IoT Lecture 2 PDF

# Internet of Things I - Lecture 2 ## IoT Enablers - The world of ubiquitous computing consists of real-life objects which are capable to **sense**, **communicate**, **analyze**, and act according to the situation. - In general, **miniaturization**, **portability**, **ubiquitous connectivity**, **...

# Internet of Things I - Lecture 2 ## IoT Enablers - The world of ubiquitous computing consists of real-life objects which are capable to **sense**, **communicate**, **analyze**, and act according to the situation. - In general, **miniaturization**, **portability**, **ubiquitous connectivity**, **integration** of a diverse range of emerging devices, and **pervasive availability** of digital ecosystems (i.e. Cloud) are the general enablers that play a significant role for enabling IoT systems. ### IoT General Enablers: Five Stages The five stages of IoT functional view or information value loop are related to: - data creation - data communication - data aggregation - data analysis - Actions to achieve set goals. | Stage | Responsible Component | |---|---| | Create | Sensors | | Communicate | Networks | | Aggregate | Integrations | | Analyze | Augmented Intelligence | | Act | Augmented | ## IoT Enabling Technologies | Technology | Description | |---|---| | Identification and Sensing Technologies | Include the development of devices (sensors) that converts any physical stimulus into an electronic signal | | Wireless Communication and Networking | Include (network) devices that are able to communicate electronic signals | | Aggregation Standardizations | Include technical standards that enable efficient data processing and allow interoperability of aggregated data sets | | Augmented Intelligence | Include analytical tools that improve the ability to describe and predict relationships among sensed data | | Augmented Behavior | Include technologies and techniques that improve compliance with prescribed action | ## Identification and Sensing Technologies - **Identification** methods offer unique identity of objects within the network, and public IP addressing provides the unique identity to the smart things over the Internet. - **Sensing** in IoT involves the originating of data from interrelated smart things through the use of sensors and actuators. - A **sensor** is basically an electronic device responsible to **Produce** electrical, optical, and digital data deduced from the physical environment that further electronically transformed into useful information for intelligent devices or people. - **Actuators** are the technological complement of sensors that are responsible to convert an electric signal to nonelectric energy. ## Wireless Communication and Networking - **Wireless communication** and **wireless networking** are the core of Wireless Identification and Sensing Technologies (WIST), which play a vital role in the IoT. - WIST refers to **RFID-based** sensors and **WSNs**. ### RFID System A typical RFID system consists of the following three components: - An **RFID Tag**, also known as Transponder or Smart Label, composed of an antenna, (optional) battery, and semiconductor chip. - An **Interrogator**, also known as Reader or read/write device, having an RF module, a control module, and an antenna. - A **Controller**, also known as Host or Workstation, to store required information in a database. **Building blocks of RFID system:** - RFID Tag - Interrogator - Controller - The RFID tag and the interrogator in the RFID system do **not require line of sight** and **communicate** with each other through **radio waves**. Within the transmission range, the Interrogator reads the required information (i.e. serial number, manufacturer, location, usage history, maintenance schedule, etc.) stored **on the RFID tag** and **directs this information toward the Controller** that ultimately uses this information for various purposes. - RFID tags **can be of two types**: - **Active Tags** (Tags having on-board power source) - **Passive Tags** (Tags without an on-board power source). - **Active RFID Tags** have greater capabilities (i.e. **large memory**, **long read range**, **high data transmission rate**, **lower infrastructure cost**, etc.) than Passive Tags but are more complex and expensive. - **RFID, sensors, and RFID sensor** are connected to the Internet through heterogeneous network devices, i.e. **Bluetooth, Access Points (APs), Wi-Fi routers, Gateways**, etc. Therefore, a **unique IP address** is required for all smart things on the Internet. Smart things require **continuous connectivity** and need to be connected to various heterogeneous networks. - Sensors in the **WSN have cooperative capabilities** to sense and transfer data. - **In fact**, sensed data **is required to be collected** to store it in a database, data warehouse, or Cloud for analysis to make **in-time right decisions**. - Typically, this kind of transmission of data in IoT involves **multiple types of wireless communication** and network technologies. Wireless communication in IoT **emphasizes the way how** heterogeneous devices are able to communicate with each other in a sustainable way that they can understand. - Depending on the range and/or rate of data transmission, a number of network technologies are available. ### Network Technologies | Technology | Connectivity type | Network type | |---|---|---| | USB | Wired | Personal Area Network | | Ethernet | Wired | Local Area Network | | Bluetooth/Bluetooth Low Energy | Wireless | Personal Area Network | | ZigBee | Wireless | Personal Area Network | | Near Field Communication (NFC) | Wireless | Personal Area Network | | Wi-Fi | Wireless | Local Area Network | | WiMax | Wireless | Metropolitan Area Network | | 2G/3G/4G, LTE/LTE-Adv. | Wireless | Wide Area Network | ## Aggregation Standardization - Aggregation refers to the **gathering of sensed data** in a way that eases the process of handling, processing, and storage of data. Aggregation, besides providing the ease of handling, is also helpful to **extract meaningful conclusions** for **future decision-making**. Within the context of data aggregation in IoT, Standardization is one of the most important issues. - Standardization is **inevitable** within the domain of IoT and is **essential to guarantee interoperability, scalability, alike data semantics, security, and privacy**. - Several standards are required to be followed to realize data aggregation in IoT. However, **Technology Standards** and **Regulatory Standards** are two broad categories of standards, which are related to the process of aggregation. - Technology standards include **network protocols** (set of rules dealing with the identification and connectivity among devices), **communication protocols** (set of rules with the provision of a common language for devices' communication), and **data-aggregation protocols** (set of rules that assist the aggregation and processing of sensed data). - Regulatory standards are important in the **evolution of IoT** and deal with the **ownership, use, and sale of the data**. ## Augmented Intelligence - Analysis of **collected data** demands the practice and advancements of different augmented cognitive technologies. - Augmented intelligence **enables the automation of systems to perform descriptive** (amenable representation of data to recognize insights), **predictive** (to foresee future consequences), and **prescriptive** (related to optimization) analysis. - Computer vision, natural-language processing, and speech recognition are a few examples of cognitive technologies that are playing an important role in predictive and prescriptive analytics. ## Augmented Behavior - Augmented behavior involves the **actions that are required to perform while considering all phases** of the information value loop, i.e. from sensing to data analysis. Following the changes in people's behavior and organizational processes, augmented behavior **supports the manifestation of suggestive actions with the use of advanced technologies** (i.e. M2M and **Machine** to **Human** [M2H]). - At this phase of the information loop, IoT concerns transferred from **data science to behavioral sciences.** Advancements in M2M and M2H are main driving forces that support the **cognitive and actuation abilities** of machines to **understand the environment** and act logically, respectively. # IoT Architectures - Considering basic IoT elements, the IoT essentially connects a **diverse range** of hardware devices to a plethora of application domains. - The heterogeneity of application and hardware domains imposes varied significant challenges that are essential to meet for the **successful deployment** of simple and complex IoT systems. - In addition to **heterogeneity**, considering all time ubiquitous connectivity, IoT needs to address a **diverse range of issues** including scalability, interoperability, and security/privacy, for high traffic/storage needs that **ultimately affect the architecture** of IoT systems. ### IoT Elements - Sensing - Digital Identification - Communication - Semantic - Services - Computation ## Three-layer IoT architecture The simplest IoT architecture consists of **three layers**, i.e. **perception**, **network**, and **application layers**. | Layer | Responsibilities | |---|---| | Application | Provides Services; Semantic Analysis | | Network | Enroutes the received data to distant application | | Perception | Information Collection; Data Transmission | ### Perception Layer - The perception layer at the **bottom** of IoT architecture **is responsible** for the collection of various types of information through physical sensors or components of smart things (i.e. **RFID sensors, objects with RFID tags or sensors, etc.**). - Moreover, the perception layer **transmits the processed information** to the upper network layer **via service interfaces**. - The main challenge at the perception layer is related to the **recognition and perception** of environmental factors through the **use of low-power and nanoscale technology** in smart things. ### Network Layer - The network (also known as **transmission**) layer **accepts processed information** from the perception layer and **forward Internet** and other communication technologies. - A number of communication technologies (i.e. **Wireless Local Area Networks (WLAN), Wi-Fi, LTE, Bluetooth Low Energy [BLE], Bluetooth, 3G/4G/5G**, etc.) are **integrated with IoT gateways** that handle heterogeneous types of data to or from different things to applications and vice versa. - In addition to network operations, the Network layer in some cases **enhances to perform information operations within the Cloud**. ### Application Layer - The application layer **is responsible for the provisioning of services requested by the users**, e.g. temperature, moisture, humidity, air pressure, light intensity measurements, etc. - In addition to the user-requested services, *the application layer provides data services (i.e. **Data warehousing, BigData storage, data mining**, etc.) to perform semantic data analysis. ## Five-layer IoT architecture - **Object** (Perception), **Object Abstraction** (Network), **Service Management** (middleware), **Application**, and **Business** are the names of the five layers in five-layer IoT architecture. | Layer | Responsibilities | |---|---| | Business | Business Model; Flow Charts; Graphs | | Application | Provides Services; Semantic Analysis; Generate Analytic Reports | | Service Management | Information Processing; Store Data in Database | | Network | Switching and Routing; Data Generation; Reliable Transmission; Protocol Translation | | Perception | Smart Thing/Object; Information Collection; Data Transmission | ### Perception Layer - The Object layer (perception layer or device layer) **primarily deals with the identification, collection, and processing** of object specific information through a diverse range of physical sensors. - Physical sensors at this layer are based on **different sensing principles (i.e. capacitance, induction, piezoelectric effect, etc.)** and are responsible to **digitize and transfer sensed data to Object Abstraction layer through secured channels**. ### Network Layer - **Object Abstraction Layer or Network layer is responsible** for secure data transmission from physical sensors to information processing systems by using various technologies. ### Middleware Layer - The **smart things in IoT implement a diverse range of services** and each smart thing is connected and capable to communicate with smart objects that have implemented the same type of services. - The service management layer **provides pairing of services with its requesters' applications and enables IoT application programmers to deal with heterogeneous data created** by smart things with different hardware specifications. - This layer includes **the processing of received data before transmitting to the application layer.** ### Application Layer - The application layer **is responsible for the provisioning** of services requested by the users. - In addition to the user-requested services, the application layer **provides data services** (i.e. Data warehousing, BigData storage, data mining, etc.) to perform semantic data analysis. ### Business Layer - The **business layer is responsible to manage overall activities/services of the IoT system** through the creation of flowcharts, business models, and graphs on received processed data from the application layer. - In addition, based on Big Data analysis, this layer supports **automatic decision-making as well as the making of smart business strategies.** ## Six-layer IoT architecture - The six-layer architecture comprises of **Focus Layer**, **Cognizance Layer**, **Transmission Layer**, **Application Layer**, **Infrastructure Layer**, and **Competence Business Layer**. - This architecture model is proposed to design the integration of more than one IoT system (**focusing on different subject areas**) and analyzing their implications on business value. | Layer | Responsibilities | |---|---| | Competence Business | Analysis of Business Models | | Infrastructure | Cloud; BigData | | Application | Categorization of Received Information | | Transmission | Secure Data Transmission from sensors to Application layer | | Cognizance | Sensing Information Collection | | Focus | Smart Object Identification; Aspects of IoT Systems | ### Focus Layer - The modules at this layer **are responsible for the identification** of smart objects while focusing on the aspects of IoT systems under consideration. ### Cognizance Layer - This layer consisting of **sensors, actuators, and data monitoring modules is responsible** for the collection of sensing information from smart objects (identified in the Focus layer). ### Transmission Layer - This layer **is responsible for the transmission of sensed data** from the cognizance layer to the application layer. ### Application Layer - This layer **is responsible for the categorization of received information** on the basis of application modes. ### Infrastructure Layer - This **deals with the availability** of service-oriented technologies, i.e. Cloud, BigData, datamining, etc. ### Competence Business Layer - This layer **includes the analysis of business models of IoT systems.** ## Seven-layer IoT architecture - Seven-layer IoT architecture comprises **seven layers** including **Things**, **Edge Computing**, **Data Accumulation**, **Data Abstraction**, **Application**, **People collaboration and processes layer**. - The architecture **provides the simplest way to understand the functionality of IoT systems**. The functionality of each layer is described in the following: | Layer | Responsibilities | |---|---| | Collaboration and Processes | Consideration of People and Business Process | | Application | Perform Analytics; Generate Reports | | Data Abstraction | Combine Data from Sources; Application-oriented Creation of Schema/Views of Data | | Data Accumulation | Take Data from Network Packets; Store Data in Database | | Edge/Fog | Data Filtration; Data Aggregation; Data Analysis; Packet Content Inspection | | Connectivity | Switching and Routing; Data Generation; Reliable Transmission; Protocol Translation | | Things | Data Generation; Data to Signal Conversion | ### Things Layer - The Things layer **comprises endpoint devices** of IoT systems including smart things (with sensors and controllers) and smart mobile devices (i.e. smartphones, tablets, **Personal Digital Assistant** [PDA], etc.) to **send and receive information**. - The Things layer supports a diverse range of devices in terms of form, size, and sensing principles; the layer **is capable to gather data and conversion of analog observations to digital signals.** ### Connectivity - Considering a **diverse range of communication and networking protocols**, the connectivity layer **is responsible for the in-time transmission of observed data** within and between smart things and across different networks. - In other words, **horizontal communication between smart things and switching/routing** and secure data transmission at different network levels are the basic functionalities of this layer. - Although, **communication and connectivity through existing IP-enabled network standards** is the main focus of IoT reference architecture, however, the involvement of **non-IP-enabled devices demands gateway standardization.** ### Edge/Fog Computing - **Edge/Fog Computing layer is responsible for the conversion** of heterogeneous network data flows into information that is appropriate in terms of storage and analysis. - According to the notion of **early information processing in intelligent IoT systems**, this layer initiates **limited processing on the received data at the edge of the network**, which is mostly referred to as Fog computing. - **Data formatting, reduction, decoding, and evaluation are the basic functionalities** of this layer. IoT gateway is an example device at this level. ### Data Accumulation - **Data accumulation or placement of moving data on disk is done at this layer.** - In other words, at this layer,** event-based data is converted to query-based data for processing.** - Considering the interests of higher layers in available accumulated data, *this layer performs *filtering or selective storing to reduce data*. ### Data Abstraction Layer - Data abstraction layer **is related to the rendering and storage of data in such a way** that reconciles all the differences in data formats and semantics for the development of simple and performance-enhanced applications. ### Application Layer - Considering the application requirements, **the interpretation of level 5 data is done at this layer.** - Applications are diverse in nature (including system management and control applications, business applications, mission-critical applications, analytical applications, etc.); therefore, **relevant data interpretation demands vary from application to application.** - If **data is efficiently organized at layer 5**, then information processing overhead gets reduced at this layer, which ultimately supports parallel activities at end devices. ### Collaboration and Processes - **In IoT, different people with different aims use the same application.** - Therefore, in IoT, **the ultimate objective is not the creation of applications but the empowerment of people to do work in a better way.** - **In collaboration and communication for business, processes mostly transcend multiple IoT applications.**