Why audio encoding for DAB+ requires more than just picking a codec profile.
DAB+ uses the HE-AAC v2 profile, the same one used in MPEG-4 applications like streaming and television broadcasting. So why isn’t it just “plug-and-play”?
In reality, integrating HE-AAC v2 into a DAB+ system requires strict adaptations. From frame structures and bitrate caps to reconfiguration constraints and encoder behavior, DAB+ brings its own set of rules.
This article breaks down what makes DAB+ encoding unique and why encoder behavior plays a critical role in delivering consistent audio quality.
DAB+ is just HE-AAC, isn’t it?
DAB+ audio encoding uses the HE-AAC v2 profile (High-Efficiency Advanced Audio Coding version 2), a standard also found in MPEG-4 applications such as television broadcasting and streaming. However, to integrate this codec within the Digital Audio Broadcasting (DAB+) system architecture, several specific adaptations and restrictions were introduced. These changes are crucial for synchronization, reconfiguration, and overall system compatibility.
Decoders Are Ready – Encoders Decide
HE-AAC v2 is formally a decoder profile, defined in the MPEG-4 standard. HE-AAC v2 is a superset of AAC-LC and HE-AAC. This means that a HE-AAC v2 decoder is always backward compatible and can correctly decode:
- AAC-LC (Low Complexity) only
- AAC-LC + SBR (equivalent to HE-AAC v1 profile)
- AAC-LC + SBR + PS (equivalent to HE-AAC v2 profile)
Whether SBR (Spectral Band Replication) and PS (Parametric Stereo) are used depends entirely on the encoder configuration, not the decoder. For example, for bitrates below 96 kbit/s the usage of SBR usually provides a quality benefit compared to using AAC-LC without SBR.
System Constraints in DAB+
To integrate HE-AAC v2 cleanly into the DAB system, a number of constraints were applied. These include:
- A maximum subchannel bitrate of 192 kbit/s
- Support for only two audio sampling rates: 48 kHz or 32 kHz
- A DAB frame length of 24 ms, which differs from native AAC frame durations
These limitations not only define the boundaries of what the codec can do within DAB+, but also drive the need for deeper structural changes in encoding.
Frame Length and Audio Super Framing
A key challenge in adapting AAC to DAB+ lies in frame alignment. A frame represents a short chunk of audio that the encoder processes at once. AAC-LC usually processes 1024 audio samples per frame, corresponding to about 21.3 ms, while HE-AAC and HE-AAC-v2 (which includes SBR) use 2048 samples per frame – approximately 42.6 ms. Neither fits well with DAB’s fixed 24 ms frame structure.
To resolve this mismatch, DAB+ mandates the use of shorter frame lengths:
- 960 samples per frame for AAC-LC (20 ms)
- 1920 samples per frame for HE-AAC and HE-AAC v2 (40 ms).
These adapted frame lengths are then grouped into audio super frames to match the transmission structure. Each super frame has a duration of 120 ms, regardless of the codec profile used.
Depending on the sampling rate and whether AAC-LC or HE-AAC is used, the number of audio frames per super frame varies. For example:
- HE-AAC at 48 kHz sampling rate groups 3 frames per super frame.
- AAC-LC at 48 kHz groups 6 frames of 20 ms each into a super frame.
These super frames are distributed across five DAB frames, enabling time-aligned, frame-exact transmission.
Why Super Frames Must Be Self-Contained
In traditional AAC encoding, bits can be shifted between adjacent frames to better represent complex audio content. However, DAB+ forbids such inter-frame dependencies. Each super frame must be 100% self-contained to support:
- Seamless switching between services or configurations
- Error resilience
- Clean reconfiguration boundaries.
This design is critical in broadcast environments where dynamic reconfiguration or service-following is required—for example, when users switch stations or when broadcasters adapt content on the fly.
Trade-Offs and the Need for Encoder Tuning
These structural adaptations – shorter frame lengths and rigid super framing – do affect coding efficiency. The encoder has less room to manage bursts of audio complexity, and fewer tools to smooth over transitions. As a result, achieving high audio quality, especially at low bitrates, requires careful encoder optimization.
The 2wcom Approach
At 2wcom, we place special emphasis on audio fidelity and system compatibility. Our DAB+ encoders – featured in the IP-4c, IP-8e, and MoIN platforms – are built around the Fraunhofer Professional DAB+ Encoder library. This ensures:
- Best-in-class audio quality, verified through extensive listening tests,
- Full compliance with the DAB+ system specifications
- Reliable performance under real-world broadcasting conditions
At 2wcom, we understand that delivering consistent audio quality in DAB+ environments requires more than just using the codec – it demands expertise in how to apply it. By building on the Fraunhofer Professional DAB+ Encoder library and focusing on encoder behavior under real-world conditions, we ensure that our solutions provide stability, synchronicity and adaptability. Whether broadcasters are operating at high or constrained bitrates, our systems are designed to maintain performance without compromising on audio integrity.