As the number of Internet of Things (IoT) devices increase, on a network connectivity level, scaling relies on a communication framework that will allow devices created by different brands to communicate with each other. The current state of the Internet part of the Internet of Things is controlled by which company makes your smart device. If a user has an Alexa smart assistant made by Amazon, it won’t communicate with your HomePod by Apple or Google Home by Google (Bohouta & Kepuska, 2018). Either way, customers are tied into a company’s ecosystem of products that work only with products made by that brand or that connects to that brand’s products via application programming interface (API).

The Internet of Things strives to have connected devices communicate with each other autonomously, without any human interaction (Kamilaris & Pitsillides, 2013). IoT should be considered in a social context, a paradigm focused on inter-device relationships that allows for seamless interactions between devices (Atzori, Iera, & Morabito, 2011). IoT devices need to be interoperable, meaning proper identification requirements for devices are necessary to facilitate communication and cooperation.

Research Problem

There are various roadblocks to scaling the Internet of Things including sharing data between devices and interoperability between IoT protocols. The current communication framework is disjointed and makes it difficult for consumers to purchase multiple devices that are work together. Ultimately, this will hinder the growth and scale of IoT, however current technologies can be leveraged to solve these problems and novel solutions aim to improve on current techniques. This communication issues expands beyond various branded devices, but also includes other intelligent devices such as security systems, mobile phones, smart grids, and environmental sensors. Different protocols were used to create different IoT systems and building a universal ecosystem for IoT is a challenge. Intelligent devices can have a social component, S-IoT, that defines device relationships and eases communication, allowing for seamless human-to-device and device-to-device interactions.

This research problem is worthy of discussion since scaling IoT will drastically impact the lives of many. Users who interact with any device, sensing network, or actuator at home or in the office will see the impact of a good communication framework for the IoT. In scenarios where humans are not in the loop, consequences of interoperable devices can be found out after mistakes have been made in vital fields like healthcare and criminal justice. This paper will review research topics covering current communication frameworks that can potentially scale to accommodate billions of sensors

Purpose Statement

This paper will cover several novel and existing technologies that allow intelligent devices to communicate. This study will provide a better understanding of current wireless network technologies and overview past research pertinent to large-scale IoT. This paper will review research topics covering various types of technologies and how they impact future research on interoperability and data dissemination frameworks.


This review will highlight many popular technologies that can be leveraged to scale IoT. This study overviews current research regarding device communication in IoT and the architectures currently used and their limitations. Some researchers believe an IoT framework for device communication will be similar to a utility like heat or electricity currently (Atzori et al, 2011).

This paper covers multiple applications of current device-to-device (D2D) communication frameworks and how their contribution to the body of knowledge that will anchor research towards new frameworks. This paper can be potentially beneficial for those interested in research about frameworks that can enable device communication in large-scale IoT deployments.

Literary Review

The present state of the Internet of Things is one that is a sparse INTRAnet in that many devices in use today have no way to communicate with each other directly, described by Zorzi, Ghulak, Lange, and Bassi (2010). Outside of their connection to the internet, wireless products made by Google, Apple, and Amazon are not interoperable with each other. The lack of interoperability is one of the biggest drivers of this study; providing companies a communication framework that would allow brands to have open, interoperable products without having to work directly with their competitors. Similar to how web Application Programming Interfaces or APIs allow companies to securely access data on different websites, leading makers of smart devices and sensors can utilize one or more of the many existing technologies to enable a truly interconnected Internet of Things.

Some researchers believe net neutrality, the concept that no Internet service provider (ISP) can hinder sharing online content by prioritizing information, is the basis of scalability and interoperability in the IoT (Ma, Wang, & Chiu, 2017).  The Internet is the foundation of IoT and the United States is divided regionally by areas Internet Service Providers (ISPs) cover. Limited ISP options are a major concern in the growth of IoT. The largest ISPs in the United States include Verizon, AT&T, Comcast, Time Warner Cable, and Charter Communications; allowing 96% of Americans at most two options for ISPs. In 2014, Verizon Communications, one of America’s largest ISPs sued the Federal Communications Commission (FCC) in an effort to repeal net neutrality. As of a December 2017 vote in favor of ISPs, the FCC is moving forward with plans to repeal net neutrality. The debate between ISPs and content providers (CP) such as Netflix, Hulu, and YouTube is whether there should be a price prioritization for transporting data. While on the edge of ethics issues around ISPs hiking rates for CPs, the repeal of net neutrality opens the door for paid prioritization of Internet services (Ma et al., 2017).  Monopolistic ISPs have a direct monetary incentive to expand network capacity to accommodate more demand. A reliable, high-capacity internet framework can prove beneficial to large-scale IoT, however the consequences and trade-offs of paid prioritization should be considered, with particular focus on the impact to social welfare.

There are many challenges to scaling IoT systems, one hurdle being large-scale deployment of sensor networks. While use of mobile phones grew as people purchased personal devices, sensing platforms are more likely to be owned and operated by a utility company than an individual. With this in mind, further investigation on the topic using the SPHERE (Sensor Platform for Healthcare in a Residential Environment) platform gave insights for what makes for successful deployment of IoT sensors and how they communicate. The sensors at work in SPHERE included one wrist-worn wearable sensor, one environmental sensor and one SPHERE gateway (Fafoutis et al, 2017). Insights found from the study showed deployment is a matter of time, there are tradeoffs between versatility and efficiency, and user acceptance is key.

Defining a social structure between IoT devices helps to address the issue of communication between sensors and personal devices. In SIoT: Giving Social Structure to the Internet of Things the authors introduce a novel paradigm that explores the social relationships among objects including sensors and devices irrespective of brand or operating system (Atzori et al, 2011). The first relationship outlined in the study is the parental object relationship, in which objects made by the same manufacturer are structured by production batch. This relationship is inherent when a new object is produced and will not change except for device obsolescence. Co-location relationships and co-work object relationships are created when either a location-based or situation-based application profile is created for the device during initialization. Objects that have a co-location relationship constantly reside in the same place, such as in a home or in a factory providing industrial automation services. Co-work relationships are defined when objects cooperate to provide a IoT utility such as emergency response systems or connected street lights. The relationships between the public and private devices are important for interoperability. The authors also described a relationship that helps define objects owned by the same user. The ownership-object relationship associates devices like mobile phones, game consoles, and wearables to a single user while allowing secure data dissemination to public sensors and devices (Atzori et al, 2011).

Radio frequency identification (RFID) is currently the most popular technology for item identification and tracking (Solic, Blazevic, Skiljo, Patrono, Colella, & Rodrigues, 2017). RFID tags allow users to quickly identify sensors and devices in multiple use cases. For instance, repairmen can easily identify malfunctioning sensors for service and consumers can access services like libraries, restrooms, and public transportation regardless of communication protocols. Some researchers believe that RFID will be the key to scaling IoT because of its low cost, reading range, and emerging popularity. RFID solutions that use the ultra-high frequency (UHF) band, called Gen2 RFID, offer the capability to work worldwide, and has the best price-performance ratio in comparison to barcode-based identification systems. Gen2 RFID has been built into many brands of smartphone devices as well as ID badges, ski passes, and animal microchips. Gen 2 offers a highly reachable reading range of about 10m and low cost of $0.1 US per tag. While RFID tags are beneficial for device identification, there are limitations in sensing proximity between tags compared to other high proximity sensors with longer battery lifetimes.

Several existing technologies can be leveraged to allow massive scale communication in the Internet of Things. Jiang, Gao, Duan and Huang (2011) detail how Mobile Crowdsensing (MCS) costs can be reduced to address the increasing demand for data dissemination in IoT. In traditional MCS, sensor data collected from a large amount of user’s mobile devices is sent to a centralized server, analyzed, then sent to a data requester (Jiang, Gao, Duan & Huang, p1, 2018). MCS applications currently rely on the server-client architecture and are unsuitable for a large number of users and requests. In a peer-to-peer MCS system, sensor data can be processed on a user’s mobile device and distributed via a mobile app and shared directly with users. Scenarios for using MCS in IoT include measuring pollution levels, monitoring traffic congestion, parking availability, and public works outages. (Ganti, Ye, & Lei, 2011). Server management and usage cost are serious considerations for accommodating millions of connected devices (Jiang et al, 2018). The P2P-based MCS architecture presented by the authors focuses on incentive design and economic analysis. With a focus on providing a quality-aware data sharing market, rewarding users for sharing their data, and charging data requesters for the desired sensor data, this work was integral to understanding a transparent framework for scaling sensor data with buy-in from users.

Ganti, Ye, and Lei (2011) demonstrate how sensor data is capable of measuring and mapping phenomena of common interest. Mobile devices are already deployed and using mobile sensing data can help efficiently build large-scale applications (Ganti et al., 2011). The authors explain the challenges of scaling current MCS applications. A major setback to scalable MCS applications is the existing application silo approach, in which application development is difficult because each individual app is built from scratch. Ganti et al. (2011) compares the different sensors equipped on the iPhone 4 including GPS, accelerometer, gyroscope, compass, proximity sensor, ambient light, in addition to camera and audio. This study also covered considerable limitations of mobile devices. The sensor data collected by individual devices is highly dynamic in availability and capacity due to various factors including device battery, non-sensor device usage, and device environmental conditions. MCS is one way to collect and share data from a large deployment of sensors, mobile phones.

An IoT gateway is a device that connect sensors and intelligent devices to the cloud. Since the only thing devices use to connect to the gateway is an Internet connection, gateways have emerged as a key element in bringing legacy devices to IoT. Legacy devices are technologies that are out of production or obsolete. These devices are still used by many consumers as media centers and smart device controls using gateways (Kang, Kim, & Choo 2017). To enable these obsolete technologies to communicate with devices using updated protocols, gateways open wireless communication between legacy devices and IoT devices. This allows IoT to scale without requiring a complete overhaul of devices with upgraded software and hardware. While gateways are important now, current gateways require users to install the device manually. The authors propose a self-configurable gateway that automatically detects and registers new devices. This uses principles of Social IoT to communicate with various smart objects using 4G, LTE, Wi-Fi, Bluetooth, and ZigBee. Kang et al., (2017) states that “IoT is being termed as the future of internet communication”. With this in mind, IoT gateways provide a framework for multiple generations of devices to communicate, including outdated Mp3 players, gaming consoles, and tablets that are no longer receiving software updates.

Lippi, Mamei, Marianim and Zambonelli (2017) discusses how connected devices will communicate directly. Devices with hearing and speaking capabilities may have a co-work object relationship and can interact via natural language. In the example of traffic management, autonomous vehicles would be able to communicate directly with traffic lights, cameras, and other cars using human understandable speech. The social relationship categories proposed by Atzori et al. (2011) for the SIoT vision are relevant here. Below is an example inquiry and negotiation interaction between an autonomous vehicle and a traffic light from Lippi et al., 2017.

Autonomous Car: Hi Traffic Light, how long will the light remain green?

Traffic Light: Hello A, it would last 30 seconds.

Autonomous Car: Could you keep the green on 30 seconds more? I’m a bit late to work.

This study offered insight to how machines may bargain and even argue with each other to solve issues in real-time using natural language. Devices using human-understandable speech is an area where research is needed to more clearly understand the limitations of natural speech communication.  

The paradigm of IoT intends to connect billions to trillions of sensors communicating with each other (Shit, Sharma, Puthal, & Zamoya, 2018). The location data of wireless sensor network (WSN) is critical, as a single localization technique is insufficient in every scenario. Global positioning system (GPS), while the most popular location technology is insufficient since it cannot always localize sensors in close proximity. This article provided a solid taxonomy of localization techniques separated by capability in order to review the current localization technologies. Smart devices such as phones, wearables and tablets have Wi-Fi, Bluetooth, and GPS to sense location, but achieving accuracy and precision is a challenge. Shit et al., (2018) describes how smart devices that have RFID chips must overcome a lack of power efficiency in localization techniques (Solic et al., 2017). Smart monitors like temperature and humidity sensors face challenges in strong signal coverage and target positioning. Moving smart devices, like autonomous vehicles, use speedometers, optical measuring meters, and cameras as sensors. These devices must overcome the complexities of Simultaneous Localization and Mapping(SLAM). These challenges impact many of the billions and trillions of “connected” devices. Location data is one of many data types that must be communicated efficiently for IoT to scale. Understanding the limitations of current localization techniques as described by Shit et al, (2018) is an important start for future research.

Employing both social networking and device-to-device (D2D) communication is vital to potentially provide IoT services more efficiently. Ma et al. (2017) discusses in detail how small coded fragments or complete items can be shared amongst devices using cooperative D2D communications. While wirelessly transferring the large amount of data from sensors can cause congested wireless backhaul, transferring cached content in a distributed manner can potentially reduce spectrum overcrowding. This wireless communication framework is an alternative to traditional internet network communications. A major difference between D2D and device communication using the Internet is that D2D communication doesn’t require information to be processed by the internet to flow between devices. This reduces at least one access point for cyber-attacks. Caching data into smaller, encrypted packets is one way of sending information between sensors, personal devices, and the cloud at a low cost.

A complex issue in IoT, analytics plays a key role in large-scale applications. Sharing sensor data faces challenges with heterogeneous data streams and the massive volume of data. Analytics for large-scale IoT are vital to these devices usefulness. In order for IoT to scale, devices must provide value to the user or another stakeholder. The authors detail a novel, two-layer approach for analyzing IoT data real-time as well using Bayesian networks, a probability-based model for representing uncertainty. A point this paper made was that any proposed solution should be generic enough to handle all types of data formats, and the solution should be able to store, manage, and communicate complex analytics on such data. (Akbar, Kousiouris, Pervaiz, Sancho, Ta-Shma, Carrez & Moessner, 2018) To perform analytics on data from a sensor or intelligent device, it must be transferred from the device to the cloud as the vast majority of sensors do not have the capability to perform data analysis. There are many proposed methodologies to perform real-time analytics, however solutions must be considered in the current context wherein most IoT devices are unable to analyze, then transmit sensing data efficiently.

Interoperability is a key research issue in IoT due to the heterogeneous nature of data produced by devices. Various communication capabilities, sensory data collected, and information processing must be managed in a system that provides interoperability between all components. This work was integral to defining SIoT and how social networks allow human-to-device interactions. Ortiz, Hussein, Park, Han, Crespi, (2014) state “In order to practically integrate the ubiquitous computing in our future daily life with high Quality of Experience (QoE), we need to improve the connectivity of all the relationships between users and things.” S-IoT goes beyond data shared between co-location or co-work devices to incorporate physical elements and their context in reality (Ortiz et al., 2014). Space sensors that track the movement and object location will be vital in including non-intelligent devices like bookshelves and coat racks in IoT. The social paradigm is a key to offering consumers interoperability and a high QoE.

Another method of allowing data dissemination in IoT is a neighbor-based broadcast protocol for mobile devices. This proposed protocol by Liu, Nakauchi, and Shoji (2018) uses localization provided by the neighbor based knowledge of mobile nodes to allow devices to communicate with others in close proximity. Using broadcasting techniques addresses issues of resource-constrained nodes and the dynamic nature of mobile nodes like autonomous vehicles. A naïve type of broadcast protocol called simple flooding (SF) allows a wireless node to rebroadcast a packet of data and deliver it to all nodes in the network. An example scenario of this type of device communication architecture is alerting vehicles of children rushing into the street. Assuming the child has at least one sensor device they carry such as a sensor-embedded backpack, sharing this real-time data with nearby drivers can help avoid an accident. This would involve the use of many kinds of devices such as the child’s sensor, tracking location relative to a street or intersection. Sensors in nearby traffic lights can communicate the potential danger to nearby vehicles. This technology will be important for passive device communication, or device-to-device interactions where humans are not in the loop. In order for IoT to scale devices should interact autonomically and without intrusion to a consumer’s life. The proposed broadcast protocol will improve upon past resource-consuming approaches.

In conclusion, there are several existing and novel wireless communications frameworks that can enable the IoT to scale with seamless interactions for users and devices. Further research must compare the overall benefits and how multiple frameworks can be interconnected to strengthen the growth of IoT. IoT will impact the lives of everyone who interacts with smart devices as simple as a mobile phone. This research hopes to fill the gap in the body of literature surrounding using current wireless communication technologies to grow and improve IoT at massive scale.





Akbar, A., Kousiouris, G., Pervaiz, H., Sancho, J., Ta-Shma, P., Carrez, F., Moessner, K. (2018)

Real-time Probabilistic Data Fusion for Large-scale IoT Applications. IEEE Access PP(99). Doi: 10.1109/ACCESS.2018.2804623

Atzori, L., Iera, A., Morabito, G. (2011) SIoT: Giving social structure to the Internet of Things. IEEE Communications Letters, 15(11) pp. 1193-1195. Doi: 10.1109/LCOMM.2011.090911.111340

Bohouta, G., Kepuska, V. (2018) Next-generation of virtual personal assistants (Microsoft Cortana, Apple Siri, Amazon Alexa and Google Home).  Computing and Communication Workshop and Conference. doi: 10.1109/CCWC.2018.8301638

Crespi, N., Han, S., Hussein, D., Ortiz, A., Park, S. (2014) The Cluster Between Internet of

Things and Social Networks: Review and Research Challenges. IEEE Internet of Things Journal. (1)(3) pp. 206-215. Doi: 10.1109/JIOT.2014.2318835

Fafoutis, X., Ests, A., Piechocki, R., Craddock, I., Experiences and lessons learned from making IoT sensing platforms for large-scale deployments. IEEE Access. (6) 3140-3148. doi: 10.1109/ACCESS.2017.2787418

Ganti, R., Ye, F., Lei, H. (2011) Mobile crowdsensing: current state and future challenges. IEEE

Communications Magazine (49)(11) doi: 10.1109/MCOM.2011.6069707

Kamilaris, A., & Pitsillides, A. (2016), Mobile phone computing and the internet of things: A survey. IST- Africa Conference and Exhibition, IEEE, 2014.

Kang, B., Kim, D., Choo, H. (2017) Internet of Everything: A Large-Scale Autonomic

IoT Gateway. IEEE Transactions on Multi-Scale Computing Systems. (3)(3) doi: 10.1109/TMSCS.2017.2705683

Lippi, M., Mamei, M., Mariani, S., Zambonelli, F. (2017) An Argumentation-based Perspective

over the Social IoT. IEEE Internet of Things Journal. PP(99) doi: 10.1109/JIOT.2017.2775047

Liu, K Nakauchi, K, Shoji, Y. (2018) A neighbor-based probabilistic broadcast protocol for data

dissemination in mobile IoT networks. IEEE Access. PP(99) doi: 10.1109/ACCESS.2018.2808356

Ma, T.B.R., Wang, J., Chiu., M. (2017) Paid prioritzaiton and its impact on Net Neutrality. IEEE Journal on Selected Areas. 35(2) pp. 367-379. Doi:  10.1109/JSAC.2017.2659020

Yu, J., Bang, H., Lee, H., Lee, Y. (2016). Adaptive Internet of things and web of things

convergence platform for Internet of reality services. Journal of Supercomputing, 72(1), pp.84-102. doi:10.1007/s11227-015-1489-6

Shit, R., Sharma, S., Puthal, D., Zomaya., A. (2018) Location of things (LoT): A review and

taxonomy of sensors localization in IoT infrastructure. IEEE Communications Surveys & Tutorials. PP(99) doi: 10.1109/COMST.2018.2798591

Solic, P., Blazevic. Z., Skijo. M., Patrono, L., Colella, R., Rodrigues. J., (2017) Gen2 RFID as IoT enabler: Characterization and performance improvement. IEEE Wireless Communications. (24)(3) pp.33-39. Doi: 10.1109/MWC.2017.1600431

Zorzi, M., Gluhak, A., Lange, S., & Bassi, A. (2010). From today’s INTRAnet of things to a

future INTERnet of things: A wireless and mobility related view. IEEE Wireless Communication, pp. 44-51. doi:10.1109/MWC.2010.5675777