IoT ARCHITECTURE

IoT
“A global infrastructure of the information society, enabling advanced services by
interconnecting (physical and virtual) things based on existing and evolving interoperable
information and communication technologies.”

“The internet of things in the internetworking of physical devices, vehicles, building, and
other items – embedded with electronics, software, sensors, actuators, and network
connectivity that enable these objects to collect and exchange data.”
 FEATURES OF IoT
⚫ Fundamental features of a sustainable IoT architecture include
⚫ Artificial Intelligence: IoT makes virtually anything “smart”. Knowledge can be extracted from generated data.
⚫ Interconnectivity: Anything can be interconnected, any time and anywhere using a communication infrastructure.
⚫ Distributivity: Data is gathered from different sources and processed by several entities in a distributed manner.
⚫ Heterogeneity: The devices in IoT are heterogeneous as based on different hardware platforms.
⚫ Interoperability: Devices from different vendors cooperate, in order to achieve common goals. Also, systems and
protocols will have to be designed in a way that allows objects (devices) from different manufacturers to exchange
data and work in an interoperable way.
⚫ Scalability: The number of devices that need to be managed and that communicate with each other is of a great
magnitude.
⚫ Security: Users feelings of helplessness and being under some unknown external control could seriously hinder
IoT's deployment.
⚫ Dynamic changes: The state of devices change dynamically, eg sleeping or waking up, connected or disconnected,
as well as context of devices including location and speed. The number of devices can also change dynamically.
⚫ Everything as a service: All resources are consumed as services.
⚫ Without addressing these conditions, the result of the IoT architecture is a failure.
 BASIC ARCHITECTURE

Figure 1: Basic Architecture of IoT

*BOSS: Business Operating System
 HARDWARE
Wireless Sensor Networks (WSN) and radio frequency identification (RFID) are considered as the two main building blocks of sensing
and communication technologies for IoT.

SENSOR NETWORK:
• Consists of sensing nodes (sensors) communicating in a wireless multi-hop fashion for gathering data.
• Sensor network can exist without IoT, but IoT cannot exist without a senor network.
• Sensors provide information, knowledge, or data about the Physical Entity (the identity of the Physical Entity and state of the
Physical Entity) they monitor.
• A sensor can measure the physical property and convert it into signal that can be understood by an instrument.
• The sensors enable the interconnection of the physical and digital worlds allowing real-time information to be collected and
processed.

Figure 2: Sensor Network

• There are various types of sensors for different purposes.
• Typical sensors could detect light, acceleration, air quality, movement, temperature, their location or humidity.
• Current research is investigating new small sensors to analyze even liquids or gases.
• The miniaturization of hardware has enabled powerful sensors to be produced in much smaller forms which are integrated into
objects in the physical world.
• E.g: miniaturised gas chromatography system for detecting volatile organic compounds developed at the Institute for Microsensors,
-actuators and systems (IMSAS) at the University of Bremen.
• Location-sensing techniques like triangulation, proximity or scene analysis can be combined with different transmitting technologies
based on radio, infrared light, magnetism, ultrasound or vision for designing location-aware objects.

Virtual Sensors:
• The data acquired by a set of sensors can be collected, processed according to an application-provided aggregation function, and then perceived as
the reading of a single virtual sensor.

Figure 3: Flow of information between real devices and virtual sensors

• A virtual sensor may not have any real sensor’s physical properties such as manufacturer or battery power information, but does
have other properties, such as: who created it; what methods are used, and what original sensors it is based on.
• Two degrees of complexity:
• The combination of a limited number of related sensors or measurements to derive new virtual data (usually done at the sensor
node or gateway level).
• The complex process of deriving virtual information from a huge space of sensed data (generally at the application level).

• Valuable higher-level knowledge is being derived from lower-level events and can be approached using different technologies from
many independent research fields (such as, discrete event simulation active databases, network management, or temporal reasoning),

Figure 4: Different levels for sensor virtualization.

 GATEWAY
• A gateway is a hardware device that acts as a "gate" between two networks. It may be a router, firewall, server, or other device that
enables traffic to flow in and out of the network.
• Gateway is also a node itself.
• The gateway node is considered to be on the "edge" of the network as all data must flow through it before coming in or going out
of the network.
• Traditional network gateways have mostly performed protocol translation and device management functions.
• They are not intelligent, programmable devices that could perform in-depth and complex processing on IoT data.
• Today's "smart" IoT gateways -- delivered by companies such as Dell Technologies, Wind River/Intel, Nexcom and others -- are
full-fledged computing platforms running modern operating systems (for example, Linux or Windows).
• These systems are sometimes also called intelligent gateways or edge gateways.
• The advantages of having a fully programmable and manageable platform closer to the IoT devices they support is immense:

• Trusted connectivity and security -- ensuring the integrity of the network and system in both directions
• Protocol and data bridge -- being able to translate and transfer data among and between systems operating with different
communications protocols and data formats, often requiring bidirectional communication capabilities
• Storage and analysis -- onboard application development platforms and storage to drive intelligence and decision-making closer to
the edge device
• Management -- the ability to provision, update and control access of connected devices to the system as well as policy-based
permissions
 TYPES OF VIRTUAL SENSORS BASED ON THE SERVICES OFFERED
1. Singular: This type of virtual sensor provides one to one mapping so that a single physical sensor can be virtualized, shared among
multiple applications.

2. Accumulator: It provides many to many mapping.

3. Aggregator: The function of Aggregator is to act on top of physical sensors, to read and do the simple computations like averaging,
finding maximum and minimum, sorting, counting as per application need.

4. Selector: It can select a single value from a group of values, based on specific criteria.

5. Qualifier: It is categorized to work on events based on priority. A Qualifier will be activated if a threshold for a single observable
quantity is breached. It checks the highest or lowest value from a group of observed sensor data.

6. Context-Qualifier: Virtual sensor designed for prioritizing events when threshold levels of multiple variables are breached at same
time. It is a typical example of many to one mapping, where multiple physical sensors are grouped and monitored together to alert a
SaaS level application.

7. Predictor and Compute: Physical sensors are unable to detect the events in priory. It may detect the occurrence of a present event,
but not the future event. A virtual sensor like predictor can be deployed to guess the occurrence of event in future, which may provide
sophisticated solution to patient monitoring like averting cardiac arrest. Also when decision making is involved in such scenarios,
compute virtual sensor can be used.
 List of some of the measurement devices used in IoT:
Accelerometers Temperature sensors
Magnetometers Proximity sensors
Gyroscopes Acoustic sensors
Light sensors Pressure sensors
Gas RFID sensors Humidity sensors
Micro flow sensors Image sensors

COMMUNICATION:

1. CHALLENGES
a) Addressing and identification: Things need to be identified through a unique address, on the basis of which they communicate with
each other. For this, we need a large addressing space, and a unique address for each smart object.
b) Low power communication: Communication of data between devices is a power consuming task, specially, wireless communication.
Therefore, we need a solution that facilitates communication with low power consumption.
c) Routing protocols with low memory requirement and efficient communication patterns.
d) High speed and nonlossy communication.
e) Mobility of smart things.
2. ENABLING TECHNOLOGIES FOR COMMUNICATION:

1. RFID (radio-frequency identification):

• Used for short range low power communication.
• To identify and track the data of the things.
• It allows the design of tiny microchips (called tags),which can be appended to an object of our daily life.
• Stored data in these tags can automatically be used to identify and extract useful information from the object.
• Tag acts as an electronic barcode.
2. Sensor:
• To collect and process the data to detect the changes in the physical status of things.
3. Smart Tech:
• To enhance the power of the network by devolving processing capabilities to different parts of the network.
4. Nano Tech:
• To make the smaller devices have the ability to connect and interact.
3. WIRELESS TECHNOLOGIES:
A) Telecommunication systems
• Initial/primary service: mobile voice telephony B) WLAN
• Large coverage per access point (100’s of meters –10’s •Initial service: Wireless Ethernet extension
of kilometers) •Moderate coverage per access point (10’s –100 meters)
• Low/moderate data rate (10’s of Kbit/s –10’s of •Moderate/high data rate (100’s Mbits/s)
Mbits/s) •Examples: IEEE 802.11(a-g), Wimax
• Examples: GSM, UMTS, LTE

C) Short Range: D) Other examples:

•Direct connection between devices – sensor •Satellite systems
networks
•Broadcast systems
•Typical low power usage.
•Examples: Bluetooth, Zigbee, Z-wave,
• One of the essential challenges in IoT is how to interconnect “things” in an interoperable way while taking into account the
energy constraints, knowing that the communication is the most energy consuming task on devices.
• Several low power communication technologies have been proposed from different standardisation bodies. The most common
ones are:

❑ IEEE 802.15.4 has developed a low-cost, low-power consumption, low complexity, low to medium range communication
standard at the link and the physical layers for resource constrained devices.

❑ Bluetooth low energy (Bluetooth LE, is the ultra-lowpower version of the Bluetooth technology that is up to 15 times more
efficient than Bluetooth.

❑ Ultra-Wide Bandwidth (UWB) Technology is an emerging technology in the IoT domain that transmits signals across a much larger
frequency range than conventional systems.
UWB, in addition to its communication capabilities, it can allow for high precision ranging of devices in IoT applications.

❑ RFID/NFC proposes a variety of standards to offer contact less solutions.

Proximity cards can only be read from less than 10 cm and follows the ISO 14443 standard and is also the basis of the NFC
standard.
RFID tags or vicinity tags dedicated to identification of objects have a reading distance which can reach 7 to 8 meters.
MIDDLEWARE:
SOFTWARE:
• Software layer that stands between the networked operating system and the application and provides well known reusable solutions
to frequently encountered problems like heterogeneity, interoperability, security, dependability.
• In charge of processing and managing the obtained raw data.
• Provides an abstraction level to users and developers.
• IoT requires stable and scalable middleware solutions to process the data coming from the networking layers.

Challenges addressed by IoT Middleware:

(1). Interoperability and programming abstractions: For facilitating collaboration and information exchange between heterogeneous devices,
different types of things can interact with each other easily with the help of middleware services.
Interoperability is of three types: network, semantic, and syntactic.
Network interoperability deals with heterogeneous interface protocols for communication between devices. It insulates the applications
from the intricacies / complication of different protocols.
Syntactic interoperability ensures that applications are oblivious of different formats, structures, and encoding of data.
Semantic interoperability deals with abstracting the meaning of data within a particular domain.

(2) Device discovery and management: This feature enables the devices to be aware of all other devices in the neighborhood and the services
provided by them.
Here the infrastructure is mostly dynamic.
The devices have to announce their presence and the services they provide.
Finally, any IoT middleware needs to perform load balancing, manage devices based on their services, capabilities and levels of battery .
3) Scalability: Middleware makes the required changes when the infrastructure scales, as a large number of devices are expected to
communicate in an IoT setup.

(4) Big data and analytics: IoT sensors typically collect a huge amount of data.
It is necessary to analyze all of this data in great detail.
As a result a lot of big data algorithms are used to analyze IoT data.

(5) Security and privacy: The middleware has built-in mechanisms to address security and privacy issues, as the IoT applications are mostly
related to someone’s personal life or an industry.
The middleware should have built-in mechanisms to address such issues, along with user authentication, and the implementation of
access control.

(6) Cloud services: The cloud is an important part of an IoT deployment.

Most of the sensor data is analyzed and stored in a centralized cloud.
It is necessary for IoT middleware to seamlessly run on different types of clouds and to enable users to leverage the cloud to get better
insights from the data collected by the sensors.

(7) Context detection: The data collected from the sensors needs to be used to extract the context by applying various types of algorithms.
The context can subsequently be used for providing sophisticated services to users.
 Middlewares can be classified as follows on the basis of their design;

(1) Event based:

•All the components interact with each other through events.
•Each event has a type and some parameters.
•Events are generated by producers and received by the consumers.

(2) Service oriented:

•Is based on Service Oriented Architectures (SOA), and have independent modules that provide services through accessible interfaces.
•A service oriented middleware views resources as service providers.
•Service oriented middleware must have runtime support for advertising services by providers and support for discovering and using services by
consumers. e.g HYDRA

(3) Database oriented:

•In this approach, the network of IoT devices is considered as a virtual relational database system.
•The database can then be queried by the applications using a query language.

(4) Semantic:
•Semantic middleware focuses on the interoperation of different types of devices, which communicate using different formats of data. It
incorporates devices with different data formats and ties all of them together in a common framework.
•The framework is used for exchanging data between diverse types of devices.
•This technique allows multiple physical resources to communicate even though they do not implement or understand the same protocols.

(5) Application specific:

•This type of middleware is used specifically for an application domain for which it is developed because the whole architecture of this middleware
software is fine-tuned on the basis of requirements of the application.
•The application and middleware are tightly coupled.
 APPLICATIONS:
• Applications are the interface to the Internet of Things and provide tools for data-entry and retrieval, analysis, planning, forecasting
and more.
• This part of the software layer constitutes the front end of the whole IoT architecture through which IoT potential will be
exploited.
• It provides the required tools (e.g. actuating devices) for developers to realize the IoT vision. In this vision, the range of possible
applications is impressive.
 A NETWORK ENGINEER’S VIEW vs
DESIGNER’S VIEW OF IoT

Figure 5: A network engineer’s view of IoT Figure 6: A designer’s view of IoT

Designers look at IoT tech as a front end and back

Engineer’s look at IoT Stack as a protocol map. end with enabling infrastructure in between.
THREE LAYER ARCHITECTURE
The client side.

A pathway for connecting clients and

operators.

Operators on the server side

Figure 7: Three layer IoT Architecture

 1. PERCEPTION LAYER:

• Includes objects to be detected (from physical moving objects, such as humans, cars, ) or factors to be observed ( temperature, or
humidity) .
• Includes sensors, other hardware such as; embedded systems, RFID tags and readers and others.

2. NETWORK LAYER
• The network layer is responsible for processing the received data from the Perception Layer.
• It is in charge of transmitting data to the application layer through various network technologies, such as wireless/wired
networks and Local Area Networks (LAN).
• The main media for transmission include 3G, 4G, Wifi, bluetooth, Zigbee, UMB, infrared technology, and so on.
• Huge quantities of data will be carried by the network.
• Crucial to provide a sound middleware to store and process this massive amount of data.

3. APPLICATION LAYER
• It is the layer at the top of the architecture and is responsible for delivery of various applications to different users in IoT.
• It realizes the use of smart objects by a set of functions to users to meet defined requirements.
• The applications can be from different industry segments such as: manufacturing, logistics, retail, environment, public safety,
healthcare, food and drug etc.
• With the increasing maturity of RFID technology, numerous applications are evolving which will be under the umbrella of IoT.
Device Data Procesing Thing/Devic
Edge
and platform e/Cloud
Figure 8: The 4 stag IoT Solutions Architecture
Stage 1: Networked Things (Wireless sensors and actuators)

Sensors convert information obtained in outer world into data for analysis.
Actuators can intervene the physical reality, e.g switching off lights

Stage 2: Sensor data aggregation systems and analog- to-digital data conversion

This stage makes data both digitalized and aggregated. Enormous amount of information collected on previous stage is processed and
squeezed into optimal size for further analysis.

Stage 3: The appearance of edge IT systems

The prepared data is transferred to the IT world. So basically the analytics and pre-processing of data is done .
Also some additional processing might happen here , prior to the stage of entering the data center.

Stage 4: Analysis, management, and storage of data

The main processed on the last stage of IoT happen in the data center or cloud. It enables in-depth processing, along with a follow-up
revision for feedback. Data from other sources might also be included to ensure the in-depth analysis.
 REFERENCE ARCHITECTURE
⚫ This architecture comprises of service layers and presentation layers.
⚫ Service layers include Event Processing and Analytics, Resource Management and Service Discovery, as well as Message
Aggregation and Enterprise Service Bus (ESB) services built on top of communication and physical layers.
⚫ API management, which is essential for defining and sharing system services and web-based dashboards (or equivalent) for
managing and accessing these APIs, are also included in the architecture.
⚫ Due to the importance of device management, security and privacy enforcement in different layers, and the ability to
uniquely identify objects and control their access level, these components are pre-stressed independently in this architecture.

Figure 9 : Reference Architecture for IoT

 REFERENCE ARCHITECTURE CONTINUED
Reference
Architecture
▪Web APIs and Representational State
Transfer (REST)- based methods have
SOA based API Oriented been introduced as promising
Architecture Architecture
alternative solutions to RMI.
▪This helps in using the communication
channel and processing the power of
Sensing layer
Network
Service layer
Interface devices more efficiently.
layer layer

o Sensing layer is integrated with available hardware objects to sense the status of things.
o Network layer is the infrastructure to support over wireless or wired connections among
things.
o Service layer is to create and manage services required by users or applications.
o Interfaces layer consists of the interaction methods with users or applications.
The IoT Layered Architecture

Figure 10: IoT layered architecture

THE DETAILED IOT LAYERED ARCHITECTURE

Figure 11: Detailed IoT layered architecture

THE SEVEN LAYER IOT ARCHITECTURE

Figure 12: The seven layer IoT Architecture

1. Physical Devices & Device Controllers (The “Things” in IoT)

IoT “devices” are capable of:

• Analog to digital conversion, as required
• Generating data
• Being queried / controlled over-the-net
• Devices are diverse, and there are no rules about size, location, form factor, or origin. Some devices will be the size of a silicon
chip. Some will be as large as vehicles. The IoT must support the entire range.

2. Connectivity (Communication & Processing Units)

Connectivity includes:
• Communicating with and between the Level 1 devices and the network.
• Communication across the network.
• Communicating with and between the Level 2 and low-level information processing occuring at Level 3.
• Reliable delivery across the network(s).
• Implementation of various protocols.
• Switching and routing.
• Translation between protocols.
• Security at the network level.
 3. Edge (Fog) Computing (Data Element Analysis & Transformation)
•The functions of Level 3 are driven by the need to convert network data flows into information that is suitable for storage and higher
level processing at Level 4.
Level 3 processing can encompass many examples, such as:
o Evaluation: Evaluating data for criteria as to whether it should be processed at a higher level
o Formatting: Reformatting data for consistent higher-level processing
o Expanding/decoding: Handling cryptic data with additional context (such as the origin)
o Distillation/reduction: Reducing and/or summarizing data to minimize the impact of data and traffic on the network and
higher-level processing systems
o Assessment: Determining whether data represents a threshold or alert; this could include redirecting data to additional destinations

•Functions Include:
o Data filtering, cleanup, aggregation
o Packet content inspection
o Combination of network and data level analytics
o Thresholding
o Event generation
 4. Data Accumulation (Storage):
• Makes the network data usable by applications.
• Converts data-in-motion to data-at-rest.
• Converts format from network packets to database relational tables.
• Achieves transition from ‘Event based’ to ‘Query based’ computing.
• Dramatically reduces data through filtering and selective storing.

5. Data Abstraction (Aggregation & Access)

• Abstracts the data interface for applications
• Create schemas and views of data in the manner that applications want
• Combines data from multiple sources, simplifying the application
• Filtering, selecting, projecting, and reformatting the data to serve the client applications
• Reconciles differences in data shape, format, semantics, access protocol, and security

6. Application (Reporting, Analytics, Control)

7. Collaboration & Processes (Involving people and business processes)