Why Production Grade DVB-I Clients Matter

As the DVB-I standard moves from specification to real deployment, manufacturers are starting to face a practical question: how do you integrate DVB-I into production firmware without destabilising your platform?

On paper, the problem looks straightforward. DVB-I defines how devices discover services, obtain service lists and access programme metadata over IP networks. But modern TV platforms are already complex systems. Devices typically run layered stacks based on embedded Linux, Android TV or RDK, each with its own architecture, dependencies and memory constraints.

Introducing a new service discovery layer into that environment is not trivial.

Manufacturers must consider memory usage, integration effort and long-term maintainability. SDRAM remains a high cost in consumer hardware, and software stacks are already under pressure. Adding poorly integrated components can quickly create instability or unpredictable behaviour.

For some early implementations, DVB-I has been handled primarily at the application or browser layer. While that approach can work for demonstrations or prototypes, it often becomes problematic in production devices. Service discovery, channel lists and programme metadata are platform-level data structures. Treating them purely as application logic can lead to duplication across the stack and inconsistent behaviour between applications.

At Ocean Blue Software, we approached the problem differently. Instead of building a browser-centric implementation, we developed a native DVB-I client designed to run directly in firmware.

Our implementation, libdvbi, is written in C++ and compiled directly into the device firmware. It operates independently of the UI layer and does not rely on any particular browser or JavaScript engine. This allows the platform to treat DVB-I as a core service rather than an application feature.

The library handles key parts of the standard, including service list discovery, regionalisation, logical channel numbers, EPG parsing, accessibility descriptors and HTTP cache management. These elements are exposed as structured data models that applications can consume directly or via APIs such as JSON-RPC.

This architecture keeps the platform in control. Native applications, browser-based HbbTV environments and operator-specific interfaces all work from the same underlying data model, reducing engineering effort and ensuring consistent behaviour across the system.

Resource efficiency was also a key design goal. The compiled binary footprint of the library is around 2.5 MB, while a typical configuration with 100 channels and a 24-hour EPG consumes between 5 and 10 MB of RAM. Heap usage remains tightly bound and predictable, helping platform teams avoid unexpected behaviour under memory pressure.

The client integrates across multiple ecosystems, including RDK platforms, Android TV through JNI bindings, embedded Linux distributions and custom OEM operating systems. This flexibility allows manufacturers to introduce DVB-I into existing architectures without major changes to their application frameworks.

To support reliable deployment, the library includes 460 automated test cases covering roughly 90% of the current DVB-I specification, helping ensure consistent behaviour as platforms evolve.

DVB-I represents an important step forward for IP-delivered television services. But like any standard, its success will depend on how effectively it can be implemented in real devices.

For manufacturers building platforms that must scale to millions of units, native implementations provide a practical way to integrate DVB-I while keeping systems efficient, predictable and ready for production.

Get in touch with me, if you have any questions!

By Stamatis Kapetanios