Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its...

77
Computer Engineering and Networks Technische Informatik und Kommunikationsnetze Design for IoT Lothar Thiele ETH Zurich, Computer Engineering and Networks - Trustable Things -

Transcript of Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its...

Page 1: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

Computer Engineering and NetworksTechnische Informatik und Kommunikationsnetze

Design for IoT

Lothar ThieleETH Zurich, Computer Engineering and Networks

- Trustable Things -

Page 2: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

2

Landscape

Page 3: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

3

© www.braingrid.org© www.openpr.com

Page 4: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

4

Smart World

physical/biological/socialprocesses

observing influencing

reasoningdecidingbig data

CYBER

•WORLD

•PHYSICAL

•WORLD

Nature

Hardware &Software

Computation

Communication

Page 5: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

5

Endeavors

in long-term autonomous embedded systems

Page 6: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

6

Building Automation - Monitoring

© Siemens Building Technologies

Characteristics:

• Certified wireless multihop firedetection system

• High dependability requirements(availability, real-time guarantees)

• Battery-powered with long lifetime

Experience:

• new algorithms, models, methods

• tremendeous engineering effort

Page 7: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

7

High-Resolution Air Pollution Maps

Application Goal

Trustable high-resolution air quality maps.

Long-term operation.

•O3

•NO2

•CO

•PM0.1

•Public transport vehicles equipped with air quality sensors

•High-resolution air pollution maps

•wearable air qualitysensors

Page 8: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

8

Mobile stations installed on top of trams in Zurich

• Continuously in operation since April 2012

• Monitor main pollutants and environmental parameters

• PM0.1, O3, CO, NO2, temperature, humidity

• more than 100 million measurements and counting …

High-Resolution Air Pollution

Page 9: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

9

Wireless Monitoring Technology

allows to quantify mountain phenomena at diverse modalities and scales,

provides information for processunderstanding and modeling

can be used for safety critical applications in an hostile environment.

PermaSenseMonitoring in the Extreme

[Jan Beutel et al.: PermaSense]

Page 10: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

10

PermaSenseMonitoring in the Extreme

Development of early warning systems

New scientific knowledge

Page 11: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

11

Camera

Page 12: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

12

Page 13: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

13

Camera

5 field sites in surveillance

> 6 years autonomous operation

> 1 Billion data points

Page 14: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

14

Dependability and Trust

Page 15: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

15

© http://www.satiztpm.it/

© nkonnect

Page 16: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

16

Dependability and Trust

Society will increasingly depend on the Internet of Things.

Dependability and trust are keyfor societal acceptance.

Page 17: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

17

Predictability and Dependability

natural hazards desaster management

health factory

surveillance

care

mobility

automation

Page 18: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

18

Scalingto Billions of Devices

Data

Edge-Fog-Cloud ComputingNetworksStandardsBig Data & Analytics

Energy

Zero Power SystemsEnergy HarvestingApproximate ComputingNear-Threshold ArchitecturesBrain-Like Computing

Trust

Security and Privacy

This Talk

Information

Page 19: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

19

What are the reasons for low dependability

(or tremendous engineering effort)?

Page 20: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

20

Reason 1: Resource Constraints

Communication Bandwidth

Memory

Computation Energy

Low Cost Sensors© Adrian Ionescu

Page 21: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

21

An IoT System View© Luca Benini

50

Page 22: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

22

Near-Threshold Multiprocessing – PULP© Luca Benini

Page 23: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

23

Near-Threshold Multiprocessing – PULP© Luca Benini

803 GOPS/W 60k GOPS/W

Page 24: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

24

Reason 2: Adaptivity

• Adaptivity and feedback due to

• changes in environment

• user interaction

• switching between (power saving) modes, …

• Typical for applications in

• surveillance

• early warning

• personalized medical and health

• industrial control, …

© Mark Dixon

Event-DrivenBehavior

Page 25: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

25

Adaptive Behavior

Example: Acoustic sensing of mountain phenomena

1. acoustic events are detected

2. host decides on relevance

3. switching on other sensor modalities

4. high rate sensor data communication

Page 26: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

27

Resources Dependability

The closer the system works at its resource limits, the less reliable it operates.

conflicting

Page 27: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

28

Resources Dependability

The closer the system works at its resource limits, the less reliable it operates.

conflicting

computation

communication

queue

task energy

memory

Page 28: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

29

Resources Dependabilityconflicting

computation

communication

queue

task energy

memory

The closer the system works at its resource limits, the less reliable it operates.

Page 29: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

30

Resources Dependabilityconflicting

computation

communication

task energy

memoryqueue

The closer the system works at its resource limits, the less reliable it operates.

Page 30: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

31

Resources Dependabilityconflicting

overprovisioningis not an option

The closer the system works at its resource limits, the less reliable it operates.

Why is this a challenge?

Page 31: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

32

Resources Dependabilityconflicting

overprovisioningis not an option

The closer the system works at its resource limits, the less reliable it operates.

Why is this a challenge?

adaptivity increases non-determinismand requires system-wide coordination

Adaptivity

Page 32: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

33

System Challengesdependable

Page 33: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

34

System Challenges

adaptive

low resources

µW

sleep welloperate efficiently

wake-up quicklyswitch gears quickly

dependable

Page 34: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

35© http://www.systematic-paris-region.org

Some Design Principles

Page 35: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

36

Interference

Page 36: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

37

Resources Predictabilityconflicting

Interference is one of the prominent mechanisms.

The closer the system works at its resource limits, the less reliable it operates.

Adaptivity

memory demanddepends on total sum

application1

application2

application1

application2

application1

application2

Page 37: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

38

Resources Predictabilityconflicting

The closer the system works at its resource limits, the less reliable it operates.

Adaptivity

memory demanddepends on total sum

application1

application2

CPU1 CPU2

shared resource

shared resources and

communication may couple

time and power domains

with cyclic dependencies

Interference is one of the prominent mechanisms.

Page 38: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

39

Resources Predictabilityconflicting

The closer the system works at its resource limits, the less reliable it operates.

Adaptivity

memory demanddepends on total sum

application1

application2

CPU1 CPU2

shared resource

shared resources and

communication may couple

time and power domains

with cyclic dependencies

interference from other

modalities lead to cross-

sentivities and necessitate

calibration / sensor fusion

CO2CO

NOx

T

H2O

VOC

O3

Interference is one of the prominent mechanisms.

Page 39: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

40

Remove unwanted interference by design.

Interference

Page 40: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

41

PartitioningTime

Page 41: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

42

PartitioningTime Space

Page 42: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

43

What about other properties?

Energy

©Cypress

Clock and Power

© Mentor Graphics

Page 43: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

44

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

Classic IoT/Sensor Node Architecture

memory

interference

time-

interference

BOLT

Partitioning in Space

clock & power-

interference

Page 44: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

45

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

memory

interference

time-

interference

BOLT

firewall for

BOLT Architecture

clock and power

Classic IoT/Sensor Node Architecture

Page 45: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

46

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

Classic Architecture

memory

interference

time-

interference

BOLT

BOLT Architecture

•Application Processor

•Communication

Processor

•BOLT

[Sutton et al. "Bolt: A stateful processor interconnect.]

Total Sleep Power ≈ 8.1 µW

BOLT Sleep Power ≈ 1.3 µW

Page 46: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

47

Remove unwanted interference by design.If you can not, bound it.

Interference

Page 47: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

48

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

Classic Architecture

memory

interference

time-

interference

BOLT

firewall for

BOLT Architecture

clock and power

Page 48: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

49

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

BOLT

BOLT Architecture

REQ

ACK

SPI

INDA

MSP430FR5969

DCO

FRAM

R/W

REQ

ACK

SPI

INDA

R/W

SP

I A

GP

IO

PO

RT

3D

MA

0

SP

I C

DM

A1

GP

IO

PO

RT

4

MSP430

CORE

INDC INDC

Page 49: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

50

Example: Wireless Node Platform

BOLTSingle Processor APP COM

Sensor

Interface

Network

Protocol

Sensor

Interface

Network

Protocol

BOLT

BOLT Architecture

REQ

ACK

SPI

INDA

MSP430FR5969

DCO

FRAM

R/W

REQ

ACK

SPI

INDA

R/W

SP

I A

GP

IO

PO

RT

3D

MA

0

SP

I C

DM

A1

GP

IO

PO

RT

4

MSP430

CORE

INDC INDC

Use formal methods to bound

interference.

• Static Timing Analysis

• Hardware Verification

Page 50: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

51

•UPPAAL Model Checker

Page 51: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

52

Example: Wireless Node Platform

Use formal methods to bound

interference.

• Static Timing Analysis

• Hardware Verification

Predictability

Compostionality

Page 52: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

53

Remove unwanted interference by design.If you can not, bound it.

If you can not, use it to your advantage.

Interference

Page 53: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

Computer Engineering and NetworksTechnische Informatik und Kommunikationsnetze

Page 54: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

55

Let us look at wireless communication

Long Range Network (LoRa)

[Orestis Georgiou et al: Can LoRa scale? 2016.] [Ferran Adelantado et al: Understanding the Limits of LoRaWAN.

IEEE Comm. 2016]

Page 55: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

56

Let us look at wireless communication

Wireless Mesh Networks

Page 56: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

57

Can we expect that this construction works

reliably and efficiently?

Page 57: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

58

Can we expect that this construction works

reliably and efficiently?

No, it doesn't !

Complex and non-deterministic interference

No chance to provide tight bounds

Page 58: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

59

Use interference to your advantage

Page 59: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

60

Synchronous Transmissions

Transmissions are overlappingin time and space.

Nevertheless, the receiver decodesuseful information.

Do not avoid interference but use it to your advantage!

Page 60: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

62

Interference Management

Make use of interference Partitioning in time

Low Power Wireless Bus

• energy (resource) efficient

• adaptive

• robust and predictable

• provable guarantees[Ferrari et al: Low Power Wireless Bus, 2012.]

Page 61: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

63

But is the underlying principle competitive?

Page 62: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

64

Page 63: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

65

Evaluation Scenario

Page 64: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

66

Evaluation Scenario

Page 65: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

67

Evaluation Scenario

Page 66: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

68

Results

The best protocolsuse the design principles just

presented.

This year and the year before.

Page 67: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

69

Reality or Fiction

Page 68: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

71

Page 69: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

72

packet losstime synchronization

communication disconnect

buffer overflow

energy harvesting

data analysis

sensor fault

failing interconnects

computing faults

Page 70: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

73

But: «Sensing in the Extreme» is very special

Page 71: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

74

High-Resolution Air Pollution Maps

Application Goal

Trustable high-resolution air quality maps.

Long-term operation.

•O3

•NO2

•CO

•PM0.1

•Public transport vehicles equipped with air quality sensors

•High-resolution air pollution maps

•wearable air qualitysensors

Page 72: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

75

Can We Trust Sensor Readings?

as recommendedby manufacturer

Example: Widely used commercial low-cost NO2 Sensor

Page 73: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

76

Can We Trust Sensor Readings?

• EPA (United States Environmental ProtectionAgency)

• Community Air Sensor Network (CAIRSENSE) project (2016)

Page 74: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

77

Can We Trust Sensor Readings?

Example: Widely used commercial low-cost NO2 Sensor

after sensor array calibration usinglow-cost ozon, humidity and temperature sensors

[Maag et al: Pre-Deployment Testing, Augmentation and Calibration of Cross-Sensitive Sensors, 2016.]

Page 75: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

78

We will rely on the IoT Infrastructure.

It should be trustable.

We are in demand of appropriatemodels, design, testing and verification methods.

Page 76: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

79

ETH Zurich

Computer Engineering and Networks Lab

Geodesy and Geodynamics Lab

Micro and Nanosystems

Federal Office for the Environment

University of Zurich

Department of Geography

EMPA

Natural Resources & Pollutants

Acknowledgements

Current and Previous Contributors:

Jan Beutel, Olga Saukh, Zimu Zhou

Reto da Forno, Andres Gomez, Tonio Gsell, Romain Jacob, Balz Maag, Matthias Meyer, Lukas Sigrist, Mina Soliman, Felix Sutton, Roman Lim, Marco Zimmerling, Bernhard Buchli, David Hasenfratz, Federiuco Ferrari, Matthias Keller, Matthias Woehrle, Andreas Meier, Clemens Moser, Lennart Meier, Philipp Blum

Page 77: Design for IoT - ETH Z...38 Resources conflicting Predictability The closer the system works at its resource limits, the less reliable it operates. Adaptivity memory demand depends

80