Expectations and Considerations for the Identity of Things

Searching for new revenue streams within the IoT sphere in order to beat the competition has become a crucial objective for both established and new market entrants. Thing identity and management services is one of the emerging battlefields and entails three primary components: a) device management, b) communication security, and c) data exchange. A clear definition is of paramount importance for this topic since research interviews with IoT vendors, industry organizations, and even certification authorities have revealed that there is a serious issue with clarifying exactly what “identity” in IoT actually involves.

The recent ABI Research webinar Identity of Things: Rethinking IoT Platform Services as Connections Exceed 50 Billion dealt exactly with that demanding challenge of quantifying and closely examining the issues revolving around the Identity of Things (IDoT). This Executive Foresight aims to give some top-level and easy-to-digest guidance for IoT players and to help them find their place in the value chain by outlining some of the most important recommendations that they need to keep in mind when dealing with the IDoT.

• Connecting *Everything* Is Not the Future of IoT Anymore: As IoT solutions begin to take more solid form compared to the fledgling approaches of the past decade, IoT vendors need to look towards finding true value in the form of meaningful implementations. Zealously connecting every device and user to each other is not the objective. A more realistic portrayal and goal of the current ecosystem includes the connection of hardware elements (such as devices, gateways, sensors, and actuators) to software elements (such as applications, APIs, systems, and platforms) and users, merging these three elements together to produce meaningful insights and outcomes.
• True Value Will Be Found in the Fine Lines That Separate One Market from the Other: Industry demands push these aforementioned expectations even further for the near future to facilitate high-value data exchange between these elements, fueling machine learning algorithms to IoT platforms, and, perhaps most importantly, enable cross-vertical communication between different IoT channels. We are traversing an era where technology that goes far beyond M2M and device-to-user interactions enables spanning across verticals and also in-between verticals. Due to de-siloing initiatives and the rise of middleware, IoT vendors will start to see market barriers break down, and they must be ready to tap into the nooks and crannies of their chosen verticals. True value will be found in the fine lines that separate one market from the other, with the primary markets revealing valuable cross-application insights found in smart cities, automotive, and smart home markets.
• Smart Cities, Automotive, and Smart Home Leading in Implementations: Highly volatile but also greatly prone to innovation, these three markets are forecast to be the very first to evolve under this transformative IDoT paradigm. Just in the last 2 years alone, the automotive market (which is generally considered the most stable of the three) has made immense strides towards other markets and state-of-the-art technologies. These include smart home and smart cities, wearables, biometrics and sophisticated authentication processes, computer vision, machine learning and artificial intelligence, virtual assistants, cognitive navigation, and many others.

Based on current infrastructure investments, the road towards systemic maturity of the IDoT will not be an easy one, and there are multiple considerations to take into account. Three of the most important ones, however, concern IP migration.

• Support for Legacy Systems: Legacy support for systems and devices in the IoT environment is one of the first implementation hurdles faced by companies when they are considering the IDoT route since their form factors belong to a pre-IoT era. Many potential implementers seem to forget that from a security standpoint, it is a battle that should not be waged in the first place. Legacy devices a) lack secure elements and have to be managed via software-based security, b) lag behind in interoperability due to the diversity of communication protocols originating out of many verticals and particularly from the industrial domain, and c) some of them are not even designed to be connected via IP and thus are a major threat for IoT security and, subsequently, thing ID management. The good news is that more and more IoT vendors are starting to support an easy transition from legacy systems to modern ID management infrastructure (e.g., featuring greater system and device transparency for IT, security orchestration, and automation), which will make future implementations much easier to achieve.

• Application Programming Interfaces (APIs): The added challenge of device and systemic heterogeneity places additional, albeit still technically manageable, pressure on API development. Implementers are advised to go that extra mile with added rounds of testing when developing APIs destined for the greater IoT ecosystem. This is because APIs are key enablers of software services and are some of the unsung heroes that allow the transition from the silicon to the cloud. When discussing IoT applications, there are two major architecture categories: REST APIs and MQTT APIs. Both are key enablers of software services, and generally tend to be vendor-agnostic. They allow for further communication and security while also remaining a vital part of identity and management services. Note that most API-rich platform products should be characterized by device and network agnosticity, allowing support for a wider array of protocols.

• Prepare for the IP Migration and Intelligent Gateways: The IP migration from the current dominant IPv4 to the elaborate IPv6 is already in the pipeline. In short, this means that one device will have one unique IP address instead of multiple devices sharing just one, a transition that will further fuel this novel transition into the IDoT. An important note at this point is that not all versions of communication protocols (from established ones like Zigbee to emerging ones like EnOcean) allow for IP connectivity. This means that certain devices and sensors are irrevocably tethered to a gateway that will allow them an IP connection, thus affecting the network architecture under which they were originally designed to be. This may be seen as a limitation (and in some ways, it is), but also an opportunity since it also gives greater emphasis towards the need for building better, more secure, IoT-focused gateways featuring enhanced memory, security, computing power, and protocol support. This is a trend that, as expected, coincides with the advent of edge computing.

These findings are part of the ABI Research report Thing Identity and Management Services: Enabler and Opportunity Analysis (AN-2416).