1

COMPARATIVE ANALYSIS OF IoT ARCHITECTURES

Authors: Ravi Teja Guthikonda Sai Srikar Chitta Shraddha Tekawade Tripti Attavar Advisor: Dr. Lijun Chen TLEN 5710 Capstone April 25, 2014

2 Abstract: The Internet had been considered as a form of human-to-human communication. However, due to recent developments in technology, intelligence is embedded in devices. These “smart” devices have the ability to interact with humans and other smart devices, as well, which led to the development of “Internet of Things” (IoT). This project explores the impacts of deployment of IoT on network architecture: how will IoT change the Internet architecture and what would be the right architecture for IoT. It focuses on a comparative analysis of the Internet architecture and a few application specific IoT architectures that are currently proposed, based on a set of design goals and design principles. Based on the results of a comparative study, we propose a new IoT architecture with elements borrowed from the existing architectures and proposals, formulate a timeline for its deployment, and briefly discuss its policy implications.

3

Table of Contents: I.

Introduction

i.

Statement of Problem

4

ii.

Research Question

4

II.

Literature Review

5

III.

Research Methodology

8

IV.

Research Results

8

V.

Discussion of Results

10

VI.

Conclusion and Future Research

22

VII.

References

24

4

I.

Introduction: i.

Statement of the problem: Internet of Things (IoT) will consist of billions of devices in the near future. There have

been predictions that, by the year 2020, about 8 billion devices will have connectivity to the Internet. The deployment of this technology will need some changes to the current network infrastructure, protocols, and services. Creation of compound applications that are part of different industries can be facilitated using IoT. These applications will be part of everyday life and the impact to our society is significant. The expectations for revenue opportunities are around $2.7 trillion. Companies such as Cisco, Microsoft, Oracle are already investing heavily in IoT. Many issues such as no standardized architecture, privacy and safety concerns, and interoperability between vendors plague IoT. Our research aims to bridge the gap in several architectures and propose a multi-industry and multi-vendor architecture that will be ubiquitous. An ideal architecture makes the usage of IoT more desirable and easier to deploy [16].

ii. •

Research Questions:

What are the defining features of IoT and how is it different from the current Internet? The purpose is to identify the significant differences between the current Internet and IoT, in

particular, what the defining features of IoT are in terms of physical layer technologies and applications, as well as the associated performance requirements in communications. •

What are the design goals for the IoT architecture and the design principles that allow the IoT to achieve these goals?

5 Architecture is about the protocol stack that impedance matches the physical layer technologies and the applications. Guided by this, the purpose is to identify a set of design goals for the IoT architecture and the design principles that allow the system to achieve these goals. •

What are the different architectures of IoT currently being proposed/used, focusing on a few use cases? The purpose is to carry out a comparative analysis of the existing proposals for IoT architecture and evaluate them according to the design goals and principles identified in the above.

•

What will be a good architecture, based on the existing proposals/Internet, as well as our own recommendation? Proposing or recommending architecture that can be standardized for universal use. A timeline for adopting/deploying the proposed architecture and the associated cost, as well as its impact on policy.

II.

Literature Review Internet of things is a term coined by Kevin Ashton in one of his presentations [13]. The term

describes a technology of the future based on the Internet and involves sharing of information [13]. IoT is revolution in the world of technology and it is the next big thing in the world of computing and communication. IoT allows the communication between all the things we see around us apart from the human-machine interaction that already exists. Applications of IoT range from various fields from the obvious IT to the surprising, saving energy using smart grids [11]. IoT can be considered as an extension of traditional wireless sensor networks (WSN) that makes the object-to-object communication possible by use of technology called radio frequency identification (RFID). RFID enables the object to identify other objects. RFID has long been used as a replacement to

6 barcode. The objects use this technology to identify other objects and so that they can connect to them. This technology also detects objects in real time and provides important information such as location and status [1]. IoT is enabled by a robust RFID system. IoT also uses sensors to link the physical and information worlds) [4]. The sensors are used to collect data about the surroundings this data can be analyzed according to different circumstances and factors to bridge the gap [11]. IoT also makes the use of nanotechnology and miniaturization to place intelligence in various devices. These devices with intelligence are called smart devices and have an important place in the architecture of IoT. These devices can configure themselves and take a decision on their own. After which real object-to-object communication will be accomplished [11]. The architecture of Internet was developed in late 70’s to use that architecture for IoT, which has billions of devices sharing data with each other, will not be practical. The amount of data produced by IoT cannot be handled by the current Internet architecture The new architecture of IoT should address several issues such as reliability, quality of service (QoS), security, and interoperability. The new architecture also needs to be universal i.e. it must be adopted by everyone so that it can be used for any application. The architecture must also be flexible so that it can be modified to change according to the future needs [11]. The fact that Internet of Things is the inevitable future of the Internet has led to extensive research in trying to solve or minimize the challenges that could hinder its deployment. A number of new network architectures have been proposed to manage the communication between the IP-speaking devices [8]. Khan et.al. [6] proposed the generic architecture, existing development trends, and possible applications of IoT. As the number of devices gets connected to the Internet, proposed IoT architecture should address many concerns such as scalability, reliability, and QoS [6]. Miao Wu et.al. [13]has explained the current 3-layer architecture of IoTand why is not enough, they also proposed a new 5-

7 layer architecture to perceive its essence. Miao Wu et.al.[13]explained the need for new IoT architecture, which includes features of both Internet and Communications Network. Increased traffic volume with deployment of IoT poses a serious threat to the security of IoT. Multimedia applications are one among the many applications, which lead to increased traffic volume. Liang Zhou, and Han-Chieh[9] proposed a Framework for Security architecture. Liang Zhou, and HanChieh[9] also explained the study of different multimedia traffic in the interest of IoT. [16].Rolf H. Weber [17] discussed the security challenges for architecture of IoT. Some of the concerns for security in IoT include resilience to attacks, data authentication, and access control. IoT generates a massive amount of data from various sensors. This data will be used to analyze and provide the needed solutions. The amount of data generated by IoT provides an opportunity for data storage and cloud computing companies like EMC, Microsoft, etc. Microsoft is investing heavily in IoT for enterprises. Microsoft has created a cross-platform cloud based IoT managing service to enterprise customers..This service also includes analytics part of the data. The cloud and data management services of IoT are few of the glaring issues that can be addressed with an appropriate architecture [14]. The architecture is like the backbone of IoT if it is not robust and flexible, deploying IoT will take more time than required. Thus, our research is prominent as it makes IoT easier to deploy. Although a number of architectures are present, it is important that any architecture be accepted universally addressing the several issues faced by IoT. Standardization has many benefits such as easy deployment, manageability, troubleshooting, etc. Our research includes solutions to various issues such as interoperability, performance and security issues. Security has also been one of the concerns of IoT. By improving security within the architecture, IoT would be installed smoothly without any skepticism.

8 III.

Research Methodology:

During this project, we planned to carry out detailed study of the available literature and current research in order to propose a new architecture for IoT. We began by studying the differences between the traditional Internet and the IoT. We identified the distinguishing features of IoT. After identifying these features we defined the design requirements of IoT and identified a set of principles that will aid in fulfilling these design requirements. After identifying the design principles we studied various use cases for IoT applications and determined the four most popular ones to use in our comparative analysis. We then studied the most appropriate architecture used in these cases based on the set of design principles, which we identified. A detailed study of these use cases helped us in identifying the advantages and the shortcomings of the various architectures. After conducting a detailed comparative study of these architectures and the current research in the industry, we have tried to propose a modular architecture for IoT that will facilitate interoperability between various vendors and standards. The suggestions in this architecture were confined to the set of design principles, which we identified. IV.

Research Results:

What’s special about IoT and how is it different from the current Internet? The Internet now is no longer used as a communication between people using computers instead it will be used in IoT to enable the communication between various devices [2]. The difference in the number of devices connected now and when IoT is in use will be colossal. 8 billion devices are expected to be connected by 2020 and the current Internet may not be able to handle it. IoT will be using the existing infrastructure as part of its connectivity. Scalability is a significant difference between the two but the current Internet will be able to handle them by a few protocol changes like using IPv6 instead of IPv4. The other difference is the devices used in IoT will have intelligence embedded in them and may

9 not use the same protocols used now. The access technologies used are in IoT will be different from the access technologies used in the current Internet. Short-range wireless technologies that consume low power are popular with IoT. The applications in IoT will be more interactive, user friendly and require less intervention from the user [2]. What are the design goals and the design principles that enable to achieve these goals? The following are the design goals that we have identified for an ideal IoT architecture. 1.

Manageability

Manageability describes the existence of intelligence in the architecture. The common types of manageability include centralized and distributed based control. 2.

Security and Privacy

Security and privacy deal with the ability of how immune the architecture would be to outside attacks. It deals with various issues such as authentication, encryption, etc. 3.

Mobility

Mobility should be considered in the architectures when the end nodes move from one place to another. 4.

Cost-effectiveness

Cost-effectiveness determines the affordability of the architecture. 5.

Efficiency

Efficiency is described in terms of power management of the different devices connected to the architecture. 6.

Quality of Service (QoS)

QoS is a performance management technique for the prioritization of different data traffic from devices.

10 The following are the design principles that we have identified for an ideal IoT architecture. A layering approach in which one-layer implements the service offered by the layer below and provides services to the layer above is identified. The existence of intelligence in the network is identified. This can be achieved either by making the core or the end-devices intelligent. What are the different architectures of Internet of things currently being proposed/used, focusing on a few use cases? We have identified the popular use cases from Healthcare, Automotive, Home automation, and Electrical Power industries. The use cases are remote healthcare monitoring, connected cars, smart home and smart grids. We have also identified the architecture that best fits the each use case according to the design goals. We determined that remote healthcare has a best fit with community health service architecture based IoT. IBM’s smart grid architecture is best suited for the smart grids use case. “Stratecast” architecture proposed by IBM is befitting the needs of smart home application. The ITS connected vehicle's architecture is suitable for connected cars. V.

Discussion of Results:

Remote Healthcare: One of the motives behind using IoT in healthcare industry is to provide remote health care. By using IoT, doctors can monitor the health of patients remotely and in real-time to provide timely assistance whenever required [18]. We found “A Community Health Service Architecture Based on the Internet of Things on Health-Care ” by Wu-Zhao, Liu Lei-Hong, Huang Yue-shan and Wu Xiao-ming to be a strong according our design goals. The community health service technical architecture consists of three layers

11 as shown. Information perception layer, network transmission layer, and application layer constitute the architecture. Perception layer includes different types of sensors such as wearable sensors to detect the physical condition of the various organs and special sensors to detect the position of the user [18].

Figure 1. A Community Health Service Architecture Based on the Internet of Things on Health-Care [18]. Information collected from these sensors is sent to the wireless receiver and from there is transferred to the computer. From the PC data is transferred to remote health care monitoring center via the Internet. Application service layer combines the technology of IoT and medical health care center

12 resulting in remote health service. Remote monitoring health care service improves cost savings and prevents overcrowding of hospital beds [18]. The general architecture for community health service as shown. Basic blocks of the architecture include family client and community health care center. The physiological conditions of the patients are collected by the wireless health sensors and then transmitted to the health care center. The health care center doctors can analyze, test the collected data and provide medical service through text/email/video/voice [18].

Figure 2. Health Care architecture [18]. The family client includes medical sensors, Zigbee wireless network, and Home Computer. Wireless sensors monitor physiological, environment parameters of the patients continuously and then transmit the information to the base stations/monitoring stations through Zigbee wireless network. Data is transferred from the base station to the computer and data is transferred to the health care center via Internet. Community Health Care center consists of doctors, pc machines database servers, and medical works. Database servers consist of personal information, monitoring data and doctor’s information [18]. The patients can use the database servers to query the doctor information and choose the right doctor for diagnosis. The patients can also input their daily life information, health reports, and disease records into the database making it easy for the doctors to provide diagnosis. Doctors can also query for

13 patient’s records to check their health status. Apart from all these the users can consult doctor’s real time through voice, video, and text [18]. The problem with this architecture is that there is no support for privacy and security of user data, QoS support for time sensitive or critical, data and mobility of the user. All these parameters are important for the timely and accurate response in case of emergencies. Smart Grids: The Smart Grid is termed as the ‘the central nerve system’ of the power system. Smart Grids can improve energy efficiency, reduce the environmental impact, improve the safety and reliability of electricity supply, and thus, reduce the electricity transmission of grid [8]. Traditional users such as residential users, commercial and industrial users along with electric vehicle charging systems, are using the smart grids for their applications. Smart grids with Internet of Things (IoT) act as an upgrade to the existing electrical grids. The IoT architecture must include the four major modules of transmission, distribution, smart substations and the efficient usage of electricity for the smart grids [10]. General smart grid architecture should have characteristics such as centralized management, very high privacy and security, high reliability, low latency, high scalability. The physical layer technologies used in smart grids include Wi-Fi, Zigbee, RFID, etc. [10]. The basic and the most common network model for Smart Grid application is a three-layer network architecture proposed by the authors Ling Zheng, Shuangbao Chen, Shuyue Xiang, YanxiangHu from the North China Electric Power University. The architecture is divided into the perceived extension layer, network layer and the application layer. This approach of architecture uses a centralized control of the smart grids through the use of Information and Communication Technology(ICT) platform [10].

14

Figure 3. Three layer architecture for Smart Grids [10]. The perception extension layer is similar to the physical layer of the OSI model in network architecture. This layer provides all the meteorological environment information, the device status information, and information related to the transmission, substation, distribution and usage of electricity. It is composed of sensors working on Wi-Fi, Zigbee; smart terminals, RFID tags and other input and output entities that are deployed in the actual monitoring environment. These sensing elements collect the device status information, line status information, weather environmental information and distribution of electricity-integrated information of the processes of electricity transmission, substation and distribution [10]. The network layer consists of wireless sensor networks, gateway nodes, access networks and core networks. The wireless sensor networks use the sensing elements in the perception extension layer to collect information to deliver it to the gateway node. The gateway node transmits the collected data to the access network, which is finally transmitted into the power communication core network. Access

15 networks include power fiber-optic access networks and broadband wireless access networks. The core side of the network contains Ethernet, ADSL, 3G, xPON, etc. [10]. The application layer is built to satisfy the business needs of smart grid. This layer builds a variety of power application platforms and serves as a management and control mechanism to the large amounts of data received from the sensing networks [10]. This architecture also considers the existing electrical power communication network into consideration for interoperability. Smart grids with this architecture have better interaction and information processing capabilities. Security can be incorporated in it by using some of the existing authentication and encryption techniques. Major companies such as Intel are heavily investing and working with Austin-based energy consortium Pecan Street Inc. to test the future of energy with smart grids and IoT [10]. The main shortcomings of this architecture are lack of security mechanisms and support for QoS. Connected/Smart Home: Smart homes consist of a number of day-to-day devices with embedded intelligence that can communicate with other similar smart devices. There are a number of companies that are proposing their own architectures for connecting these smart devices. The main problem with these current architectures is interoperability between devices manufactured by different vendors. These architectures are more or less similar in many aspects [15]. “Stratecast” is one such architecture proposed by IBM. What makes this architecture unique is that it addresses the problem of integration and delivery of different services along with standardizing the communication between various devices. This architecture uses the principle of integrating different services in the cloud as against the traditional approach of integrating them at the edge device on the user's network such as a wireless gateway. IBM refers this functionality as “home in the cloud” [15].

16 This architecture uses a common service platform to integrate various services accessed by the users and to facilitate seamless communication between smart devices. Rather than having all the devices try to communicate with each other directly, these devices can communicate with a single point/ platform that will take care of the communication between various devices and services. This platform will accommodate all types of devices irrespective of the vendor. Instead of having the devices connect to the cloud directly, a better approach is to have these devices communicate to a local device such as a local router/gateway that will act a mediator between the end devices and the cloud. This gateway can be managed and controlled from the cloud. It will just act a point of integration between the end devices and the cloud [15]. The following diagram shows the common service integration platform:

Figure 4. IBM “Stratecast” architecture for Smart Home [15]. Using this architecture, the service providers deliver various services to the communication service providers (CSP) who will then integrate these different services and deliver them to the end users

17 according to the user requirements. This architecture design will help in simplifying the effort on the user side to a great extent [15]. IBM is working with CSPs such as Verizon and device vendors such as Philips to try and implement this architecture. This architecture focuses on the needs of both machine to machine (M2M) services and Over the Top (OTT) services such as online video streaming. Since the user can access all these types of services without any additional efforts on the user side, it provides a perfect platform for seamless integration of different technologies and standards. This is what the end user ultimately demands [15]. This architecture takes care of most of the design goals except for the support for QoS, user data privacy, and security requirements. Connected Cars:

Figure 5. ITS architecture for connected cars [5].

18 Connected car is an application of Internet of things that are gaining vast popularity rapidly. As the name suggests every car is connected to another and to the Internet through various wireless technologies such as wireless local area networks (WLAN), Long term evolution (LTE), etc. There are various sensors inside the car that provide data about the functionality of the car, communicate with other cars and a centralized server, etc. The data can be used for road safety, improving traffic efficiency, other trivial issues such as navigating, finding open parking spots, video content delivery, etc [5]. There are various architectures proposed for this popular application. Companies like Cisco, Oracle, Broadcom, and Volkswagen have proposed various architectures but architecture called intelligent transport system (ITS) is more feasible universally as it is not vendor specific, it is more functional as it supports more applications. The figure of the architecture is shown. It is based on the OSI protocol stack. The access layer is concerned with both the physical and data link layers. The shortrange communication can use 802.11 technologies more specifically wireless access in a vehicular environment and dedicated short-range communications. For long-range communication existing infrastructure such as 3G, LTE and WiMAX can be used. The next layer combines both the networking and transport layer of OSI model. IPv6 is used in the network layer for scalability, higher network efficiency, and in built security features. The transport layer consists of either TCP or UDP [5]. Each networking protocol can have its own dedicated existing transport protocol such as TCP, UDP or ITSC transport protocols. IPv6 over geonetworking can be used for more efficiency. It uses the IPv6’s multicast according to the geographical region that makes the architecture more efficient as it does not use the regular IP metrics such as hop, cost, etc. The facilities layer consists of application and presentation support. The facilities layer also supports of the maintenance of various applications. Service messages and requesting for repetition messages are also part of the facilities. The management layer is a cross functional that supports all other layers. It has various uses as regulatory management,

19 application management, and station management. The security layer is also a cross functional layer that is used for authentication, authorization, firewall, intrusion management, etc [5]. The ITSC architecture consists of a station that every vehicle can connect and exchange data. It also consists of handheld subsystem, vehicle subsystem, a central hub and roadside subsystem. This architecture is developed in collaboration between ITS Europe, ITS America and ITS Japan making the ITS architecture universal without any interoperability issues [5]. The concerns about security and network management are taken care by cross-functional security and management layers. However, this application has strict constraints on the timely delivery of data irrespective of dynamic network conditions. This is one design goal that needs to be better addressed. Proposed Architecture: After studying the various IoT use cases and their architectures it is safe to say that IoT and the traditional Internet architectures are similar in many aspects. So we do not need to propose an entirely new architecture, but only need to suggest some improvements in the existing layered architecture to address the needs of IoT applications.

User Interface Performance Management Transport Networking Enhanced Link Layer

Device Management

•  Provudes an API or understandable interface to the user •  Converts the user defined network parameters and provides them to transport layer •  TCP, UDP, QoS •  Interpretation of different data, data forwarding, security •  Data integrity, efficient point to point connection, interference management, channel managemnt •  Connection of devices, collection of data from sensors, and authentication.

20 Figure 6. Enhanced Architecture for IoT applications. The first module is the device management module. It performs various functions such as connection of devices, collection of information from sensors, and the authentication and authorization of different devices. We strongly suggest that all the devices support IPv6 with geonetworking for more efficient addressing and security. The devices should also support short-range wireless technology such as Zigbee, Wi-Fi and long-range communication such as Long Term Evolution (LTE). The second module is the enhanced link layer. This layer carries out all the functions of the traditional data link layer. This layer needs to address the issues related to reliability of point-to-points links with greater detail for IoT applications. It also needs to implement a stronger solution for interference avoidance and management. Efficient channel access mechanisms need to be implemented that will prioritize devices in order of the critical data that they transmit. The third module is the Network module. It performs the functions related to internetwork communication. This module implements a translation between data formats of different vendors and protocols. It also implements security practices such as data encryption and data filtering. It also supports adaptive routing in accordance with the available network resources. This module executes adaptive routing using software defined networking (SDN) [7]. The fourth module is the transport module. Similar to the traditional transport layer this layer supports TCP or UDP for different applications. It also has support for Integrated Services and Differentiated services. However, for many IoT applications Differentiated Services may be implemented for certain applications or certain devices within an application whereas some applications or devices may require Integrated Services. There are certain applications that may require both the applications. Thus a new hybrid model that will support both integrated as well as differentiated services

21 for the same application is required. This module implements appropriate QoS policies to fulfill the desired network performance restrictions such as latency, jitter, throughput, etc. The fifth module is the performance management module. It acts as an interpreter between the user-interface module and the transport module. The primary function of this module is to convert the user defined network performance parameters to appropriate QoS policies. It relays these policies to the transport layer for implementation. This module also monitors the network performance to determine the best QoS policies for a particular application. If the network conditions are not favorable to support the minimum performance requirements this layer informs and negotiates with the user. Implementation of this module will also require capabilities similar to SDN [7]. The sixth and final module is the user-interface module. It acts a mediator between the user application and lower modules. It provides performance parameters to the lower module and presents the data to the user in an understandable format. The distinguishing features of this architecture are flexibility using modular approach and also efficient network performance management using SDN. Policy Considerations for IoT: Large-scale deployment of IoT will require certain regulatory changes. First, policies will be required to regulate the spectrum sharing between existing IoT devices. Second, the limits need to be defined to protect the devices from interference with each other. Third, if reallocation of the existing spectrum or new spectrum range allocation is needed to accommodate the large number of devices in IoT. Sub 1Ghz channels are being expanded in the new IEEE 802.15.4 standard to be utilized by key technologies of IoT such as ZigBee, 6LoWPAN, etc. Wireless controlled devices and home automation applications use the 2.4Ghz band that come under 802.15.4 standard. Devices from Wi-Fi routers to

22 microwave ovens also use the 2.4Ghz spectrum overcrowding the spectrum completely. The 802.15.4 standard evolved in 2003 comes with 16 channels in the 2.4Ghz band termed as ISM (Industrial, Scientific, and Medical) band. The number of Sub 1Ghz channels was then increased to 3 in Europe and 30 in North America. The latest version of the 802.15.4 standard provides support for countries such as China and Japan in the new Sub 1Ghz bands. These include the ranges of 779-787MHz for China and 915-930MHz for Japan that provide reliable and higher throughputs inside buildings and homes [12]. VI.

Conclusions and future research: In conclusion, it is safe for that IoT is the inevitable future of the Internet. IoT connects billions

of devices and enables machine-to-machine communication. It can transform daily life with everyday objects connected to each other and the Internet. We have identified the design principles for an ideal architecture from the shortcomings of other architectures. Various issues such as interoperability, performance and security issues currently plague IoT. Our research tries to provide solutions to all these issues by a flexible semi-flat architecture. Comparative analysis of the proposed architectures is the basis of our architecture. Our architecture is based around the design principles that were identified earlier. We also address certain policy issues related to IoT and suggest certain steps for cost analysis of IoT deployment. Despite providing substantial potential, the business, technology and policy challenges need to be addressed. The business challenges need to be addressed by including the costs of implementing IoT. It should also include the cost analysis of the smart devices, technologies implemented etc. The policy foundations should be created so that the spectrum can be shared without any harmful interference and acquire further spectrum for IoT. A collaborative effort from different industries such as Internet service providers, device vendors, software providers, and cloud providers is required for efficient implementation of IoT. The future research should focus on standardized protocol stack and networking

23 technologies used for a seamless flow of data between different devices. Streamlined data analytics for the massive amounts of data generated by IoT must also be an area of emphasis.

24 VII.

References:

[1] Castellani, A.P.; Bui, N.; Casari, P.; Rossi, M.; Shelby, Z.; Zorzi, M., "Architecture and protocols for the Internet of Things: A case study," Pervasive Computing and Communications Workshops (PERCOM Workshops), 2010 8th IEEE International Conference, pp.678-683, March 29 2010-April 2 2010 doi: 10.1109/PERCOMW.2010.5470520 [2] Daniel Castro, Jordan Misra, "the internet of things,"Internet: http://www2.datainnovation.org/2013internet-of-things.pdf, November 2013 [Apr. 25, 2014] [3] Dieter Uckelmann, Mark Harrison, Florian Michahelles (Ed.), "Integrated Billing Solutions in the Internet of Things" in Architecting the Internet of Things, Heidelberg, Germany: Springer-Verlag, May 2011, pp.231-236 [4] HuanshengNing; Ziou Wang, "Future Internet of Things Architecture: Like Mankind Neural System or Social Organization Framework?," Communications Letters, IEEE, vol.15, no.4, pp.461-463, April 2011 doi: 10.1109/LCOMM.2011.022411.110120 [5] "Intelligent Transport Systems (ITS); Communications Architecture." Retrieved March 2014, Internet: http://www.etsi.org/deliver/etsi_en/302600_302699/302665/01.01.01_60/en_302665v010101p.pdf

[6] Khan, R.; Khan, S.U.; Zaheer, R.; Khan, S., "Future Internet: The Internet of Things Architecture, Possible Applications and Key Challenges," Frontiers of Information Technology (FIT), 2012 10th International Conference, pp.257-260, 17-19 Dec. 2012

25 doi: 10.1109/FIT.2012.53

[7] KwangtaeJeong; Jinwook Kim; Young-Tak Kim, "QoS-aware Network Operating System for software defined networking with Generalized OpenFlows," Network Operations and Management Symposium (NOMS), 2012 IEEE, pp.1167-1174, 16-20 April 2012 doi: 10.1109/NOMS.2012.6212044

[8] Li Li; Hu Xiaoguang; Chen Ke; He Ketai, "The applications of WiFi-based Wireless Sensor Network in Internet of Things and Smart Grid," Industrial Electronics and Applications (ICIEA), 2011 6th IEEE Conference, pp.789-793, 21-23 June 2011 doi: 10.1109/ICIEA.2011.5975693

[9] Liang Zhou; Han-Chieh Chao, "Multimedia traffic security architecture for the internet of things," Network, IEEE, vol.25, no.3, pp.35-40, May-June 2011 doi: 10.1109/MNET.2011.5772059 [10] Ling Zheng; Shuangbao Chen; Shuyue Xiang; Yanxiang Hu, "Research of Architecture and Application of Internet of Things for Smart Grid," Computer Science & Service System (CSSS), 2012 International Conference, pp.938-941, 11-13 Aug. 2012 doi: 10.1109/CSSS.2012.238

26 [11]Lu Tan; Neng Wang, "Future internet: The Internet of Things," Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference, vol.5, pp.V5-376,V5-380, 20-22 Aug. 2010 doi: 10.1109/ICACTE.2010.5579543 [12] Magnus Pedersen, "Utilizing the sub-1GHz spectrum for the Internet of Things," Internet: http://www.embedded.com/design/real-world-applications/4415110/1/Utilizing-the-sub-1GHzspectrum-for-the-Internet-of-Things, May 25, 2013 [Apr. 25, 2014]

[13] Miao Wu; Ting-Jie Lu; Fei-Yang Ling; Jing Sun; Hui-Ying Du, "Research on the architecture of Internet of Things," Advanced Computer Theory and Engineering (ICACTE), 2010 3rd International Conference, vol.5, pp.V5-484,V5-487, 20-22 Aug. 2010 doi: 10.1109/ICACTE.2010.5579493

[14] Michael Endler, "Microsoft Azure Intelligent Systems: 4 Facts," Internet: http://www.informationweek.com/big-data/software-platforms/microsoft-azure-intelligent-systems-4facts/d/d-id/1204521, [Apr. 25, 2014]

[15] Mike Jude, "IBM: Working Towards a Smarter Connected Home". Retrieved March 2014 Internet: http://docs.caba.org/documents/IBM-Smart-Cloud-Home-SPIE2012.pdf [16] "What Exactly Is The "Internet of Things"?" Internet: http://postscapes.com/what-exactly-is-theinternet-of-things-infographic, [Apr. 25, 2014]

27 [17] Rolf H. Weber, "Internet of Things – New security and privacy challenges," Computer Law & Security Review, vol. 26, no. 1, pp.23-30, January 2010, http://dx.doi.org/10.1016/j.clsr.2009.11.008. [18] Wu-Zhao, Liu Lei-hong , Huang Yue-shan and Wu Xiao-ming, “A Community Health Service Architecture Based on the Internet of Things on Health-Care,” World Congress on Medical Physics and Biomedical Engineering, vol 39, pp.1317-1320