UDS ) ) + max ( P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r ) · P ( f UDS ) .

In the following, the bit error probability of BSC, p BSC , is varied between 0 and a maximum value p max BSC :

(11) P r e BSC / UDS 1 / UDS 2 ≤ max 0 ≤ p BSC ≤ p max BSC P r e Binom ( p = p BSC ) · ( 1 − P ( f UDS ) ) + max ( P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r ) · P ( f UDS ) ≤ ? max 0 ≤ p BSC ≤ p max BSC P r e Binom ( p = p BSC ) · ( 1 − P ( f UDS ) ) + max 0 ≤ p BSC ≤ p max BSC ( P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r ) · P ( f UDS ) .

Consider any 2 values of the bit error probability, p 1 BSC and p 2 BSC , of the interval [ 0 , p max BSC ], and the following abbreviations:

P r e Binom ( p = p 1 BSC ) · ( 1 − P ( f UDS ) ) = a ( 1 ) , P r e Binom ( p = p 2 BSC ) · ( 1 − P ( f UDS ) ) = a ( 2 ) , max ( P r e Binom ( p = 0.25 + 0.5 · p 1 BSC ) , 2 − r ) · P ( f UDS ) = b ( 1 ) , max ( P r e Binom ( p = 0.25 + 0.5 · p 2 BSC ) , 2 − r ) · P ( f UDS ) = b ( 2 ) ,

then statement (11) can be denoted by

(12) max ( a ( 1 ) + b ( 1 ) , a ( 2 ) + b ( 2 ) ) ≤ ? max ( a ( 1 ) , a ( 2 ) ) + max ( b ( 1 ) , b ( 2 ) ) .

Case 1. a ( 1 ) ≥ a ( 2 )

max ( a ( 1 ) + b ( 1 ) , a ( 2 ) + b ( 2 ) ) ≤ ? a ( 1 ) + max ( b ( 1 ) , b ( 2 ) ) , max ( b ( 1 ) , a ( 2 ) − a ( 1 ) ︸ ≤ 0 + b ( 2 ) ︸ ≤ b ( 2 ) ) ≤ ? max ( b ( 1 ) , b ( 2 ) ) .

Case 2. a ( 1 ) ≤ a ( 2 )

max ( a ( 1 ) + b ( 1 ) , a ( 2 ) + b ( 2 ) ) ≤ ? a ( 2 ) + max ( b ( 1 ) , b ( 2 ) ) , max ( a ( 1 ) − a ( 2 ) ︸ ≤ 0 + b ( 1 ) ︸ ≤ b ( 1 ) , b ( 2 ) ) ≤ ? max ( b ( 1 ) , b ( 2 ) ) .

In both cases, eqn. (12) is true. Therefore, the upper estimation

P r e BSC / UDS 1 / UDS 2 ≤ max 0 ≤ p BSC ≤ p max BSC P r e Binom ( p = p BSC ) · ( 1 − P ( f UDS ) ) + max 0 ≤ p BSC ≤ p max BSC ( P r e Binom ( p = 0.25 + 0.5 · p BSC ) , 2 − r ) · P ( f UDS )

and finally, eqn. (10) hold.

References

1. Bertsekas, D.P. and R.G. Gallager. 1992. Data Networks. Prentice Hall.Search in Google Scholar

2. Castagnoli, G. 1989. On the Minimum Distance of Long Cyclic Codes and Cyclic Redundancy Check Codes. PhD thesis, ETH Zurich, Dissertation No. 8979.Search in Google Scholar

3. Chun, D. and J.K. Wolf. 1994. Special Hardware for Computing the Probability of Certain Binary CRC Codes and Test Results. IEEE Transactions on Communications 42(10): 2769–2772.10.1109/26.328943Search in Google Scholar

4. Di Natale, M., H. Zeng, P. Giusto and A. Ghosal. 2012. Understanding and Using the Controller Area Network Communication Protocol. Springer.10.1007/978-1-4614-0314-2Search in Google Scholar

5. Gilbert, E.N. 1960. Capacity of Burst-Noise Channel. The Bell System Technical Journal 39(5): 1253–1265 Nokia Bell Labs.10.1002/j.1538-7305.1960.tb03959.xSearch in Google Scholar

6. Horch, A., H. Hannen, H. Schween and K. Wagner. 2019. Verschlüsselung sicherer Kommunikation. atp magazin 06-07/2019: 93–99 (in German).10.17560/atp.v61i6-7.2412Search in Google Scholar

7. IEC: Functional Safety of Electrical/Electronic/Programmable Electronic Safety-related Systems, Part 1, IEC 61508-1, 2010.Search in Google Scholar

8. IEC: Safety of machinery: Functional safety of electrical, electronic, and programmable electronic control systems, IEC 62061, 2021.Search in Google Scholar

9. IEC: Functional safety fieldbuses – General rules and profile definitions, IEC 61784-3, 2021.Search in Google Scholar

10. Lin, S. and D. Costello. 2004. Error Control Coding. Pearson Prentice Hall, Pearson Education.Search in Google Scholar

11. Peterson, W.W. and D.T. Brown. 1961. Cyclic Codes for Error Detection. Proceedings of the IRE 49: 228–235.10.1109/JRPROC.1961.287814Search in Google Scholar

12. Peterson, W.W. and E.J. Weldon. 1961. Error-correction Codes. MIT Press.Search in Google Scholar

13. Schiller, F. and T. Mattes. 2006. An Efficient Method to Evaluate CRC-Polynomials for Safety-Critical Industrial Communication. Journal of Applied Computer Science 14(1): 57–80. Technical University Press, Łódź, Poland.Search in Google Scholar

14. Schiller, F. and T. Mattes. 2006. Analysis of CRC-polynomials for Safety-critical Communication by Deterministic and Stochastic Automata. In: 6th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, SAFEPROCESS 2006, Beijing, China, pp. 1003–1008.10.1016/B978-008044485-7/50159-7Search in Google Scholar

15. Schiller, F., T. Mattes, H. Büttner and J. Sachs. 2007. In: A New Method to Obtain Sufficient Independency of Nested Cyclic Redundancy Checks. 5^th Int. Conference on Safety of Industrial Automated Systems, SIAS 2007, Tokyo, Japan, pp. 149–154.Search in Google Scholar

16. Schiller, F. and F. Wieczorek. 2020. Safety-Analyse für Security-geschützte Kommunikation. atp magazin 04/2020: 86–92 (in German).10.17560/atp.v62i4.2474Search in Google Scholar

17. Wacker, H.D. and J. Boercsoek. 2008. Binomial and monotonic behaviour of the probability of undetected error and the 2 − r -bound. WSEAS Transactions on Communications 7: 188–197.Search in Google Scholar

Received: 2021-07-01

Accepted: 2021-11-19

Published Online: 2022-01-13

Published in Print: 2022-01-27

You are currently not able to access this content.

Articles in the same Issue

https://doi.org/10.1515/auto-2021-0098

Keywords for this article

functional safety; safety communication; communication error; residual error probability; binomial distribution; uniform distribution