23
Dynamic Data Analytics for the Internet of Things: Challenges and Opportunities Payam Barnaghi Institute for Communication Systems (ICS) University of Surrey/CityPulse Consortium Guildford, United Kingdom IoT Large-Scale Analytics Workshop IoT Week Lisbon, June 2015

Dynamic Data Analytics for the Internet of Things: Challenges and

Embed Size (px)

Citation preview

Page 1: Dynamic Data Analytics for the Internet of Things: Challenges and

Dynamic Data Analytics for the

Internet of Things: Challenges and

Opportunities

1

Payam Barnaghi

Institute for Communication Systems (ICS)

University of Surrey/CityPulse Consortium

Guildford, United Kingdom

IoT Large-Scale Analytics Workshop

IoT Week Lisbon, June 2015

Page 2: Dynamic Data Analytics for the Internet of Things: Challenges and

Contextual Challenges

2

AnyPlace AnyTime

AnyThing

Data Volume

Security, Reliability,

Trust and Privacy

Societal Impacts, Economic Values

and Viability

Services and Applications

Networking and

Communication

Page 3: Dynamic Data Analytics for the Internet of Things: Challenges and

IoT Data- Challanges

− Multi-modal and heterogeneous

− Noisy and incomplete

− Time and location dependent

− Dynamic and varies in quality

− Crowed sourced data can be unreliable

− Requires (near-) real-time analysis

− Privacy and security are important issues

− Data can be biased- we need to know our data!

3

Page 4: Dynamic Data Analytics for the Internet of Things: Challenges and

4

“Relying merely on data from sources that are

unevenly distributed, without considering

background information or social context, can

lead to imbalanced interpretations and

decisions.”

“It’s also about automation in addition to insight

and information extraction.”

?

Page 5: Dynamic Data Analytics for the Internet of Things: Challenges and

Data Lifecycle

5

Source: The IET Technical Report, Digital Technology Adoption in the Smart Built Environment: Challenges and opportunities

of data driven systems for building, community and city-scale applications,

http://www.theiet.org/sectors/built-environment/resources/digital-technology.cfm

Page 6: Dynamic Data Analytics for the Internet of Things: Challenges and

IoT environments are usually dynamic and (near-)

real-time

6

Off-line Data analytics

Data analytics in dynamic environments

Image sources: ABC Australia and 2dolphins.com

Page 7: Dynamic Data Analytics for the Internet of Things: Challenges and

IoT Data

7

Page 8: Dynamic Data Analytics for the Internet of Things: Challenges and

Deep IoT

8

Page 9: Dynamic Data Analytics for the Internet of Things: Challenges and

9

“The ultimate goal is transforming the raw data

to insights and actionable knowledge and/or

creating effective representation forms for

machines and also human users and creating

automation.”

This usually requires data from multiple sources,

(near-) real time analytics and visualisation

and/or semantic representations.

Page 10: Dynamic Data Analytics for the Internet of Things: Challenges and

10

“Data will come from various source and from

different platforms and various systems.”

This requires an ecosystem of IoT systems with

several backend support components (e.g.

pub/sub, storage, discovery, and access services).

Semantic interoperability is also a key

requirement.

Page 11: Dynamic Data Analytics for the Internet of Things: Challenges and

Search on the Internet/Web in the early days

11

Page 12: Dynamic Data Analytics for the Internet of Things: Challenges and

IoT discovery engines?

12

“Working across different systems and various

platforms is a key requirement. Internet search

engines work very well with textual data, but IoT

data comes in various forms and often as

streams.”

This requires an ecosystem of IoT systems with

several backend support components (e.g.

pub/sub, storage, discovery, and access services).

Page 13: Dynamic Data Analytics for the Internet of Things: Challenges and

IoT discovery engines?

13

“To make it more complex, IoT resources are

often mobile and/or transient. Quality and trust

(and obviously privacy) are among the other key

challenges”.

This requires efficient distributed index and

update mechanisms, quality-aware an resource-

aware selection and ranking, and privacy control

and preservation methods (and governance

models) .

Page 14: Dynamic Data Analytics for the Internet of Things: Challenges and

Accessing IoT data

14

“The internet/web norm (for now) is usually

searching for the data; the search engines are

usually information locators – return the link to

the information; IoT data access is more

opportunistic and context aware”.

This requires context-aware and opportunistic

push mechanism, dynamic device/resource

associations and (software-defined) data routing

networks.

Page 15: Dynamic Data Analytics for the Internet of Things: Challenges and

Web search is already adapting this model

15

Image credits: the Economist

Page 16: Dynamic Data Analytics for the Internet of Things: Challenges and

A discovery engine for the IoT

16

A. HosseiniTabatabaie, P. Barnaghi, C. Wang, L. Dong, R. Tafazolli, "Method and Apparatus for Scalable Data Discovery in

IoT Systems”, US Patents, May 2014.

Page 17: Dynamic Data Analytics for the Internet of Things: Challenges and

CityPulse demo

17

Page 18: Dynamic Data Analytics for the Internet of Things: Challenges and

KAT- Knowledge Acquisition Toolkit

http://kat.ee.surrey.ac.uk/

Page 19: Dynamic Data Analytics for the Internet of Things: Challenges and

The future: borders will blend

19Source: IEEE Internet Computing, Special issue on Physical-Cyber-Social Computing

Page 20: Dynamic Data Analytics for the Internet of Things: Challenges and

In conclusion

− IoT data analytics is different from common big data analytics.

− Data collection in the IoT comes at the cost of bandwidth, network,

energy and other resources.

− Data collection, delivery and processing is also depended on multiple

layers of the network.

− We need more resource-aware data analytics methods and cross-layer

optimisations (Deep IoT).

− The solutions should work across different systems and multiple platforms

(Ecosystem of systems).

− Data sources are more than physical (sensory) observation.

− The IoT requires integration and processing of physical-cyber-social data.

− The extracted insights and information should be converted to a feedback

and/or actionable information.

20

Page 21: Dynamic Data Analytics for the Internet of Things: Challenges and

Smart city datasets

21

http://iot.ee.surrey.ac.uk:8080

Page 22: Dynamic Data Analytics for the Internet of Things: Challenges and

IET sector briefing report

22

Available at: http://www.theiet.org/sectors/built-environment/resources/digital-technology.cfm

Page 23: Dynamic Data Analytics for the Internet of Things: Challenges and

Q&A

− Thank you.

− EU FP7 CityPulse Project:

http://www.ict-citypulse.eu/

@pbarnaghi

[email protected]