Salona Choudhury (16100049) Sanvida Tulam (16100050) Dept. of Computer Science & Engineering Dept. of Computer Science & Engineering IIIT Naya Raipur, Chattisgarh, India IIIT Naya Raipur, Chattisgarh, India [email protected] [email protected]
in Abstract – Internet of Everything (IoE)is the fastest growing technological trend in the today’s world. Broader thenthe trending concept of Internet of Things (IoT), IoE is not only limited toconnecting electronic devices and people to Internet, but also providing smart& knowledge based solutions and services. The future technology will becompletely dominated on IoE, it will be all around us and will bringrevolutionary changes in our lives. IoEcomprises of 3 basic components: millions of everyday objects which need to becustomised to connect to network, information-centred networks, and providingsmart and knowledge-based solution through real time analysis of collecteddata. IoE is one of the most promising future technologies, for it integratesthe daily life with technology, making everything smart and artificiallyintelligent. Index Terms – Internet ofThings, Artificially Intelligent, information-centric I. Introduction Internetof Everything is the advancement and integration of technologies, like Cloud,Internet of Things, Artificial Intelligence (AI), Deep Learning, MachineLearning etc. The systems won’t only provide services, but will self-learn togain knowledge, will analyse the situation and provide solution based on that.
But each component to reach IoE has its own difficulties and limitations, whichhave to be dealt with. Fig.1 Internet of Everything (IoE) Smart Objects The 1stpillar of IoE consists of millions of everyday objects which have to beconnected with network. From industrial machines, electrical appliances, towearable items, packaging technologies, medical equipments etc. all are beingintegrated to the net. At present, traditional silicon chips are used to connectdevices, but to truly become ubiquitous, electronics has to be shaped in such away that it can be integrated with all type of objects – miniature sized, soft,flexible and wearable. Recent developments inPrinted Electronics are paving the way towards low-cost and low-performanceelectronics.
It defines a technology for creating electronics on top of somesubstrate using organic and inorganic inks. It gives the possibility of preparing stacks of micro-structuredlayers and thus, helps in manufacturing thin-film devices. Many Packaging companies are using Printed Electronics for developingsmart labels, for Anti-CounterfeitingPackaging techniques. Especially, pharmaceuticals, healthcare products, beautyproducts, and food and beverage companies are increasingly using thistechnology, which improves security as well as is disposable. The concept of System in Packages (SiP) has led to miniaturized and morepowerful systems, where various components are stacked on top of another, tocreate thin but highly efficient electronics. Information-Centered Networks Atpresent, Internet is built on Internet Protocol (IP) based concept, wherepoint-to-point communication takes place on the basis of IP nomenclature ofsource and destination.
But with billions of smart objects connected toInternet, there will be a burst of data flowing through net in the comingyears, which the present technology will not be able to handle. Also,information and facts based search on a search engine is entirely differentthing from searching for sensor data from smart objects. Forthis, Information-centric Network is being developed which makes data contentdirectly addressable and routable, also known as Content-centric network (CCN)or Named-data networking. In the CCN security model, instead of usingadditional layers for security features, individual units of data are madesecure using encryption. When user requests data by name, CCN transmits namedcontent to the user from the nearest cache, therefore less number of hops istraversed, redundant requests gets eliminated, and ultimately less resourcesare consumed. Automated Real Time Insights After connecting differentdevices to net and making the network suitable for IoE, the real use is whenthese devices generate real time insights and take action accordingly.Developments in the fields of Machine Learning, Deep Learning, ArtificialIntelligence, Big Data, Cloud etc.
have made possible efficient and secureprocessing of vast amounts of data to provide low-power and smart services. Butthe objective is not only to make these devices smart, but also capable ofself-learning, so that they can use their previous experiences to generate moreaccurate and better solutions and services. Fig.2 Self Learning Mechanism Fig.
2 shows the proposedarchitecture for self-learning and knowledge enhancement. It contains threemodules: 1) The data collection module collects real time data, and processesit using the previous knowledge and training. Then the data is forwarded toMachine Learning Module (MLM).
2) Using machine learningmodel from Knowledge Processing Module (KPM), learning process is initiated. Ifit satisfies the standards set by the designer, the learning results andparameters are transferred to KPM for generating knowledge. If the learningresults aren’t up to the standards, re-learning process is initiated underdifferent learning conditions.
3) After updating knowledge,this new knowledge is used to process the future inputs. II. CONCLUSION Thus, the IoE concept stands on threepillars: Smart devices, Information based Networks, and self-learning devicesfor smart and real time services. But toimplement it possesses a number of challenges, both in hardware and softwareparts. For IoE applications, flexible, miniaturized, very low-power, low-costand extremely fast chips have to be designed. Many difficulties are arising:like downscaling CMOS technology for miniaturization has led to compromisedamplifier performance.
Designing techniques for chips with self-learningalgorithms, without compromising with its speed, size, and at low costs, haveto be explored. Acknowledgment First ofall, we express our gratitude towards the Almighty, by whose grace we were ableto complete the task given on time. We are indebted to DR. RAMESH VADDI Sir,for his guidance and supervision in completing this paper. References 1 A Study on the Virtuous Circle Self-Learning Methods for KnowledgeEnhancement – Jaehak Yu, Young-Min Kim, SoonHyun Kwon, Kwihoon Kim, Nae-SooKim, Sun-Jin Kim, * Cheol-Sig Pyo 2 S. V.
Vandebroek, “Three pillars enabling the Internet ofEverything: Smart everyday objects, information-centric networks, and automatedreal-time insights,” 2016 IEEE International Solid-State CircuitsConference (ISSCC), San Francisco, CA, 2016, pp. 14-20.3 Named Data Networking Matthew N.O.
Sadiku1 ,Adebowale E. Shadare2 , Sarhan M. Musa34 J. Gubbi, R. Buyya, S. Marusic, and M. Palaniswami, “Internet of Things(IoT): A vision, architectural elements, and future directions,” FutureGeneration Computer Systems, Vol. 29, No.
7, pp. 1645-1660, September. 20135 L, Atzori, A, Iera, and G, Morabito, “The Internet of Things: A survey,”Computer Networks, Vol. 54, No. 15, pp. 2787-2805, October. 2010.6 Internet of Everything, Available online: http://ioeassessment.
cisco.com/7 M. Mosko, I. Solis, “CCNx Semantics”, IRTFdraft-mosko-icnrg-ccnxsemantics01, ICNRG June 2015.8 L. Zhang et al, “Named Data Networking”, ACM SIGCOMM ComputerCommunications Review, Volume 44, Number 3, July 2014, 66-73.
9 Mark Weiser, “The Computer for the 21st Century” – Scientific American,September 1991, pages 94-104.