Interoperability in IoT

Introduction

Interoperability is defined as the ability of different technology systems, system components or software applications to establish communication between them, exchange data, and interpret properly the received information for its use (ETSI, 2013). This property applies to interactions within a system (i.e. the internal communication of its different components), but also to the interaction between two or more systems.

In IoT interoperability plays an essential role; there is probably no other technology area in which interoperability becomes as critical and relevant as in the case of IoT (World Economic Forum, 2015). Interoperability is the key that allows any set of devices to exchange information and work together in concert, acting as an actual IoT system. For example, without interoperability lights would not respond to remote switches, sensors could not be read by smartphones, and devices in general would be unable to connect to accessible networks. Moreover, according to a study by the McKinsey Global Institute in 2015 (Mckinsey Global Institute, 2015), without interoperability, at least 40% of the potential benefits of IoT cannot be achieved. This is evident considering that a transparent integration and interconnection of different IoT systems and system components would critically simplify their implementation, maximize performance and facilitate their operation with other systems. This interconnection of systems propitiates to share relevant data and to establish significant synergies, improving the quality of the information, the quality of service and the experience provided to the user. Moreover, interoperability enriches Big Data analytics using 3v model (Tanque, M., & Foxwell, 2014) through the integration of a variety of data formats, models and definitions, in a common data model to increase its effectiveness. Indeed, one of the main challenges in Big Data is to handle this data diversity properly (Chen et al., 2014).

To highlight the aforementioned advantages, let us consider an application of a bus company that calculates its optimal route. This application could benefit from interoperability with other transportation services. For instance, it could consider links with trains using the real time information that they provide. The application could also benefit from the interoperation with the traffic monitoring service of the city, capable of indicating the less congested routes. Thus, the service provided by the bus application would be more precise, complete and useful for the user.

Let us also consider some examples of IoT systems in the e-Health domain. In this area, interoperability among sensors and medical devices permit the remote monitoring of different bodily vital signals such as heart rate, blood pressure or breath rate using wearable sensors. Through this monitoring, it is possible for an IoT health system to detect any abnormality of vital signals remotely, at any moment, and automatically alert health services and caregivers. In both examples, it is necessary the interconnection and integration of systems in order to collect, analyse and use big amounts of data from heterogeneous sources.

To achieve a high degree of interoperability in an IoT system is therefore desirable, but regrettably it is still one of the most difficult and important challenges to solve in IoT. As a matter of fact, currently the different IoT systems are typically unable to communicate with each other or to interoperate in general (Diallo, Herencia-zapana, Padilla, & Tolk, 2011). The main cause of this lack of interoperability is the highly heterogeneous nature of IoT systems. The Internet of things covers a wide range of devices, protocols, technologies, networks, middleware, applications, systems and data that present a vast diversity. The heterogeneity of the underlying technologies can prevent the interoperation of smart objects and systems, as they follow different rules and standards. This diversity also affects the process of extracting value from the IoT Big Data due to its inherent heterogeneity, variety of data formats, and rapid growth (Chen et al., 2014). In this sense, the existence of a global reference standard for IoT would notably facilitate interoperability by giving rules and certain homogeneity to this heterogeneous universe. Though, currently there is no de facto reference standard, posing a significant problem when designing new IoT systems (Ganzha, Paprzycki, Pawlowski, Szmeja, & Wasielewska, 2016).