Below is a guest contribution from David Rowe VK5DGR of FreeDV. If you’re a grantee and are interested in submitting a guest contribution, please reach out to giving@ardc.net.

The FreeDV project has recently published a paper on their Baseband FM (BBFM) technology. This technology provides unprecedented speech quality and robustness for VHF/UHF land mobile radio (LMR) applications, with experimentally verified 10 dB performance gains over current analog FM and digital systems.

Use cases for LMR include VHF/UHF walkie talkies, and two-way radios in vehicles for applications such as amateur radio, public safety, mining, and recreational vehicles.

Existing protocols for LMR include P.25, Tetra, DMR, M17, D-STAR and C4FM. Classic analog FM also remains popular, as its speech quality is competitive with digital systems and is low cost. While a variety of LMR protocols continue to emerge, the speech quality and physical layer performance of LMR systems hasn’t evolved since the mid 1990s.

BBFM technology presents a modern, machine learning (ML) approach for the LMR physical layer using open source software. This can be considered a drop-in replacement for the classical vocoder/FEC signal processing on radios using the BBFM architecture and can be applied to existing or emerging LMR protocols.

After 12 months of building this technology, the FreeDV project has recently released a paper, RADE for Land Mobile Radio: A Neural Codec for Transmission of Speech over Baseband FM Radio Channels, which is now available on arxiv.org. This paper is written at a professional technical level and has a description of the BBFM technology, including design, derivation of our linearised FM model, training, ASR evaluation results, and a demo over UHF radio hardware.

A more general overview of BBFM, which includes demo samples and a GitHub branch with source code, can be found on the FreeDV website: https://freedv.org/rade-bbfm/.

This work was a collaboration between David Rowe of the FreeDV project with Tibor Bece and George Karan, veterans of the LMR industry with 30 years experience and 2 million radios manufactured. David’s work was kindly supported by our grant from ARDC. While the main focus of our grant is HF, we realised RADE technology could also be applied to VHF/UHF to provide high quality, robust, open source, patent free digital speech.