Internet of Things is promising to be a set of technologies able to have a high impact on how people live, produce, modify and interact with the environment. Such a transformation is driven by increasing technologies capabilities of sensors/actuators, communications, general-purpose hardware, availability of software and programmability of devices. The integration of so different technologies is a problem in itself and IoT is also trying to solve cogent issues of specific problem domains, such as e-health, transportation, manufacturing, and so on. Large IoT systems (e.g., smart cities) stand on their own because the smartness requires integration of different technologies, processes and different administrative domains creating the needs to deal with a complex system. In addition to technological and problem domain specific challenges, there exist further challenges that fall in business, social and regulation realms. They can greatly impact the deployment and the success of IoT deployment. The speech aims at providing a view on some major technologies challenges of IoT and to cover a few critical business and social issues that could hamper the large deployment of IoT systems by providing some examples of implementation.

1. Thanks to the IEEE IoT Initiative
iot.ieee.org
1
IoT Challenges: Technological,
Business and Social aspects
Roberto Minerva
Telecom SudParis
Institut Mines-Telecom
ICNSC 2017
14th IEEE International Conference on
Networking, Sensing and Control
May 16-18, 2017, Calabria, Southern Italy

2. Outline
2
 The IoT Context
 Definition and challenges of IoT
 The Communications challenge
 The Software Challenge
 Data and Identity Challenges
 Business Challenges
 Societal Challenges
 Virtual Continuum

3. Let’s talk about … Tennis ?
My First Tennis Racket
My Next one ?
3

4. Put a Sensor in it …

5. The IoT Context5

6. The Rise of Softwarization
Softwarization instantiations
Key drivers towards softwarization
Commoditization of HW,
i.e., general purpose HW is becoming more and more powerful and cheap.
Cloud computing evolving towards the a Fog of very powerful terminals
(smartphones)
Commoditization of communications,
i.e., the ubiquitous availability of communications means
Virtualization,
i.e., the capability to execute
functions and services on virtual
computational environments
Autonomics and Self-Organization
i.e., the ability of large system to
adaptively and autonomously optimize
their behavior
Big data,
i.e. the capability to collect data in real time that describe a phenomenon
associated with a resource or a person (or groups of them)
Availability of Application Programming Interfaces
for several resources and functionalities (pertaining to the Comm, Stor, Proc,
Sens/Acting realms)
Open Source,
i.e., the ability to model resources and functions by means of software
communities that share results and tools
Softwarization of
the Telcos
• Software Defined
Networks (SDN)
• Network Functions
Virtualization (NFV)
• Integration of SDN,
NFV with Cloud
Emergence of new
Services
paradigms and Biz
Models
• Servitization: Anything
as a Service (e.g., IoT,
IwT)
• Pervasive sensoring
and actuation
Virtual Continuum
• Creating new Virtual
Worlds bridging the
Physical
• WorldMetaverse:
Integrating of the
Physical and Virtual
Worlds
• MicroManufacturing:
3D Printers
Big Data
• Real Time Data
management
• The Bank of User Data
• Electronic Money
Processing, Storage and Communication resources will be interchangeable. Their composition will allow to provide high quality
services, while virtualization and autonomics will allow for system optimization (aggregating resources where they are needed the most)
Edge as Point of
Intelligence
Accumulation
• Smart Terminals
• Different connectivity
options
• Smart environment
• ….

7. Do It Yourself Paradigm ... A paradigm
shift

8. Source: Beecham Research
IoT: Application Domains (the high expectations)

9. The Vision: the Programmable World
9
In the Programmable
World, All Our
Objects Will Act as
One
Wired - BILL WASIK 05.14.13.
available at:
https://www.wired.com/2013/05/int
ernet-of-things-2/
We are talking about a gigantic
network of all things (millions of
systems). It has to be:
• Open
• Programmable (API based)
• Secure
• Accountable
• Manageable
• Reliable
• Always available
• Usable
• Democratic
• Inclusive

10. And its challenges
An IoT Definition10

11. What is Internet of Things ?
 “Machine-to-Machine (M2M) communications is the communication between two or more entities that do
not necessarily need any direct human intervention. M2M services intend to automate decision and
communication processes.” - ETSI oneM2M
 IoT as “A global infrastructure for the information society, enabling advanced services by interconnecting
(physical and virtual) things based on existing and evolving interoperable information and
communication technologies” – ITU
 “Internet of things” as: “A network of items—each embedded with sensors—which are connected to the
Internet.” IEEE Institute. March 2014
 “The basic idea is that IoT will connect objects around us (electronic, electrical, non-electrical) to provide
seamless communication and contextual services provided by them. Development of RFID tags, sensors,
actuators, mobile phones make it possible to materialize IoT which interact and co-operate each other to
make the service better and accessible anytime, from anywhere.” – IETF
 “Cyber-physical systems (CPS) – sometimes referred to as the Internet of Things (IoT) – involves
connecting smart devices and systems in diverse sectors like transportation, energy, manufacturing and
healthcare in fundamentally new ways. Smart Cities/Communities are increasingly adopting CPS/IoT
technologies to enhance the efficiency and sustainability of their operation and improve the quality of
life.” – NIST
 “A global network infrastructure, linking physical and virtual objects through the exploitation of data
capture and communication capabilities. This infrastructure includes existing and evolving Internet and
network developments. It will offer specific object-identification, sensor and connection capability as the
basis for the development of independent cooperative services and applications. These will be
characterized by a high degree of autonomous data capture, event transfer, network connectivity and
interoperability.” - CASAGRAS
• Internet of Things
is a buzzy phrase
• It has to be
interpreted
according to the
needs and assets
of the proponents

12. What Internet of Things is (a simplistic view)
Aggregator/Gateway
Sensors
Internet
Service Service Service
Aggregator/Gateway
Sensors
Events
Comm.
Usage

13. What Internet of Things is
Aggregator/Gateway
Sensors &
Actuators
Internet
Service Service ServiceEvents
Aggregator/Gateway
Sensors &
Actuators
Events
Commands
Commands
Interworking
Interworking
Comm.
Comm.
Usage

14. What Internet of Things is
Service Service Service
Networking
Usage
Virtualization
Data harmonization
Data Distribution
Different Administrative Domains

15. What IoT is: two different views
 Internet of Things envisions a system comprising
sensors/actuators, aggregators and gateways,
service control. These components use Internet
protocols and/or specific sensor protocols to
communicate.
 These systems could be quite large in size and
complex in technologies (even if they will tend to
use a few of them), but they are homogeneous
from a management perspective and in ownership
 IoT envisions the integration of several heterogeneous systems
(i.e., networks of networks), each one using different
technologies, interfaces and protocols and
governed/managed by different Actors by means of several
processes and managements functions.
 IoT in a multi domain envisions a self-configuring and
adaptive complex system made out of networks of sensors
and smart objects.
 Single Administrative Domain  Multiple Administrative Domains
The Internet of Things (IoT) envisions systems made out of networked sensors and smart objects whose
purpose is to measure/control/operate on an environment in such a way to make it intelligent, usable, and
programmable and capable of providing useful services to humans.

16. An “IoT in the Large” Definition
(http://iot.ieee.org/definition.html)
 Smart Objects and
Sensors
 Networks of Things
 Self-organizing systems
The Internet of Things (IoT) envisions a self-configuring and
adaptive complex system made out of networks of sensors and
smart objects whose purpose is to interconnect “all” things,
including every day and industrial objects in such a way to make
them intelligent, programmable and more capable of interacting
with humans by providing useful services.
Intelligence at the edge
Massive Data
New communication
paradigms
Servitization
Virtual Continuum

17. IoT Challenges
 5 Challenges The IoT Presents To Manufacturers – Forbes
 http://www.forbes.com/sites/ptc/2014/09/10/5-challenges-the-iot-presents-to-manufacturers/
 IoT Challenges - Rob van Kranenburg and Alex Bassi
 http://www.muxjournal.com/content/1/1/9
 Five challenges for the Internet of Things (IoT) - Rolph Haspers
 http://blog.leaseweb.com/2014/07/17/five-challenges-internet-things-iot/
 IoT Challenges – Texas Instruments
 http://www.ti.com/ww/en/internet_of_things/challenges.html
 Opportunities: Back To The Future – IoT & Smart Systems Evolution
Challenges – Harbor Research
 http://harborresearch.com/iot-evolution-challenges/
 5 Challenges of Internet of Things Connectivity – PubNub
 http://www.pubnub.com/blog/5-challenges-of-internet-of-things-connectivity/
 Also the Challenges are pretty related to assets and specific views

18. IoT implies a lot of Challenges
• Definition of Things and
«Identity of Things»
• Complexity
• Communication Paradigms
• Data
• The Software Platform
• Silos vs. Horizontal
Application Domains
• The revenue
• Per device
• Connectivity
• Data
• The Value Chain
• New Biz Model
• Privacy
• Data Ownership
• Security
• Easiness of Use
• Social Cooperation
Regulation

19. What are things?19

20. What are the Things in «IoT» ?
A Measure
An Identity
An action
Properties of a Thing

21. Things communicate (maybe through the
Internet)
Message
What networks and nodes ?
What protocols ?
What formats?
What communication paradigm?
…
The http
The xml
The Client – Server model
The technology for things
?

22. What Comes with Distribution ?
Identity is a primary constituent of Things!!!
IPv6 Address
QR Code
EPC Global
http://www.impinj.com/resources/about-rfid/how-do-rfid-systems-work/
http://an.authority.org/sensorlist/sensorX
http://www.slideshare.net/arneb/meaningful-ur-isforlinkedsensordataforpublic
1
2
3
4
5
6
7
8
9
10
11
12
13
<!-- ================================================= -->
<!-- System Description -->
<!-- ================================================= -->
<gml:description>A solid state Gamma measurement module</gml:description>
<gml:identifier codeSpace="UID">urn:heath:gamma2070</gml:identifier>
<gml:name>Health Physics Instruments 2070 Gamma Detector</gml:name>
<sml:keywords>
<sml:KeywordList>
<sml:keyword>CBRNE</sml:keyword>
<sml:keyword>gamma</sml:keyword>
<sml:keyword>insitu</sml:keyword>
</sml:KeywordList>
</sml:keywords>
http://www.sensorml.com/sensorML-2.0/examples/description.html
SensorML

23. ….
Authentication and Authorization Attributes
Location and Other Objects Relations Attributes
Identity and its properties
Basic Object Attributes (what
type of sensor, what
measures,…)
Object Id

24. What are “Things” ?
https://ibmcai.files.wordpress.com/2014/06/iot-network.jpg
Smartness refers to the ability of the object to provide some forms of
Sensing/Actuation together with processing, storage and communications

25. What are Internet-Connected Things?
Active/passive, with/without context
Generic Info Contextualized Info
Passive Objects A Tag,
A Pointer to some
information
Info + a location
Reactive Objects A switch at home (turn it
on/off),
A smart meter
Home Automation (when
temperature reaches 20 C
stop heating)
Autonomous Objects A Vending Machine,
An Intelligent Fridge
A Cleaning Robot

26. How Many Things?
http://tarrysingh.com/2014/07/fog-computing-happens-when-big-data-
analytics-marries-internet-of-things/

27. Characterization of Things
Thing
Actuator
Sensor
Smartness
Simple
ripetitive task
Complex and
adaptive
intelligence
Specialized Hw + Task Oriented Software +
System Software (communication and management)
Communication
Communication,
autonomics,
communication
paradigm
Data
Single Data Stream of Data
Identity
Mobile

28. The Communications Challenge28

29. Communications and Distribution

30. Characterization of Communication in
IoT
GatewaySensor Server
Short Range Long Range
Long Range
Machine to machine (M2M) refers to the ability of autonomous machines to exchange
data through a Mobile Network in order to control some actuator or sensor
A wireless sensor network (WSN) is a wireless network consisting of spatially
distributed autonomous devices using sensors to monitor physical or environmental
conditions. A WSN system incorporates a gateway that provides wireless connectivity
back to the wired world and distributed nodes
http://www.ni.com/white-paper/7142/en/

31. One network fits all?
 Wireless Sensor Networks
 M2M
 Capillary Networks and LPWAN
 Mobile Networks (5G)
 Edge Networks (i.e., the FOG)

32. And Networks of Networks
Networks of Things

33. Three Communications Environments
Personal Device
Smart Environment
Communication
(capillary network)
Services and Applications
(Cloud and Internet)
Long Range Communication
(3G, 4G, 5G, …)
Short Range Communication
(WiFi, BlueTooth, NFC, ZigBee,
6LoWPAN, …)
Control/Mngmt
Communication
• Network as a smart fast Pipe
• Distributed Edge Networked
Platform
• Value of UpLink
• Control of Spikes of
Information
• Virtualization in the Cloud of
Resources
• Transactional
Communication with
Guarantees
The Network is a commodity
Intelligence aggregates at the Edges Pervasiveness and high distribution of functions
Complete decentralization
New communication paradigms
Autonomic behavior
Opportunistic and dynamic usage of resources and networks
Integration of processing, storage, communication and “sensing”

34. Communications Technologies for IoT
(some)
Requirements associated with M2M and IoT:
• low cost,
• low power,
• compact form factors,
• rapid connection setup times,
• massively scalable deployments
See more at: http://www.wi-fi.org/beacon/craig-mathias/wi-fi-and-the-internet-of-
things-much-more-than-you-think#sthash.rzxKXMiJ.dpuf
http://d3uifzcxlzuvqz.cloudfront.net/images/stories/content/handbooks/iot-handbook/communication-iot1.jpg

35. IoT Protocol Landscape
35
https://entrepreneurshiptalk.wordpress.com/2014/01/29/the-internet-of-thing-
protocol-stack-from-sensors-to-business-value/

36. IP or not IP: ZigBee vs. 6LoWPAN
http://www.hindawi.com/journals/ijdmb/2011/926363.fig.007.jpg

37. WIFI in IoT ?
A video is available at:
https://www.youtube.com/watch?v=9QZkCQSHnko
See more at: http://www.wi-fi.org/beacon/craig-mathias/wi-fi-and-
the-internet-of-things-much-more-than-you-think#sthash.rzxKXMiJ.dpuf

38. The protocol challenge in the small
range
https://wsnblog.files.wordpress.com/2010/08/feature.png

39. Longer Range Networks39

40. Low Power WAN
http://image.slidesharecdn.com/lpwan-150313025707-conversion-gate01/95/lpwan-technologies-for-internet-of-things-iot-and-m2m-scenarios-5-
638.jpg?cb=1426299913

41. LPWAN
http://www.slideshare.net/zahidtg/lora-introduction

42. 5G

43. 5G in a nutshell
 The move is from a Mobile broadband network (e.g., 4G) to a lifeline network
very similar in certain characteristics and capabilities to the fixed network. A
few foreseen features could give the flavor of the differences:
• More than 50 Mbps everywhere
• Support to dense areas and crowds (up to 150.000 people/km2)
• Support to fast moving vehicles (cars, high speed trains, and airplanes)
• Coverage of Indoor areas with shared bandwidth of up to 1Gbps
• Ultra low-latency (latency less than 1ms) and Ultra-High Reliability
• Resilience and support to surge of traffic
• Support to massive low-cost/long-range/low-power Machine Type Communication
• And many more…
More on: http://iot.ieee.org/newsletter/september-2015/internet-of-things-and-the-5th-
generation-mobile-network.html

44. Two Issues in 5G
http://www.slideshare.net/zahidtg/thinking-networks-by-prof-simon-saunders
 Ultradense Network
 Many Antennas
 Interworking with legacy
 Heterogeneous access
 Softwarization of the Network
 By means of SDN and NFV
 Reuse of existing architectures or new approaches?
 IMS or evolution to other software architectures?

45. Slicing Concept (from Next Generation Mobile Network Forum)
https://www.ngmn.org/uploads/media/NGMN_5G_White_Paper_V1_0.pdf

46. Virtualization Infrastructure Other InfrastructuresEdge Infrastructures
Core Resources and Networks
Southbound API
North bound API
IoT Services and Applications
5G Slicing for IoT (Telcos view)
IoT Slice
VirtualResources
(IoTSlice)
5GResourcesandInfrastructureIoTServiceLayer
Service API
Virtualized Functions
Possibly defined according to a standard
Architecture (e.g., oneM2M, P.2413)
External Clouds
Sensors
and actuators

47. Virtualization Infrastructure Other InfrastructuresEdge Infrastructures
Core Resources and Networks
Southbound API
North bound API
5G Slicing for IoT (Providers view)
IoT Slice
VirtualResources
(IoTSlice)
5GResourcesand
Infrastructure
IoTServiceLayer
Virtualized Functions
Possibly defined according to a standard Architecture
(e.g., oneM2M, P.2413) and highly distributed
Service API
IoT Services and Applications

48. Edge Computing

49. The Philosophy of an All-IP Network (and 5G is an all-IP
network)
Autonomous
SubSystems
E-E protocol
Intelligent EndPoints
Stupid Network
“mechanisms should not be enforced in the network if they can be deployed at
end nodes, and that the core of the network should provide general services,
not those tailored to specific applications” Clarks & Reed from MIT
With All-IP Networks, intelligence (and
services) moves to the edge
End – to – End Principle for IP networks

50. Anything will be a node !
• Intel has unveiled a WiFi sliver of silicon that can
be part of a normal microprocessor chip.
• We can expect that wherever we find a
microprocessor (e.g. in over 70% of toys, to name
just one area) we will find embedded connectivity.
Roberto Saracco
http://www.blog.telecomfuturecentre.it/
A trend in devices: integration of communication,
processing, storage and sensing/actuation capabilities
This a is a big challenge from devices: how can we take
advantage in terms of services and application of this
power in a single node ?

51. Nodes will connect each other in unpredictable ways
http://muxware.net/sol_mesh.php
1
2
3
Node Aggregation at time t1
Aggregation 1
Aggregation 2
Aggregation 3
1
2
3
Node Aggregation at time t2
Aggregation 2
Aggregation 1
Network
• Increasing richness and complexity at
the edge of networks
• D2D Communications

52. Example of Egde Communications - Fog
Computing
http://www.slideshare.net/Angelo.Corsaro/20141210-fog
http://www.slideshare.net/Angelo.Corsaro/20141210-fog
Fog and Cloud

53. Bonomi, Flavio, et al. "Fog computing and its role in the internet of
things." Proceedings of the first edition of the MCC workshop on
Mobile cloud computing. ACM, 2012.

54. Mobile Edge Computing
(moving back intelligence to the edge)
54
http://www.etsi.org/images/files/ETSIWhitePapers/etsi_wp11_mec_a_key_technology_towards_5g.pdf
Source: ETSI White Paper No. 11
Mobile Edge Computing - A key
technology towards 5G
First edition – September 2015
ISBN No. 979-10-92620-08-5
Authors: Yun Chao Hu, Milan Patel,
Dario Sabella, Nurit Sprecher and
Valerie Young
Why ?

55. Mobile Edge Computing (2)
Network EdgeUser Edge
And the User Edge is based on a lot of
resources!!!

56. Disruption in (User) Edge Networks
▪ Edge networks can be built and programmed in the same way as today’s end systems!
▪ Edge will look like a federation of Clouds…and Cloud Computing will become “commodities”
▪ This will be a new space for innovating in the ICT-Telco biz
▪ A step towards “Smart Environments”, deployed by several actors (e.g., Operators in conjunction with Public
Administrations or customers)
▪ New business models (like Resource as a Service) based on federation and brokering of resources
▪ Examples of potential tasks for an Operator:
▪ Broker (auctioneer) of available resources
▪ Virtualization of resources in the Network (servitization)
▪ Anchor point for allowing dispersed local environments to communicate
▪ Managing and orchestrating a “complex system of virtual resources” while workloads and customer
demands change in real time.
Edge Complexity,
Brokering, Smart
Environments

57. Network: a short recap57

58. Challenge: A Network of Networks
Data Center Networks
LoRA
WiFi
Zigbee
Dash7
5G
Telecommunications
Network A
Internet
…
Things
• Different Access Technologies and Networks
• Different Core Networks
• Different Clouds (networks)
Telecommunications
Network B

59. The Software Challenge59

60. The Software Challenge of IoT
Middleware
Things
Applications
What Platforms?
What Mddleware ?
What software approaches?
What software tools?
What Virtualization?
…
IDEs and other
Tools
Virtualization
Instances
Protocols Layer

61. Communication between remote systems:
Protocols vs APIs
 Protocols
 No visibility of internal structure of communicating systems  Applications
and services are not visible and/or programmable
 Protocol can be
 text-based (e.g., HTTP, LDAP, …)
 binary (e.g., H.323, …)
 APIs or Sw libraries (.dll, Java APIs, …)
 They support levels of abstraction of system features to programmers
 They support portability of applications
 Platform APIs (OpenStack, …)
 They allow the invocation of operations on remote applications/services
 Communication is supported by a software layer
 The intricacies of the transport layer are hidden to programmers

62. Protocol Dependence of the Middleware
 There is a strong dependence of middleware from the
underlying protocols
 Protocol expressive power can be abstracted by the middleware,
i.e., part of protocol possibilities is hidden inside the middleware
 Middleware can add functionalities to the protocols (i.e.,
functionalities that are not provided by the protocol)
 A plain mapping between a protocol and a component not
necessarily introduces a different state machine for the sw
component (supporting the API)
 Middleware can aggregate two or more protocols into a
component, usually a composite state machine is created that is
different from the two state machines of the single protocols. This
may cause a loss in expressiveness

63. Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
SubArchInstance SubArchInstance SubArchInstance SubArchInstance SubArchInstance SubArchInstance
LinkInstance
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Component
Interface
Instance
Connector
Basic
CommunicationLayer
LinkInstanceLinkInstance
Component
Interface
Instance
Connector
LinkInstance
Aggregated
Layer
Upper
Layer
Application Application Application
Protocol Dependence of the Middleware (cont.)
Aggregation of
protocols can lead to
loss of functionalities
Abstraction should support different levels of granularity

64. Control Paradigms for IoT
Client – Server model disregards the network
aspects and can lead to a tragedy of commons
(misuse of common networking resources)
Other mechanisms (message
based like PubSub) can be more
attractive for IoT
This is a reason for different IoT protocols
…
Is it there a single communication paradigm
for IoT ?
Triggering Model

65. IoT Protocol Landscape
https://entrepreneurshiptalk.wordpress.com/2014/01/29/the-internet-of-thing-
protocol-stack-from-sensors-to-business-value/

66. A Bit of Protocols
http://blog.zuehlke.com/en/iot-for-tiny-devices-lets-talk-mqtt-sn/
Constrained Application Protocol (CoAP) uses a REST-
like approach offering resources with unique URI’s for
consumption by CoAP clients or the setting and update
of such resources.
MQTT (formerly Message Queue Telemetry
Transport) is a publish-subscribe based "light
weight" messaging protocol for use on top of the
TCP/IP protocol. It is designed for connections
with remote locations where a "small code
footprint" is required and/or network bandwidth
is limited.

67. From Protocols to Platforms67

68. IoT Protocols Stack and initial Platform
Available inhttp://postscapes.com/internet-of-things-protocols
Source: Butler project

69. ARM Operating System for IoT
http://hexus.net/tech/news/software/75357-arm-looks-supercharge-internet-things-mbed-os/

70. IoT Middleware70

71. A blueprint for an IoT Platform
Sensor as a small computer  it needs an
Operating System providing for basic functions
Mobile Sensor API is an example of
middleware for Wireless Sensor Networks
Apps
Obviously there are many OSes for IoT (e.g., Contiki, …)
Many European Projects are working towards this vision
Sensors /
Actuators/ Smart Objects
Distributed OS (comprising Local
OSes)
IoT Platform Services
Application Framework
Apps
Apps
A Framework for developing
Applications
A choice between vertical and specialized platforms vs.
horizontal and general purpose ones

72. Middleware for IoT
https://www.cs.virginia.edu/wsn/vigilnet/
• Layering
• Components and APIs

73. IoT-A Architecture
http://iot-datamodels.blogspot.it/2012/09/data-models-for-internet-of-things-5.html
• Virtualization
• Management
• Security
• Orchestration

74. Oracle’s view on Middleware for IOT
http://www.slideshare.net/junsukseo946/0-2-oracle

75. Web of Things
http://webofthings.org/2011/12/01/phd-web-of-things-app-archi/

76. Barcelona City OS
http://www.christianestay.com/2015/07/de-estandares-patters-frameworks.html

77. Toward a New Architecture based on
Virtualization, Softwarization, and Autonomics
77
Physical Resources
Virtualized and Autonomic Resources
Processing
Resource
Control Plane
Storage
Resource
Control Plane
Communication
Resource
Control Plane
Things and
Other Resources
Control Plane
Interfacing and Aggregation
ApplicationApplication
Request
Allocation
Negotiation
Usage
Application
Application
Infrastructure
Layer
ResourceAware
Layer
Application
Layer
Request
Allocation
Negotiation
Usage
Application Resources
Virtualized Application Resources
Application Resources Control Plane
Interfacing and Aggregation
KnowledgePlaneKP

78. The IoT Conundrum: Data and Identity

79. Where is the Value then ?
Knowledge
• Aggregation
• Personalization
• Inference
• Relationships
Information
Data
79
The Big Data Challenge …

80. Where are the Data? Wireless Sensor
Network Aggregation
http://www.easinet.cn/photo/coalmine.jpg
http://www.mdpi.com/1424-8220/9/6/4845/ag
Plenty of Data … distributed locally
and remotely
Aggregation function and dispatching
functions are fundamental

81. How much Data (and traffic)?
http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-networking-index-vni/white_paper_c11-520862.html#Trend_3_Measuring_Mobile_IoE
= 2.33 MB /Day
= 27 Byte /s
= 12.2 MB /Day
= 141 Byte /s

82. A simple Communication Perspective
Sensors
Gateway /
Mediation
Servers
GW
Low Range
Technologies
Wide Area Network
Technologies
How much bandwidth for IoT ?

83. A few parameters to consider
How many Sensors ?
GW
How many Gateways ?
(or how many edge networks?)
How many Messages?
How large?

84. IoT Devices with Cellular
GW
GW
Cellular
IoT Devices directly connected to
Cell Net
IoT Devices connected to Cell Net
by gateway
• Ericsson predicts 1.5 billion
IoT devices with cellular
subscriptions by 2021
• Does it make sense to
postulate
• 0.5 Billion of individual
Sensors directly
connected to Cellular Net
• 1 Billion of gateways
directly connected

85. The Gateway Effect
1 B Gateways Pbytes /dayPbytes/Month
1 mes per day length 1 byte 0,000001 0,00003
1 mes per min length 64 bytes 0,09216 2,7648
1 mes per 0.5 sec length 8 bytes 1,3824 41,472
1 mes per ms length 1000 bytes 86400 2592000
1 message per sec 1000 bytes 86,4 2592
1 message per sec 256 bytes 22,1184 663,552
Western Europe 0,8
USA 1,3
World Total 5,4 5400
Total Mobile Traffic in Ebytes per month in 2015
Source: Ericsson

86. A few Considerations
86
 Data Sources
 Sensor data rate generation strongly depends on the sensor and the specific application. When the number of
sensors grows the aggregate data generation rate grows
 Data load increases with the size of the payload (some sensor can generate small ammout of data with a high
rate)
 Multimedia sensors (e.g., cameras) tends to generate large streams of data
 When the Dt between two messages  0 then the sensor generates a stream of data
 Streams of sensor data can exhibit similar patterns as multimedia data, but sometimes they have different
requirements
 Number of «streaming IoT devices» can have a huge impact on network
 Mediation
 When many sensors are mediated by a Gateway, then
 Messages may be processed and reduced in quantity (aggregation)
 The rate of outgoing messages from GW to Server may be high and may generate a stream
 Sensor behaviour
 Some IoT systems exhibit impulsive behaviours
 A Container Ship with M2M devices entering in a harbor generates a spike of signalling and data
 An alarmgenerated by several sensor in a specific are
 Malfunctioning: a sensor that generates many messaging and stresses the nearby resources
 Data Collection (data sinks)
 Data generated by sensor may be stored and processed locally (edge computing) or transfered as raw data
over the network
 Data sets of raw data can be very useful to study particular phenomena, but transfer them to cloud systems
could br «expensive»

87. ... But87

88. Aggregating Data … means collecting
Information
“OUR”SmartThings
Raw data to be
transformed into
Info
Personal
Profiling
Functional
Profiling
Who, Where, When, What, Why, …
+
Events and commands
* = Bigger
DATA
• Who is the Owner
of all these Data ?
• Who has the right
to extract info ?
50 B Devices *
(Average Aggregated Traffic of M2M Devices)
~ 2MB/day = ~ 88.81 petabytes
/day

89. Identity of Things and User Profiling
• Each Resource is
addressable
• Each resource is
CONNECTED
– Connectivity must be
guaranteed in a variety of
environments
• Each Resource can be
associated to a User
(Identity)
• The relationship
between a User and an
Object carries
information
Identity of objects and their relationships with other
objects and other identities (Humans)

90. IoT Data and … Identity of Things
Things have Identities (and Owners) People have Identities and use Things
Me
“My” Smart Thing
Identity Relation
Functional Relations
(events and commands)
Personal
Profiling
Who, Where, When, What, Why, …
Sensors
Identity Relation
Service Provider

91. What to do with data ?
Based on https://securityledger.com/2014/04/will-ot-big-data-create-darwinian-
struggle-for-insurance-carriers/

92. Dealing with Data in a Fair Way: Data Anonimization, Contracts
and Contextualization - Help the User to take control on his data
TSP/MSC Communication
Networks and Services (ComNETS)
https://www.oasis-open.org/committees/xdi/faq.php

93. Complexity Challenge

94. Are we ready to deal with Billions of
smart and independent things?
Billions of smart
objects cannot be
managed in a
traditional manner
There is the need to
move towards zero-
configuration and
autonomic systems

95. Dealing with Complexity at the edge:
Self-Organization of Networks
http://innovation.gsa.gov/blogs/OCIO.nsf/dx/Management-
Innovators-Bookshelf-Small-Pieces-Loosely-Joined-A-Unified-Theory-of-
the-Web-by-David-Weinberger-2002
• Management of complex and
dynamic “Networks of Networks”
will be critical
• No human intervention possible
• Competition on resources
Requires
• Self-organization
• Game theory techniques for
highly distributed systems

96. Autonomic Computing
► What is it? (from wikipedia)
► IBM has defined the following four functional
areas:
► Self-Configuration
► Self-Healing
► Self-Optimization
► Self-Protection
► Autonomic Networking follows the concept of
Autonomic Computing. Its ultimate aim is to apply
autonomic concepts to Networks.
► Functions of an Autonomic Network:
► Autognostic, Configuration management, Policy
management, Autodefense, Security
► Connection Fabric (i.e., The connection fabric
supports the interaction with all the elements and
sub-systems of the autonomic system)
► Principles of Autonomic Networking
► Compartmentalization (how to implement the
operational rules and administrative policies for
a given communication context)
► Function re-composition
► Atomization
► Closed control loop
http://www-03.ibm.com/servers/eserver/zseries/zos/sm/autonomic.html
Why? Millions of pervasive nodes!! There is the need of a new approach
Users should be “bothered” by such a cumbersome task!

97. A fast pace technological world
dominated by Software
Software Platforms
Protocol Availability
Data Usage and Privacy/Security
Processes and management of IoT
Reliability / Autonomics

98. Business Challenges

99. Source:Forrester
WW connectivity revenues in 2016
WW connectivity revenues in Million $
Connectivity revenues are forecasted to be
about 10-15% of total revenues
The Bandwidth Challenge …

100. «Per devices» effect
Source: Forrester
Average revenue per devices changes
very much according to applicative area
The Revenue Challenge …

101. Different Roles for a Distributed
Objects Ecosystem
Data Aggregator Info Aggregator
P
P
P
P
P
P
Data Producers Dispatcher Consumers
The Value Chain Challenge

102. An Example of an important Data
Aggregation Functionality for IoT
Sensor 1
Ingress
Queue A
Aggregator a
Aggregator b
Aggregator c
Ingress
Queue B
Events
Events
Security
Monitoring
In aggregation
nodes at the
edge of the 5G
Nework
Egress
Queue Y
Egress
Queue Z
Policing
Control and Management
Sensor 2
Sensor 3
Sensor n
Sensor n+1
Sensor n+2
Sensor n+3
Sensor
n+m
Aggregator d
Aggregator e
Aggregator f
• Intelligent Routing of Events and
Messages thanks to SDN
• Transaction Management
• R.T. extraction of Knowledge
Data Flow
Events
Events
Control Layer
need to bring Intelligence at the Edge of the Network
5G NetworkMEC MEC
The Tweeter of Things

103. A long value chain opens up
opportunities for many Actors
Source: Nokia Siemens Networks
«New»
Markets
Traditional
Markets
The Ecosystem Challenge

104. Which Business Model For the Internet
of Things?
 Thesis 2: Sensor information
should be made available on
an open platform to allow
everyone to offer higher level
services
 Thesis 3: Intermediaries are
needed as coordinating
central structures on the IoT
market
 Thesis 8: The value of the IoT market
grows more than linearly with the
number of consumers
 Thesis 9: Intermediaries should
consider subsidizing micro providers
to create an additional incentive for
service provisioning and enable the
intermediaries' business in the first
place
 Thesis 14: Incentives will be needed
to stimulate participation of a large
number of (micro) providers
Initial Observations on Economics, Pricing, and Penetration of the Internet of Things Market
Jens-Matthias Bohli, Christoph Sorge, and Dirk Westhoff
ACM SIGCOMM Computer Communication Review
The Biz Model Challenge …

105. Servitization is the capability of creating a link between a (physical) product and
a set of services and enriched functionalities that extend, complement, and add
value to the product itself
Servitization as a viable Business
Model for IoT

106. Source: Beecham Research
Siloing Challenge
Vertical Application domains vs.
Horizontal Platforms Challenges

107. Business Issues
Value Chain
Viable Business
Models
Prosumers and
Users

108. Societal Challenges

109. Identity of Things
• Things can collect user related actions and data
• Things can be used for tracking Users
• Owners of Things can collect a lot of data
The Privacy, Trust, and Ownership
Challenges
Big Brother Syndrome

110. Fair Management of Personal Data
• Volunteered data: created and explicitly shared by individuals
• Observed data: captured by recording the actions of individuals
• Inferred data: based on analysis of volunteered or observed information
Source:Bain&Company
Excerptform:“PersonalData-TheEmergenceofaNewAssetClass”(WEF)
Personal Data should be
properly regulated and
managed
A “user-centric personal data eco-system” (WEForum)

111. The Usability and Effectiveness Challenges
 How Many Sensors !! What
a complicated system
 What useful Services ?
 How to use them?
 How to deploy them
 How to maintain them?
www.bwired.nl

112. The Socialization Challenge of IoT
 A fundamental aspect of all adaptive systems is cooperation.
 Natural selection favors cooperation, if the benefit of the altruistic
act, b, divided by the cost, c, exceeds the average number of
neighbors, k, which means b/c > k.
 It is necessary enforcing altruistic behaviors in IoT networks (social
aspects on it)
Hisashi Ohtsuki, “A simple rule for the evolution of cooperation
on graphs and social networks”, Nature, Letters, Vol 441|25 May 2006|doi:10.1038/nature04605

113. Social Issues
The Big Brother Syndrome
Fairness to Users
Easiness of usage
Security and Privacy
Disappearing Interfaces
Providers responsibility
Customers’ acceptance

114. Towards a Virtual Continuum, an enabler for
the Programmable World

115. Evolutionary roadmap for key
functional elements
M2M IoT
Virtual
Continuum
Key elements:
Separated applications
Ad-hoc designed modules
Ubiquitous connectivity
SIM management
International agreements
Embedded SIM
Key elements:
Low-cost standard sensors
Short range communication
Capillary and macro netw.
Horizontal Services
Data aggregation in cloud
Third party development
Key elements:
“Virtual Objects”
Mirroring Things in cloud
Object Semantics
Data integration,
federation and portability
Cloud as developing platf.

116. The Virtual Continuum
Entanglement

117. An example: Amazon Dash
117

118. Virtual Continuum is Virtualization of
Resources
ICT
Resource
Local
Services
Virtualization
in the “Net”
Interface
Global
Services
Extended
Functions /Interface
Resource
ICT Resources
Physical
Resource
Virtualization
in the “Net”
API
API
Global
Services
Extended
Functions /Interface
Resource
API
APIPhysical/Logical
Resources
• Each Resource is
representable
• Each resource is
programmable
• Each Resource can be
functionally
augmented
Virtual Continuum

119. Virtualization and APIs as a means to
enter into the Virtual Continuum
Sensors Things Actuators
Real World
Objects
Each Virtual Object will have:
- A unique identifier
- A contract
- An API
- A communication mechanism
(PubSub)
Proprietary Protocols
Normalized
Interfaces
Virtual
Objects
Applications
• Decoupling of physical objects and
virtual ones (Virtual Continuum)
• Normalized applications can be built on
Virtual Objects (having an impact on
Physical Objects)
• Servitization
Terminals
Cloud

120. The Context for Virtual Continuum
ComputingCloud
images of
terminals
and objects
(Proxying and Security
Functions)
applications
& services
physical
world
smart
objects
mobile
devices
Virtualized
resources
Secure,
always best
connected
virtualized
channel
Brokering and
Aggregation Function
Virtual
Environment
network resources
processing resources
storage resources

121. An IoT Middleware View
API
Always Best
Connected Comm.
Sensors
Things
T
a
g
Tags Others
App Ecosystem
Platform Value
Ecosystem Value
Service/Apps
Value
Programmability
Value
Processing
Storage
Communications
Comm
Value
Communication Engine
(e.g., event based)
Autonomics and Self
Organization
Brokering of Virtual
Objects
Data Management
Objects
Registry
Objects
management
Extensive Objects Virtualization
API
Telco
Building
Blocks
MobileDevicePlatform
Native
Operating
System
Middleware
Functions
Terminal
to Cloud
Relationship
Terminal
to Capillary
Relationship
API
API

122. Back to the challenges

123. IoT Challenges

124. IoT Challenges categories
IoT
Challenges
Technical Business Societal
• Market Value
• New Business Models
• New Ecosystem
• Applications Domains
• New Business Processes
• User Needs
• Market and solutions
fragmentation
• …
• Security
• Privacy
• Trust
• Usability
• Effectiveness
• Social Control
• …
A Major Challenge of IoT
Global Cooperation !!!
• Platform
• Security
• Signaling and Protocols
• Data Management
• Cloud
• Large System Management (autonomics)
• Power Consumption
• Connectivity
• Programmability
• Complexity
• Standardization
• Virtualization
• Smart Objects
• Communication Paradigms (cooperation
and gossiping)
• …
Regulation

125. Grazie
Roberto TSP
roberto.minerva@telecom-sudparis.eu