WHITE PAPER

IT’S TIME TO FOCUS ON QUALITY IN THE IOT ECOSYSTEM

WHAT YOU WILLLEARN

nn

WHAT DOES QUALITY MEAN IN IOT?

nn

THE SPECIFICS OF MODERN IOT SERVICES

nn

WHY REAL-TIME ACTIONABLE IOT ANALYTICS?

nn

THE VARIOUS EXPECTATIONS OF IOT CUSTOMERS

nn

MODERN IOT SOLUTIONS – A WIN-WIN FOR IOT PROVIDERS AND CUSTOMERS

Mobile customers have become very demanding. They constantly search for a better price, a more reliable connection, or higher quality. Machines, in a way, are similar. Maybe the price factor is not as important to them as it is to people, but high quality is a must. Without appropriate quality of service (QoS), data transmission can’t be established and the whole device system can collapse. For example, a battery can get lost or rebooting can take too long, resulting in loss of connection and leading to a serious device or process malfunction. In the energy sector, devices are spread out over large geographical areas making service trips very costly. Metering devices also are highly sensitive to service disruption resulting from loss of connectivity. Quality also differs depending on the standard and type of connectivity used by customers. In many verticals, mobile connectivity still outmatches new wireless standards (LoRa, SIGFOX, etc.) because it delivers the highest quality for IoT services.

WHAT DOES QUALITY MEAN IN IOT? For some industry verticals these potential issues can be marginalized, but not for all. The biggest problem is that there is no universal definition of quality. Neither is there a single point of its measurement. From the customer perspective quality may be a subjective issue, making it impossible to establish one understanding of its meaning. On top of that, quality of service in the Internet of Things (IoT) has not yet been sufficiently standardized, so it is still unclear which KPIs should be measured and when. Although some work has been done in the area of IoT standardization1, quality of service has not been welladdressed. One of the reasons for this is the multitude of connectivity standards in IoT, which means that QoS would have to be measured differently for each. A classic definition by ITU2 says that QoS is “a collective effect of service performances which determine the degree of satisfaction of a user of the service”. Toni Janevski, the author of “Internet Technologies for Fixed and Mobile Networks (Mobile Communications)”, says that quality of service “refers to measurable parameters and techniques to select, control, measure, and guarantee the required quality for a given service”. QoS parameters (otherwise called KPIs or key performance indicators) are the key factors for evaluating whether technologies, services, devices and applications meet the expectations related to their quality, availability, efficiency and reliability. Such KPIs can be measured from two main perspectives: the network perspective and the application/device/service perspective, always within a specific vertical. The second perspective is much more comprehensive. The IoT is very diverse, with dozens of industry verticals, each of them with different customer expectations and requirements, and different numbers of devices in use, which makes it impossible to apply a “one size fits all” approach. For example, the healthcare sector will have very different needs to those of the automotive industry. The first uses a device by device approach, while the latter installs them in mass numbers. Differences can also be seen in these verticals’ requirements related to service efficiency and availability, as well as in the potential influence of any malfunction on their operations.

The European Telecommunications Standards Institute (ETSI) and the 3rd Generation Partnership Project (3GPP) have made efforts to standardize IOT.

1

ITU Recommendation ITU-T E.800 (09/2008), http://www.itu.int/ITU-T/recommendations/rec.aspx?rec=E.800

2

WHITE PAPER

3

IoT service quality means something different for each industry, but in general it comes down to a reliable, constant connection, real-time operation of the device, and efficient online monitoring and fault detection. Business customers operating in the IoT with specific devices often do not even know that measuring service quality is possible, that so many KPIs can be monitored, or that any device can be fixed before serious problems occur (for example, lack of information about increasing temperature could cause a device to burn out, or a device that monitors traffic signals might take too long to boot up resulting in an accident, etc.).

IoT service quality means something different for each industry, but in general it comes down to a reliable, constant connection, real-time operation of the device, and efficient online monitoring and fault detection.

What is most important for an enterprise that sells industrial pumps? Aside from profits, other success factors include good brand perception among their customers, low failure ratio, low frequency of necessary technician visits, etc. Customers who buy industrial pumps will rely on the instructions and manuals to know how to use the equipment safely, but they are usually not aware that they could also benefit from online maintenance, predictive alerts, or remote parameter checking. All those elements boost service quality, and better quality means higher profit. That is why an IT solution for telecom operators and enterprises offering IoT services to various vertical markets should be able to answer the specific needs of each vertical.

THE SPECIFICS OF MODERN IOT SERVICES A service in IoT is composed not only of connectivity (such an approach is long gone), so IoT service providers need to build service bundles composed of applications, devices, support, consultancy and connectivity. A modern, end to end IoT proposition must take into account some way of measuring real experience and performance at the point of connectivity, as well as the device and application usage, which together form the foundation of the IoT service. The KPIs collected from a given device SIM or an embedded SIM card can then be correlated with network measurements, ensuring the highest quality of monitoring. This results in the IT system performing the right actions, with the aim of improving and ensuring the quality of the IoT service and the related SLA, and of providing an actionable analytics capability to IoT customers.

A service in IoT is composed not only of connectivity (such an approach is long gone), so IoT service providers need to build service bundles composed of applications, devices, support, consultancy and connectivity.

WHY REAL-TIME ACTIONABLE IOT ANALYTICS? Actionable, real-time analytics provide IoT suppliers with thorough knowledge about their customers, the services they use, their devices, processes, and much more. Such functionality can show real performance as statistics associated with correct or erroneous functioning of the equipment, mobility analytics (including the density and trajectories of SIM card movements), consumption analysis of different service types presented in a variety of settings (including temporal and spatial), and KPIs in relation to SLAs. An innovative and unique way to collect KPIs related to devices and applications is to deploy device application software (an “agent”) and SIM / embedded SIM (eSIM) applets, acting as collection applications and allowing

4

WHITE PAPER

service providers to retrieve the relevant KPIs that complement network-based measurements (such as probes, service control modules or other data sources). In some verticals, applets are the best way to collect all the necessary KPIs, when there is no means of accessing the devices. The applet can be preinstalled, or installed via an over the air (OTA) interface when needed, and can collect data on-request (issued by the application). It can also send data automatically, even using retry mechanisms in cases of periodical loss of connectivity . Such applets are already vertical specific, with separate functionality for use cases such as automotive, elevators, alarms, and energy meters. Such software can turn any IoT device into an intelligent, self-manageable and self-monitored IoT service. A modern IoT platform should enable this kind of monitoring and analysis, to allow service providers to monitor the cellular connectivity of static and mobile objects within different verticals in real time from the device perspective. Such an IT system should also provide instant access to information about the network status, and perform analyses in order to highlight problems immediately, by ensuring a rich set of data is available and accessible at any time. The way in which anomalies can be monitored is very interesting. An anomaly is defined as a data pattern that differs from the rest of the data, or from what is expected. Examples might include unexpected device mobility changes, authentication problems, excessive use of network signaling and absence of use of paid services, or unexpectedly high device activity in a specific location. Telecom operators have limited knowledge about the purpose of each particular device connected to the network. While they know a lot about each vertical, they do not know exactly how each device behaves, what its configuration is, or what happens if it is faulty. Therefore, a prerequisite for detecting anomalies in M2M data is segmentation of devices into groups with similar properties or parameters. Using information about past device behavior, it’s easy to tell whether its current behavior is typical. The devices are segmented into groups with similar features, and then their activity is subjected to further analysis. This method, called machine learning, is used for developing complex models and algorithms that serve for predictive analytics purposes. The advantage of anomaly detection based on machine learning is that it allows the identification of previously unknown issues that may be responsible for quality problems and security threats. This enables service providers to eliminate negative anomalies and prevent them from happening in the future. As the process is based on streaming data, potential problems can be detected and solved pro-actively and fast. By analyzing data collected through diagnostic applets installed on SIM cards, a service provider can perform much deeper analysis of potential problems, which can uncover issues related to signal strength, battery function, and environment (these are only a few examples, problems can be a result of a combination of factors).

WHITE PAPER

5

THE VARIOUS EXPECTATIONS OF IOT CUSTOMERS Customers in the healthcare sector expect their IoT solutions to provide information about anomalies, especially in situations when device behavior differs from expectations, when the time to service after a device reboot is too long, or even when the battery level is low. We can imagine a patient holding a remote Holter device, which is sending ECG (electrocardiogram) analysis data. Such analysis can be interrupted if the Holter monitor takes too long to reboot, or if reboots happen too often. In such cases, a healthcare application platform that receives and analyzes medical data from remote devices has to be notified about the detected anomaly and appropriate actions have to be taken as soon as possible. By comparison, the needs of the automotive industry are very different. For these customers, the most useful data would be related to the location of a given SIM card (including geolocation beyond cell ID), its mobility, information on battery level, or the software version installed on a given device being changed without permission. The automotive vertical is different, especially because of massive device deployment. When more devices are deployed, instead of sending technical teams to check the parameters on each separate device regularly, the service provider would rather control such devices remotely, including reconfiguration of parameters, KPI checking and performing corrective actions, all from a single application. In smart metering, useful information delivered by an IoT solution could contain details of the meter itself, the number of reboots occurring per day, the time to service after each reboot, the placement of the meter and the KPIs related to connectivity (for example, zero sessions, sessions that were too long, or even erroneous behavior, such as when a hundred machines are active but one of them is taking too long to connect). Like the automotive vertical, the smart metering sector also uses mass deployment and requires bulk remote control to minimize the cost of field service operations.

6

WHITE PAPER

MODERN IOT SOLUTIONS – A WIN-WIN FOR IOT PROVIDERS AND CUSTOMERS In all verticals, an IoT real-time actionable analytics solution enables mobile network operators to enhance their network service quality and reduce the risk of connectivity issues for their IoT customers. Such an IT tool enables the correlation of data collected from connectivity, IoT devices and applications, and includes analytics capabilities, which brings unique value to the service providers and to their IoT customers. The operators can generate additional revenues by selling premium services with guaranteed highest quality, offer reliable connectivity to prevent customer churn, and provide best in class customer service. On the other hand, such an IoT solution empowers IoT customers by giving them access to information about real device experience and performance, as measured at the point of connectivity, with predictive actions and maintenance and the opportunity to limit the cost of service level agreements significantly. In summary, measuring and monitoring quality of service in the IoT is still a green field and many mobile operators sell various quality levels as options in their tariff plans. Such SLA proposals usually offer access to 24-hour consulting services, but do not typically include real-time analytics and monitoring. Solutions such as Comarch M2M Platform with Quality of Service in IoT enable communication service providers to get one step ahead of the competition and to meet the expectations of today’s IoT customers. On top of connectivity management, the whole IoT ecosystem also requires device management and application enablement. These improve the competitive position of the operators, letting them sell a complete IoT service. Customers will not buy a service composed only of a SIM card, they need an end to end solution that also includes a monitoring device, alarm tracking, applications for visualization, support services in 24h mode, insurance, etc. Only such a comprehensive service that also includes QoS measurement and analysis forms a complete IoT product. For more about the Comarch approach to the IoT, e-mail [email protected]

ABOUT COMARCH Comarch is a provider of complete IT solutions for telecoms. Since 1993 the company has helped CSPs on 4 continents optimize costs, increase business efficiency and transform BSS/OSS operations. Comarch solutions combine rich out-of-thebox functionalities with high configurability and are complemented with a wide range of services. The company’s flexible approach to projects and a variety of deployment models help telecoms make networks smarter, improve customer experience and quickly launch digital services, such as cloud and M2M. This strategy has earned Comarch the trust and loyalty of its clients, including the world’s leading CSPs: Vodafone, T-Mobile, Telefónica, E-Plus, KPN and MTS. Copyright © Comarch 2016. All Rights Reserved

[email protected]

|

telecoms.comarch.com