Why IoT needs an intelligent ecosystem
Although 5G networks are still in their infancy, IoT is well established and growing exponentially. Consider that in 2018, there were about 7 billion IoT devices, and that number almost quadrupled to 26 billion in 2019. This year, it’s estimated the number will climb to as many as 31 billion IoT devices, according to a Statista report.
While the continued deployment of 5G will certainly boost the number of IoT devices on the network, the reality now is that IoT is growing because it isn’t bandwidth constrained. 5G and edge computing will enable the creation of additional bandwidth for specific IoT use cases that will provide specialized network functionality to enhance IoT use and improve access.
Currently, the majority of IoT is OTT-based, meaning operators aren’t yet reaping the rewards of carrying IoT traffic across their networks. However, operators will want enterprises to leverage 5G services directly from their networks. This will require a complex, intelligent ecosystem. Why? Because operators will need to allow enterprise applications to gain access to their networks, and this will require a variety of APIs and simplified access to utilize 5G network slices.
Applications are at the heart of the ecosystem. And the goal is to avoid redesigning everyone. Instead, applications should leverage commercial middleware solutions from providers such as Red Hat, IBM or VMware to make smarter middleware components.
Most application developers won’t have the skills to take full advantage of the 5G-specific features and functionality, such as network slicing, or understand the capabilities of the edge computing platform to make the most of the computational performance. An advanced ecosystem will provide the necessary framework and expertise to link all this together.
Connected cars offer a roadmap
Take the example of a connected car. The advanced automotive application could use orchestration technology in a car and within the operator’s cloud as part of a multi-access edge-based solution. Edge-based 5G processing would provide the performance and computational power to enable advanced services such as collision avoidance and early warning of traffic conditions ahead. Unlike existing services, these advanced services would utilize different APIs.
None of this can happen without the application providing not just the user experience, but the operational feedback to quite literally drive the services.
Just as with connected cars, other IoT use cases will be propelled by how well 5G networks are designed. These 5G-connected networks will need a mobile core and advanced middleware for communication, synchronization and data virtualization to take full advantage of network slicing over a WAN. Multi-tier, API integration to 5G platform components will facilitate the deployment of these distributed applications and allow them to access the capabilities of the networks without changing their design and intelligence. There has never been a more pressing need for an ecosystem.
An ecosystem such as this could work as such: Platform-level microservices and middleware components would open the door to network software application providers (NSAPs) to interface with the network APIs and middleware. NSAPs that were traditionally network equipment providers (NEPs) or independent software vendors (ISVs) have already expanded their product suites to include new technologies.
At the application layer, sector-specific application providers would need to create industry solutions for new use cases. NSAPs would then supply the middleware to connect industry applications to 5G-enabled platform components. Additionally, microservices-based components would provide the software-based functionality to control and leverage 5G network APIs, including the configuration of network slicing.
What makes this approach effective? Within the context of the automotive industry, this all-inclusive ecosystem helps automotive application developers design the use case, the operational flow and the user experience without having to understand the technical details of the 5G network.
Consider two distinct automotive applications: a self-driving app and another for car insurance. These apps will soon rely on a common platform that uses reusable microservices and API middleware to provide data that’s required for these use cases. Advanced driver-assistance systems will use spatial recognition for the self-driving app, while both applications use telemetrics. These microservices can provide functionality for content processing, command-and-control systems, location intelligence, data analytics and other functions that underpin use cases such as in-car driver support systems.
Network vendors’ role in the new ecosystem
NSAPs bridge the 5G network and the platform services that power this value chain. As part of the ecosystem, NSAPs will provide the microservices that enable the applications for the edge and 5G. They will offer robust carrier-grade software solutions that not only herald the next wave of networking capabilities, but also create a monetizable set of products and services for operators and ISVs.
Given the complex nature of accessing microservices via the cloud, there is a need for a three-tier layering of APIs to make the connection simpler. The first layer consists of 5G networking and connectivity services. This layer has the intelligence to connect to the network services required and meet the service-level agreement criteria.
Next, the orchestration layer provides the intelligence to manage the use case at the edge and coordinate with the application on the enterprise cloud. This layer contains vertically independent microservices used by the application. In the automotive example, these microservices would tap into the data transferred between the car and the network edge, as well as variables such as driver preference settings.
The third layer interfaces directly with the application and is independent of any 5G-based services or platform microservices. It focuses on the application functionality as experienced by the end user. That can take the shape of information services, assisted driving or operational vehicle metrics collected by the manufacturer. Critical information from the platform microservices required to deliver the application functionality will be part of this layer.
Think of these three tiers as a pyramid. The base of the pyramid is composed of network components, the middle layer is semi-aware of a broad set of transportation-related requirements and the top layer is a set of specific applications, such as fleet tracking and self-driving. The higher up the pyramid, the more application or use case-specific the layer gets.
Time to hit the ground running
This next wave of 5G-focused applications can start providing value now and ahead of a full network deployment. System integrators can work directly with enterprises or industry verticals looking to deploy advanced IoT architectures or access new capabilities, intelligence and automation to modernize their operations. System integrators can also help enterprises create a clear business case, define requirements, and detail the solution and application architecture.
An intelligent ecosystem will accelerate time to market for this next wave of edge and 5G-aware applications, resulting in greater efficiency and a stronger commercial model for IoT solutions. NEPs can provide edge capabilities and 5G-enabling components for application providers. ISVs and middleware providers can offer network-aware integration patterns and components as an expansion of their own portfolios. Operators will then be able to leverage this ecosystem to provide differentiating applications to battle OTT deployment scenarios.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.