First Study of f the Proactive Transmission of Replicated Frames - - PowerPoint PPT Presentation

▶
first study of f the proactive
SMART_READER_LITE
LIVE PREVIEW

First Study of f the Proactive Transmission of Replicated Frames - - PowerPoint PPT Presentation

Dec 23, 2022 562 likes •907 views

First Study of f the Proactive Transmission of Replicated Frames Mechanism over TSN Ins lvarez, Drago avka , Julin Proenza, Manuel Barranco 2 Introduction Time-Sensitive Networking (TSN) Task Group. Developing a set of standards

slide-1
SLIDE 1

First Study of f the Proactive Transmission of Replicated Frames Mechanism over TSN

Inés Álvarez, Drago Čavka, Julián Proenza, Manuel Barranco

slide-2
SLIDE 2

Introduction

  • Time-Sensitive Networking (TSN) Task Group.
  • Developing a set of standards to provide Ethernet:
  • REAL-TIME GUARANTEES,
  • ONLINE MANAGEMENT,
  • RELIABILITY
  • on the layer 2.

2

slide-3
SLIDE 3

Introduction

  • Time-Sensitive Networking (TSN) Task Group.
  • Developing a set of standards to provide Ethernet:
  • REAL-TIME GUARANTEES,
  • ONLINE MANAGEMENT,
  • RELIABILITY
  • on the layer 2.

3

slide-4
SLIDE 4

Reliability in TSN

  • Qci: Per-Stream Filtering and Policing.
  • Error containment.
  • Detect babbling idiot.
  • Detect delayed frames.

4

slide-5
SLIDE 5

Reliability in TSN

  • Qca: Path Control and Reservation
  • Stablish multiple paths between nodes.
  • CB: Frame Replication and Elimination for Reliability
  • Stablish logical links over the created paths.
  • Send replicated messages through the redundant links.

5

slide-6
SLIDE 6

Reliability in TSN

  • TSN does not have time redundancy on layer 2.
  • What are the options to tolerate temporary faults?

6

slide-7
SLIDE 7

Reliability in TSN

  • Using spatial redundancy to tolerate temporary faults.
  • Not a suitable solution:
  • High cost.
  • No efficient solutions (specially in highly critical systems).

7

slide-8
SLIDE 8

Reliability in TSN

  • Using ARQ-based protocols.
  • Not a suitable solution (for HRT systems):
  • High jitter.
  • Non-deterministic bandwidth consumption.
  • ACK/NACK messages introduce new fault scenarios.

8

slide-9
SLIDE 9

Reliability in TSN

  • We propose to use proactive retransmissions.
  • Lower cost than adding more paths.
  • Lower jitter than with ARQ.
  • Deterministic bandwidth consumption.
  • More efficient than ARQ in the worst case scenario.

9

slide-10
SLIDE 10

End-to-end estimation and replication

10

B1 B2 T L

Proactive Transmission of Replicated Frames

slide-11
SLIDE 11

End-to-end estimation and replication k=3

11

B1 B2 T L

Proactive Transmission of Replicated Frames

slide-12
SLIDE 12

End-to-end estimation and replication k=3

12

T L B1 B2

Proactive Transmission of Replicated Frames

slide-13
SLIDE 13

End-to-end estimation and replication k=3

13

T L B1 B2

Proactive Transmission of Replicated Frames

slide-14
SLIDE 14

End-to-end estimation and replication k=3

14

T L B1 B2

Proactive Transmission of Replicated Frames

slide-15
SLIDE 15

End-to-end estimation and replication

15

T L B1 B2

Proactive Transmission of Replicated Frames

slide-16
SLIDE 16

End-to-end estimation, link-based replication

16

B1 B2 T L

Proactive Transmission of Replicated Frames

slide-17
SLIDE 17

End-to-end estimation, link-based replication k'=3

17

B1 B2 T L

Proactive Transmission of Replicated Frames

slide-18
SLIDE 18

End-to-end estimation, link-based replication k'=3

18

T L B1 B2

Proactive Transmission of Replicated Frames

slide-19
SLIDE 19

End-to-end estimation, link-based replication k'=3

19

T L B1 B2

Proactive Transmission of Replicated Frames

slide-20
SLIDE 20

End-to-end estimation, link-based replication k'=3

20

T L B1 B2

Proactive Transmission of Replicated Frames

slide-21
SLIDE 21

End-to-end estimation, link-based replication

21

T L B1 B2

Proactive Transmission of Replicated Frames

slide-22
SLIDE 22

Link-based estimation and replication

22

Proactive Transmission of Replicated Frames

T L B1 B2

slide-23
SLIDE 23

Link-based estimation and replication

23

T

k''=3

B1

k''=2

B2

k''=3

Proactive Transmission of Replicated Frames

L

slide-24
SLIDE 24

Link-based estimation and replication

24

Proactive Transmission of Replicated Frames

T

k''=3

B1

k''=2

B2

k''=3

L

slide-25
SLIDE 25

Link-based estimation and replication

25

Proactive Transmission of Replicated Frames

T

k''=3

B1

k''=2

B2

k''=3

L

slide-26
SLIDE 26

Link-based estimation and replication

26

Proactive Transmission of Replicated Frames

T

k''=3

B1

k''=2

B2

k''=3

L

slide-27
SLIDE 27

Link-based estimation and replication

27

Proactive Transmission of Replicated Frames

T

k''=3

B1

k''=2

B2

k''=3

L

slide-28
SLIDE 28

IN THIS WORK WE VALIDATE AND COMPARE THE APPROACHES OF THE TIME REDUNDANCY MECHANISM THROUGH SIMULATION

28

slide-29
SLIDE 29

OMNeT++ simulation model

29

  • Implement the PTRF mechanism over OMNeT++.
  • TSimNet model as starting point [1].
  • Additions:
  • Creation of replicas.
  • Identification and elimination of replicas.
  • Frame structure specification.

[1] P. Heise, F. Geyer, and R. Obermaisser. TSimNet: An Industrial Time Sensitive Networking Simulation Framework Based on OMNeT++. In 2016 8th IFIP International Conference on New Technologies, Mobility and Security (NTMS), Nov 2016.

slide-30
SLIDE 30

OMNeT++ simulation model

30

slide-31
SLIDE 31

Experiments

31

  • We validated and compared the three approaches.
  • We used exhaustive fault injection.
  • Inject all the possible combinations of frame losses.
  • We used an automotive use case.
  • Study the behaviour of each approach in a realistic scenario.
slide-32
SLIDE 32

32

Questions? Please meet me at the poster!

Fir irst Study of the Proactive Transmission of Replicated Frames Mechanism over TSN

Abstract

Time-Sensitive Networking (TSN) is a task group from the IEEE working to provide Ethernet with flexibility, real-time and reliability services. For these reasons, TSN represents an appealing technology for the networks of Cyberphysical Systems. Nevertheless, TSN does not cover some reliability aspects that are important to reach the reliability levels required by certain Cyberphysical Systems. Specifically, TSN does not devise any time redundancy mechanisms in the layer 2 to tolerate temporary faults in the channel. Thus, we proposed a time redundancy mechanism, called Proactive Transmission of Replicated Frames, to increase the reliability of TSN-based networks. In this work we describe two previous designs of PTRF and we present a new design. We also describe the simulation model used to compare the designs. Specically, we carried out exhaustive fault injection to validate the mechanism and a case study to compare the three designs.

In this work we evaluate time redundancy through exhaustive fault injection and an automotive use case

Inés Álvarez, Drago Čavka, Julián Proenza, Manuel Barranco

Departament de Matemàtiques i Informàtica, Universitat de les Illes Balears, Spain Ines.Alvarez@uib.es, drago.cavka@etf.unibl.org, julian.proenza@uib.es, manuel.barranco@uib.es

Exhaustive fault injection Time-Sensitive Networking Overview

TSN is a set of standards that aims at providing Ethernet with hard real-time, on-line management and reliability services. To provide timing guarantees and enable on-line management of the network TSN relies, among others,
  • n the SRP.
SRP enables the reservation of resources along the path between two nodes that want to communicate to guarantee availability and bounded transmission times. The communication is done through virtual communication channels called streams and the resource reservation is done in a per-stream manner. Talker Listener 1 Listener N … m1 m2 mn ··· This work is supported in part by the Spanish Agencia Estatal de Investigacin (AEI) and in part by FEDER funding through grant TEC2015- 70313-R (AEI/FEDER, UE). Ines Alvarez and Drago Čavka was supported by a scholarship of the EUROWEB+ Project, which is funded by the Erasmus Mundus Action II programme of the European Commission.

Problem

TSN does not provide any time-redundancy mechanisms in this level of the architecture specifically designed to tolerate transient faults. Although TSN can use higher level protocols, such as those based in Automatic Repeat Request (ARQ), this solution is not good enough in real-time systems. Using spatial redundancy to tolerate temporary faults is not adequate:
  • The communication channel is specially vulnerable
to transient faults.
  • Spatial redundancy has high impact in the cost and
size of the system.
  • When permanent faults cause the attrition of the
spatial redundancy, it may not be possible to tolerate transient faults any more.

Proactive Time Redundancy

Use Proactive Transmission of Replicated Frames (PTRF) to tolerate temporary faults and TSN spatial redundancy to tolerate permanent faults in the links. L T k=3 B1 k=2 B2 k=3 L T k=3 B1 k=2 B2 k=3 L T k=3 B1 k=2 B2 k=3 L T B1 B2 L L T B1 B2 T B1 B2 L L L T B1 B2 T B1 B2 T B1 B2 E2E estimation and replication of frames (A) E2E estimation, link-based replication of frames (B) Link-based estimation and replication of frames (C) Approach Replicas Combinations
  • Max. Delay (µs)
A 3 169 92.08 B 3 823543 212.18 C 2342342 297675 202.13 Inject all the possible combinations of errors where at least one replica traverses each link. ฀ 𝑓1= 0 𝑙−1 ··· ฀ 𝑓𝑚= 0 𝑙−𝑓1−···𝑓𝑚−1−1 ෑ 𝑛= 1 𝑚 𝑙 − σ 𝑗= 1 𝑛−1 𝑓𝑗 𝑓𝑛 ฀ 𝑓′ = 0 𝑙′ −1 𝑙′ 𝑓′ 𝑚 ෑ 𝑛= 1 𝑚 ฀ 𝑓′ ′ = 0 𝑙𝑛 ′ ′ −1 𝑙𝑛 ′ ′ 𝑓′ ′

(A) (B) (C)

𝒍, 𝒍′ , 𝒍𝒏 ′ ′ : number of replicas in the link 𝒇, 𝒇′ , 𝒇𝒏 ′ ′ : number of errors in the link 𝒎: number of links in the path The goal of these experiments is twofold:
  • Verify the correct operation of the mechanism.
  • Compare the approaches in terms of number of
scenarios that can be tolerated. The network parameters used are:
  • 6 hops (6 bridges between talker and listener).
  • 100 Mbps.
  • No interfering traffic.

Automotive use case OMNeT++ simulation model

We used simulation to evaluate and compare the proposed approaches. Transmission Stream Identification Upper layers / Forwarding Frame Replication Identifier Generation PTRF Encapsulation MAC Stream Replication Upper layers / Forwarding Replica Elimination Replica Identification Fault Injection MAC Reception Modules on transmission and reception Frame Structure Topology L1,1 L2,2 L4,4 L3,3 L1,2 L2,4 L3,4 L1,3 Type Priority Size (B) Period (ms) Sender Control 7 72 10 N1 ADAS 5 1526 30 N2 Video 3 1400 0.28 N3 Audio 2 1400 1.4 N4 BER # Selected replicas Control ADAS Video Audio 10−12 2 2 1 1 10−11 3 2 1 1 10−10 4 3 1 1 Experiment 𝑴𝟐,𝟐 𝑴𝟑,𝟑 𝑴𝟒,𝟒 𝑴𝟓,𝟓 𝑴𝟐,𝟑 𝑴𝟐,𝟒 𝑴𝟑,𝟓 𝑴𝟒,𝟓 1 10−12 10−11 10−12 10−11 10−12 10−12 10−11 10−11 2 10−12 10−10 10−11 10−10 10−12 10−12 10−10 10−11 3 10−11 10−10 10−11 10−10 10−11 10−11 10−10 10−10 Experiments parameters Traffic parameters, with 100Mbps and all nodes receive all streams. Number of replicas transmitted depending on the BER . Network configuration for each experiment. The variance on the BER represents the changing environmental conditions. Approach Traffic type
  • Exp. 1
  • Exp. 2
  • Exp. 3
A Control 4 2 ADAS 8 2 Video 14 148 176 Audio 6 25 56 Total 53 400 619 B Control 2 2 ADAS 1 8 Video 17 163 173 Audio 7 38 60 Total 58 436 624 C Control 2 3 ADAS 1 3 5 Video 11 142 172 Audio 8 37 73 Total 55 412 626 Results Lost frames in the longest link and lost in total in all the links.
slide-33
SLIDE 33

First Study of f the Proactive Transmission of Replicated Frames Mechanism over TSN

Inés Álvarez Vadillo, Drago Čavka, Julián Proenza, Manuel Barranco