To survey on architecture of iot platform Essay

To Survey on architecture of IOT Platform


Matoshri College of Engineering


[email protected]

Ms. P.R.Dholi

Matoshri College of Engineering


Abstract—. Internet of Things can be presented as a network of surrounding things which are connecting to the Internet such as various sensors, vehicles, devices which can be monitored, detected, controlled. The things are embedded with the sensors to sense the environment and communicate with other things. Current IoT platforms have a market share and offer for the customers some competitive advantages and features to make them select it and encourage their choice. It provides several services and applications such as data acquisition and analytics, device management, integration, security, insight to users on the operations and ability to identify and manage devices. Finding an appropriate IoT platform for a given field of application is a challenge any company is facing when wanting to select the appropriate platform from the mess of different IoT platforms.

Keywords—IOT, Platform, M2M",

I. Introduction

The term Internet of Things (IoT) was first defined by Kevin Ashton in 1999. Since then, a variety and number of devices connected to the internet has increased exponentially, specially in the recent years. Internet of Things (IoT) has been growing rapidly due to recent advancements in communications and sensor. The Internet of Things(IoT) refers to the type of the network which connect anything i.e. physical objects- devices, buildings, vehicles and other items embedded with software, sensors and network connectivity based on stipulated protocols that enables these objects to collect and exchange data technology. The Future Internet enables us to have immediate access to the information of the physical world and its objects. As such, Internet of Things (IoT) has been adopted to incorporate the digital information and the real world of devices. The accelerated growth of IoT industry requires robust IoT platforms which address the renewal requirements such as in the smart cities applications, an enormous amount of data has to be handled.

IoT platforms provide various capabilities in the industry, Emerging industrial IoT and the fourth industrial revolution (Industry 4.0) provides the flexibility for the planners and implementers and leads to better decision-making. Moreover, the machine monitoring and the cloud services in addition to provided applications contribute to growth and production.

Although the functionality provided by IoT platforms is similar or even equal, their implementation and the underlying technologies are different. Sometimes, platform selection process is done without a detailed analysis of requirements [2].

The IoT is gaining increasing attention. The overall aim is to interconnect the physical with the digital world. Therefore, the physical world is measured by sensors and translated into processible data, and data has to be translated into commands to be executed by actuators. As a result of different approaches, standards, and use cases, there is a wide variety and heterogeneity of IoT platforms. This leads to difficulties in comprehending, selecting, and using appropriate platforms. Cloud based platforms are used for processing and analyzing larger data sets.

The vision of the Internet of Things (IoT) describes a future where many everyday objects are interconnected through a global network. They collect and share data of themselves and their surroundings to allow widespread monitoring, analyzation, optimization, and control [27]

The internet of things is a trending subject. The things when connected to internet, users can access the things through the internet and has a control over the things. The things are able to sense and be identified uniquely. By giving things access to the internet, the environment is monitored, gets notified and taken action to control the environment. The application of the IOT is in many fields including home automation, logistics and transportation, agriculture, healthcare, banking, energy management system, smart city, smart parking system, smart traffic control system.IOT is a combination of sensor network technology, cloud computing and big data concepts.

II. literature survey

Few researchers showed some results for the study and comparisons of IOT platforms , which is given below:

Hamdan Hejazi, Husam Rajab et. al.",[1] addressed about IOT reference platform as a simpler four-part architecture as shown in fig. (1) sensing component which include sensors, actuators and devices",(2)a communication and identification component represents the communication protocols (and a gateway if needed), (3) a computation and cloud component represent the tasks of “processing unit” of IoT and finally (4) the services and applications component that represents the provided services and features offered to the user to connect and to control the environment through the IoT platform. There might be direct communication between “sensing component” and “services and applications component” without the computation and cloud component. However, then we have no IoT platform in the strict sense.

[image: image1.png]

Fig 1:Components of IOT

Akshat Savaliya, Aakash Bhatia et. al., [2]

Much research in recent years has focused on data mining in Internet of Things (IoT) which connects physical objects, person to person, person to machine or machine to machine via internet and manages information . Across various network infrastructures, IoT allows sensing of the objects and remotely access which in turn enables opportunities for a better integration between real and computerized world. It results into an improved effectiveness, exactness and better economic outcomes.

They define issues in data mining in IOT.

A. Efficiency in data gathering

B. Data abstraction and aggregation

C. Distributed data processing and mining

Jasmin Guth, Uwe Breitenbucher, et. al.",[3]

A lack of standardization and development in silos has led to a heterogeneous platform landscape. As a result of this heterogeneity, comparing and selecting one of these platforms is a difficult task. Not only do they use different concepts and technologies, platforms are described with synonyms or homonyms, or differ in granularity. To help with these problems, they introduced an IoT reference architecture which is based on existing platforms.They defined each component and described the communication between them. These components do not necessarily have to stay separated, but can be combined.They compared reference architecture to eight existing platforms, four of which are open-source. They showed that the components of their architecture map to those of the existing platforms. When comparing or evaluating these different platforms, their IoT reference architecture can be a useful tool. Besides, it may be useful by providing a common basis on which to base new IoT platform designs.

Tetsuya Yokotani [4]

Requirements of lightweight IoT protocols can be ummarized as follows.


(1). Short protocol headers

(2). Simplified communication sequences

(3). Optimized communication path

(4). Capable of QoS control for every IoT application

(5). Independent of TCP connections for reliable transfer

(6). Compatible with data format of TCP/IP or UDP/IP

Currently, various lightweight IoT applications have been discussed. These are categorized into two groups. One group uses simple communication sequences based on legacy architecture. For example, CoAP is one such candidate. The other group intends to be independent of IP addressing and “information” base access, and is referred to as ICN” technology. MQTT.

Jahoon Koo, Young-Gab Kim [5] address difference between heterogeneous IoT platforms where device ID interoperability in diverse IoT platforms hence to overcome this problem, thee analyze the existing device ID systems and propose the conceptual model of IoT device ID translator, named IoT DNS (Device Name System), which can translate diverse types of device ID into oneM2M device ID format. A various IoT platforms are under development, including oneM2M, OCF 'IoTivity', Apple 'HomeKit', Samsung 'ATIK', Google 'Brillo·Weave', AllSeen Alliance 'AllJoyn' and GS1 'Oliot'. In the future, it will generalize not only smart home, but also various fields incorporated IoT technology.

However, the devices currently used in the smart home environment have common limitations. In order to use smart home devices, that is IoT devices, the device registration process must be performed using the smartphone application after purchase. In this process, devices are searched through the network. Since the device identification (ID) used for each platform differs, devices of other platforms, that is smart home devices of other companies, are not searched. Therefore, the smart home environment can be configured only as IoT devices that operate the same ID system. Currently, IoT communication can be performed with devices released by one company only. each platform uses different device ID system. To identify devices between heterogeneous platforms, a translator that integrates or translates device ID is needed. Consolidating the device IDs of all platforms is not possible because all the device IDs used in each platform architecture must be modified. Therefore, a translator is needed that analyzes and translates heterogeneous platform device IDs to the appropriate platform device ID. The conceptual model, named IoT DNS proposed .

Slavko Zitnik, Marko Jankovic, et. al., [6]

They presents an interoperable and extensible IoT framework that follows oneM2M standard. Currently there are more than 300 IoT platforms available that offer different features. The current IoT software platform landscape is enormous. All big vendors have already introduced their own platforms, for example Watson (IBM), HANA (SAP), Jasper (Cisco), AWS IoT (Amazon), Azure IoT (Microsoft), HomeKit (Apple), Brillo (Google), IoTivity (Intel) or AllJoyn (Qualcomm). Each of these platforms offers its own IoT ecosystem, which is hard to interconnect with others or integrate with additional unsupported devices. Therefore, they used non-commercial and open-source IoT platforms that support at least one of the key IoT protocols. They implement a reference model.

M. Sruthi, B. R. Kavitha[7]

There are several platforms that provide services such as data analytics, business analytics, notification, etc. Each platform provides a specific service. They aims to give an overview of various platforms for IOT. This would help the users to choose their platform that is suitable for their application.They have taken a list of cloud providers for IOT, the protocols supported by them and their special features such as some engines for data analytics or a middleware to connect any things to cloud dynamically and their special services such as business analytics services and notification call and manually setting some threshold value and when the event happens the cloud can triggers the action in an environment.

III. iot platform

IoT platforms consist of a huge number of objects connected around the world. It connects the edge of devices, gateways, and data networks to cloud services and applications. The objects could be surrounded or separated by long distances in different environments but controlled by the centralized management that plays the role of the processing unit of the IoT platform. To understand the behaviour of IoT platforms, there is a need to investigate and identify the elements/ components/blocks to more comprehend IoT attitude and absolute sense. In this paper, the essential blocks of IoT platforms are presented as components because every block of IoT platforms becomes an attractive field of research, they make a loosely coupled system and all the blocks are influential in the competition between IoT vendors.


To deploy various IoT services, it is important that the IoT platform provides information processing and communication network technologies.

First, IoT networks and multi-network connectivity. Recently, various types options of network technology were used to link IoT devices. However, the best choice of the networking solution depends on how and where it will be installed, as long as the required standard of service is guaranteed. Therefore, an inclusive IoT platform should support the communication and provide with all essential IoT types to display the most significant flexibility for recent and future projects.

Second, IoT service management, using an IoT platform is an important point to get better managing of the work and business and also to improve the capacity, and optimize operations. To keep on IoT solutions continually working",essential to managing the data and IoT networking

simultaneously. IoT platforms should provide dministration for accessing the user-controlled software tools to retain managing the endpoints and the connections via the networks as an aspect of IoT solution. Also, an appropriate on-demand service management implementation allows the control of the IoT network, providing adding, moving, removing, or changing IoT device reporting functions. IoT devices and platforms could be managed efficiently for enhancing the IoT deployments[8].

Third, IoT data management and application enablement, for orchestrating data. Most of IoT solutions impact various sensors that can produce a high volume of data over time, such as condition, location, and status. Data streams composed and stored from the collected information. Each data point is usually short, while the amount of information received rapidly, relying on the reporting frequency of IoT devices. An IoT platform supports the ability to secure and normalize the data from different IoT endpoints, practically any sensor and any device reading. Multiple methods are sending streams of information and after receiving information could perform fragmentation, so the collected data can be processed efficiently, used, or reacted depending on the data obtained through applying the commands.

Fourth, analytics, statistic processing and data management by various data connections and hardware have to provide a useful result with accurate data analytics. The farthest aim of collecting the data is to fuel better business outcomes

through increased visibility and insight. An IoT platform could support complete information and view of data analytics and capabilities that can extract the data and retain business from shipwright of new information that might organize weakly. IoT platform will have analysis performed, as well as the capability to rise specific third-party analytics software via secure API and services.


The most challenging question for a company is how to select the suitable IoT platform to from a massive offer of

different IoT platforms of different vendors and different

providers. Each provider has specific features and different

services distinguishing it from the others. The consideration

depends on several factors such as hardware type, protocols",

data visualization, the required service, etc. So, the company

must investigate these options before considering to invest in a particular IoT platform [12].

• The stability of the platform: Throughout plenty of platforms on the market, it has some likelihood that some will fail. It is essential to choose a platform that has high availability, i.e., the probability of being operational with no outage to be around several years, furthermore, the investment might be wasted if the platform provider fails. To be sure the chosen platform is good or not, ask about the current and past customers.

• The scalability and flexibility of the platform: Make sure that the latform will work not only for few endpoints in the beginning but also with rowing needs during the time. Along with the scalability, the platform should be flexible to keep up with speedily varying protocols, technologies, or features. It is also crucial that the platform has to be network agnostic. This means that it can integrate and work with all vital tech systems out there, rather than be locked into one vendor.

• The pricing model and the business case: A platform provider should be consequent in pricing. Some providers offer a prefatory rate and after that rise it up significantly when the contract has been retained. Also, the assigned budget for a platform to be selected is mostly the most effective factor of the consideration. There are some platforms that provide introductory prices and after selecting and subscribing to it, many features which are not needed or which are required are found. Also, platforms that are a concern on saving time have more costly operations, but platforms that will save money might not contain the whole competitive features.

IoT platforms help:

· Connect hardware, such as sensors and devices

· Handle different hardware and software communication protocols

· Provide security and authentication for devices and users

· Collect, visualize, and analyze data the sensors and devices gather

· Integrate all of the above with other web services


IOT is worldwide technology. It can used across different application from civilian to defence sectors. These domains include water, construction, manufacturing, healthcare, space and mining. Now, IoT is one of the building blocks that is considered to be of use for developing smart homes and smart cities. So, at present not only in our country, but throughout the world there is a lot of interest on developing smart cities and smart homes. There are many platforms available by each company. Each having their unique features.We need to take some parameters to select individual platform. IOT having multiple technology including real-time, sensors, embedded system and machine learning. Good IOT platform should transfer data easily. There are heterogeneous platforms available. Hence to select best one it needs to check its compatibility with the service we need parameters.


[1] Hamdan Hejazi, Husam Rajab, Tibor Cinkler, Laszlo Lengyel

Department of Automation and Applied Informatics Budapest University of Technology and Economics Budapest, Hungary

“Survey of Platforms for Massive IoT” 978-1-5386-1208-8/18/2018 IEEE

[2] Akshat Savaliya, Aakash Bhatia, Jitendra Bhatia-Computer Department, LD college of Engineering, Ahmedabad, Gujarat, India “ Application of Data Mining Techniques in IoT: A Short Review” 2018 IJSRSET | Volume 4 | Issue 2 | Print ISSN: 2395-1990

[3] Jasmin Guth, Uwe Breitenbucher, Michael Falkenthal, Paul Fremantle, Oliver Kopp, Frank Leymann, Lukas Reinfurt “A Detailed Analysis of IoT Platform Architectures-Concepts, Similarities, and Differences” Institute of Architecture of Application Systems, University of Stuttgart, Germany 2018 Springer-Verlag

[4] Tetsuya Yokotani-Kanazawa Institute of Technology “Requirements

on the IoT Communication Platform and Its Standardization” 978-1-

5386-1359-7/2017 IEEE

[5] Jahoon Koo, Young-Gab Kim, Dept. of Computer and Information

Security Sejong University Seoul, South Korea “Interoperability of

Device Identification in Heterogeneous IoT Platforms” 978-1-5386-

4266-5/17/2017 IEEE

[6] Slavko Zitnik, Marko Jankovic, Klemen Petrovcic, Marko Bajec-

“Architecture of Standard-based, Interoperable and Extensible IoT

Platform”.University of Ljubljana, Faculty for Computer and

Information Science, SI-1000 Ljubljana 978-1-5090-4086-5/16/2016


[7] M. Sruthi & B. R. Kavitha Department of SCOPE, VIT University",


ISSN (Online): 2455 – 5630

How to cite this essay: